WMO No. 488

Guide to the Global Observing System
2010 edition
Updated in 2017
WEATHER CLIMATE WATER
WMO-No. 488
Guide to the Global Observing System
2010 edition
Updated in 2017
WMO-No. 488
EDITORIAL NOTE
METEOTERM, the WMO terminology database, may be consulted at http://public.wmo.int/en/
resources/meteoterm.
Readers who copy hyperlinks by selecting them in the text should be aware that additional
spaces may appear immediately following http://, https://, ftp://, mailto:, and after slashes (/),
dashes (-), periods (.) and unbroken sequences of characters (letters and numbers). These spaces
should be removed from the pasted URL. The correct URL is displayed when hovering over the
link or when clicking on the link and then copying it from the browser.
WMO-No. 488
© World Meteorological Organization, 2010
The right of publication in print, electronic and any other form and in any language is reserved by
WMO. Short extracts from WMO publications may be reproduced without authorization, provided
that the complete source is clearly indicated. Editorial correspondence and requests to publish,
reproduce or translate this publication in part or in whole should be addressed to:
Chairperson, Publications Board

World Meteorological Organization (WMO)
7 bis, avenue de la Paix Tel.: +41 (0) 22 730 84 03
P.O. Box 2300 Fax: +41 (0) 22 730 81 17
CH-1211 Geneva 2, Switzerland Email: [email protected]
ISBN 978-92-63-10488-5
NOTE
The designations employed in WMO publications and the presentation of material in this publication do
not imply the expression of any opinion whatsoever on the part of WMO concerning the legal status of any
country, territory, city or area, or of its authorities, or concerning the delimitation of its frontiers or boundaries.
The mention of specific companies or products does not imply that they are endorsed or recommended by
WMO in preference to others of a similar nature which are not mentioned or advertised.
PUBLICATION REVISION TRACK RECORD
Part/chapter/
Date Purpose of amendment Proposed by Approved by
section
July Part III, Modification to 3.2.1.2.6 – Station CBS-Ext. (06) Resolution 10

2012 section 3.2, identification, subparagraphs (b) and and EC-LIX (EC-LIX)
3.2.1.2.6 (c), to make consistent with the Manual
on the Global Observing System (WMO-
No. 544) following Recommendation
1 (CBS-Ext.(06)) and approval by the
Executive Council at its fifty-ninth
session.
May Part III, Revised functional specifications for CBS-15 and Resolution 10
2013 Appendix III.1 automatic weather stations, following EC‑65 (EC-65)
Recommendation 2 (CBS-15) and
approval by the Executive Council at its
sixty-fifth session
October Part III, Previous 3.2.1.3.3.2 and 3.2.1.3.3.3: President of CBS Resolution 3
2017 section 3.2, deleted (EC-69)
3.2.1.2.6,
3.2.1.3.3.1, Previous 3.2.1.3.4 is now 3.2.1.3.3.2
3.2.1.3.3.2,
3.2.1.3.3.3, Changes required to include new
3.2.1.3.4; guidance material in accordance with
New section Recommendation 4 (CBS-16)
3.4;
New
Appendix
III.4
CONTENTS
Page
INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
PART I. PURPOSE, SCOPE, REQUIREMENTS AND ORGANIZATION OF THE

GLOBAL OBSERVING SYSTEM. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.1 Purpose and scope of the Global Observing System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.2 Global Observing System requirements. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.3 Organization and Implementation of the Global Observing System. . . . . . . . . . . . . . . . . 5
PART II. OBSERVATIONAL DATA REQUIREMENTS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

2.1. General. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.2 Assessment and formulation of observational data requirements . . . . . . . . . . . . . . . . . . . 8
2.2.1 Data sensitivity tests or observing system experiments. . . . . . . . . . . . . . . . . . . . . . 8
2.2.2 Observing system simulation experiments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.2.3 Theoretical studies and simulations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.2.4 Laboratory assessments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.2.5 System design and analysis activities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.2.6 Field site assessments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.2.7 End-user application areas. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.3 Evaluation of requirements against system capabilities. . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.3.1 The Rolling Requirements Review process. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.3.2 User requirements and observing system capabilities database. . . . . . . . . . . . . . . 11
2.3.2.1 User requirements. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.3.2.2 Observing system capabilities. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
2.3.3 The Critical Review. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .12
2.3.4 Statements of Guidance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
2.4 Network design and national requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
2.5 Evolution of the Global Observing System. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
APPENDIX II.1. EXTRACT FROM USER REQUIREMENTS AND OBSERVING SYSTEM

CAPABILITIES DATABASE – EXAMPLE OF GLOBAL NUMERICAL WEATHER
PREDICTION REQUIREMENTS FOR SOME VARIABLES. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
APPENDIX II.2. OUTCOME OF THE ROLLING REQUIREMENTS REVIEW PROCESS:

EXAMPLES. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
PART III. THE SURFACE-BASED SUBSYSTEM. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18

3.1 General. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
3.1.1 Design of observational networks. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
3.1.2 Planning of networks and stations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3.1.3 Management of manned station networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
3.1.3.1 General. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
3.1.3.2 Organization of the station network management unit. . . . . . . . . . . . . 21
3.1.3.3 Administrative arrangements. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
3.1.3.4 Station network management unit staff. . . . . . . . . . . . . . . . . . . . . . . . . . 22
3.1.3.5 Station network management unit operational tasks. . . . . . . . . . . . . . . 22
3.1.3.6 Logistics and supplies. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
3.1.3.7 Establishment of a new station . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
3.1.3.8 Regular inspections. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
3.1.3.9 Other station network management unit activities. . . . . . . . . . . . . . . . . 25
3.1.3.10 Procurement of instruments and equipment. . . . . . . . . . . . . . . . . . . . . . 25
3.1.3.11 Instrument checks and maintenance . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
3.1.3.12 Coordination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
vi GUIDE TO THE GLOBAL OBSERVING SYSTEM
Page
3.1.3.13 Planning and budgeting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26

3.1.3.14 Network performance monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
3.1.4 Management of automatic surface land station networks. . . . . . . . . . . . . . . . . . . . 27
3.1.4.1 General. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
3.1.4.2 Administrative arrangements. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
3.1.4.3 Operational tasks of the automatic station network supervising unit. . 27
3.2 Surface synoptic stations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
3.2.1 Organizational aspects. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
3.2.1.1 General. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
3.2.1.2 Land stations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
3.2.1.3 Sea stations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
3.2.1.4 Automatic stations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
3.2.2 Observations/measurements. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
3.2.2.1 General. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
3.2.2.2 Observations at land stations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
3.2.2.3 Observations at sea stations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
3.3 Upper-air stations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
3.3.1 Organizational aspects. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
3.3.1.1 Site selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
3.3.1.2 Planning of facilities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
3.3.1.3 Organization of the upper-air unit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
3.3.1.4 Archiving data and maintenance of records. . . . . . . . . . . . . . . . . . . . . . .70
3.3.1.5 Communications. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
3.3.1.6 Personnel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
3.3.1.7 Training. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
3.3.1.8 Quality standards . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
3.3.2 Observations/measurements. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
3.3.2.1 General. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
3.3.2.2 Pilot-balloon observations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
3.3.2.3 Radiosonde observations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
3.3.2.4 Radiowind observations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
3.3.2.5 Rawinsonde observations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
3.3.2.6 Combined radiosonde and radiowind observations. . . . . . . . . . . . . . . .76
3.3.2.7 Aerological soundings using automated shipboard or land-
based upper-air systems. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
3.3.2.8 Upper-air systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
3.3.2.9 Observational requirements. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
3.3.3 Special management considerations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
3.3.3.1 General. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
3.3.3.2 Procurement of instruments and equipment. . . . . . . . . . . . . . . . . . . . . . 82
3.3.3.3 Maintenance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
3.3.3.4 Budgetary requirements. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
3.4 Aircraft meteorological stations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
3.5 Aeronautical meteorological stations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
3.5.1 General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
3.5.2 Instrumentation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
3.5.3 Location of meteorological stations and instruments. . . . . . . . . . . . . . . . . . . . . . . . 85
3.5.4 Observing and reporting programme . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
3.5.5 Communications. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
3.5.6 Personnel and training . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
3.5.7 Quality standards. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
3.6 Research and special-purpose vessel stations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
3.7 Climatological stations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
3.7.1 Organization. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
3.7.2 Climatological station networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
3.7.3 Station classification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
3.7.3.1 Reference climatological stations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
3.7.3.2 Principal climatological stations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
3.7.3.3 Ordinary climatological stations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
CONTENTS vii
Page
3.7.3.4 Special-purpose stations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89

3.7.4 Operation of stations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
3.7.5 Quality standards. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
3.7.6 Archiving. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
3.8 Agricultural meteorological stations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
3.8.1 Organization. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
3.8.2 Station classification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
3.8.3 Operation of stations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
3.9 Special stations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
3.9.1 General tasks and purposes of special stations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
3.9.2 Types of special stations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
3.9.2.1 Weather radar stations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
3.9.2.2 Radiation stations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
3.9.2.3 Atmospherics detection stations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
3.9.2.4 Meteorological reconnaissance aircraft stations. . . . . . . . . . . . . . . . . . . 99
3.9.2.5 Meteorological rocket stations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
3.9.2.6 Global Atmosphere Watch (GAW) stations . . . . . . . . . . . . . . . . . . . . . . . 102
3.9.2.7 Planetary boundary-layer stations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110
3.9.2.8 Tide-gauge stations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
APPENDIX III.1. FUNCTIONAL SPECIFICATIONS FOR AUTOMATIC WEATHER STATIONS. 116
APPENDIX III.2. BASIC SET OF VARIABLES TO BE REPORTED BY STANDARD

AUTOMATIC WEATHER STATIONS FOR MULTIPLE USERS. . . . . . . . . . . . . . . . . . . . . . . . . . . . 122
APPENDIX III.3. AUTOMATIC WEATHER STATION METADATA . . . . . . . . . . . . . . . . . . . . . . . . 124
APPENDIX III.4. THE WMO VOLUNTARY OBSERVING SHIP SCHEME. . . . . . . . . . . . . . . . . . . 131
PART IV. THE SPACE-BASED SYSTEM. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162

4.1 GENERAL. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .162
4.1.1 History of the space-based subsystem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162
4.1.2 Relation to the surface-based subsystem. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162
4.1.3 Coordination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163
4.2 THE BASELINE SPACE SEGMENT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164
4.2.1 Sun-synchronous polar-orbiting satellites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164
4.2.1.1 Principle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164
4.2.1.2 Implementation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165
4.2.1.3 Observing missions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165
4.2.1.4 Data dissemination missions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 166
4.2.1.5 Other communications missions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168
4.2.1.6 Space monitoring missions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168
4.2.2 Geostationary satellites. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168
4.2.2.1 Observing missions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
4.2.2.2 Data dissemination missions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170
4.2.2.3 Data collection and search and rescue missions. . . . . . . . . . . . . . . . . . . 171
4.2.2.4 Space environment monitoring missions. . . . . . . . . . . . . . . . . . . . . . . . . 171
4.2.3 Research and development (R&D) satellites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171
4.2.3.1 Primary purpose of R&D satellite missions. . . . . . . . . . . . . . . . . . . . . . . . 171
4.2.3.2 Relevance of R&D satellite missions to the Global Observing System . 172
4.2.3.3 Transition to operational status. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172
4.3 DATA CIRCULATION AND USER SERVICES. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173
4.3.1 General ground segment features. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173
viii GUIDE TO THE GLOBAL OBSERVING SYSTEM
Page
4.3.2 Integrated Global Data Dissemination Service . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174

4.3.3 User services. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175
4.3.4 Satellite meteorology user training. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176
4.4 DERIVED PRODUCTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176
4.4.1 Calibration issues. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176
4.4.2 Product categories. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177
4.5 TRENDS IN SPACE-BASED SUBSYSTEMS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179
PART V. REDUCTION OF LEVEL I DATA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181

5.1 GENERAL. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .181
5.2 REDUCTION PROCESS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181
5.3 AVERAGING OF MEASURED QUANTITIES. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182
PART VI. DATA QUALITY CONTROL. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183

6.1 GENERAL. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .183
6.1.1 Levels of application of quality control procedures. . . . . . . . . . . . . . . . . . . . . . . . . . 184
6.1.2 Observational errors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185
6.2 PROCEDURAL ASPECTS OF QUALITY CONTROL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 186
6.2.1 Responsibility and standards. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 186
6.2.2 Scope of quality control. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .187
6.2.3 Implementation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187
6.2.3.1 Manual methods. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188
6.2.3.2 Automated methods. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188
6.3 OTHER QUALITY CONTROL PROCEDURES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189
6.3.1 Availability of statistics on variables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189
6.3.2 Use of accepted abbreviations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189
6.3.3 Pictorial representations and diagrams. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189
6.3.4 Simplified mathematical checks. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 190
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 190
APPENDIX VI.1. DATA QUALITY CONTROL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192
APPENDIX VI.2. GUIDELINES FOR QUALITY CONTROL PROCEDURES APPLYING

TO DATA FROM AUTOMATIC WEATHER STATIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 198
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207
PART VII. MONITORING THE OPERATION OF THE GLOBAL OBSERVING SYSTEM . . . . . . . 208
7.1 GENERAL. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .208
7.2 IMPLEMENTATION OF THE MONITORING PROCEDURES . . . . . . . . . . . . . . . . . . . . . . . . . 208
7.2.1 Quantity monitoring of the operation of the World Weather Watch . . . . . . . . . . . 208
7.2.1.1 Annual Global Monitoring. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208
7.2.1.2 Special main telecommunication network monitoring . . . . . . . . . . . . . 209
7.2.2 Data quality monitoring. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210
7.2.2.1 Monitoring centres. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210
7.2.2.2 Procedures and formats for exchange of monitoring results. . . . . . . . . 212
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212
PART VIII. QUALITY MANAGEMENT. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213

8.1 GENERAL. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .213
8.2 QUALITY MANAGEMENT FRAMEWORK . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213
8.3 WMO TECHNICAL STANDARDS AS REFERENCE DOCUMENTATION. . . . . . . . . . . . . . . . . 214
CONTENTS ix
Page
8.4 QUALITY MANAGEMENT SYSTEM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 214

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215
INTRODUCTION
General
As laid down in the Convention, one of the principal purposes of the World Meteorological
Organization (WMO) is to facilitate worldwide cooperation in the establishment of networks of
stations for the making of meteorological observations or other geophysical observations related
to meteorology, and to promote the establishment and maintenance of meteorological centres
charged with the provision of meteorological services. Another aim of the Organization is to
promote standardization of meteorological observations and to ensure the uniform publication
of observations and statistics. With a view to ensuring the required standardization of practices
and procedures in meteorology, the World Meteorological Congress adopts Technical
Regulations laying down the meteorological practices and procedures to be followed by the
Member countries of the Organization. The Technical Regulations include manuals relating to
various aspects of the Organization’s activities and are supplemented by a number of guides
which describe in more detail the practices, procedures and specifications which Members are
invited to follow in establishing and conducting their arrangements for compliance with the
Technical Regulations and in developing meteorological services in their respective countries.
The present Guide deals with the organization and implementation of the Global Observing
System, one of three essential components of the WMO World Weather Watch Programme
(WWW).
World Weather Watch
World Weather Watch, a core WMO Programme, combines observing systems,

telecommunication facilities, and data-processing and forecasting centres operated by Members
to provide meteorological and related geophysical information for efficient service delivery in all
countries.
WMO Members coordinate and implement by means of the WWW Programmes standardized
measuring methods and common telecommunication procedures, and present observed data
and processed information in a manner understood by all, regardless of language.
These arrangements and the operation of World Weather Watch facilities are coordinated
and monitored by WMO with a view to ensuring that every country has access to all the
information it needs to provide weather services on a day-to-day basis and to conduct long-
term planning and research activities. A key objective of the WWW Programme is to provide the
basic infrastructure for obtaining observational data and related services needed for relevant
international programmes that address global environmental issues.
World Weather Watch operates at global, regional and national levels. It involves the design,
implementation, operation and further development of the following three interconnected, and
increasingly integrated, core elements:
(a) The Global Observing System (GOS), which consists of facilities and arrangements for
making observations at stations on land and at sea, and from aircraft, meteorological
operational satellites and other platforms. It is designed to provide observational data for
use in both operational and research work;
(b) The Global Telecommunication System (GTS), composed of integrated networks of

telecommunication facilities and centres, especially regional telecommunication hubs,
for the rapid, reliable collection and distribution of observational data and processed
information;
(c) The Global Data-processing and Forecasting System (GDPFS), made up of World, Regional/
Specialized and National Meteorological Centres to provide processed data, analyses and
forecast products.
2 GUIDE TO THE GLOBAL OBSERVING SYSTEM
The implementation, integration and efficient operation of the three core elements are achieved
by means of the following support programmes:
(a) The World Weather Watch Data Management programme, which monitors and manages
the information flow within the World Weather Watch system to ensure quality and the
timely availability of data and products and the use of standard representation formats to
meet the requirements of Members and other WMO Programmes;
(b) The World Weather Watch System Support Activities programme, which provides specific
technical guidance, training and implementation support, the World Weather Watch
Operational Information Service and supports cooperative initiatives.
Further specifications and details of the functions and organization of the three World Weather
Watch core components are provided in Volume I of the respective Manuals on the Global
Observing System, the Global Data-processing and Forecasting System and the Global
Telecommunication System, which are annexes to the Technical Regulations.
Purpose of this Guide
The main purpose of the Guide is to provide practical information on the development,
organization, implementation and operation of the Global Observing System in order to enhance
both the participation of individual Members in the System and the benefits they may obtain
from it. The Guide explains and describes Global Observing System practices, procedures
and specifications and is aimed at assisting the technical and administrative staff of National
Meteorological Services responsible for the networks of observing stations in preparing national
instructions for observers.
The first edition of the Guide on the Global Observing System was completed in 1977 as the
result of a decision adopted by the WMO Commission for Basic Systems at its sixth session, held
in Belgrade in 1974. Since then, it has undergone a number of revisions and amendments. These
have been consolidated into this new revised edition.
The present edition contains the amendments adopted by the Commission for Basic Systems at
its fourteenth session held in Dubrovnik, Croatia, from 25 March to 2 April 2009, as included in
the annexes to Recommendations 2 (CBS-XIV) and 3 (CBS-XIV), and endorsed by the Executive
Council at its sixty-first session in June 2009 in Resolution 6 (EC-LXI).
The Guide supplements the regulatory material on observational matters contained in the
Technical Regulations (WMO-No. 49) and the Manual on the Global Observing System (WMO-
No. 544) and, for ease of reference, follows roughly the same structure as the Manual. The
Guide also complements the Guide to Meteorological Instruments and Methods of Observation
(WMO-No. 8), while the Guide on the Global Data-processing System (WMO-No. 305) is used to
complement the present Guide.
A list of publications which are related to and may be used in conjunction with the Guide to the
Global Observing System is given below:
Guide on Meteorological Observing and Information Distribution Systems for Aviation Weather Services
(WMO‑No. 731)
Guide on the Global Data-processing System (WMO‑No. 305)
Guide to Agricultural Meteorological Practices (WMO‑No. 134)
Guide to Climatological Practices (WMO‑No. 100)
Guide to Hydrological Practices (WMO‑No. 168)
Guide to Marine Meteorological Services (WMO‑No. 471)
Guide to Meteorological Instruments and Methods of Observation (WMO‑No. 8)
Guide to the Implementation of Education and Training Standards in Meteorology and Hydrology
(WMO‑No. 1083), Volume I – Meteorology.
Information on Meteorological and Other Environmental Satellites (WMO‑No. 411)
International Cloud Atlas (WMO‑No. 407)
INTRODUCTION 3
Manual on Codes (WMO‑No. 306)

Manual on the Global Data-processing and Forecasting System (WMO‑No. 485)
Manual on the Global Observing System (WMO‑No. 544)
Manual on the Global Telecommunication System (WMO‑No. 386)
Sixth WMO Long-term Plan (2004–2011) (WMO‑No. 962)
Technical Regulations (WMO-No. 49)
World Weather Watch: http://www.wmo.int/pages/prog/www/index_en.html.
Procedures for amending the Guide
A detailed explanation of the procedures for amending WMO Guides that are under the
responsibility of the Commission for Basic Systems can be found in the appendix to the General
Provisions of the Manual on the Global Observing System (WMO-No. 544).
PART I. PURPOSE, SCOPE, REQUIREMENTS AND ORGANIZATION OF THE
GLOBAL OBSERVING SYSTEM
1.1 PURPOSE AND SCOPE OF THE GLOBAL OBSERVING SYSTEM
The Global Observing System (GOS) provides from the Earth and outer space observations of
the state of the atmosphere and ocean surface for the preparation of weather analyses, forecasts
and warnings for all WMO Programmes and relevant environmental programmes of other
international organizations. It is operated by National Meteorological Services, national or
international satellite agencies, and involves several consortia dealing with specific observing
systems or geographic regions.
The Global Observing System is a coordinated system of different observing subsystems,

the main purpose of which is to provide in a cost-effective way high-quality standardized
meteorological and related environmental and geophysical observations from all parts of the
globe and outer space, as they are required for the real-time preparation of weather analyses and
forecasts, including warnings. The Global Observing System also provides observational data
for research purposes in support of other WMO Programmes or relevant programmes of other
international organizations, as agreed by the Organization.
Main long-term objectives
The main long-term objectives of the Global Observing System are as follows:
(a) To improve and optimize global systems for observing the state of the atmosphere and
the ocean surface in order to meet effectively and efficiently the requirements for the
preparation of increasingly accurate weather analyses, forecasts and warnings, and for the
conduct of climate and environmental monitoring activities under programmes of WMO
and other relevant international organizations;
(b) To provide for the necessary standardization of observing techniques and practices,
including the planning of networks on a regional basis to meet user requirements with
respect to quality, spatial and temporal resolution and long-term stability.
1.2 GLOBAL OBSERVING SYSTEM REQUIREMENTS
The requirements to be met by the Global Observing System are defined by Members of the
Organization through the regional associations and Technical Commissions set out in the various
WMO Programmes. Essentially, the Global Observing System must provide the basic data for the
services rendered by the National Meteorological Services or other organizations in contributing
to public safety, socio-economic well-being and development in their respective countries. These
services fall into three main categories:
(a) Weather forecasts, including reports on current weather, warnings of hazardous weather
phenomena and predictions of weather on varying timescales up to one month and
sometimes beyond;
(b) Climate information and advice on the application of meteorological data and knowledge;
(c) Hydrological services, including flood warnings.
These three categories cover a variety of specialized services and applications of meteorology
which require different types of meteorological observations and measurements on varying
PART I. PURPOSE, SCOPE, REQUIREMENTS AND ORGANIZATION OF THE GLOBAL OBSERVING SYSTEM 5
scales. These include short-, medium- and long-range weather forecasting; the provision of
severe weather warnings for the occurrence of such phenomena as tropical cyclones, polar lows,
hail storms, floods and heavy snowfalls; services for aviation, shipping and agriculture and other
diverse areas such as energy production, environmental protection, the construction industry
and tourism. In general, the requirements to be met by the Global Observing System for each of
these areas are established by the WMO Technical Commissions for Basic Systems, Climatology,
Atmospheric Sciences, Hydrology, Aeronautical Meteorology, Agricultural Meteorology, and
Oceanography and Marine Meteorology.
A number of international programmes also use World Weather Watch facilities and those of the
Global Observing System in particular, and have their own special requirements. These include:
the World Climate Research Programme of WMO, the International Council for Science, the
World Area Forecast System, the United Nations Environment Programme Global Environment
Monitoring System and the Integrated Global Ocean Services System of the United Nations
Educational, Scientific and Cultural Organization’s (UNESCO) Intergovernmental Oceanographic
Commission (IOC) and WMO.
The formulation of data requirements is an evolving process based on experiences with

observing systems and improvements in data assimilation techniques. The process balances
demands of users with the technical feasibility of data resolution. Part II provides more detailed
information on this topic.
1.3 ORGANIZATION AND IMPLEMENTATION OF THE GLOBAL OBSERVING

SYSTEM
In order to meet these requirements, the Global Observing System is designed as a composite
system consisting of surface- and space-based satellite subsystems. The former, discussed
in detail in Part III of the Guide, comprises the regional basic synoptic networks and other
networks of land, sea and airborne stations; it also includes agricultural meteorological
stations, climatological stations and special stations. The space-based subsystem (see Part IV)
is composed of a space segment including meteorological operational polar-orbiting and
geostationary satellites and environmental research and development satellites, and a ground
segment for satellite data reception and processing.
The composite system provides observational information that falls broadly into two categories:
(a) Quantitative information, derived directly or indirectly from instrument measurements;
(b) Qualitative, descriptive information.
Examples of quantitative information, which specify the physical state of the atmosphere, are
measurements of atmospheric pressure, humidity, air temperature and wind speed, whereas
qualitative or descriptive information includes such observations as the amount and type of
clouds and types of precipitation.
At the behest of the World Meteorological Congress, the Commission for Basic Systems studied
the evolution of the Global Observing System and issued the Implementation Plan for Evolution of
Space- and Surface-based Subsystems of the Global Observing System (WMO/TD-No. 1267). A key
aim of the Plan is to help Members prepare for anticipated changes in the Global Observing
System over the next two decades. Implementation of the new Global Observing System should
facilitate cooperation among Members at national, regional and global levels. The future Global
Observing System in developing countries will have to address issues such as infrastructure,
training, equipment and consumables.
Implementation activities
Global Observing System implementation components are aimed at achieving the following
objectives:
(a) Enhanced standardization of observing techniques and practices and their improvement,
including the redesign, optimum planning and the implementation of redesigned
observational networks on a regional basis;
(b) Improved performance of the global network to meet in the most efficient manner the
stated requirements in terms of uncertainty, temporal and spatial resolution, and timeliness
of meteorological observations;
(c) Assessment of cost-effectiveness, long-term sustainability and innovative collaborative

arrangements between National Meteorological and Hydrological Services related to
operations of the upgraded Global Observing System;
(d) Analysis of evolving observational data requirements from various application programmes
and preparation of guidance for the further development of the Global Observing System.
With respect to the implementation of the Global Observing System, the guiding principle is that
all activities and facilities connected with the establishment and operation of the System in the
territories of individual countries are the responsibility of the countries themselves and should be
met to the extent possible with national resources. Where this is not possible, assistance may be
provided by the United Nations Development Programme, through other multilateral or bilateral
assistance programmes or by the WMO Voluntary Cooperation Programme.
Implementation of the Global Observing System outside the territories of individual countries,
such as outer space, the oceans and the Antarctic, is based on the principle of voluntary
participation of countries that are willing and able to contribute by providing facilities and
services individually or jointly from their national resources or through collective financing.
The Global Observing System is a flexible and continuously evolving system with a choice and
mix of observing components that may be adjusted to take full advantage of new technology
or to meet new requirements. As a general rule, however, the evolution of the system should be
based on proven techniques and represent the best mix of observing components which fully
meet the following requirements:
(a) Agreed-upon data requirements in respect of required uncertainty, temporal and spatial
resolution and timeliness;
(b) Operational and technical feasibility;
(c) Members’ cost-efficiency requirements.
Throughout the Global Observing System, standardized quality control procedures are applied
(see Part VI of the Guide) to all observing system components in order to ensure high quality and
compatible data.
Certain levels of redundancy are required for quality assurance purposes and to ensure against
catastrophic failure in any single component; multi-purpose components or stations are
encouraged in order to comply with cost-efficiency requirements.
PART II. OBSERVATIONAL DATA REQUIREMENTS
2.1. GENERAL
Weather prediction and other environment-related activities involve observational data analyses.
Weather prediction, in particular, is based on precise meteorological analyses. All analyses
require highly reliable observational data received in a timely manner at analysis centres from a
sufficiently dense network or other observation source. In meteorological analyses, the required
uncertainty, temporal and spatial resolution and timeliness of these data are dependent on the
following factors:
(a) The different scales of meteorological phenomena to be analysed;
(b) The resolution and other characteristics of the techniques used to perform the analyses and
modelling based upon them.
Observational data requirements are always specific to purpose and change with time as
techniques improve. In general, they have become more demanding as the power of computers
has increased and operational numerical weather prediction and associated models have
become better able to represent smaller-scale phenomena.
Varying scales of meteorological phenomena co-exist in the atmosphere. For example, one cell
of a thunderstorm is only several kilometres on a horizontal scale and has a lifetime of several
hours, whereas a tropical cyclone is about 1 000 km on a horizontal scale and has a lifetime of
10 days or more; many thunderstorm cells appear and disappear during the life cycle of a tropical
cyclone. Therefore, the frequency and spacing of observations should be adequate to obtain
observational data that describe temporal and spatial changes of the meteorological phenomena
with sufficient resolution to meet user requirements. If the spacing of observations is more than
100 km, meteorological phenomena with a horizontal scale of less than 100 km are not usually
detectable. The classification of horizontal scales of the meteorological phenomena given in
Volume I of the Manual on the Global Observing System (WMO‑No. 544) is as follows:
(a) Microscale (less than 100 m) for agricultural meteorology, for example, evaporation;
(b) Toposcale or local scale (100 m–3 km), for example, air pollution, tornadoes;
(c) Mesoscale (3 km–100 km), for example, thunderstorms, sea and mountain breezes;
(d) Large scale (100 km–3 000 km), for example, fronts, various cyclones, cloud clusters;
(e) Planetary scale (more than 3 000 km), for example, long upper tropospheric waves.
Horizontal scales are closely related to the timescales of the phenomena. Larger horizontal-scale
perturbations are likely to survive for a longer time period (Figure II.1). Therefore, short-range
weather forecasts require more frequent observations from a denser network over a limited
area in order to detect any small-scale phenomena and their development. As the length of
the forecast period increases, so does the area over which observations are required. Owing to
dynamic interaction between the meteorological phenomena on different scales, it may not be
possible to provide definitive requirements for individual scales.
There are generally three types of requirements:
(a) Global requirements refer to observational data needed by WMO Members for a general
description of large- and planetary-scale meteorological phenomena and processes;
(b) Regional requirements are related to observations needed by two or more WMO Members
to describe in greater detail the large- and planetary-scale atmospheric phenomena, as well
as to describe smaller ones on the mesoscale and small scales as may be agreed by regional
associations;
Planetary scale
Long upper
trophospheric
10
6 Large scale
Mesoscale Cycloneswaves
Front Depressions
10
5
Cloud clusters Anticyclones
Toposcale Hurricanes Long-range
(day) Squall lines forecast
10
4 Thunderstorms
Clear air Katabatic winds
(hour) Short-range
turbulence Tornados
forecast
10
3 Microscale Convective clouds
Mountains waves
Whirlwinds Local forecast (aerodrome,
Strong local winds
10
2 take-off, landing forecast)
(min)
Small eddy
(Turbulent motions)
1
10
1 2 3 4 5 6 7
10 10 10 10 10 10 10
(1 km) (10 km) (100 km) (1 000 km) (10 000 km)
Figure II.1. Horizontal scales (metres) and timescales (seconds) of meteorological phenomena
(c) National requirements are determined by each WMO Member in the light of its own
interests.
Although the above discussion has been focused upon processes taking place in the atmosphere
and meteorological uses of the data, similar considerations apply to processes occurring on
the Earth’s surface and for applications such as hydrology and agricultural meteorology. The
controlling physical and chemical processes set the scale that analyses must resolve, and
interactions between them determine the domain from which data are required.
2.2 ASSESSMENT AND FORMULATION OF OBSERVATIONAL DATA

REQUIREMENTS
The assessment of data requirements is an ongoing process based on the need for information
services and an evolving level of experience with actual and candidate observing systems.
A number of techniques and tools are available to carry out the assessments. Some require
substantial resources and are best deployed to test specific hypotheses.
2.2.1 Data sensitivity tests or observing system experiments
Such tests require actual observations, from operational, pilot, demonstration or research
networks and systems and a numerical weather prediction capability. Most straightforward
experiments may be characterized as data-denial or data-inclusion experiments. Typically, an
assimilation and prediction system is run with a control data set, and then with one or more data
types withheld, or reduced in quantity. Analyses or forecasts are verified against observations.
Comparison of the two runs indicates the effect of data denial, or, equivalently, the value of the
observing system when included. Observing system experiments are valuable, for example, in
examining the impact of temporal or spatial changes in the network configuration or of adding
or deleting existing observing systems, without actually making operational changes.
2.2.2 Observing system simulation experiments
Those experiments deal with hypothetical or simulated data sets and are useful for estimating
how an entirely new observing system might affect forecast accuracy. A historical forecast
is usually designated as the control run; it describes the “true” atmosphere. Hypothetical
observations with plausible error characteristics are then manufactured from the control run at
designated locations and times. The observational data set to be examined is then assimilated
PART II. OBSERVATIONAL DATA REQUIREMENTS 9
by a prediction model, and a new forecast is generated along with the control forecast. The
impact of the simulated observing system is approximated by the difference between the two
forecasts. Despite their limitation to hypothetical observations, observing system simulation
experiments are an important part of assessing the potential utility of data from a system before
it is implemented.
2.2.3 Theoretical studies and simulations
Theoretical studies and simulations of expected data utility from potential future sensor systems
can be important in planning changes to the existing observing system. For example, substantial
theoretical studies and simulations were conducted prior to the launch of the first geostationary
operational environmental satellite (GOES) 1-M series to predict sensor performance. The results
provided an important part of the basis for designing the ground data processing system and
many other aspects of the support facilities required. As systems become increasingly complex
and costly, the need for well-planned theoretical studies and simulations should increase. They
are an important way of reducing the risk in making decisions to develop and implement systems
still in the conceptual or research stage.
2.2.4 Laboratory assessments
Some assessments, particularly of data processing and display techniques, are best and/or more
economically conducted in a controlled laboratory environment. Several WMO Members have
the capability to develop and test techniques for data processing and display. In the past, results
derived from their work have been instrumental in designing both individual sensor suites and
networks.
2.2.5 System design and analysis activities
System design and analysis activities are concerned primarily with identifying the cost and
operational impact of recommended changes stemming from the scientific studies. These
activities also include the design and coordination of any field and/or pilot projects that may be
needed.
2.2.6 Field site assessments
Existing field sites offer the opportunity to examine the impact new data sets could have on
forecasting and on the generation of products and services. Such evaluations become especially
important in both the early and late stages of development and deployment to ensure that
operational support is defined properly, is in place when needed and that field personnel are
trained to obtain the best results from new systems.
2.2.7 End-user application areas
Observational data requirements are specific to the end-user application areas for which services
are being provided, beyond weather forecasting. They include the following areas:
(a) Agriculture and food production;

(b) Aviation;
(c) Land transport;
(d) Marine resources and shipping services;
(e) Hydrology and water resources;
(f) Industry;
(g) Environmental monitoring;
(h) Disaster mitigation and prevention, emergency response;
(i) Energy;
(j) Public weather services, health and safety;

(k) Climatology and climate services.
2.3 EVALUATION OF REQUIREMENTS AGAINST SYSTEM CAPABILITIES
It is a challenging exercise to bring together the expertise as described above and develop a
consensus view on the design and implementation of composite observing systems. This is
particularly true where the need and implementation occur on global or regional scales. The
Commission for Basic Systems has encouraged the development of a procedure or process to
accomplish this as objectively as possible: the Rolling Requirements Review. It applies to each of
the application areas covered by WMO Programmes, namely:
(a) Global numerical weather prediction;

(b) Regional numerical weather prediction;
(c) Synoptic meteorology;
(d) Nowcasting and very short-range forecasting;
(e) Seasonal and inter-annual forecasts;
(f) Atmospheric chemistry;
(g) Aeronautical meteorology;
(h) Climate variability;
(i) Climate change;
(j) Marine meteorology;
(k) Hydrology;
(l) Agricultural meteorology.
2.3.1 The Rolling Requirements Review process
The process jointly reviews users’ evolving requirements for observations and the capabilities of
existing and planned observing systems. Statements of Guidance describing the extent to which
such capabilities meet requirements are produced as a result. Initially, the process was applied to
the requirements of global numerical weather prediction and the capabilities of the space-based
subsystem, but more recently the range of requirements has been expanded and the technique
has begun to be applied successfully to surface-based observing systems and other application
areas.
The process consists of four stages (see Figure II.2):
1. A review of user requirements for observations, within an area of application covered by

WMO Programmes;
2. A review of the observing capabilities of existing and planned observing systems;
3. A Critical Review of the extent to which the capabilities (2) meet the requirements (1);
4. A Statement of Guidance based on (3).
The aim of the Statement of Guidance and the Critical Review is as follows:
(a) To inform WMO Members about the extent to which their requirements are met by
present systems, will be met by planned systems, or would be met by proposed systems.
It also provides the means whereby Members, through the Technical Commissions, can
check whether their requirements have been correctly interpreted and can update them if
necessary, as part of the Rolling Requirements Review process;
(b) To provide resource materials useful to WMO Members for dialogue with observing system
agencies as to whether existing systems should be continued, modified or discontinued;
whether new systems should be planned and implemented; and whether research and
development is needed to meet unfulfilled user requirements.
Clearly, the Rolling Requirements Review process needs to be repeated periodically as

requirements change and further information becomes available. Figure II.2 indicates the
anticipated interactions with observing system agencies and user groups.
2.3.2 User requirements and observing system capabilities database
To facilitate the Rolling Requirements Review process, the World Weather Watch Department
has been collecting the observation requirements to meet the needs of all WMO Programmes,
based on techniques such as those listed in 2.2, and has been cataloguing the current and
planned provision of observations, initially from environmental satellites and now extended to
in situ observing systems. The resulting database is called the Database on User Requirements
and Observing System Capabilities and is accessible via the WMO Space Programme web page:
http://www.wmo.int/pages/prog/sat/index_en.php. For example, Appendix II.1, extracted
from this database, tabulates a part of the observations required currently for global numerical
weather prediction.
2.3.2.1 User requirements
The user requirements are system independent: they are intended to be technology-free in that
no consideration is given to what type of measurement characteristics, observing platforms or
data processing systems are necessary, or even possible, to meet them. A 2005–2015 time frame
has been set to meet those requirements. The database has been constructed in the context of
a given application, or use. The observation requirements are stated quantitatively in terms of a
set of relevant parameters, of which the most important are horizontal and vertical resolution,
frequency (observation cycle), timeliness (delay in availability), and uncertainty (acceptable
root mean square error and any limitations on bias). For each application, there is usually no
abrupt transition in the utility of an observation as its quality changes; improved observations,
in terms of resolution, frequency or uncertainty, for example, are generally more useful, whereas
degraded observations, although less useful, are on the whole not useless. Moreover, the range
of utility varies from one application to another. The requirements for each parameter are
expressed in terms of two values, a maximum or goal and a minimum, or threshold requirement.
The maximum requirement or goal is an optimal value: if exceeded, no significant improvement
2
Review Description of proposed New
and systems “System specifications” initiatives
update
Space and in situ observing system operators
2
Summary of Plan and
Description of present/ Other
present/planned/proposed implement
planned systems inputs
system capabilities systems
3 4
Statement of conformance Review 4
3
of present/planned/proposed and update Statement of Guidance
Critical
system capabilities to Guidance on on feasibility of meeting
Review
maximum/minimum feasibility requirements
requirements
1
User requirements
(technology free)
For each application: Users
Statement of maximum
and minimum requirements
1
Review Feedback to users and
and Expertise on each application technical commissions
update
Note: The four stages of the Rolling Requirements Review process are 1, 2, 3 and 4.
Figure II.2. Rolling requirement review process

in performance is expected for the application in question. Therefore the cost of enhancing the
observations beyond this maximum requirement would not be matched by a corresponding
increased benefit.
Maximum requirements are likely to change as applications progress; they develop a capacity
to make use of better observations. The minimum requirement is the threshold below which the
observation is not significantly useful for the application in question, or below which the benefit
derived does not compensate the additional cost involved in using the observation. Assessment
of minimum requirements for any given observing system is complicated by assumptions
concerning which other observing systems are likely to be available. It may be unrealistic to try
to state the minimum requirement in an absolute sense because the very existence of a given
application relies on the existence of a basic observing capability. The observations become
progressively more useful in the range between the minimum and maximum requirements.
2.3.2.2 Observing system capabilities
Initially, attention was focused on the Global Observing System space-based subsystem
capabilities. Each of the contributing space agencies has provided a summary of the potential
performances of their instruments expressed in the same terms as the user requirements,
together with sufficiently detailed descriptions of the instruments and tasks to support
performance evaluations. Assessment of service continuity is based on the programmatic
information supplied. Particular care has been taken to establish a common language in the
form of agreed definitions for the geophysical parameters for which observations are required or
provided and agreed terminology to characterize requirements and performances.
At present, the performance Global Observing System surface-based subsystem components

have also been characterized in a similar manner, taking into account their uneven distribution
across the board in 34 homogeneous regions.
2.3.3 The Critical Review
The comparison of requirements to capabilities utilizes the database. As the database changes to
reflect more effectively user requirements and existing and planned observing capabilities, the
Rolling Requirements Review must be performed periodically.
The process compares user requirements with observing system capabilities and records the
results in terms of the extent to which the capabilities of present, planned and proposed systems
meet stated requirements. This is a challenging process, and considerable work has been done to
develop a process and presentation ensuring that the critical review meets the following criteria:
(a) The presentation must be concise and attractive, and understandable to senior managers
and decision makers, whilst retaining sufficient detail to represent adequately the full range
of observational requirements and observing system capabilities;
(b) The presentation of user requirements must be accurate; although it is necessarily a

summary, it must be recognizable to experts in each application as a correct interpretation
of their requirements;
(c) The presentation of the observing system capabilities must be accurate; although it is also
a summary, it must be recognizable to expert data users as a correct interpretation of the
system’s characteristics and potential;
(d) The results must accurately reflect the extent to which current systems are useful in practice,
whilst drawing attention to those areas in which they do not meet some or all of the user
requirements;
(e) The process must be as objective as possible.

An example of the output of the Rolling Requirements Review of the space-based and surface-
based subsystems capabilities to meet the requirement to measure wind profiles for the
numerical weather prediction application is shown in Appendix II.2. This is a single parameter
for a single applications area. The process produces hundreds of these charts, but software tools
have been developed which provide the required subsets of charts to experts involved in the
Rolling Requirements Review.
2.3.4 Statements of Guidance
Statements of Guidance are designed to provide an interpretation of the output of the Critical
Review, draw conclusions and identify priorities for action. The process of preparing such
a Statement is necessarily more subjective than that of the critical review. Moreover, whilst
a Review attempts to provide a comprehensive summary, a Statement of Guidance is more
selective, drawing out key issues. At this stage, judgements are required concerning, for
example, the relative importance of observations of different variables.
Since the Preliminary Statement of Guidance was published by WMO in 1998, several updates
and additions have been completed in order to extend the process to new application areas, take
into account the evolving nature of requirements and include the capabilities of surface-based
sensors (WMO, 1999, 2001).
The latest Statements of Guidance can be found via the WMO Space Programme web page:
http://www.wmo.int/pages/prog/sat/spaceweather-intro_en.php.
2.4 NETWORK DESIGN AND NATIONAL REQUIREMENTS
In addition to the Global Observing System, observing networks may be required at the national
level for the derivation of local weather parameters from forecast fields, verification of the quality
of issued forecasts and warnings and other real or non-real-time applications. The observational
data required for this purpose include surface and upper-air data obtained from land stations
and ships, aircraft and buoys, as well as weather radar data and satellite information.
National observing networks are designed by Members to meet their individual needs or in
agreement with other Members, in accordance with WMO regulatory and guidance material.
Network design should take into account special observational data requirements and forecast
products of the end-user groups for whom the services are being provided. Much of the data
requirements for individual services may often require additional data, denser networks or
higher observation frequency.
2.5 EVOLUTION OF THE GLOBAL OBSERVING SYSTEM
The Global Observing System gradually evolves to address global, regional and national
requirements for observational data. Many of the requirements stated cannot be met without
space-based observing systems. In most cases, a combination of satellite and in situ data is
needed to obtain adequate resolution and to ensure remote sensing system calibration stability.
The Global Observing System will, therefore, continue to be composed of surface-and space-
based subsystems. However, owing to resource constraints, careful judgements should be
made regarding the value of increased quality of Global Data-processing and Forecasting
System output products weighed against the costs of additional observations. The definition of
requirements and Global Observing System design are largely influenced by cost and countries’
ability to operate Global Observing System components and facilities. It is, therefore, important
to define realistic and achievable goals for Members in respect of the composite Global
Observing System.
Based on the Statements of Guidance mentioned above, a Vision for the Global Observing
System in 2015 and beyond was developed and agreed by the Commission for Basic Systems
at its extraordinary session in December 2002. The evolution for the Global Observing System
was framed in the 42 recommendations of final report CBS/IOS/ICT-2 (2002). Twenty‑two
recommendations relating to the Global Observing System surface-based component
address more complete and timely data distribution; enhanced aircraft meteorological data
relay (AMDAR), especially over data-sparse areas; optimized rawinsonde launches; targeted
observations; inclusion of a ground-based global positioning system (GPS), radars and wind
profilers; increased oceanic coverage through expanded Automated Shipboard Aerological
Programme (ASAP) observations, drifting buoys and ARGOS, the data relay and platform
location system; and possible use of unmanned aerial vehicles. The 20 recommendations
for the Global Observing System space-based component state the need for six operational
geostationary and four optimally spaced polar-orbiting satellites complemented by satellites for
research and development; they call upon rigorous calibration of remotely sensed radiances and
improved spatial, spectral, temporal and radiometric resolution. The wind profiling and global
precipitation missions are singled out for their importance to the Global Observing System.
The Implementation Plan for Evolution of Space and Surface-based Subsystems of the GOS
(WMO/TD-No. 1267) was endorsed by Commission for Basic Systems at its thirteenth session
in February 2005. The Implementation Plan is subject to regular review and provides essential
guidance for the Global Observing System as it advances towards its vision for 2015. Following an
approval of the new Vision for the Global Observing System in 2025 by the Commission for Basic
Systems at its fourteenth session in April 2009 and the Executive Council at its sixty-first session
in June 2009, the Commission is developing a new Implementation Plan for the evolution of the
global observing systems taking into account a new vision as well as the WMO Integrated Global
Observing System.
References
Abridged Final Report with Resolutions and Recommendations of the Extraordinary Session (2002) of the
Commission for Basic Systems (WMO-No. 955), general summary, 6.1.14 to 6.1.24.
Implementation Plan for Evolution of Space and Surface-based Subsystems of the GOS Developed by the CBS Open
Programme Area Group on the Integrated Observing Systems (OPAG/IOS) (WMO/TD-No. 1267)
Preliminary Statement of Guidance Regarding How Well Satellite Capabilities Meet WMO User Requirements in
Several Application Areas (WMO/TD-No. 913, SAT-21)
Statement of Guidance Regarding How Well Satellite and In Situ Sensor Capabilities Meet WMO User Requirements
in Several Application Areas (WMO/TD-No. 1052, SAT-26)
Statement of Guidance Regarding How Well Satellite Capabilities Meet WMO User Requirements in Several
Application Areas (WMO/TD-No. 992, SAT-22)
APPENDIX II.1. EXTRACT FROM USER REQUIREMENTS AND OBSERVING SYSTEM CAPABILITIES DATABASE –
EXAMPLE OF GLOBAL NUMERICAL WEATHER PREDICTION REQUIREMENTS FOR SOME VARIABLES
Geophysical Horizontal Vertical Observing Delay of Uncertainty Confidence Remarks Identifier Requirement
variable resolution resolution cycle availability source
Threshold (km)
Threshold (km)
Goal (km)
Goal (km)
Threshold
Threshold
Threshold
Goal
Goal
Goal
20/10/2003, ET
Sea-surface bulk WMO_
50 250 3h 360 h 3h 180 h 0.5 K 2K Firm ODRRGOS, Geneva,
temperature Sfc_006C
November 2003
20/10/2003, ET
WMO_
Sea-ice thickness 15 250 1d 7d 1d 7d 50 cm 100 cm Speculative ODRRGOS, Geneva,
Sfc_021
November 2003
20/10/2003, ET
Significant wave WMO_Sfc_
100 250 1h 12 h 1h 4h 0.5 m 1m Firm ODRRGOS, Geneva,
height N059
November 2003
20/10/2003, ET
WMO_
Soil moisture 15 250 1d 7d 0.25 d 1d 10 g kg–1 50 g kg–1 Reasonable ODRRGOS, Geneva,
Sfc_012A
November 2003
Specific humidity 20/10/2003, ET
3 Uncertainty WMO_
profile – higher 50 250 1 1h 12 h 1h 4h 5% 20% Firm ODRRGOS, Geneva,
5% in RH UA_006A
troposphere November 2003
Uncertainty WMO_
profile – lower 50 250 0.4 2 1h 12 h 1h 4h 5% 20% Firm ODRRGOS, Geneva,
5% in RH UA_006
troposphere November 2003
WMO_Sfc_
profile – total 50 500 1h 12 h 1h 4h 1 kg m–2 5 kg m–2 Firm ODRRGOS, Geneva,
N044A
column November 2003
Wind profile
20/10/2003, ET
(horizontal WMO_
50 500 1 10 1h 12 h 1h 4h 1 m s–1 8 m s–1 Firm ODRRGOS, Geneva,
component) – UA_001A
November 2003
higher troposphere
RH Relative humidity
APPENDIX II.2. OUTCOME OF THE ROLLING REQUIREMENTS REVIEW

PROCESS: EXAMPLES
Below is an assessment of the adequacy of space-based and surface-based observing system

capabilities compared with the requirement for a specific variable (horizontal wind component in the
high troposphere) in a specific application area (global numerical weather prediction).
Wind profile 500–100 hPa (HT)

Analysis for global numerical weather prediction (in situ and space-based observing systems)
Requirement summary and assessment key
Colour Horizontal Vertical Observation

Delay Uncertainty
key resolution resolution cycle
km km h h m s–1
Optimum 50.0 1.0 1.0 1.0 1.0
107.7 2.2 2.3 1.6 2.0

Median
232.1 4.6 5.2 2.5 4.0
Threshold 500.0 10.0 12.0 4.0 8.0
Instruments for wind profile 500–100 hPa (HT)
Horizontal Vertical Observation

Delay Uncertainty Mission
resolution resolution cycle
Instrument Orbit
km km h h m s–1 name rating
ACARS P RA-VI WE 175.0 0.1 2.0 1.0 2.00 WWW G

ACARS FL RA-VI WE 38.0 0.6 8.0 1.0 2.00 WWW G
SEVIRI 100.0 5.0 1.0 1.0 4.00 WWW G
ACARS FL RA-VI EE 159.0 0.6 8.0 1.0 2.00 WWW G
ACARS FL RA-V SW 167.0 0.6 12.0 1.0 2.00 WWW G
IMAGES 150.0 5.0 1.0 1.0 5.00 WWW G
IMAGES/MTSAT 150.0 5.0 1.0 1.0 5.00 WWW G
SOUNDER 150.0 5.0 1.0 1.0 5.00 WWW G
ACARS FL RA-II S 310.0 0.6 12.0 1.0 2.00 WWW G
ACARS FL RA-IV N 318.0 0.6 12.0 1.0 2.00 WWW G
ACARS FL RA-IV C 380.0 0.6 12.0 1.0 2.00 WWW G
ACARS FL RA-II W 429.0 0.6 12.0 1.0 2.00 WWW G
MVIRI 150.0 5.0 1.0 2.0 5.00 WWW G
VISSR (GMS-5) 150.0 5.0 1.0 2.0 5.00 WWW G
VHRR 150.0 5.0 1.0 2.0 6.00 WWW G
WND P 449 RA-IV C 700.0 0.3 1.0 0.5 1.50 WWW G
Horizontal Vertical Observation

Delay Uncertainty Mission
resolution resolution cycle
Instrument Orbit
km km h h m s–1 name rating
ACARS FL NAO CST 50.0 0.6 24.0 1.0 2.00 WWW G

WND P 915 RA-IV C 1000.0 0.1 1.0 0.5 2.00 WWW G
ACARS P RA-VI EE 692.0 0.1 2.0 1.0 2.00 WWW G
ACARS FL MED 156.0 0.6 24.0 1.0 2.00 WWW G
ACARS FL NAO OPN 223.0 0.6 24.0 1.0 2.00 WWW G
ACARS P RA-V NW 3821.0 0.1 6.0 1.0 2.00 WWW G
ACARS P RA-V SW 644.0 0.1 6.0 1.0 2.00 WWW G
ACARS FL ARC 270.0 0.6 24.0 1.0 2.00 WWW G
ACARS FL RA-I S 330.0 0.6 24.0 1.0 2.00 WWW G
ACARS FL NIO CST 334.0 0.6 24.0 1.0 2.00 WWW G
ACARS FL RA-I N 375.0 0.6 24.0 1.0 2.00 WWW G

ACARS FL RA-I T 402.0 0.6 24.0 1.0 2.00 WWW G
ACARS FL SAO CST 414.0 0.6 24.0 1.0 2.00 WWW G
RAOBS RA-VI WE 218.0 0.3 16.0 1.5 2.00 WWW G
ACARS FL RA-III N 455.0 0.6 24.0 1.0 2.00 WWW G
ACARS FL NIO OPN 498.0 0.6 24.0 1.0 2.00 WWW G
ACARS FL RA-II E 998.0 0.6 12.0 1.0 2.00 WWW G
ACARS FL RA-II N 614.0 0.6 12.0 1.0 2.00 WWW G
ACARS FL RA-IV S 690.0 0.6 12.0 1.0 2.00 WWW G
ACARS FL RA-V NW 550.0 0.6 12.0 1.0 2.00 WWW G
PART III. THE SURFACE-BASED SUBSYSTEM
3.1 GENERAL
The surface-based subsystem is divided into primary and secondary components. The
subsystem’s primary components are as follows: surface synoptic stations, upper-air synoptic
stations and aircraft meteorological stations. For a detailed composition of the subsystem, see
the Manual on the Global Observing System (WMO‑No. 544), Volume I, Part III, section 1. The
WMO Regional Associations define regional basic networks of surface and upper-air stations
adequately to meet the requirements of Members and the World Weather Watch. A listing and
description of all surface and upper-air stations and the corresponding observing programme are
published in Weather Reporting (WMO‑No. 9), Volume A – Observing Stations.
The primary components of the surface-based subsystem also include additional surface and
upper-air synoptic stations, in particular, fixed and mobile, manned and automatic sea stations
and aircraft meteorological stations, the latter generally operating at non-synoptic hours.
Stations on ships and aircraft are especially important for the provision of information from data-
sparse areas.
The other subsystem components comprise a variety of more or less specialized observing
stations and include aeronautical meteorological stations, research and special-purpose vessel
stations, climatological stations, agricultural meteorological stations and special stations.
3.1.1 Design of observational networks
The following criteria apply to observational networks:
(a) The location of each station should be representative of conditions in space and time;
(b) The station spacing and interval between observations should correspond with the desired
spatial and temporal resolution of the meteorological variables to be measured or observed;
(c) The total number of stations should, for reasons of economy, be as small as possible but as
large as necessary to meet requirements.
In principle the air-mass properties should be sampled at a station covering the smallest
possible area, although instruments should be positioned so that they do not affect each other’s
measurements. In selecting station sites, the intention is to obtain data that are representative
of a larger area. Ideally, all measurements and visual observations at all stations would be made
at the same time, that is, at a predetermined synoptic hour. However, as this is not practical,
measurements should be made within the shortest possible time.
For purposes of uniformity, the following terms are used in this Guide:
(a) Standard time of observation (see the Manual on the Global Observing System
(WMO‑No. 544), Volume I, Part III, 2.3 and Appendix (Definitions, part A);
(b) Scheduled time of observation.
In addition to these times, there will be an “actual time of observation”, that is, the time when an
observation is actually carried out at the station. This time must not deviate by more than a few
minutes from the “scheduled time of observation”. Wherever variables may change considerably
within the period normally required for completing an observation, arrangements should be
made to obtain information on critical variables as close as possible to the scheduled time.
PART III. THE SURFACE-BASED SUBSYSTEM 19
Permanent real-time and near-real-time monitoring of availability of observational data on

completeness and timeliness should be performed by the National Meteorological Service.
In addition, availability of data from regional basic synoptic network stations is monitored
by quantity monitoring coordinated by the WMO Secretariat within the framework of the
World Weather Watch Programme. When the National Meteorological Service does not follow
regulations regarding the standard times of observation determined by the Manual on the Global
Observing System (WMO‑No. 544), the National Meteorological Service can expect negative
results of the monitoring provided by WMO.
Station spacing should be such as to provide sufficiently accurate values for the meteorological
variables required at any point between two stations by means of visual or numerical
interpolation, taking due account of the effects of topography on the variation of variables of
interest. The same consideration applies to time series of observations obtained at the same
location that require a relatively short distance between observing sites and an accuracy of
measurement better than that to be obtained by interpolation. On the other hand, a very dense
network or high frequency of observation could lead to more data than are necessary and thus to
unnecessarily high cost.
Variations in space and time differ for individual meteorological variables and depend on the
topography of the area. If any information is available or can be obtained on spatial or temporal
variations, it can be used to decide upon the network configuration which is necessary to provide
data of the required uncertainty (see The Planning of Meteorological Station Networks (Technical
Note No. 111, WMO‑No. 265)). For certain variables, such as precipitation, a separation of 10 km
between stations may be required in some areas for several purposes, such as very-short-period
forecasting, climatology and hydrological forecasting, although, in the case of rainfall, data from
more widely separated weather radars can also satisfy many requirements. For variables such
as atmospheric pressure and upper winds, a separation of 100 km between stations will suffice.
In general, a fairly homogeneous distribution of observing stations is most suited to support
numerical analyses and forecasts. However, a relatively higher station density may be necessary
to support local or area forecasting, for example, to reflect the differences between coastal
and inland conditions or valley and mountain weather, whereas a lower density is likely to be
sufficient in regions of low population and little topographical variation.
It is generally not practicable within one network to achieve an optimization of such

vastly differing requirements without severely prejudicing either operational and scientific
requirements or economic considerations. The solution to this problem is to establish different
types of networks within the subsystem, such as the regional basic synoptic network and its
selected stations for global exchange, as well as stations established on a national level and
the special networks observing “other variables”. For details, refer to 3.2 to 3.9 below covering
individual types of network and stations.
3.1.2 Planning of networks and stations
When a National Meteorological Service has difficulties in solving a problem due to a lack of
observations within its own area of responsibility, it should first assess which data are required
and from which area, location or height. The next step in decision-making is to determine the
type of network or station most suitable to provide the required data.
If a station is to be integrated into a network, its site has to be chosen primarily based on network
configuration. This can be done by adding a task to or shifting an existing station, or by the
establishment of a new station.
Key considerations for spacing stations within an optimized station network should also be borne
in mind when developing a system of station indicators with consecutive numbers or letters. It
is never practicable to install all the stations required for a network at once, and some indicators
should be kept in reserve to fill the remaining gaps. If no such provision is made, new stations
may create increasing chaos in the system.
For the study of small-scale phenomena, arrangements of a non-network type will sometimes
also prove to be adequate and at the same time more economical. These may apply to
agricultural meteorological observations at a single representative station or to precipitation
measurements along a more or less straight line across a mountainous barrier providing typical
values for the amounts of precipitation along the windward and leeward slopes.
Decision-making must include cost and benefit considerations. The most suitable method of
achieving the highest cost–benefit ratio is normally the co-location of stations. This can be
achieved by establishing a station of another type close to an existing one, or by gradually
augmenting the task of a one-variable observing station to a multi-variable one. This may begin
with the measurement of precipitation only and end up with a round-the-clock programme of
a fully equipped surface and upper-air synoptic station requiring larger facilities and additional
personnel.
Before setting up a new station, and if there is a possible choice of site, the following questions
will be helpful in the decision-making process:
(a) Is the site representative of the required meteorological data?
(b) Will the site remain representative in view of existing or anticipated construction plans or
change of vegetation, for example?
(c) Can steps be taken to improve or safeguard the site’s representativeness, such as cutting
down trees or reserving rights to building and planting limits in the vicinity?
(d) Is the site sufficiently accessible to personnel who operate the station or carry out
inspection and maintenance?
(e) Does the site provide housing and storage facilities, or can they be made available if
necessary?
(f) Are facilities such as electric power, telecommunications and running water available if
required?
(g) To what extent are security measures against lightning, flooding, theft or other
interferences required and how can they be taken?
(h) Can difficulties of posting personnel be overcome by part- or full-time automation or by

locally available staff? Part-time staff from the public services sector are especially suitable
for certain work at weather stations, as the continuity is ensured, regardless of staff changes.
Several points should be taken into account when planning a new observing station or its
network. During this phase, the management of the National Meteorological Service responsible
for the development of the observing network should ask itself the following questions:
(a) Which system to choose for the required observation?
(b) What is the representativeness of meteorological observations of an area in accordance with

application for which they are used?
(c) What are the standards for and definitions of measurements?
(d) What are the standardization procedures?
(e) What instruments are required?
(f) What are the required uncertainties and achievable accuracy?
(g) What are a station’s or network’s general requirements for siting and exposure, inspection
and maintenance, system performance monitoring and availability and quality of data?
(h) How to carry out meteorological observations?
(i) How to establish effective liaison procedures between those responsible for monitoring and
maintenance in order to facilitate prompt remedial actions?
(j) What type of metadata related to meteorological observations is needed?
(k) What measures have been provided for training?
Furthermore, it is wise to choose land that is in public or governmental hands, for there will be
less chance of having to move the station later. A long-term contract should be established with
the authorities concerned or with the landowner, if necessary, with the help of an estate agent.
The validity of the contract should be based on the usual international standard period for
climatological measurements and have a duration of at least 30 years. It should prohibit changes
such as the construction of buildings near the measuring site. The contract should provide for
the installation and operation of instruments and other necessary equipment, transmission and
power lines, and a regulation governing the right of access.
There is an understandable tendency to select a station site which cannot otherwise be utilized
and whose cost is therefore relatively low. Only in very rare circumstances will such a site
correspond with meteorological requirements, which should primarily determine its suitability. It
must be borne in mind that nothing is more costly and frustrating than long observation records
that subsequently prove to be useless or even misleading. Therefore, the following rule should be
observed: “Keep the quality standard as high as necessary and the cost as low as possible”.
More detailed guidance on the location of the observing site can be found under 3.2.1.2.
3.1.3 Management of manned station networks
3.1.3.1 General
The responsibility for the management of a meteorological station network, the primary
task of which is the production of data of the best possible quality, lies with the Member
concerned. The Member should establish an appropriate organizational unit or units within the
Meteorological Service in charge of the operation, maintenance and supervision of the stations,
logistics, procurement, supply and the repair of equipment and other material necessary to
ensure uninterrupted operation. The unit should function operationally within the Service, be
responsible for national standards and have an appropriate status. This unit will also need to
liaise and coordinate its activities with data users at the national level and support services,
administration and finance. A continuous survey of new technological developments is required
with a view to introducing improved types of instruments, equipment and techniques. Further
information on the management of the observational network can be found in the Guide to
Meteorological Instruments and Methods of Observation (WMO‑No. 8), Part III, Chapter 1.
3.1.3.2 Organization of the station network management unit
The organization of the unit should take into account the size of the network. In the case of a
country with very large networks it may be necessary to have a central unit with sub-centres. The
location of such sub-centres depends on the Member’s needs. Economic considerations as well
as problems of a technical and logistic nature, such as personnel, communications and transport
facilities, should be taken into account.
A different approach to station management may be based on the specific functions of

the stations that comprise the network: synoptic, aviation, climatological and agricultural
meteorological stations.
The unit must have means of transportation to carry out its various activities.
3.1.3.3 Administrative arrangements
The unit should have a filing system containing all relevant and up-to-date documentation of a
scientific, technical, operational, and administrative nature (metadata documentation). A station
gazetteer with information on geographical conditions, staff and activity programmes should be
available.
The instruments at the station play a major role in the system. Particular attention should be paid
to maintaining appropriate records relating to the instruments in use, including an up-to-date
equipment inventory. The technical specifications of an instrument, its movement and periodical
test certificates should be available and carefully maintained. For more information, see the Guide
to Meteorological Instruments and Methods of Observation (WMO‑No. 8), Part I, Chapter 1, 1.3.4,
and Part III, Chapter 1.
Station breakdowns, defects, requests for repairs, supply needs and other matters requiring
prompt action, should be stated in brief on an “action card”. Card entries should be prioritized
and action taken on that basis. In addition, the metadata should include a log of instrument
faults, exposure changes and remedial actions taken. Depending on the type of instrument—
mechanical, electric, electronic or mixed—and the nature of the fault, various types of workshops
and laboratories may be involved.
3.1.3.4 Station network management unit staff
Unit staff members must be qualified and specially trained for their work in meteorology. In
addition, the personnel must be sensitive to the human dimension within the Meteorological
Service and in their contacts with voluntary observers, private institutions or other government
agencies outside the Service.
An experienced meteorologist1 specialized in observing work should be in charge of the unit. He

or she should also be a good administrator and an able organizer, whose main responsibility will
be to produce the best observational information for the users, in the most economical way.
The unit may be divided into smaller sections as necessary, for example, when the network is
managed on a geographical or functional basis (see 3.1.3.2). Each section head should also be
a highly qualified, seasoned meteorologist, or hydrologist or engineer and should be capable of
providing direct fieldwork supervision.
Depending on the size of the station network, there should be one or more inspectors who are
members of the meteorological staff—at least qualified as a meteorological technician—with
experience in running observing stations.
Provision should be made for technical staff: station network technicians and technical assistants.
The former are specially trained to cope with all technical problems and activities connected with
station management, involving tasks to be performed both in the field and at the duty station.
The latter should be responsible for carrying out technical tasks involving logistics and links with
the stations.
Finally, the necessary clerical staff should be avail-able for administrative work.
3.1.3.5 Station network management unit operational tasks
The operational tasks are based on the activities and the performance of individual stations. The
unit will carry out the following tasks:
(a) Formulate plans and policy for network development, maintenance and operation;
1
For a classification of meteorological personnel and a description of their duties, see the Guide to the Implementation
of Education and Training Standards in Meteorology and Hydrology (WMO-No. 1083), Volume I.
(b) Maintain network functions;
(c) Monitor network performance, recommend and implement improvements;
(d) Monitor and review network efficiency and effectiveness;
(e) Develop and define performance and calibration standards, procedures and functional
requirements for observations, instruments and equipment, and issue relevant instructions;
(f) Exercise functional control and inspection of the network;
(g) Provide liaison between users of meteorological observations and data and equipment
suppliers;
(h) Advise on technical training for all those involved in the network;
(i) Produce and maintain observation specifications which detail installation specifications for
network observations;
(j) Provide consumables for network measurements;
(k) Advise on long-term re-equipment plans.
Station activities are set forth in the prescribed programme which must be carried out according
to a routine day-to-day schedule. The unit should issue instructions relating to the correct
implementation of standard procedures, the operation of the instrumentation, including
reliability tests, and the use of official communication. It should provide relevant tables, forms
and manuals, and issue directives regarding relationships with local weather data users.
The unit should appoint an inspector responsible for the activities of a group of stations, the
quality of their observations and the smooth working of the instruments.2 A scheme should be
worked out with the users to ensure that incoming observational data and all relevant charts and
forms from a station are routinely checked for errors and the inspector responsible for the station
is notified accordingly. Information about malfunctioning instruments or requests for remedial
action must be evaluated by the inspector so as to enable the unit to rectify discrepancies and
ensure that the stations are operating properly.
Periodic activity reports from the stations should be sent to the unit.
Station personnel must be kept informed about the organization of the National Meteorological
Service, and in particular about the station network. This can be done through a circular letter or
a printed bulletin, which would also be a medium for the dissemination of communications or
messages to and from the stations. Special attention should be given to recognizing events such
as anniversaries, distinguished service and retirements.
3.1.3.6 Logistics and supplies
Each type of station should have its own standards for activities, equipment, instrumentation
and operational procedures; these must be in accordance with the regulations of WMO and of
the relevant Member. An up-to-date inventory of instruments, office equipment and other types
of material at the station should be available.
An efficient communication system must be worked out within the organization to allow the
smooth transmission of messages and information, with more than one type of communication
medium if possible.
2
See Guide to Meteorological Instruments and Methods of Observation (WMO‑No. 8), Part I, Chapter 1, 1.3.5, for the main
purpose of inspections.
The meteorological assistant in charge of the station is responsible for its principal activities
and upkeep of the instrument site. Vegetation around the station and within its perimeter must
not interfere with the operation of the instruments. Installation, repair and major maintenance
work on the equipment is the responsibility of a maintenance staff from the station network
management unit.
A system for ordering forms, charts and other expendables for the stations, preferably on a half-
yearly basis, should be developed and implemented. Necessary supplies should be forwarded on
demand to the station by means of a reliable system, bearing in mind that most of the material
is fragile; special packing providing adequate protection such as boxes, cardboard, cushions or
pads should be used as needed.
3.1.3.7 Establishment of a new station
The first step following the decision to establish a station is to visit the site. All the requirements
must be assessed so as to ensure that the instruments to be installed can operate unhampered.
It must be ascertained whether appropriate working conditions for the observer, office
accommodation and other required facilities such as running water, electricity and
communications are available.
The unit should prepare well in advance all instruments, equipment, supplies and material
required for a new station.
The task of setting up a new station is assigned to a team that includes an inspector, a technician
and assistants. The team should be trained for the specific job to be carried out in the most
efficient manner according to a detailed standard plan.
During the installation of the equipment, the necessary explanations should be given to the
meteorological assistant to be placed in charge of the station to enable him/her to assume full
responsibility for its operation.
A detailed report is to be written on the new station. This should include, preferably in the form
of a checklist, a description of the site and its surroundings, accompanied by a drawing and an
extract from a detailed map of the area. A visibility chart should be prepared for a surface station.
The report should include details about instruments, their operation, test results, tables to be
used and an inventory. It is recommended that pictures taken from the four main directions be
included.
The operation and performance of a newly established station should be closely monitored by
the unit. The documentation that arrives after the first month of operation must be carefully
reviewed. Following the checking of data and evaluation of any deficiency, further visits to the
station may become necessary. Thereafter, a system of regular inspections should be adopted.
3.1.3.8 Regular inspections
Regular inspections, including routine maintenance activities at automatic stations, help to

ensure the smooth functioning of a meteorological station. A detailed schedule is to be worked
out by the unit, spacing the inspections according to national practices. The inspection should
follow a standardized checklist whereby information accumulated from the previous inspection,
relevant station files, notification by other users and, if necessary, from special inquiries made
before departure, will provide additional guidance to the inspector. Field tests of instruments at
the station should be included among the items requiring the inspector’s attention (see 3.1.3.10
and 3.1.3.11). See the Manual on the Global Observing System (WMO‑No. 544), Part III, section 3,
for the range and frequency of regular inspections.
The findings of regular inspections should be documented in an inspection report which can
be less elaborate than the report referred to in 3.1.3.7. Copies of the report should be circulated
to the users of the observational data within the organization, the administration and others
involved in the meteorological station’s activities.
3.1.3.9 Other station network management unit activities
The unit may provide technical assistance to other bodies outside the National Meteorological
Service if requested. Such assistance may be provided in writing or by active participation in
various projects involving the performance of instruments and the application of meteorology
and operational hydrology.
3.1.3.10 Procurement of instruments and equipment
The equipment used in the observing station network of a Member should be in accordance with
the general requirements of meteorological instruments as laid down in 1.4, Chapter 1, Part I, of
the Guide to Meteorological Instruments and Methods of Observation (WMO‑No. 8). The instruments
should be standardized and suitable for operation under prevailing climatic conditions. The
standard instrument should be carefully selected, taking into account both economic and
technical considerations, so as to ensure introduction of the best possible type of instrument into
the system.
Instruments should be introduced only after a series of intercomparison tests and other checks
have been performed. Intercomparisons of instruments should subsequently be carried out
during each regular station inspection. Portable standard instruments used by inspectors should
be checked against the relevant standards used by the Service before and after each tour of
inspection.
Once a decision to procure a certain type of instrument has been made, the necessary
administrative steps should be taken. Testing procedures should be instituted following the
arrival of the shipment to determine if the instrument deviates in any way from the national
standard, particularly in the operational scale range. Test certificates will be issued for each
instrument. An instrument that falls short of the required uncertainty should not be introduced
into the system. A separate record card will be established for each new instrument (see 3.1.3.3).
A minimum stock of instruments to be used must be established; the personnel in charge of

procurement must make sure that it is maintained at the required level. An emergency reserve
is recommended, especially for items difficult to replace. The system for keeping the spare
instruments should be well organized.
An ordering and issuing system should be introduced within the Service. It should be applied to
all instruments delivered to the organization from an outside supplier and allocated to individual
stations in the network through the station network management unit.
Constant efforts must be made to introduce improvements in the quality, performance and price
competitiveness of the various supplies. In the case of equipment, the search for improved ideas
and means is very important.
Perishable items should be stocked properly and used regularly. Where items such as
meteorological balloons or batteries are concerned, quality tests should be made from time to
time.
A computerized information system may be of great advantage in managing the equipment.

In organizations where these facilities do not exist, a manual follow-up system must be
implemented.
3.1.3.11 Instrument checks and maintenance
A system for checking the instruments at a station regularly should be introduced so that faults
can be discovered at an early stage. The system should include regular reliability tests. If faults
are discovered or suspected, the unit should be notified immediately. Depending on the nature
of the fault and the type of station, the unit will decide whether the instrument is to be changed
or a field repair carried out.
The station inspector shall assist the unit in keeping the instruments in the best possible working
order and in carrying out periodic intercomparisons with national standards. (See 3.1.3.8 and
3.1.3.10.)
3.1.3.12 Coordination
In addition to the circulation of inspection reports within the divisions or sections concerned
and the notification of discrepancies or likely errors in the observational data, close coordination
between the various users of the observational data in other branches of the organization and
the unit should be arranged. Periodic meetings should be held to discuss and decide on desirable
improvements or changes. Appropriate working arrangements within the unit for different types
of repairs, such as electrical or mechanical repairs, for example, including familiarization with
new equipment, will also be necessary.
3.1.3.13 Planning and budgeting
Planning, which should be short-term (one to two years) as well as medium- and long-term (five
years or more), is concerned mainly with changes and improvements in the system, priorities to
be set, development and new technology. Owing to financial implications, the cost-effectiveness
of any new type of equipment will be an important factor to be taken into consideration.
Planning-related decisions may have important effects on the organizational structure for
management of the station network and staff and training requirements.
3.1.3.14 Network performance monitoring
As real-time quality control procedures implemented by the National Meteorological Service

have their limitations and some errors can go undetected, National Meteorological Centre
(NMC) quality control monitoring at the network level is required by network managers at the
Centre. Real-time quality control monitoring should include checks of the following items:
(a) Completeness of observations at the observing station;

(b) Quality of data transmitted from the station;
(c) Completeness and timeliness of observational data at the Centre.
Quality control monitoring is intended to identify deficiencies and errors, monitor them and
activate appropriate remedial procedures.
Quality control monitoring requires the preparation of summaries and various statistics.
Therefore, it is necessary to build up a quality control monitoring system to collect statistics on
observational errors of individual meteorological variables through a series of flags indicating the
results of each check, and generate hourly, daily, weekly, monthly and yearly statistics. Stations
with a large percentage of failed observations probably experience hardware or software failures
or inappropriate maintenance. These should be referred back to the network manager.
The quality control monitoring system should keep station monitoring statistics on the frequency
and magnitude of observation errors encountered at each station. The statistics provide
information aimed at:
(a) Monitoring station performance;

(b) Locating persistent biases or failures in observations;
(c) Evaluating improvement of observation data quality, performance and maintenance of

station/network.
3.1.4 Management of automatic surface land station networks
3.1.4.1 General
Since automatic meteorological land stations are generally used to expand a basic manned
station network, the management of automatic station networks should, in principle, follow
the same general rules and practices as for the management of manned station networks (see
3.1.3). This is to guarantee the acquisition of an observational data set with comparable quality
and accuracy, as can be achieved by a manned station network. According to the Manual on the
Global Observing System (WMO‑No. 544), Volume I, Part III, 3.1.10, automatic stations should be
inspected not less than once every six months.
Detailed information on automatic stations can be found in 3.2.1.4 of the present Guide and
in the Guide to Meteorological Instruments and Methods of Observation (WMO‑No. 8), Part II,
Chapter 1.
For reasons of compatibility and homogeneity of data generated by automatic stations with
similar data from manned stations, the responsibility for the management of an automatic
station network should rest with the same organizational unit or units within the meteorological
authority in charge of the management of manned station networks. The main aim should be to
implement a composite observing system of consistent quality at global, regional and national
levels.
3.1.4.2 Administrative arrangements
The station network management unit should have access to all technical details of the
configuration and the sensor files for each automatic station installed in an operational network.
Experience in operational system evaluations and scientific networks studies have shown that
the preparation of national operating instructions for weather stations equipped with devices for
automatic data acquisition is essential for the satisfactory employment of new components such
as automatic weather stations.
In view of the special position of an automatic weather station within the flow of data from the
observing site to the national data-processing centre, many system features must be taken into
account when preparing the necessary guidance material.
Since the technology used at automatic weather stations is changing rapidly, more emphasis
should be placed on new areas of automation, for example, in the field of data acquisition,
processing and local archiving techniques for meteorological measurements. In an automated
system, a large number of different algorithms are used to define the quality control routines:
to evaluate with appropriate smoothing the physical quantities from the digital measurements
and translate the resulting list of measured quantities into the WMO code format. International
standardization remains to be achieved.
3.1.4.3 Operational tasks of the automatic station network supervising unit
The operational tasks of the network supervising unit may vary according to the type of
automatic station used.
(a) Supervision of a semi-automatic station network
As with manned station networks, instructions on the observation of standard procedures must
be prepared and strictly followed by the personnel in charge. The instructions should include
guidance on the operation of instrumentation and on preventive maintenance measures and,
where feasible, may include small repairs of some automatic instrumentation or sensors carried
out at observing sites. The unit should perform regular inspections of these stations in order to
check the operation of automatic instruments or sensors.
Where appropriate, a diagnostic check of the operation can be carried out together with the data
quality control at the national data-collection centre. Information on possible malfunctioning
should be transmitted as soon as possible to the maintenance experts (see also 3.1.3.14).
(b) Supervision of a fully automatic station network
Since the technology used in automatic surface observing systems is complicated, the unit
may need to consult specialists in order to address various problems in electronics, software,
telecommunications and sensor engineering. It is useful for the unit to be involved in network
management from the initial stages of its deployment starting with delivery, site preparation,
check-out and activation. The unit should have access to all documentation concerning
equipment, system configuration, site specification, software system and engineering services.
To ensure the reliability of sensors, data-acquisition systems and data quality, the staff should
be provided with guidance material on manned and automatic test requirements. Equipment
control procedures for remote automatic tests may include duty-checks performed on a
daily basis. Nevertheless, regular on-site instrument field tests and inspections are needed to
guarantee proper functioning of the automatic station network.
The unit should provide engineering support for the operation of the network and guidance
material to the technical staff. Likely system modifications, additions and site relocations in the
future also require engineering support and in some cases revised operational software versions.
Operational tasks of the automatic station network supervising unit also include the organization
of training courses. (See 3.1.3.5.)
3.2 SURFACE SYNOPTIC STATIONS
3.2.1 Organizational aspects
3.2.1.1 General
Surface synoptic stations may be on land or at sea, manned or automatic. For the purpose of
the present Guide, surface synoptic stations are dealt with under three categories, namely land
stations, sea stations and automatic stations.
The establishment of a network of stations, their operation in accordance with prescribed

standards and their maintenance involve many questions of an organizational nature of varying
degrees of complexity, depending on the type of station, its location, functions, instrumental
equipment, communication links for data transmission and requirement for trained personnel of
different levels. The broad aspects of such questions, as they apply to each type of station falling
under the three surface synoptic station categories referred to above, are discussed in 3.2.1.2,
3.2.1.3 and 3.2.1.4 below.
3.2.1.2 Land stations3
3.2.1.2.1 Station siting or location
Each station making surface synoptic observations should be located at a site where the
meteorological data obtained are representative of the state of the atmosphere over a large
region. The dimensions of this region, or area of representativity, may range from 2 000 km2 to
10 000 km2 for a plane or homogeneous relief.
The station should have a plot of land specially assigned to it. The optimum area is approximately
1 ha.
The location of the observing posts, or meteorological instrument areas, should be typical of
geographical conditions of the surrounding area and protected from the influence of industry.
Therefore, a meteorological instrument area should be located in an open site far from any
constructions or woods. The minimum distance from constructions and groups of trees should
be greater than 10 and 20 times their heights, respectively. The site should also be farther than
100 m from bodies of water, except where coastal measurements are required.
3.2.1.2.2 Meteorological observing area
The meteorological observing area is where most of the instruments and devices are situated.
Where there are many installations, the observing area should ideally be no smaller than 25 m
x 25 m, but in cases where there are relatively few installations, as in Figure III.1, the area may
be considerably smaller. The sides of the observing area should be oriented north-south and
east-west. An adequate north-south dimension is very important for measurements which can
be strongly influenced by shadow, for example, radiation, sunshine duration and temperature
gradients just below and above the ground.
The instruments and equipment should be set out in a definite order, in several rows or lines. In
the northern hemisphere the sensors are arranged as follows: wind-measuring equipment on the
north side, along with temperature and humidity equipment, then a row of precipitation gauges
with soil temperature measurement taking place in the southern part of the observing area.
Figure III.1 provides an example of the layout of an observing station in the northern hemisphere
showing minimum distances between installations.
The meteorological observing area should be surrounded by open fencing or palings to keep out
unauthorized persons. In the Arctic, desert and certain other regions the observing area may not
need to be fenced and may simply be marked.
The observing area surface must be left in its natural state, and grass should be kept down to
20 cm. The area should not be walked on except along paths or tracks. Paths should not be of
asphalt or concrete. In the interest of safety, the electrical voltage supplied to the equipment shall
not exceed 24 or 36 volts. Installations should preferably be painted white; any other colour may
be used for masts and fencing.
If the area covers one or more hectares, special protected zones extending about 200 m in all
directions from the boundaries of the station area should be provided around the plots. Where
possible, these zones should be left unchanged and their use shall be agreed with the National
Meteorological Service.
Particular attention should be given to the following points in selecting the site for the
measurement of precipitation:
3
The requirements for station siting and instrument exposure given in this section apply to an “ideal” situation, which
should be attained where possible. It is understood that these requirements cannot always be fully met for one
reason or another.
(a) Any precipitation-measuring method should aim at obtaining a sample representative of

the true amount falling over the area that the measurement is intended to represent. The
site choice and systematic measurement error are therefore important;
(b) The systematic wind field deformation above an elevated gauge orifice, as well as the
effects of the site itself on the air trajectories, should be considered when choosing a site;
(c) For each site, the average vertical angle of obstacles should be estimated, and a site plan
made. Sites on a slope or roof should be avoided. The surface surrounding the precipitation
gauge can be covered with short grass or gravel or shingle, but hard flat surfaces, such as
concrete should be avoided to prevent excessive in-splashing;
(d) In areas where there is homogeneous dense vegetation, such vegetation should be clipped
regularly to keep it at the same level as the gauge orifice;
(e) Sites selected for measurement of snowfall and/or snow cover should be sheltered from
the wind as much as possible. The best sites are often found in clearings within forests or
orchards, among trees, in scrub or shrub forests, or where other objects act as an effective
windbreak for winds from all directions.
Further information on siting and exposure can be found in 1.1.2 and 1.3.3.1, Chapter 1, Part I, of
the Guide to Meteorological Instruments and Methods of Observation (WMO‑No. 8).
1.5 m 1.5 m
1.5 m 1.5 m
Cup-counter
Thermometer anemometer
screen on slender
2 m pole
100 cm
2m Soil thermometer
Rain-gauge 1
1.5 m 1.5 m
W Rain-gauge 2 1.5 m Recording E

3m
Soil thermometer rain-gauge
30 cm
60 cm
Mini- Grass minimum thermometer
Concrete slab mum
1.4 m thermo- 1 m
meter 75 m
2m
Soil thermometers
5m
1.25 m 20 cm 10 cm 5cm
Sunshine Bare-soil minimum 2m

thermometer
recorder
1.5 m on 2-m 1m
Bare patch to be
pillar kept weeded
1.5 m
1m
Figure III.1. Layout of an observing station in the northern hemisphere showing minimum
distances between installations
Source: The Observer’s Handbook, Meteorological Office, United Kingdom, 1982
3.2.1.2.3 Observatory premises
To ensure normal operations, each station should be provided with premises suitable for the
working staff, with optimum floor space, heating and/or cooling system as required, safety and
fire-fighting equipment and an emergency electricity supply.
3.2.1.2.4 Observing station staff
Every station must be provided with personnel whose number and functions are established in
conformity with the Member’s regulations and standards, taking into account the observation
programme and other work carried out by the particular station. The work at land stations
should preferably not be interrupted between the observation times.
A station working around the clock to collect and transmit emergency information on dangerous
weather phenomena, in addition to the standard observations carried out at the eight synoptic
times, usually has a staff of five. For a station that only carries out observations at the eight
synoptic times and is not continuously attended, three are sufficient.
Official staff titles, such as senior technician, technician, senior observer and observer, are
determined by the type and importance of the data gathered by the station, the degree of
complexity of the measuring equipment used, staff duties and National Meteorological Service
practice.
Observers who are not full-time officials of a National Meteorological Service but are designated
to make meteorological observations at any synoptic station shall be certified by the appropriate
Service as having a sound knowledge of observing instructions and being capable of observing
meteorological variables with the accuracy required. Similarly, the National Meteorological
Service should certify the competence of any other observers who are responsible for
meteorological observations.
3.2.1.2.5 Station staff training
Each station should be provided with trained personnel according to the WMO classification
scheme; for more details, see the Guidelines for the Education and Training of Personnel in
Meteorology and Operational Hydrology (WMO‑No 258). Training of meteorological staff and other
specialists for work at the station is arranged by the Member within the country, or appropriate
courses are sought abroad. In addition to the original training in the area of specialization, staff
should undergo periodic refresher courses to maintain their efficiency. General and specific
staff training guidelines are contained in relevant WMO publications, for example, the Guide to
Meteorological Instruments and Methods of Observation (WMO‑No. 8), Part III, Chapter 5.
In order to ensure reliable observations and information, it is recommended that the observing
personnel be trained at the following levels:
(a) Chiefs of meteorological stations making standard observations (see 3.2.2): intermediate
specialized training (completion of a technical college or equivalent);
(b) Technicians selected amongst the most experienced of the junior technicians or observers:
same as above;
(c) Junior technicians or observers: special training or courses at special schools lasting no less
than six months.
Notes:
1. Up to two observers can be trained on the job at the station itself (for no less than one month), preferably with
subsequent training provided by courses at special training centres or by correspondence.
2. For a description of the classification of meteorological personnel and their duties, see Guide to the Implementation
of Education and Training Standards in Meteorology and Hydrology (WMO‑No. 1083), Volume I – Meteorology.
Stations should be provided with all necessary documentation, manuals, guides, other
instructions and guidelines to which all staff should have access, and which they should study
regularly.
3.2.1.2.6 Station identification
All stations contributing to WMO observing systems are to be uniquely identified by a WIGOS
station identifier. Further information about WIGOS station identifiers can be found in the
Manual on the WMO Integrated Global Observing System (WMO-No. 1160), section 2.4.1 and
Attachment 2.1, as well as in the Guide to the WMO Integrated Global Observing System (WMO-
No. 1165). As indicated in the Manual on the Global Observing System (WMO-No. 544), Volume I,
Part III, section 2.3.2, some of the now-expired identification requirements for synoptic stations
are reproduced below because they may be adopted by an issuer of identifiers as a convention to
be followed in defining local identifiers for new stations:
A surface station included in the regional basic synoptic network shall be

identified by a station index number assigned by the relevant Member in
compliance with the scheme set out in the Manual on Codes (WMO‑No. 306),
Volume I.1, Part A. The list of station index numbers, observing programmes
and other relevant information is published by the WMO Secretariat in Weather
Reporting (WMO‑No. 9), Volume A – Observing stations.
Each Member operating synoptic stations is required to provide the WMO

Secretariat with the necessary information for this purpose in compliance
with the regulations laid down in the Manual on the Global Observing System
(WMO‑No. 544), Volume I, Part III, 2.3.2.
Each Member should keep an up-to-date list, or directory, of the synoptic

stations on its territory, providing the following information for each station:
(a) Name, and where appropriate, station index number;
(b) Geographical coordinates in degrees, minutes and integer seconds of arc;4
(c) Elevation of the station in metres (up to two decimals) above mean sea
level;4
(d) Geopotential of the datum level in whole metres to which the pressure is
reduced, or the reference isobaric surface the geopotential of which is reported;
(e) Category of the station and observing programme;
(f) Times at which synoptic observations are made and reported;
(g) Brief description of surrounding topography;
(h) Instrument exposure, in particular height above ground of thermometers,

rain-gauges and anemometers;
(i) Station history: date regular observations were initiated, transfers,

interruptions in observations, name changes and any substantial changes made
to the observing programme;
(j) Name of supervising organization or institution;
4
See Guide to Meteorological Instruments and Methods of Observation (WMO-No. 8), Part I, Chapter 1, 1.3.3.2.
(k) Any other information required for completion of the entries in Weather
Reporting (WMO‑No. 9), Volume A – Observing stations.
3.2.1.2.7 Telecommunications
All stations shall be provided with means of telecommunication to transmit their data as fast as
possible to meet the needs of forecasting services (global, regional and national requirements),
and of local users, on a permanent basis and on request. The equipment used at the stations to
transmit and receive information may be of various kinds, including telephone, telegraph and
radio. General and specific guidelines on the collection and transmission of information are
contained in the Manual on the Global Telecommunication System (WMO‑No. 386).
Each synoptic station whose reports are included in the list for international exchange shall
be provided with such telecommunication equipment as will guarantee regular and reliable
transmission of the necessary reports and other information to the established addressees of
messages.
3.2.1.2.8 Quality standards
Reference should be made to:
(a) Manual on the Global Data-processing and Forecasting System (WMO‑No. 485), Part II, 2.1.3 –
Minimum standards;
(b) Guide on the Global Data-processing System (WMO‑No. 305), Chapter 6.
3.2.1.3 Sea stations
3.2.1.3.1 General
About 70 per cent of the Earth’s surface is covered by the oceans. Obtaining regular and
adequate meteorological and oceanographic information from these vast areas is an important
task because timely and precise weather forecasts and services for marine interests depend
heavily on observations from the oceans.
3.2.1.3.2 Fixed sea stations
3.2.1.3.2.1 Ocean weather stations
(a) General
Ocean weather stations are the most sophisticated of the meteorological sea stations. Owing
to high costs, ocean weather station networks are generally organized as a joint project of
participating Members, individual Members being responsible for operating the ships from
national harbours. An example is the North Atlantic Ocean Stations network, which is operated
on this basis under WMO auspices.
(b) Station design
An ocean weather station consists of a ship specially built or refurbished for that purpose. To keep
a continuous observing programme at a certain location, more than one ship is needed. The ship
must have deck space for launching balloons for upper-air observations and adequate space for
meteorological instruments. There must also be space for supplies and expendables for 30 to
40 days with the safety measures necessitated by the use of hydrogen. The main storage area
should, however, be located in the harbour from which the ship operates. The ship must have
sufficient accommodation for the crew and meteorological personnel.
The variables comprising a surface synoptic observation from an ocean weather station are listed
in 2.3.3.11, Volume I, Part III, of the Manual on the Global Observing System (WMO‑No. 544) and
many of them are the same as for land stations given elsewhere in this Guide (see 3.2.2.2). For
sea stations, there are different ways of obtaining the meteorological variables in some cases.
Generally, meteorological instrument exposure may be more difficult for sea stations because of
the limited area and the influence of the superstructure of a ship or other installations. Figure III.2
offers information on where to expose different instruments.
(c) Site selection
Station positions should be chosen in the best interests of the National Meteorological Services
and the Global Observing System. The harbours from which the ships operate should be chosen
in such a way as to minimize the distance to the positions they occupy during operation at sea.
(d) Operations
National Meteorological Services operating the ships should be responsible for technical and
scientific standards and calibration and maintenance of the instruments on board. A supervisor
from the National Meteorological Service should ensure that all observational work is done
efficiently and in accordance with the regulations. He or she should also ensure that the
personnel are properly trained and that all relevant manuals and other documents are available
to the personnel.
(e) Identification
The ocean weather stations—ships—shall be identified by an alphanumerical name assigned to

the station position, not the name of the individual ship, for example, C7R.
(f) Communications
Foremast:
wind speed
wind direction
Aftermast:
wind speed
wind direction
Bow boom:
Port: wind speed
temperature wind direction
humidity temperature
Solar
radiation
Pressure
Starboard:
temperature
Water tunnel:
Central humidity
Operator terminal unit Water temperature conductivity
Ship direction
Ship speed
Ship data processing
Video monitors
220V 50Hz Battery
Figure III.2. Instrument exposure on an ocean weather station in Vaisala Oy, Finland a
The types of equipment suitable for timely data transmission from ocean weather stations can
include the following:
(i) Interface with public communication networks;

(ii) Telegraphs;
(iii) Telex typewriters;
(iv) Radioteleprinters;
(v) Radio facsimile broadcasts;
(vi) Radios;
(vii) Global system for mobile communication;
(viii) Satellites.
There should be at least one alternative in case of failure or disruption of the primary link.
(g) Personnel and training
Three types of personnel are needed to operate ocean weather ships:
(i) Ship’s crew;

(ii) Meteorological personnel (observers and technicians);
(iii) Telecommunication personnel.
The number of staff in (ii) and (iii) depends upon the equipment used and the level of expertise
required. Observers can be given the responsibility to carry out the data dissemination
procedure over the GTS. If given adequate training, they can also be responsible for the
operation and maintenance of the equipment on board.
Using the ordinary crew as observers or operators under an experienced meteorological

supervisor has proved, at least for one of the North Atlantic Ocean Stations, to be a very efficient
way to reduce operating costs. Some of the crew members must then be properly trained to
make the observations. The number of people required to operate an ocean weather ship may be
reduced considerably in this manner.
(h) Quality standards
Reference should be made to the Guide to the Applications of Marine Climatology (WMO‑No. 781),
3.1.4 – Quality control, processing and archiving data, Appendix I – Minimum Quality Control
Standards; the Guide to Marine Meteorological Services (WMO‑No. 471), 3.2.9 – Quality control,
Annex 3E – Minimum Quality Control Standards; the Manual on Marine Meteorological Services
(WMO‑No. 558), Volume I, 5.6.3 – Quality control of data and Appendix I.15, Minimum Quality
Control Standards; and the Manual of Quality Control Procedures for Validation of Oceanographic
Data, Manuals and Guides, No. 26, UNESCO.
3.2.1.3.2.2 Lightship stations, island and coastal stations
(a) General
These stations may be important for the regional basic synoptic network and the global network.
Members should take this into consideration when planning and maintaining their national
network of such stations.
(b) Station design
A lightship station is a meteorological observing station on board a lightship whose primary

function is to serve as a lighthouse in coastal waters. The meteorological instruments must
be properly exposed, generally in accordance with the rules given in the section dealing with
ocean weather stations. Care should be taken to avoid influence of the lightship’s special
superstructure.
Island and coastal stations should be equipped in the same way as a land station. In addition, the
stations should be able to measure sea-surface temperature and observe the state of the sea and
sea-ice conditions. They could also be designed to make upper-air observations.
(c) Site selection
The siting of island and coastal stations should be made in accordance with the rules given
elsewhere in this Guide for land stations (3.2.1.2.1 and 3.2.1.2.2). In addition, care should be
taken to ensure the observation of the state of the sea and the sea-surface temperature.
(d) Operations
National Meteorological Services shall operate or be responsible for the stations’ technical
standards, and for instrument calibration and maintenance. A National Meteorological Service
supervisor should ensure that the personnel have proper training and that all relevant manuals
and other documents are available at the stations.
(e) Identification
Island and coastal stations shall be identified by a station index number, as for land stations (see
3.2.1.2.6). Lightship stations are moored in fixed positions and may be also identified by a station
index number.
(f) Communications
The stations shall be provided with appropriate telecommunication equipment that will
guarantee a regular and reliable transmission of the coded reports. (See 3.2.1.2.7 dealing with
telecommunications for land stations.)
(g) Personnel and training
The personnel required for surface synoptic observations at island and coastal stations are
the same as for land stations for similar observations. If, however, both surface and upper-air
observations are carried out, the staff must be large enough and properly trained for both types
of observation. A supervisor must ensure that the operational personnel have the necessary
qualifications for the service, including normal technical maintenance at the station and
communication procedures (see also 3.2.1.2.4 and 3.2.1.2.5).
3.2.1.3.2.3 Fixed platform stations and anchored platform stations
(a) General
The offshore oil industry operates more or less permanently rigs and platforms on the continental
shelf all over the globe. Platforms for oil drilling or production may serve as excellent sites
for observations of meteorological variables and Members should take advantage of them.
Observations are needed by the platform operators to monitor weather conditions on and near
the platforms during helicopter and supply ship operations. Operators of offshore platforms
are generally required by the regulations laid down by individual countries to make reliable
surface observations of at least some meteorological and oceanographic variables. Cooperative
arrangements can readily be made with this industry.
(b) Station design
Meteorological instrument exposure is very important and the most difficult part of
instrumentation on platforms, owing to the size and the structure of the platform, which may be
more than 100 m above sea level.
(c) Operations
Care should be taken that the instrumentation and control of the observations remain the
responsibility of the National Meteorological Service. It is essential that the standard practices
defined by WMO be followed. Observers must be trained by the National Meteorological Service
to make manual observations. Adequate technical expertise in automatic instruments should be
available on board. The supervisor must ensure that all observational work is done according to
WMO regulations and that relevant documentation is available.
(d) Identification
Fixed and anchored platform stations are identified as ships and included in the International List
of Selected, Supplementary and Auxiliary Ships (WMO‑No. 47) and contain appropriate explanatory
notes.
(e) Communications
The types of equipment appropriate for timely transmission of observational data from platforms
and rigs are:
(i) Interface with public communication networks;

(ii) Telegraphs;
(iii) Telex typewriters;
(iv) Radioteleprinters;
(v) Radio facsimile broadcasts;
(vi) Radios;
(vii) Global system for mobile communication;
(viii) Satellites.
There should be at least one alternative in case of failure of the primary link.
(f) Personnel and training
The number of personnel required depends on the degree of automation. A good general
education, such as that for naval officers, is required. The observers should attend a theoretical
and practical course administered by the National Meteorological Service. The course should
comprise:
(i) A general presentation of the relevant regulations and guidelines of WMO and of the
National Meteorological Service;
(ii) Instruments for use at sea;
(iii) Visual observing techniques;
(iv) Lectures on the weather and weather forecasting for the particular area of
responsibility.
3.2.1.3.3 Mobile sea stations
3.2.1.3.3.1 Voluntary Observing Ship stations
The international scheme by which ships are recruited for making and transmitting
meteorological information is called the Voluntary Observing Ship (VOS) Scheme. Mobile sea
stations of the VOS Scheme comprise selected ship stations, selected ship stations equipped with
automatic weather stations (AWS), VOSClim (VOS Climate) ship stations, VOSClim ship stations
equipped with AWS, supplementary ship stations, supplementary ship stations equipped with
AWS, auxiliary ship stations and auxiliary ship stations equipped with AWS. Ships participating in
the VOS Scheme are one of the main sources of surface observations over the oceans.
In accordance with the Manual on the Global Observing System (WMO‑No. 544),Volume I, Part III,
2.3.3.2, Members shall recruit as mobile ship stations as many ships as possible that traverse
data-sparse areas and regularly follow routes through areas of particular interest. By fulfilling this
obligation, each Member contributes to the common objective of obtaining sufficient coverage
of meteorological observations over the sea. Uniform coverage is desirable, albeit difficult to
achieve, owing to large differences in the density of shipping traffic over the oceans, which is
comparatively greater in the northern hemisphere. Consequently, greater attention should
be given to the recruitment of voluntary observing ships that operate in the tropics or in the
southern hemisphere. To meet international meteorological requirements for data density over
the oceans, successive plans under the World Weather Watch (WWW) have shown the need for
maintaining or increasing the number of voluntary observing ships.
Relevant standards and recommended practices and procedures are contained in Appendix III.4.
For details, please refer to the International Meteorological Vocabulary (WMO-No. 182), the Manual
on the Global Observing System (WMO‑No. 544), Volume I, Part III, 2.3.3, and The Voluntary
Observing Ships Scheme – A Framework Document (WMO/TD-No. 1009), JCOMM Technical
Report No. 4.
3.2.1.3.3.2 Ice-floe stations
(a) General
An ice-floe station is generally a part of a scientific base on a large ice floe drifting in the polar
regions. Such stations make an important contribution to the network in data-sparse polar
regions.
Members, individually or jointly, should arrange for meteorological observations from large ice
floes whenever possible, as part of the programme either of a scientific base or of an automatic
station. In the case of a joint undertaking, one National Meteorological Service should have the
responsibility for the station’s scientific and technical standards.
(b) Identification
Identification of ice-floe stations shall be the same as for ships.
(c) Communications
Ice-floe stations should have two-way radio connections or automatic transmission via satellite.
In the polar regions, only polar-orbiting satellites can be used. The ARGOS system operated with
some of the United States satellites offers this possibility, and the use of the Doppler effect in
the receiver signals makes it possible to locate the station fairly accurately. Using polar-orbiting
satellites as a means of communication may give synoptic reporting times.
(d) Personnel and training
A sufficient number of the staff on the ice-floe base must have adequate training for taking all
the required observations in accordance with WMO regulations. At least one trained technician
should be available to operate and maintain the instruments and be responsible for the supply
of expendables and back-up equipment. The staff must also include personnel to operate the
communications system.
3.2.1.4 Automatic stations
3.2.1.4.1 General
An automatic weather station is defined in International Meteorological Vocabulary

(WMO‑No. 182) as a “meteorological station at which observations are made and transmitted
automatically”.
The following information in this section of the Guide deals with the planning and establishment
of real-time networks of automatic stations that are part of the regional basic synoptic networks
and other synoptic station networks where the emphasis is on quick and direct access to the
data.
Complementary information can be found in Chapter 1, Part II, of the Guide to Meteorological

Instruments and Methods of Observation (WMO‑No. 8).
3.2.1.4.2 Purposes of automatic stations
Automatic stations are used for many purposes. These include:
(a) Providing data from sites that are difficult to access or are inhospitable;
(b) Providing observations at manned stations outside the normal working hours of the
observing staff, for instance during the night or on weekends;
(c) Increasing the reliability of the data and standardizing observing methods and timing for all
network stations;
(d) Cutting costs by reducing the number of manned stations;
(e) Placing sensors in meteorologically favourable sites apart from the places of residence and
work of the observer.
3.2.1.4.3 Types of automatic synoptic networks and stations
3.2.1.4.3.1 Network configuration
Automatic synoptic networks need real-time operation for data collection, transmission and
processing. Stations can be organized within a network in different ways. Data collection is
directly controlled by a single data processor at a central data collection point or by several data
processors at sub-central data collection points which periodically collect the data from the
stations and distribute them (see Figure III.3). Sub-central data acquisition processors are suited
to large networks when a regionalization of the control and processing functions seems to be
an advantage. The use of a single processor to service a network makes the entire automatic
observing system vulnerable to a failure of this processor.
The data-transmission facilities offered by automatic synoptic networks can also be used, if
necessary, by manned or partially automated stations, provided the observers have adequate
terminals for manual observation input. These terminals may be used to enter synoptic data,
coded or in parameter form, or climatological information. The network central processor
collects the observations directly or together with the automatic measurements via the
automatic stations (Figure III.4).
3.2.1.4.3.2 Data processing
The major part of data processing or coding is accomplished either at the station site or at a sub-
central or single central processor site.
AWS AWS
AWS SP AWS
AWS
AWS
Central
AWS SP
processor
AWS
AWS
Key
AWS
AWS Automatic weather station
SP AWS
SP Sub-central processor
AWS
Dedicated leased line
Switched line
Figure III.3. Network configuration

Source: Branke, W., 1978: System technology for networks, Technical Seminar for Measuring Techniques, Automation
and Data Processing for Water Control, May 1978, Bayerisches Landesamt für Wasserwirtschaft
The main advantage of a central data processing facility is that quality control, real-time
computation and data conversion are performed at a single place. Furthermore, synoptic code
changes can be implemented for all stations at once with only one modification; a single station
can be modified and maintained without altering the standard codes. Moreover, this concept
offers an important advantage to the data user, who can analyse the instrumental problems with
the raw sensor data directly from the central site and can plan repair work more efficiently.
3.2.1.4.3.3 Data transmission
Data transmission is a vital function for real-time synoptic stations. For more details see the Guide
to Meteorological Instruments and Methods of Observation (WMO‑No. 8), Part II, Chapter 1, 1.3.2.10.
3.2.1.4.3.4 Multi-purpose stations
Since the costs of automatic synoptic stations are very high, it seems judicious to use the station
facilities for other purposes as well, for instance, to meet the needs of climatology, aeronautical
meteorology, storm warnings, nuclear power safety, air and water quality surveys and flood
warnings.
For such multi-purpose stations, the data may be stored continuously on local storage units.
Thus, the data can be retransmitted to the network central processor after an interruption or
processed at a later date on a separate computer system.
3.2.1.4.3.5 Sensors
The sensors for use with automatic weather stations for the measurement of the different
variables and their performance are described in 1.2.1, Chapter 1, Part II, of the Guide to
Meteorological Instruments and Methods of Observation (WMO-No. 8).
3.2.1.4.4 Planning guidance
3.2.1.4.4.1 Determination of requirements
All disciplines involved with meteorological observations—synoptic meteorology, climatology,

aeronautical meteorology, agricultural meteorology, and hydrology—have formulated
their own functional requirements for observations to satisfy their specific service needs. All
disciplines, however, agreed that it was beneficial to apply universal rules or standard methods of
observation in order to avoid unnecessary confusion and achieve data compatibility. In line with
this policy, standardization of automatic weather stations will be beneficial if designed to fulfil
the requirements of the various disciplines.
To support present and future automatic weather station applications, functional

specifications—a list of required meteorological variables and their characteristics—for automatic
weather stations were developed (see Appendix III.1). They show current users’ requirements for
data reported by automatic weather stations, and may be used by manufacturers in designing
automatic stations and sensors. These specifications are expressed in terms of variable name,
maximum effective range, minimum reported resolution, mode of observation and present
ability to represent variables by BUFR/CREX codes. Future requirements will be incorporated in
the functional specifications upon proposal of the users.
Observer’s Regional weather Central weather Central message

site service service switching system
Alternative 1
OBSERVER
ManDat
DATA
USER
Switched 600b/s FSK

Alternative 2
telephone ATESTO
network 2400 b/s Central
OBSERVER ADAC communi- to GTS
cation
computer
ManDat
DATA
USER
ADAT
Alternative 1. Manned observing station

Sensors Alternative 2. Partly manned observing station
ADAC = Automatic data aquisition central GTS = Global Telecommunication System

ADAT = Automatically generated data ManDat = Manual data
FSK = Frequency shift keying
Figure III.4. Automatic data collection system for conventional stations and partially or fully
automatic weather station
Source: Hovberg, T. and Udin, I., 1984: Papers presented at the WMO Technical Conference on Instruments and Cost-
effective Meteorological Observations (TECEMO), Nordwijkerhout, September 1984; WMO Instruments and Observing
Methods, Report No. 15
Some of the variables listed in the functional specifications should be mandatory. A standard
automatic weather station should consist of an observing system that provides observational
data from a standard set of variables, for example, pressure, temperature, wind and humidity.
Apart from this standard set, a set of optional variables may be considered. The basic set of
variables to be measured by automatic weather stations, compiled from the Manual on the Global
Observing System (WMO‑No. 544), is provided in Appendix III.2.
The first step in planning an automatic network is to set up a list of requirements of known
and potential data users. At the outset, purely meteorological aspects are to be considered, for
example, which station distribution, measuring cycle and observing programme are necessary to
meet the requirements of the weather predictions made in the country and which are necessary
to fulfil international weather information requirements. The answer could be found by using a
table similar to the one developed for Scandinavia in Table III.1. Interdependency with other data
acquisition systems, such as radar, upper-air stations or satellites, should also be considered.
Table III.1. Meteorological data user requirements in Scandinavia
Time and space scales Observations

0–2 h • Complete regional radar coverage; continuous operations
0–100 km • Automatic stations, including buoys; regional network for wind,
Nowcasting humidity with density approximately 40 km; wind measurements in
narrow channels with density less than 20 km; wind and temperature
along popular mountain tracks; temperature, wind, humidity and
radiation along highway sections prone to icing; all values in real time
• 1–2 vertical sounding systems for wind, temperature and humidity;
measurements every hour
• Reports from civilian and military aircraft in the Region
• Airport observations, hourly synoptic observations and METAR
• In southern Sweden, METEOSTAT digital information every half hour
2–6 h • Complete radar coverage

20–300 km • Complete synoptic observations every third hour; density 80 km
• Automatic stations, including buoys; pressure measurements with
density approximately 50 km; wind, temperature and humidity with
density approximately 40 km once an hour
• Digital satellite pictures with a period of 3–6 hours
• 1–2 vertical sounding systems at least every sixth hour
• Scandinavian synoptic observations every third hour
• Acoustic sounders, instrumented masts and the like
6–18 h • Synoptic observations every third hour; density approximately 80 km

20–300 km • Automatic stations with pressure sensor every third hour; density
approximately 50 km
• Digital satellite pictures with period 3–6 hours
• Satellite vertical soundings, for example, TOVS, every sixth hour or
more frequently
• 1–2 vertical sounding systems every sixth hour
• Foreign observations (SYNOP, TEMP, PILOT, AIREP) every third or sixth
hour
• Ship observations
• Acoustic sounders, masts and the like
12–26 h • As above
150–4 000 km
Source: Ag., L., 1981: Papers presented at the second WMO Technical Conference on Instruments and Methods of
Observation (TECIMO-II), Mexico City, October 1981; WMO Instruments and Observing Methods, Report No. 9
Results obtained from manned stations are often regarded as standards by opponents of
automation, who compare the performance of automatic equipment with the performance
of ideal conventional stations. This way of thinking is often unfounded. In some cases it is
indispensable that new methods be adopted if meteorological observation is to be successfully
automated. Replacing manual observing methods by automatic means frequently leads to
a complex, expensive and unreliable outcome. In view of this problem, an automatic system
should be designed to work according to a predetermined specification rather than in terms of
“measurements” taken by an observer. Sensors whose output is consistent with automatic data
handling should be adopted.
Because of the diversity of meteorological problems, network planning should not be the
concern solely of automatic measuring system engineers or manufacturers who are often
unaware of the real problems faced by users. During the planning phase future users
must devote their time and bring their experience to bear on the issue in order to avoid
disappointments caused by an unsuitable system. Member countries without experience in that
field should seek advice from those who have been running automatic observational networks
for a number of years.
It is essential to establish detailed specifications that take into account local requirements and
the environment. These specifications should mention technical parameters such as measuring
range, uncertainty, resolution, reproducibility, response time, stability, reliability, power
consumption, exchangeability, critical dimensions (distance between sensors and transmitters/
receiver, space or weight limitations) and spare parts and maintenance requirements. Other
factors to be included are long-term compatibility requirements with attached or neighbouring
equipment, if the equipment is intended to replace a part or to be a complementary part of
another system, and possible interferences with other systems, in particular at airports.
3.2.1.4.4.2 System selection criteria
(a) Future station environment
Automatic weather stations must be able to withstand the most severe meteorological extremes.
It is essential, therefore, to analyse the station’s future environment before specifying or
designing a system. The major influences are high humidity, low or high temperature, dust, high-
frequency fields, lightning and corrosive environments. Nuclear-electromagnetic pulses should
also be taken into account. Protective measures against these influences should be planned at the
outset.
(b) Reliability
The mean time between failures of an automatic synoptic station should be more than
10 000 hours without taking into account individual sensor failures.
The reliability of automatic weather stations can be enhanced by making a partial or full
duplication of the station, in other words, by providing a backup system. Partial duplication is
defined as the duplication of critical variables by using redundant subsystems such as power
supplies, and wind and temperature sensors. Full duplication, where the second station is a less
expensive type with a lower capability which will report only basic variables such as atmospheric
pressure, wind speed, wind direction or air temperature, would require different power supplies
and communication channels, at least at the station, if all risks are to be avoided. A feature of
the duplication approach is that both the primary and the secondary systems will be working
continuously except, of course, when one of them is out of order.
Generally, partial or full equipment duplication tends to be expensive and is only worthwhile
in the absence of a suitable maintenance organization that guarantees corrective action within
acceptable time limits.
The percentage of synoptic observations that may reach the user in time to be of value is a
critical quality factor in the assessment of an operational automatic system. The point at which a
decline from 100 per cent becomes sufficient to render the system no longer cost-effective might
depend somewhat upon the circumstances of its use; in general, however, the aim is to achieve
a data availability of more than 90 per cent for successful operational systems. For Regional Basic
Synoptic Stations, a data availability of at least 95 per cent appears to be indispensable for daily
routine work.
The most important losses of reliability are generally connected with data transmission
interruptions. The safety of data transmission can be improved by overlapping star-type network
designs and rerouting communications along different communication lines. (See Figure III.5.)
(c) System architecture
The system should be flexible and modular in order to suit the most varying applications. Special
attention should be paid to expanding capability. It should be possible to connect additional
stations, new sensors and peripherals to the system at a later stage. The conception of a network
should leave the choice of data routing and diverse communication devices open, so that they
can be adapted to the latest technological developments.
The basic structure of an automatic station and its data handling should also be kept as modular
as possible. As much signal conditioning as possible should be carried out by each sensor
interface, preferably at, or very close to, the sensor.
Synoptic stations designed to be used unattended over a long period of time should be kept
as simple as possible. More elaborate solutions, including sophisticated data processing, can
be envisaged for synoptic stations which can be visited more often or which are used semi-
automatically, however.
(d) Life-length considerations
Life length is considered by manufacturers to be the time a piece of equipment remains in active
production; the user, however, is more likely to think in terms of useful life in the field. It is well
known that electronic products tend to have a short production life cycle. For the user, the useful
life of a system tends to be much longer.
In some cases, the life length of a system is limited by the rapid progress of technology.
Availability of spare parts or human know-how becomes a serious problem. It is possible that by
the time a system is designed, tested and accepted, it has already become obsolete.
Single star network Overlapping star network with

rerouting facilities
Measurement site Communication line

Data collecting sub-centre Rerouting options
Data collecting centre
Figure III.5. Single star network and overlapping star network with rerouting facilities
Source: Van den Enden, I.F.H.C.C., 1984: Papers presented at the WMO Technical Conference on Instruments and Cost-
effective Meteorological Observations (TECEMO), Noordwijkerhout, September 1984; WMO Instruments and Observing
Methods, Report No. 15.
It is, therefore, better to choose sensors which have already been successfully used in other
countries and are readily available instead of undertaking expensive research and development
in one’s own country. This is especially true with respect to the acquisition of smaller series. The
contract with the manufacturer should contain guarantees concerning spare parts maintenance
and the availability. If a system manufacturer cannot guarantee the required life length under
acceptable conditions, a commitment from the network operator is indispensable. The latter
must participate in development and maintenance work to acquire the necessary know-how and
obtain enough materials for an adequate period.
3.2.1.4.4.3 Logistics
(a) Site selection
As automatic stations are expensive, it is necessary to study site facilities carefully before making
significant installation investments. The considerations in choosing a site for a surface synoptic
station (see 3.1.2) also apply to automatic stations.
Since there should not be any difference between the performance and quality of the
observational data from manned and automatic stations, 3.2.1.2.1 and 3.2.1.2.2 outlining siting
and exposure requirements also apply to any automatic weather stations and sensors installed.
(b) Resources required
The establishment of an automatic observational network calls for considerable material

resources. If the quality and quantity of the data acquired by automation is disregarded,
establishment of an automatic synoptic network can be financially advantageous only if it
replaces many manned stations making round-the-clock observations by totally unmanned
stations or by partially manned stations with a reduced observer presence.
The total cost of an automatic synoptic network is composed of the initial costs and the
operating costs. Initial costs include costs for development, acquisition, installation, efficiency
tests, documentation and software programmes. Operating costs are staff costs, maintenance,
transmission, modification and replacement of parts, electricity consumption, land rental,
training, measurement control and processing. The costs of modifying and replacing system
parts should be estimated on the basis of the initial costs, as these can be distributed over the
years in relation to the life length of each system.
The annual operating costs of a well-maintained network are about 10 to 20 per cent of the
initial costs. Operating costs are rarely included in a realistic manner in the offers made by
the manufacturers, and are therefore often underestimated. The staff component of the initial
costs is relatively small; as for the operating costs, the corresponding parts for the staff and the
material are of similar size. It is generally more important to spend the available resources for
the infrastructure needed to maintain a small automatic network than to enlarge the network
without such support.
3.2.1.4.4.4 Time needed for the establishment of an automatic observational network
(a) Time for development
When National Meteorological Services participate in the development of new sensors or

complete automatic stations, they must ensure, through the use of prototypes and pilot series,
that the technical specifications of the instruments have been fully respected; they must
undertake compatibility tests in the field. Since complete field intercomparison of existing and
new instruments should cover all four seasons, the minimum test duration is one year. After
completing the evaluation of obtained data sets, the test results may require the redesign of the
product. It may take years to successfully develop a product for use in the field. If it takes too
long, the rapid progress of technology may overtake developments and the finished equipment
may be obsolete by the time routine operations can start.
(b) Test operations
Setting up a system as complex as an automatic measuring network requires a good working

team. The time needed for completing test operations depends on the complexity and scale of
the network, and available means. Depending upon the level of experience, it takes about six
months to one year for the team to become familiar with the system. This period becomes much
longer if the network operators have not participated in the development and construction
of the system. After the completion of an automatic network and before routine operation
and dissemination of synoptic information at the international level, a learning and testing
period should take place. Testing should also be carried out for any network station established
subsequently, especially those belonging to the regional basic synoptic networks.
(c) Parallel operation with conventional stations
If previous long-standing climatological data series have to be extended in time by data

provided by automatic synoptic stations, parallel measurements by conventional and automatic
observing methods are indispensable to ensure continuity of the records. One year of parallel
measurements does not suffice; preference is given to at least two years, depending on the
climatic region.
Once stations become fully or partially automated, it is often difficult to stimulate observers to
make parallel observations, or financial pressures may demand a reduction in the number of
operating stations. In that case, sufficiently long parallel observations should be carried out at
least at a selected number of automatic stations.
3.2.1.4.5 Operations
3.2.1.4.5.1 Time and frequency of observation
For most meteorological variables measured by automatic weather stations and for their
applications, a measuring interval of one to ten minutes is possible; in many countries a
measuring interval of 10 minutes has become usual. See the Guide to Meteorological Instruments
and Methods of Observation (WMO‑No. 8), Part III, Chapter 2, 2.4.2.
If there is an intention to use data from automatic stations for real-time monitoring, warning and
forecasting, or even nowcasting, an interval of a few minutes—one to five—is indispensable,
making it possible to follow weather developments continuously and offers some possibilities for
interpolation after a short system failure.
3.2.1.4.5.2 Surface synoptic meteorological observation variables
When partially automated stations are used along with an observer providing complementary
observations of variables which are not measured automatically, human observations can be
conducted at a separate location, for example, if the observer lives far away from the station
site. If so, the observer can be equipped with a remote data entry device allowing him or her
to contact the automatic stations at any time by telephone or high-frequency transmission.
Thus, human observations are independent of those made automatically. However, the distance
between the remote data entry device and the automatic station should not be more than 10 km,
especially in mountainous regions, in order to ensure the consistency of the observations.
3.2.1.4.5.3 Safeguards against breakdown
Failures at the network central processors can paralyse a whole network or large parts of it. For
safety reasons, it is recommended that a double central processor system should be provided.
Even for failures of a completely double system, procedures should be planned that will ensure
the continuation of some minimal real-time network functions.
At major surface synoptic stations, at least for those in the regional basic synoptic network,
failures of the automatic data acquisition from the stations must be compensated by an adequate
emergency system. Observers, with the help of some alternative instrumentation, should be able
to take the measurements themselves and code and transmit the synoptic messages until the
fault is repaired.
3.2.1.4.5.4 Monitoring and processing
To increase user confidence in the reliability of information derived from an automatic network,
it is necessary to institute a continuously operational real-time and near-real-time monitoring
programme and thereby validate the quality of the data generated by the network.
The quality requirements for pre-processing and processing control at automatic stations are
set out in a general manner and for each variable in the Guide to Meteorological Instruments and
Methods of Observation (WMO‑No. 8), Part II, Chapter 1, and Part III, Chapters 1, 2, and 3. Further
information about quality control at the observing site and at data collection centres can be
found in Parts V and VI of the present Guide.
Data quality control and correction should be carried out as quickly as possible after data
collection. Timely data processing is only possible if the site and instrument characteristics of
the parameters are permanently known. This intensive work should already be considered when
planning the network.
3.2.1.4.5.5 Maintenance
Important considerations in setting up a maintenance organization for automatic stations

and the principles to be followed in carrying out a maintenance programme are described
in 1.6, Chapter 1, Part II, of the Guide to Meteorological Instruments and Methods of Observation
(WMO‑No. 8), and in a more general way in Part III, Chapters 4 and 5 of the aforementioned
Guide.
Maintenance work should be carried out primarily by specially trained technical personnel. This
staff, however, cannot always solve the problems encountered by the observer with the non-
automatic observations or notice possible deficiencies in station performance. It is, therefore,
useful for partially automated synoptic stations to be inspected by specially trained staff,
independently of the technical maintenance work.
In a well-established system, modifications should generally be kept to a minimum. As far as

possible inspections and the bulk of preventive maintenance work should be done by a small—
and always the same—team, in order to improve the homogeneity and the continuity of an
automatic network.
3.2.1.4.5.6 Training
The more complex the equipment, the more technical knowledge is required from the people
who will maintain and use the system. Rapid strides in technical developments make regular
training courses indispensable. The staff’s technical knowledge needs to be kept up by refresher
courses from time to time, especially when the duties and responsibilities of staff members have
changed.
The Guide to Meteorological Instruments and Methods of Observation (WMO‑No. 8) outlines the
general requirements for the training of observers in 1.8, Chapter 1, Part II.
At many partially automated synoptic stations, observers will no longer have the same close
relationship with their work as they formerly had with conventional measurements. In such
cases, it is recommended that they be given instructions as to the necessity, importance and
purpose of their new work along with practical examples of the value and use of the data their
station provides.
3.2.1.4.5.7 Documentation
Detailed documentation is the basis for the international exchange of experiences with automatic
weather observation networks and should therefore be available at the time the network is
established. The documentation should be obtained from the competent authorities or the
manufacturer, together with the equipment specifications.
The actions and conditions that influence measurements at a weather station should be featured
in standardized documentation. Writing down all changes in measuring conditions provides a
complementary source of meteorological information and permits the data user to interpret the
measurements correctly. With automatic measurements made over a long period of time, the
events which should be recorded become so numerous that subsequent reconstruction is almost
impossible. Therefore, the role and importance of station metadata cannot be disputed.
Data producers are responsible for providing adequate and sufficiently detailed metadata.
For more information, see the Guide to Meteorological Instruments and Methods of Observation
(WMO‑No. 8), Part I, Chapter 1, 1.1.3 and 1.3.4, and Part III, Chapter 1, 1.6.
Two sets of metadata for automatic weather stations were developed with respect to real
time, near-real time, and non-real-time; taking into account the significance of each entry for
operational data. They are reproduced as possible guidelines for network managers in Appendix
III.3.
Real-time automatic networks that permit dialogue between stations and the network’s central
processor can also be used to provide documentation of all kinds. Observers or maintenance
staff equipped with fixed or mobile data terminals for interactive communication can, inter alia,
achieve the following:
(a) Obtain guidance for complex maintenance procedures at the station using information
called up from the central station;
(b) Record maintenance work that has been done or comments of the inspector. This
information can be transmitted to the network’s central processor for storage;
(c) Update automatically system tables containing the basic characteristics of the individual
stations or update stock management files after installation, exchange removal or sensor
calibrations;
(d) Interrogate the observer handbook. If the handbook is centrally modified it is easier to keep
up to date.
3.2.1.4.5.8 Quality standards
Reference should be made to the following publications:
The Guide on the Global Data-processing System (WMO‑No. 305), Chapter 6; the Manual on the
Global Data-processing and Forecasting System (WMO‑No. 485), Part II, Volume I, section 2; and
Guidelines for Quality Control Procedures Applying to Data from Automatic Weather Stations
which appear in Part VI, Appendix VI.2, of the present Guide.
3.2.1.4.6 Automatic sea stations
(a) General
Automatic stations for provision of meteorological data from the oceans are an important and
reliable tool for collecting data, especially from such remote areas as the polar regions. General
considerations applicable to automatic land stations are also largely valid for automatic sea
stations. Station reliability problems are also generally similar.
Moored and drifting buoys are used with automatic stations to provide data from sea areas
where few or no mobile ships operate. A striking example is the low-cost Lagrangian drifting-
buoy system operated in the global oceans. In line with the framework of the former Surface
Velocity Programme (SVP) of the World Ocean Circulation Experiment 1995–2005, standard
Lagrangian drifters are referred to as SVPs; SVPBs are Lagrangian drifters equipped with
barometers. A mobile ship observing programme may also be fully automatic, but it is advisable
to provide for manual insertion of data in the system at least for the visual observations that
cannot be made automatically. In general, automatic sea stations supervised and supplemented
by human observers, when possible, are recommended for several reasons: overall reliability is
improved, time resolution is increased, sensors and other vital parts can be replaced quickly and
efficiently and costs of manned stations are reduced, owing to smaller staff.
Stations in some parts of the globe such as the Arctic and Antarctic, isolated islands and drifting
buoys on ice and sea are difficult to visit for repair and replacement in case of failure. Reliability
is therefore even more essential than for land stations. Full duplication is the best solution, but
a rather expensive one. For drifting buoys, duplication means simply deploying two buoys
instead of one. By making the buoy very simple, with only a few sensors, such as pressure and
temperature sensors, the risk of malfunction is kept low.
(b) Site selection
Isolated unpopulated islands and relatively inaccessible coastal regions are natural sites for
automatic stations. Members can improve their national network in an efficient and inexpensive
way by establishing such stations which might also make an important contribution to the
regional and global network.
Moored buoys in fixed positions in ocean or coastal areas may also be sites for meteorological
observations and for surface flux and oceanographic sub-surface measurements. Members
should be aware and take advantage of the planning and deployment of such buoys by
others, such as oceanographic institutes; reciprocally, when such buoys are operated by a
Meteorological Service, the latter should offer to carry oceanographic sensors on board. This may
also apply to some extent to drifting buoys.
Fixed platforms may also be selected for fully automatic stations.
Coastal stations may also be automatic or semi-automatic if personnel are available for manual
observations of additional variables.
Lightship stations can be automated in the same way if they are unmanned or inadequately
manned.
Relatively large ice floes provide excellent sites for automatic stations, and ice-buoy networks
should be operated in the polar regions individually or jointly by Members.
Drifting buoys with automatic stations provide an efficient way of obtaining meteorological
information from the high seas. Members should plan deployments jointly to obtain a desirable
network.
(c) Design
An automatic sea station should consist of the following components:
(i) A number of sensors for measuring or observing the variables;

(ii) An electronic package with a microprocessor or microcontroller to sample, process

and record sensor output;
(iii) A source of electric power such as battery packs, solar panels or external sources to
provide sufficient electric power for the station to operate without interruption during
its life-time; some safety prevention measures must be taken, as dangerous explosions
have been reported and recommendations subsequently made;
(iv) A transmitter for communication purposes.
If possible, meteorological sensor exposure should be the same for automatic lightship, island
and coastal stations as for manned stations.
The exposure of instruments (sensors) on fixed-platform stations is discussed in 3.2.1.3.2.3.

Exposure should be taken into account in the planning/construction phase of a platform. It
should be negotiated between the platform owner and the National Meteorological Service.
An offshore drilling or production platform is a state-of-the-art structure with sophisticated
equipment on board, including computers. It would be wise to connect the meteorological
sensors to an on-board computer with the necessary software to handle the raw data and
convert them to meteorological variables, and to code the information in the relevant WMO
codes for transmission to a coastal collection centre.
Drifting buoys for the oceans or for deployment on ice floes (ice buoys) can have different
designs; for most meteorological purposes a simple version is sufficient. A sketch of a typical
simple drifting buoy is given in Figure III.6. Like the buoys which were developed during the first
Global Experiment of the Global Atmospheric Research Programme, i.e. FGGE drifters, these have
sensors for two variables only. A drogue is generally used to maximize the drift of the buoy and
minimize the slip with regard to the moving water parcel (Lagrangian drifters).
More sophisticated buoys may have a number of sensors, for example, to take wind
measurements. In that case, the hull must be much larger—taller—and consequently much more
expensive. Ice buoys generally resemble drifting buoys except for the hull, which in the former is
designed to rest on an ice surface.
(d) Observing programme
In accordance with 2.3.3.16, Part III, Volume I, of the Manual on the Global Observing System
(WMO‑No. 544), a surface synoptic observation from a fixed automatic sea station shall cover
atmospheric pressure; wind direction and speed; air temperature and sea-surface temperature.
If possible sea-state observations (waves) and information on precipitation (yes or no; especially
in tropical areas) should be included.
Observing programmes for a typical simple drifting buoy consist of observing two variables:
atmospheric pressure and sea temperature. Generally, surface synoptic observations should be
conducted in accordance with stipulations set out in 2.3.3.17, Part III, Volume I, of the Manual on
the Global Observing System (WMO-No. 544).
The observing programmes presented above for automatic sea stations must be regarded as
minimum requirements. Large automatic stations, especially those that are supervised daily,
should also, if possible, provide cloud base height, visibility, computed pressure tendency and
characteristic, and amount of precipitation.
The larger drifting and moored—often combined oceanographic-meteorological—buoys may

have a more extensive observing programme including, for instance, wind measurements.
(e) Network organization
It is advantageous to use automatic means when setting up a sea-station network; in many

cases automatic observing stations are the only solution. In a number of cases, and especially
with regard to automatic weather stations onboard voluntary observing ships, it is preferable to
employ “hybrid” stations where manual observations are used together with automatic sensor
output to obtain a complete set of observations, as for some ships. A network will in general
consist of both manual and automatic stations.
Fixed platforms, lightships and coastal stations may be isolated automatic stations in an
otherwise conventional network and be an integrated part of national, regional and global
networks.
Automatic ice-floe stations, composed of ice buoys, and drifting buoys are specialized and fully
automated networks that provide data from remote and otherwise data-empty areas.
By introducing automatic means for new stations or for automating conventional stations,
Members could contribute to maintaining and/or improving the total network for national,
regional and global purposes.
Members should, through suitable joint organizations or arrangements, try to establish a

network of drifting buoys in critical sea areas. When planning such a network, knowledge of the
wind systems in the sea areas in question is essential. Outside the tropical areas, it is generally
sufficient to compute the mean geostrophic wind for each month. The drift paths of free-drifting
buoys can then be determined with sufficient accuracy for planning deployments. This is
accomplished successfully by the WMO-IOC Data Buoy Cooperation Panel.
(f) Logistics
Equator of surface float

rests approximately 2.5 cm above
32 cm the water line
2.5-cm strips of ∅ 0.32 cm

webbing connect
the top ring to the
first middle ring
and the bottom
ring to the last
middle ring
15 m
Notes:
- Tether connections to drogue and surface

float are potted in rigid resins and
∅ 30.5 cm strain relieved with flexible urethane carrots
- 4 section holeysock drogue with rigid plastic

490 cm
rings included at the top and bottom and
semi-rigid rings between each section
- Cylindrical holeysock drogue comprised of

4 stacked sections
Scale drawing - Each section includes two pair of diametrically

opposed holes with a diameter equal to 30.5 cm
122 cm - Holes pairs are orthogonal in horizontal plane
- Drogue fabric resists ripping and fraying;

Drag area ratio ≥ 40 61 cm is negatively buoyant with proven history
of resisting decomposition in sea water
- Top drogue ring is filled with polyurethane foam

for buoyancy; other rings flooded or solid
- Any additional drogue ballasting installed

in bottom ring
Figure III.6. Typical simple drifting buoy

(i) Electrical power must be available, preferably by power-generating equipment like

solar cells. If batteries are used, they should last at least one year (two years for drifting
buoys and three for ice buoys); security measures aimed at preventing possible
explosions must be taken when using batteries in non-vented compartments;
(ii) Telecommunication facilities must be available. For automatic sea stations this
generally means an automatic transmitter with suitable antennae for direct
communication with land stations or via satellites;
(iii) Service, maintenance and supplies must be provided by the operating agency;
(iv) Specially trained staff must be available to plan maintenance and monitor operations
properly.
To maintain a certain number of buoys (ice buoys and drifting buoys) within a specified area,
it is necessary to make successive deployments. The effective operation of a buoy network is,
therefore, dependent on available ships, or aircraft in the case of ice buoys. Ships of opportunity
can be used for drifting buoys; drifting buoys can be deployed from low-level aircraft.
Certain types of drifting buoys floating out of a desired area or no longer functioning properly
can be recovered and re-used. However, modern versions of Lagrangian drifters are not meant
for recovery or refurbishment. An advantage of simple buoys, however, is that, owing to their
relatively low cost, they may be regarded as expendables.
(g) Coding and communications
Data processing and coding can be carried out at the automatic station itself by a
microprocessor, or at a central receiving station and processing centre. The latter method is
recommended because the automatic station can then be made very simple.
The pressure tendency, three hours, and the tendency characteristic can be provided for the
simple drifting buoys in addition to the pressure. This requires a microprocessor to do some
handling, including storage of the sensor output.
The communications for automatic coastal stations may be achieved by land line, very high
frequency (VHF) and ultra-high frequency (UHF) radio or direct satellite link, for example,
geostationary or polar orbiting. The data can be retransmitted via satellite to local users with
a receiving station, or disseminated over the Global Telecommunication System from the
large main satellite ground stations. The drifting- and ice-buoy communication link is chiefly
accomplished via polar-orbiting satellites because this makes it possible to determine the
transmitting buoy’s position at the same time. A platform telemetry transmitter is preset for
dissemination at fixed intervals, usually every 90 seconds. The satellite must have at least four
different contacts with the buoy’s platform telemetry transmitter at each pass to obtain sufficient
data for proper location. Because the Doppler shift in the frequency is transmitted together with
the sensor data, platform telemetry transmitter circuit stability must be ensured. Data obtained
in this way are essentially asynoptic in case only the most recent data are being transmitted.
New buoy systems also record past observations at synoptic and round hour times on board and
transmit them asynoptically through the satellite system.
The ARGOS system for determining the location of drifting buoys and for collecting data via
satellites provides a very effective means of taking full advantage of drifting buoys. A special
tariff is negotiated by user Countries, under the auspices of WMO and the Intergovernmental
Oceanographic Commission with the agency responsible for administering the ARGOS system
for the benefit of interested Members, permitting a reduction in the cost of acquisition of data
from buoys and other automatic stations.
(h) Personnel
The implementation of an automated network requires a considerable number of well-qualified

personnel to keep the systems operating properly. This is sometimes overlooked, with the
unfortunate result that expensive equipment becomes useless. This is the most important advice
to Members planning a network of automatic sea stations.
(i) Quality standards
In addition to the resources mentioned in 3.2.1.3.2.1, please refer to the following publications:
(i) Handbook of Automated Data Quality Control Checks and Procedures of the National Data
Buoy Center, NDBC Technical Document 03-02;
(ii) Reference Guide to the GTS-Sub-system of the ARGOS Processing System, DBCP Technical
Document No. 2;
(iii) Guide to Data Collection and Localization Services Using Services Argos, DBCP Technical
Document No. 3;
(iv) Global Drifter Programme Barometer Drifter Design Reference, DBCP Report No. 4.
3.2.2 Observations/measurements
3.2.2.1 General
3.2.2.1.1 Time and frequency of observations
Sections 2.3.1.3 and 2.3.1.4 contained in Part III, Volume I of the Manual on the Global
Observing System (WMO‑No. 544) specify the main standard times for surface synoptic
observations—0000, 0600, 1200 and 1800 Coordinated Universal Time (UTC)—and the
intermediate standard times for these observations, 0300, 0900, 1500 and 2100 UTC. The
mandatory and/or recommended times of observation at different types of surface synoptic
stations such as land stations, fixed sea stations, mobile sea stations and automatic stations,
are given in 2.3.2 and 2.3.3, Part III, Volume I, of the Manual on the Global Observing System
(WMO‑No. 544).
3.2.2.1.2 Observing programme
The variables comprising surface synoptic observations required to be made at different types of
stations, for example, land stations, ocean weather stations, mobile ship stations and automatic
stations on land and at sea, are given in 2.3.2.9, 2.3.2.10 and 2.3.3.11 to 2.3.3.16 contained in
Part III, Volume I, of the Manual on the Global Observing System (WMO‑No. 544). Additional
guidance material for the observation or measurement of each of these variables is given below.
For convenience, the information is provided separately for land stations and sea stations,
although for some variables the rules to be followed are the same in both cases.
3.2.2.2 Observations at land stations
Meteorological variables that shall be observed and recorded at a manned synoptic land station
are defined in the Manual on the Global Observing System (WMO‑No. 544), Volume I, Part III,
2.3.2.9. They are described below.
3.2.2.2.1 Present and past weather
The specifications used for present and past weather shall be those given in the Manual on Codes
(WMO-No. 306, Volume I.1, Part A, code form FM 12-XIV SYNOP). The specifications used for
atmospheric phenomena shall be those provided in the same publication under the description
of “weather”. Additional specifications and descriptions of all types of weather phenomena given
in the International Cloud Atlas (WMO‑No. 407) should also be adhered to, namely hydrometeors
(precipitation), lithometeors, electrometeors (electrical phenomena) and photometeors

(optical phenomena). See the Guide to Meteorological Instruments and Methods of Observation
(WMO‑No. 8), Part I, Chapter 14, for further information.
Observations of weather and atmospheric phenomena are mainly conducted visually.
Land stations shall make round-the-clock observations of the weather, including atmospheric
phenomena. Other surface stations should do so to the extent possible. The frequency of
atmospheric phenomena observations (in between the standard observation times) should be
such as to cover even short duration and non-intensive phenomena.
During observations, the following steps must be taken:
(a) Note the type and intensity of atmospheric phenomena (light, moderate, heavy);
(b) Record the time of the beginning, change in intensity and end of the phenomena in hours
and minutes;
(c) Observe the station’s closest surroundings.
The following steps are optional, but recommended:
(d) Monitor the changing state of the atmosphere as a composite whole (development of
clouds, wind changes, rapid changes in atmospheric pressure, visibility and the like);
(e) Correlate the type of precipitation and electrometeors with the cloud types, phenomena-
reducing visibility with the visibility value, type of snowstorm phenomenon with wind
speed and snowfall intensity, etc.
The weather phenomena observations are recorded in the appropriate part of the log-book for
surface meteorological observations. When the observations are recorded, it is recommended
that the conventional symbols given in the Technical Regulations (WMO‑No. 49) be used.
3.2.2.2.2 Wind direction and speed
The following parameters should be measured:
(a) Mean wind speed at the time of observation;

(b) Mean wind direction at the time of observation;
(c) Maximum wind speed at the time of observation;
(d) Maximum wind speed in between standard observation times.
The wind instruments to be used, their height, the averaging period for the observation and
the estimation method in the absence of instruments are given in 3.3.5, Part III, Volume I, of the
Manual on the Global Observing System (WMO‑No. 544) and in 12.2.2.3, Part A, Volume I.1, of the
Manual on Codes (WMO-No. 306).
Land stations shall read the mean wind direction clockwise from the geographical (true)
meridian as the direction from which the wind is blowing. For this purpose, the instruments shall
be oriented exactly along the geographical meridian. This orientation should be systematically
checked, as should the verticality of the equipment’s mast and instruments, and corrected where
necessary.
During observations, the following shall be strictly adhered to:
(a) Prescribed measurement times;

(b) Averaging period for wind characteristics;
(c) Uncertainty for the readings:
– speed: ±0.5 m s–1 for ≤ 5 m s–1 and ±10% for > 5 m s–1;
– direction: ≤ 5°.
All wind measurements should be recorded in the surface meteorological observations log-book.
All wind equipment should be installed on special masts allowing access to the equipment. It
must be possible either to lower the upper part of the mast, or the mast must be fitted with metal
crossbars or rungs.
A preventive check should be carried out once a year on the wind vane: the vane should be
removed from its spindle and cleaned, the weight of the fin checked (permissible error ±1 per
cent) and the vane recoated with black lacquer. If the pivot bearing (the upper part of the
spindle, screwed into the mast) is worn, it should be unscrewed and remachined.
See the Guide to Meteorological Instruments and Methods of Observation (WMO‑No. 8), Part I,
Chapter 5, for further information.
Note: The averaging of wind direction is straightforward in principle, but a difficulty arises in that a 0–360° scale
has a discontinuity at 0°. As an extreme example, the average of 1° and 359° is 180°. This presents no difficulty to an
observer making a continuous recording of wind direction, but automatic computation devices must be provided with
some means to resolve the ambiguity.
3.2.2.2.3 Cloud amount, cloud types and cloud base height
Cloud amount should be determined according to the degree of cloud over the visible celestial
dome in tenths or octas with an uncertainty of one unit.
For visual observations of cloud types, the classification tables, definitions and descriptions of
types, species and varieties of clouds as given in the International Cloud Atlas (WMO‑No. 407),
Volume I, shall be used. See the Guide to Meteorological Instruments and Methods of Observation
(WMO‑ No. 8), Part I, Chapter 15, for further information.
Cloud base height should be determined by measurement. The technical means for measuring
can be based on several methods, such as light-pulse location and lasers. Pilot-balloons launched
from the ground can also be used.
The following recommendations shall apply to cloud observations:
(a) The observation sites should be as unobstructed as possible to enable observers to see the
maximum amount of the celestial concave;
(b) In order to determine the cloud species and types correctly, their evolution should be
monitored systematically at and between the observation times;
(c) Cloud amount should be determined globally for all layers (total cloud amount) and
individually for each significant cloud layer in order to meet the requirements of code form
FM 12-XIV SYNOP set out in Volume I.1, Part A, of the Manual on Codes (WMO-No. 306);
(d) At night, the determination of cloud species should be correlated with the nature of
precipitation and optical and other phenomena.
Cloud observations should be recorded in sufficient detail in the surface observations log-book
to permit the observations to be reported in code form FM 12-XIV SYNOP (see the Manual on
Codes (WMO‑No. 306), Volume I.1, Part A).
3.2.2.2.4 Visibility
For definitions of visibility during the day and at night, see Part I, Chapter 9, of the Guide to
Meteorological Instruments and Methods of Observation (WMO‑No. 8).
Surface synoptic stations shall measure or determine the meteorological optical range. Other
visibility characteristics such as runway visual range and slant visual range can be measured at
aerodromes and from aircraft.
Visual estimation and instrumental measurement of the meteorological optical range are
described in detail in Part I, Chapter 9, of the Guide to Meteorological Instruments and Methods of
Observation (WMO‑No. 8).
The objects used to estimate the meteorological optical range during the day should be spaced
at standard distances enabling determination of the visibility value in accordance with code table
4377, Horizontal visibility at surface, Part A, Volume I.1, of the Manual on Codes (WMO‑No. 306).
The distances up to the objects (L) should be measured instrumentally.
Visibility observations (meteorological optical range) should be recorded in the surface

meteorological observations log-book in three steps as specified in code table 4377, Part A,
Volume I.1, of the Manual on Codes (WMO‑No.306).
3.2.2.2.5 Air temperature and extreme temperature
See 3.3.3, Part III of the Manual on the Global Observing System (WMO‑No. 544) for the basic
regulations relating to this point.
The methods and instruments involved in measuring air temperature at surface stations are
described in Chapter 2, Part I, of the Guide to Meteorological Instruments and Methods of Observation
(WMO‑No. 8).
Surface stations shall measure the following air temperature characteristics:
(a) Temperature at the time of observation;
(b) Maximum temperature (highest temperature over a prescribed period of time, for example,
12 or 24 hours);
(c) Minimum temperature (lowest temperature over a prescribed period of time, for example,
12 or 24 hours).
Extreme (maximum and minimum) temperatures, when required by Regional Associations, shall
be measured at a minimum of two of the standard times (main or intermediate) with a 12-hour
interval between each; these roughly correspond to the morning and evening local time at the
observing site or station.
The measurement results and corrections, shall be recorded in the surface meteorological
observations log-book.
3.2.2.2.6 Humidity
See 3.3.4, Part III of the Manual on the Global Observing System (WMO‑No. 544) for basic
regulations relating to this point.
The methods and instruments used to measure the atmospheric humidity at surface stations are
described in Part I, Chapter 4, of the Guide to Meteorological Instruments and Methods of Observation
(WMO‑No. 8).
Land stations shall measure or calculate the following parameters:
(a) Vapour pressure;

(b) Relative humidity;
(c) Dew-point temperature.
Psychrometers and hair hygrometers are the most commonly used instruments for measuring
humidity at land stations.
Instrument readings shall be recorded at the specified time of observation in the surface
meteorological observations log-book. Calculated atmospheric humidity characteristics shall also
be recorded there.
3.2.2.2.7 Atmospheric pressure, pressure tendency and characteristic of pressure tendency
The methods and instruments for measuring atmospheric pressure at surface stations are
described in Chapter 3, Part I, of the Guide to Meteorological Instruments and Methods of Observation
(WMO‑No. 8).
Requirements for measuring atmospheric pressure, procedures to be followed for reducing the
pressure to mean sea level and requirements and procedures for reporting the geopotential
height of an agreed standard isobaric surface at high-level stations in accordance with the
relevant resolution adopted by the Regional Association are outlined in 3.3.2, Part III, Volume I,
of the Manual on the Global Observing System (WMO‑No. 544), and in 12.2.3.4.2 and code table
0264, Part A, Volume I.1, of the Manual on Codes (WMO‑No. 306). Furthermore, atmospheric
pressure observations should be made exactly at the standard time for surface synoptic
observations defined in 3.2.2.1 above.
The direct reading of atmospheric pressure from a barometer should be recorded in the surface
meteorological observations log-book. The following parameters should also be recorded in the
log-book: corrected station-level atmospheric pressure, calculated sea-level pressure or isobaric
surface height, calculated pressure tendency and pressure tendency characteristic.
Atmospheric pressure can be recorded continuously by using electronic barometers or

barographs.
The pressure tendency shall be determined from the atmospheric pressure values measured
from a barometer and expressed as the difference between these values during the three hours
preceding the observation time. The pressure tendency can be calculated from the barograph
readings as the difference between the readings taken from the recorded curve (plotted around
the clock) at corresponding observing times, that is, every three hours.
Characteristic of pressure tendency shall be determined by using the appropriate sign (rise =
“+” or fall = “-”) when taken from a barometer, and with the type of curve when taken from a
barograph.
Characteristic of pressure tendency shall be designated according to code table 0200 found in
Volume I.1, Part A, of the Manual on Codes (WMO‑No. 306).
Observation of the following variables are determined by the relevant resolutions of regional
associations:
3.2.2.2.8 Amount of precipitation
See 3.3.8, Part III, of the Manual on the Global Observing System (WMO‑No. 544).
The methods and instruments for measuring precipitation at surface stations are described
in Part I, Chapter 6, of the Guide to Meteorological Instruments and Methods of Observation
(WMO‑No. 8).
Surface stations shall, when required by regional associations, measure amounts of precipitation
and determine other precipitation characteristics such as duration and intensity. The amount
of precipitation shall be measured at a minimum of two standard times (main or intermediate),

12 hours apart and roughly corresponding to the morning and evening local time at the
observing site or station.
Note: In addition, Members may establish other times for measuring precipitation and carry out continuous
recording of liquid and solid precipitation.
Rain gauges are used for measuring the amount of precipitation. The type and exposure of
the rain-gauges and the material of which they are made should be chosen in such a way as to
reduce to a minimum the influence of wind, evaporation, wetting of the glass and splashing.
The measurements and corrections shall be recorded in the surface meteorological observations
log-book.
3.2.2.2.9 State of ground
The methods of observing the state of the ground at surface stations are described in Chapter 14,
Part I, of the Guide to Meteorological Instruments and Methods of Observation (WMO‑No. 8).
Land stations shall, when required by regional associations, determine the following parameters
at the morning observation time when the minimum temperature is measured, provided that in
winter there is enough light:
(a) State of ground without snow or measurable ice cover;

(b) State of ground with snow or measurable ice cover.
The state of the ground with or without snow, or measurable ice cover, should be determined
visually in accordance with the specifications given in code tables 0901 and 0975, Volume I.1,
Part A, of the Manual on Codes (WMO‑No. 306), which are self-explanatory.
The observations should meet the following requirements:
(a) In the absence of snow or measurable ice cover, the state of the ground is determined
in the meteorological instrument area at the spot where the thermometers are installed
for measuring the temperature of the surface and the ground is free of plant cover (bare
ground);
(b) The state of the ground and the snow or ice cover shall be determined in such a way as to
characterize the station’s environment, an open, representative area. Consequently, the
observations must always be made from the same, preferably raised place through a visual
survey of the area surrounding the station, or meteorological instrument area.
The observations shall be recorded in the surface meteorological observations log-book. The
recording can be in words, abbreviated conventional signs and in code form FM 12-XIV SYNOP
found in Volume I.1, Part A of the Manual on Codes (WMO‑No. 306).
3.2.2.2.10 Direction of cloud movement
Surface stations shall, when required by Regional Associations or national decisions, determine
cloud movement direction. This can be estimated visually and can also be determined, together
with its angular velocity by using a nephoscope.
3.2.2.2.11 Special phenomena
Surface stations should observe special weather phenomena that are generally called dangerous
or extremely dangerous—catastrophic, hazardous or severe—on an uninterrupted, round-the-
clock basis whenever possible.
These special phenomena hamper industrial and other daily activities and frequently cause
significant losses to industry and the population. In order to prevent or reduce losses, appropriate
observations should be made at the stations.
Special phenomena may comprise the following variables:
(a) Large values of the usual meteorological variables, for example, strong wind, considerable
rainfall and drop in air temperature during the transitional periods below 0–frosts;
(b) Unfavourable combinations of variables, for example, high temperatures and low air
humidity leading to droughts;
(c) Exceptionally long-lasting atmospheric phenomena such as fog or snowstorms;
(d) Individual rare phenomena, for example, hail tornadoes and others.
In practice, lists of special phenomena that are dangerous or extreme in nature and relevant
criteria such as threshold values are established by the Members.
Land stations shall ensure measurement or observation of the phenomena given in the Manual on
Codes (WMO‑No. 306), Volume I.1, Part A, code form FM 12-XIV SYNOP, section 3, as specified in
code table 3778.
Other special phenomena are determined by Members, depending on local conditions.
The following recommendations are made for observing special phenomena:
(a) Measurements should be made using instruments which have a sufficient range or scale
capable of pinpointing a rarely occurring value;
(b) Observers should be extremely attentive and flexible when there are signs of a special
phenomenon;
(c) Observations may be made at the station and in its vicinity, whilst data on the
consequences of a special phenomenon may also be collected through a survey of local
inhabitants.
Observations should be recorded in the log-book for surface meteorological observations

in expanded form, preferably including a brief descriptive text written in a section specially
reserved for this purpose.
Special phenomena of a catastrophic nature should be described in detail and their sequences
should, as far as possible, be photographed and mapped. It is recommended that special
instructions for the stations be prepared by Members for this purpose.
3.2.2.2.12 Automatic measurements
The content of a surface synoptic observation at an automatic land station is described in

2.3.2.10, Volume I, Part III, of the Manual on the Global Observing System (WMO‑No. 544).
Sensors and uncertainty requirements for automatic weather stations are outlined in Part II,
Chapter 1, of the Guide to Meteorological Instruments and Methods of Observation (WMO‑No. 8).
Information on sampling and reduction methods can be found in Part V of the present Guide and
in Part III, Chapters 2 and 3, respectively, of the Guide to Meteorological Instruments and Methods of
In general, classic visual observations are difficult to replace by automated means, although
in some cases new observing technology, such as satellites or remote-sounding techniques,
can supply better information than those obtained by classic means. It is possible, however, to
approach classic visual information by combining some automatically measured variables at land
stations. Examples are provided below.
(a) Present weather and past weather
(i) Of the possible 99 code variations, some of the most significant present weather
and past weather types may, with the development of suitable algorithms, be
automatically reported by using the combination of outputs of different common
automatic sensors such as precipitation sensors, thermometers, lightning counters and
anemometers, for example, ww: 17, 18, 21, 22, 29, 51, 61, 63, 71, 73, 75, 91, 92, 95, 97;
(ii) Distinction between solid and liquid precipitation with the help of melting power
used in precipitation gauges;
(b) Cloud information
(i) Interpretation of the gradient of air temperature near the ground (for example,
difference of temperature between 2 m and 5 cm above ground) to estimate the total
clouds amount;
(ii) Evaluation of radiation and illumination measurements to derive information about

the development of cloud cover.
3.2.2.3 Observations at sea stations
Meteorological variables that shall be observed and recorded at ocean weather stations
are defined in 2.3.3.11, Part III, Volume I, of the Manual on the Global Observing System
(WMO‑No. 544) and described in 3.2.2.3.1 to 3.2.2.3.11 below.
3.2.2.3.1 Present and past weather
The rules to be followed at sea stations also apply to land stations.
3.2.2.3.2 Wind direction and speed
The methods and instruments for observing the wind at sea stations are outlined in 4.2.5,
Chapter 4, Part II, of the Guide to Meteorological Instruments and Methods of Observation
(WMO‑No. 8).
Wind speed and direction may be observed by making visual estimates or using anemometers or
anemographs.
Visual estimates are generally based on the appearance of the sea surface. Observers should be
aware that wave height in itself is not always a reliable criterion since it also depends on wind
fetch and duration, and swell presence.
The Beaufort wind scale in Table III.2 provides the criteria for the visual estimation of wind speed.
Table III.2. Beaufort wind scale (for a standard height of 10 metres above flat open ground)
Estimated mean speed (range) Specifications for use Probable Probable

Beaufort
Descriptive wave wave
number
term Knots m s–1 km/h mph On land At sea On the coast heighta heighta
(force)
(metres) (feet)
0 Calm <1 0-0.2 <1 <1 Calm; smoke rises Sea like a mirror Calm - -
vertically
1 Light air 1–3 0.3–1.5 1–5 1–3 Direction of wind Ripples with the appearance of Fishing smack 0.1 (0.1) ¼ (¼)
shown by smoke scales are formed, but without just has steerage
drift but not by wind foam crests way
vanes
2 Light 4–6 1.6–3.3 6–11 4–7 Wind felt on face; Small wavelets; still short but Wind fills the 0.2 (0.3) ½ (1)
breeze leaves rustle; ordinary more pronounced; crests have sails of smacks
vanes moved by wind a glassy appearance and do not which then travel
break at about 1–2
knots
3 Gentle 7–10 3.4–5.4 12–19 8–12 Leaves and small Large wavelets; crests begin to Smacks begin to 0.6 (1) 2 (3)
breeze twigs in constant break; foam of glassy appearance; careen and travel
motion; wind extends perhaps scattered white horses about 3–4 knots
light flag
4 Moderate 11–16 5.5–7.9 20–28 13–18 Raises dust and loose Small waves, becoming longer; Good working 1 (1.5) 3½ (5)
breeze paper; small branches fairly frequent white horses breeze, smacks
are moved carry all canvas
with good list
5 Fresh 17–21 8.0–10.7 29–38 19–24 Small trees in leaf Moderate waves, taking a more Smacks shorten 2 (2.5) 6 (8½)
breeze begin to sway; pronounced long form; many sails
created wavelets form white horses are formed (chance
on inland waters of some spray)
6 Strong 22–27 10.8–13.8 39–49 25–31 Large branches in Large waves begin to form; Smacks have 3 (4) 9½ (13)
breeze motion; whistling the white foam crests are more double reef in
heard in telegraph extensive everywhere (probably mainsail; care
wires; umbrellas used some spray) required when
with difficulty fishing
7 Near gale 28–33 13.9–17.1 50–61 32–38 Whole trees Sea heaps up and white foam Smacks remain 4 (5.5) 13½ (19)
in motion; from breaking waves begins to in harbour and
inconvenience felt be blown in streaks along the those at sea lie to
when walking against direction of the wind
wind
Estimated mean speed (range) Specifications for use Probable Probable

Beaufort
Descriptive wave wave
number
term Knots m s–1 km/h mph On land At sea On the coast heighta heighta
(force)
(metres) (feet)
8 Gale 34–40 17.2–20.7 62–74 39–46 Breaks twigs off trees; Moderately high waves of greater All smacks make 5.5 (7.5) 18 (25)
generally impedes length; edges of crests begin to for harbour, if
progress break into the spindrift; the foam near
is blown in well-marked streaks
along the direction of the wind
9 Strong 41–47 20.8–24.4 75–88 47–54 Slight structural High waves; dense streaks of foam – 7 (10) 23 (32)
gale damage occurs along the direction of the wind;
(chimney pots and crests of waves begin to topple,
slates removed) tumble and roll over; spray may
affect visibility
10 Storm 48–55 24.5–28.4 89–102 55–63 Seldom experienced Very high waves with long – 9 (12.5) 29 (41)
inland; trees overhanging crests; the resulting
uprooted; foam, in great patches is blown
considerable in dense white streaks along
structural damage the direction of the wind; on
occurs the whole the surface of the sea
takes a white appearance; the
“tumbling” of the sea becomes
heavy and shock-like; visibility
affected
11 Violent 56–63 28.5–32.6 103–117 64–72 Very rarely Exceptionally high waves (small – 11.5 (16) 37 (52)
storm experienced; and medium-sized ships might
accompanied by be for a time lost to view behind
widespread damage the waves); the sea is completely
covered with white long patches
of foam lying along the direction
of the wind; everywhere the
edges of the wave crests are
blown into froth; visibility affected
12 Hurricane 64 and 32.7 and 118 and 73 and – The air is filled with foam and – 14 (–) 45 (–)
over over over over spray; sea completely white
with driving spray; visibility very
seriously affected
a
This table is intended to serve as a guide to provide a rough indication of what may be expected in the open sea, remote from land. It should never be used to log or report the state of
the sea. In enclosed waters, or when near land, with an off-shore wind, wave heights will be smaller and the waves steeper. Figures in brackets indicate the probable maximum height
of waves.
GUIDE TO THE GLOBAL OBSERVING SYSTEM 63
Smoke d
e win
Tru
A
Apparent wind
D
Sh
ip
’s
he
ad
Figure III.7. Velocity parallelogram

Source: Marine Observer’s Handbook, Meteorological Office, United Kingdom, 1995
Wind direction is determined by observing the crest orientation of the wind-driven sea waves.
Wind measurements using anemometers and anemographs are made in the same way as for
land stations, but it may be difficult to avoid local effects, for instance those produced by a ship’s
superstructure. Therefore, the instrument should be sited on board of the mobile ship as far
forward and as high as possible.
When measurements are made on a moving ship, it is necessary to distinguish between the wind
relative to the ship and the true wind. For meteorological purposes, the true wind should always
be reported. It may be obtained from the apparent wind by using the velocity parallelogram as
shown in Figure III.7.
The apparent wind speed measured on board a moving ship is to be corrected for the ship’s
course and speed in order to obtain the speed of the true wind, which be reported. The
correction can be made by using the velocity parallelogram or special tables (see 12.2.2.3.3,
Part A, Volume I.1, of the Manual on Codes (WMO‑No. 306)).
Special rules must be applied for determining the wind on fixed and anchored platform stations
because their heights may be more than 100 m above sea level, whereas surface wind is defined
as the horizontal component of the wind vector that is measured 10 m above the ground or the
sea surface. If the wind sensor is exposed at a higher elevation, the readings must be corrected.
The following scale should be applied for reporting mean wind speed over 10 minutes:
Elevation (height in meters) Coefficient (r)

20 1.10
30 1.15
40 1.20
50 1.23
60 1.26
70 1.29
80 1.31
90 1.33
100 1.35
Example:
Sensor at 75 m, observed speed 50 kt
Coefficient 1.30 (interpolated)

Wind reduced to a standard 10 m height = 50/1.30 = 38.46 or 38 kt
3.2.2.3.3 Cloud amount, cloud type and cloud base height
The rules for land stations are generally followed for sea stations as well, but estimating the
cloud base without landmarks such as mountains may be difficult. Ordinary methods using
a searchlight are of limited value because of the short baseline available on a ship. The best
solution is probably a pulse-light cloud searchlight which does not require a baseline and which
electronically measures the reflection time from the cloud base. This instrument, however,
is rather sophisticated and expensive and therefore not widely used. Observers should take
every opportunity to check their cloud height estimates against known heights, for example,
mountains near the coast.
3.2.2.3.4 Visibility
In the absence of suitable landmarks, it is not possible for visibility observations at sea stations
to attain the same uncertainty as those made at land stations. The requirements for accurate
visibility observations at sea stations are therefore set low, as shown in decade 90–99 of code
table 4377, Volume I.1, Part A, of the Manual on Codes (WMO-No. 306).
When not uniform in all directions, visibility should be estimated or measured in the direction of
least visibility. A suitable entry should be made in the log-book, except in situations of reduced
visibility due to the ship’s exhaust.
Methods of observing visibility at sea stations are described in 4.2.8, Part II, Chapter 4, of the
Guide to Meteorological Instruments and Methods of Observation (WMO‑No. 8).
3.2.2.3.5 Air temperature and humidity
Temperature and humidity observation requirements for use at sea stations are described in
4.2.9, Part II, Chapter 4, of the Guide to Meteorological Instruments and Methods of Observation
(WMO‑No. 8).
When reporting air temperature and humidity observations made at fixed and anchored
platform stations higher than 100 m above sea level, it is not necessary to take into account
variations of temperature and humidity with height.
3.2.2.3.6 Atmospheric pressure, pressure tendency and characteristic of pressure tendency
Requirements and instruments used for observing atmospheric pressure at sea stations are
described in 4.2.6, Part II, Chapter 4, of the Guide to Meteorological Instruments and Methods of
3.2.2.3.7 Ship’s course and speed
A ship’s position, course and speed is taken from its navigation system or computed
independently using a satellite navigator, for example, a global positioning system.
The mean speed shall be reported. (See group 10 in the code form FM 13-XIV SHIP contained in
Volume I.1, Part A, of the Manual on Codes, WMO‑No. 306).
3.2.2.3.8 Sea-surface temperature
The measurement of sea-surface temperature is described in 4.2.11, Part II, Chapter 4, of the

The method used for measuring sea-surface temperature at manned sea stations shall be
recorded in the relevant meteorological log-book.
3.2.2.3.9 Ocean waves and swell
The observation of waves and swell is described in 4.2.12, Part II, Chapter 4, of the Guide to
The characteristics of a simple wave are shown in Figure III.8.
Observations of waves made from island and coastal stations are not representative owing to
shallow water, effects of shelter by the shore and other factors.
3.2.2.3.9.1 Use of wave measurement instruments
In recent years suitable sea-wave recorders have been developed for measuring the height and
period of waves. The following instruments are used:
(a) Surface-following buoys that measure acceleration;
(b) Ship-borne wave recorders that measure pressure and acceleration;
(c) Wave staffs that measure electrical resistance or capacity;
(d) Radar measurement instruments mounted on a platform or on land. It is strongly

recommended that such recording instruments be used at ocean weather stations, research
ships and fixed platforms.
Further information on wave observations is given in the Handbook on Wave Analysis and
Forecasting (WMO‑No. 446).
3.2.2.3.10 Sea ice
The observation of ice is described in 4.2.13, Chapter 4, Part II, of the Guide to Meteorological
The following four features of sea ice should be observed:
(a) Ice thickness;

(b) Amount—concentration (estimates expressed in eighths of sea surface covered by ice);
(c) Types of sea ice, such as fast ice or pack ice;
(d) Ice movement.
C H C
Figure III.8. Characteristics of a simple wave

Reports on ice accretion at sea may be made in plain language or in code form (FM 13-XIV SHIP)
as indicated in Volume I.1, Part A, of the Manual on Codes (WMO‑No. 306).
3.2.2.3.11 Special phenomena
WMO has been requested to provide observations of a special nature through its VOS Scheme.
Examples of these are as follows:
(a) Observations of locust swarms in sea areas around Africa, the Middle East, Pakistan and
India;
(b) Observations of freak waves which present a great danger to shipping;
(c) Sea-surface currents, which can be determined from measurements of a ship’s set and drift,
and are of value to research and climate studies.
Further details on the reporting of these observations can be found in 6.4.5, Chapter 6, of the
Guide to Marine Meteorological Services (WMO‑No. 471) and its Annexes.
Waterspouts should be reported as a special observation. When describing a waterspout, the

direction of rotation should always be given as if seen from above.
3.3 UPPER-AIR STATIONS
3.3.1 Organizational aspects
An upper-air observation is a direct or indirect meteorological observation made in the free

atmosphere. Pilot-balloons, radiosondes, radiowinds, combined radiosondes and radiowinds,
or rawinsondes are used for in situ direct measurements. Sodars, wind profilers, radio-acoustic
sounding systems, lidars and other observing technologies can be used for remote sensing of the
troposphere. See measured and calculated variables listed in 3.3.2.6.
3.3.1.1 Site selection
Once the general area for siting a station has been chosen, it is necessary to select a specific site
for the facility. It is recommended that the following criteria be applied:
(a) Government-owned land should be considered as the first choice, since there is less chance
of having to relocate, and future encroachment would be minimized;
(b) The optimum site area should be approximately 40 000 m2;
(c) The site must be accessible by an all-weather road to facilitate delivery of supplies and
proper station maintenance;
(d) The site should not be located in a flood plain; it should have good drainage;
(e) The site should be free from natural or man-made obstructions that would interfere with
the launch, path or tracking of balloons;
(f) Utilities such as electric power, water, sewerage and communications should be available;
(g) The site must be surveyed to ensure that electronic equipment or telecommunications are
free from any interference.
See below for an example of a site survey questionnaire.

Part I, Chapters 12 and 13, of the Guide to Meteorological Instruments and Methods of Observation
(WMO‑No. 8) provides further details.
upper-air site survey questionnaire
Location: Date:
1. (a) Describe the proposed upper-air site and enter latitude and longitude coordinates.
(b) Attach an obstruction diagram plotted from data obtained by theodolite

measurements showing direction, distance and angular elevation of all obstructions
to tracking equipment above zero elevation. Attach a copy of photos arranged to
form a 360º panoramic view of the horizon. If this report is being prepared for
co-location with radar, panoramic photos prepared for the radar can serve the purpose.
2. (a) Where will tracking set be located? Describe location: on roof of building, on top of
inflation shelter, on tower, on ground and with respect to office and inflation shelter.
(b) Height in feet or metres:
(c) Elevation in feet or metres above mean sea level:
3. Length of the cable run between the tracking set and the recorder:
4. Estimated costs. The estimates prepared in the Meteorological Department

show the following:
(a) Land – purchase or rental costs:
(b) Site preparation (roads, walks, utilities):
(c) Building construction or modification:
(d) Communications:
(e) Inflation shelter or support:
(f) Conduits and cables:
(g) Other:
TOTAL:
5. Remarks:
Primary release
Prevaling wind Theodolite area
100 metres (minimum)

15 cm stabilized gravel
Parking Grade 1/25 or flatter
Figure III.9. Upper-air facility with station office
3.3.1.2 Planning of facilities
The basic buildings on the site are the station office (Figure III.9) and the balloon inflation shelter
(Figure III.10). In many cases, the radar or radiotheodolite are located above the main station
building.
200 m
Prevailing wind Fence
Theodolite
Balloon Primary release area
inflation (15-cm stabilized gravel)
shelter
m 10
10 m
7m
1 m-wide
200 m
stairs
All-weather road 5m
7m
Office Underground
12 m cable
9m
Plan
Clear space
No obstructions
46 m
Minimum
9m
Elevation Balloon inflation shelter
Figure III.10. Upper-air site

Design considerations for the station office are as follows:
(a) Operational functions;

(b) Area limitations;
(c) Severe weather protection;
(d) Heating and cooling equipment;
(e) Emergency electric power;
(f) Fire protection;
(g) Lightning protection;
(h) Communications;
(i) Safety techniques.
Design considerations for the balloon inflation shelter and release area are the following:
(a) Storage of expendable supplies;

(b) Orientation;
(c) Area lighting;
(d) Ventilation;
(e) Explosion-proof electrical system;
(f) Door openings;
(g) Fire protection;
(h) Pit equipped for waste;
(i) Safety techniques.
Design considerations should also include a location for the following equipment:
(a) Observing equipment;

(b) Inflation equipment;
(c) Power-generation equipment;
(d) Communication equipment;
(e) Hydrogen-generation equipment or helium supply;
(f) Balloon-preparation equipment.
Station design should be in the hands of qualified architects or engineers who are familiar with
the functional requirements of the activity programme at the station, and in close collaboration
with the Meteorological Service.
Several sites should be considered, site surveys should be conducted and the results submitted
to the relevant authorities for official approval. Plans including specifications and other contract
documents should also be prepared. Procurement of necessary equipment, design of buildings
or rental of new or existing property should be undertaken.
Provisions necessary for the day-to-day operation of the station must include the following:
(a) Procurement and storage of expendable supplies:
(i) Inflation gas and supplies;

(ii) Radiosondes, targets and balloons;
(iii) Engine generator fuel;
(iv) Office materials;
(b) Adequate documentation such as the Technical Regulations (WMO-No. 49), Manuals and
Guides;
(c) Complement of spare parts;
(d) Maintenance and supplies for buildings and grounds;
(e) Space for electronic technician for on-site or on-call equipment maintenance.
3.3.1.3 Organization of the upper-air unit
The upper-air unit consists of those components required to carry out upper-air observations.
It includes all aspects of facilities, personnel, equipment and maintenance for any upper-air
observations made at the station: pilot-balloon, radiosonde, radiowind, rawinsonde, combined
radiosonde and radiowind observations.
The upper-air unit may or may not be in the same location as other weather services. It may also
provide the only type of observation made by a participating Member at a particular station.
Co-location of weather services and observations is often a cost-effective approach. Usually, the
unit is merely a part of another observational office. Observers usually perform other functions
in addition to upper-air observations. However, in some cases it may be necessary to maintain
the unit apart from the other weather services. Observers may staff the upper-air unit alone, or
perform duties at both locations. If upper-air observations are the only service provided at a
station, observers would require upper-air training only.
Regardless of the unit’s location, certain organizational relationships must be developed

to ensure its efficiency. When the upper-air unit is co-located with other weather services,
upper-air observations become an integral part of the organization. Personnel skills, work
schedules, training and the like must be expanded or modified to meet upper-air observational
requirements. When the unit is not co-located with the main weather office, it may or may not
be an integral part of the latter. With few exceptions, the upper-air unit should ideally have an
organizational relationship with another weather office.
If applicable, the upper-air unit should have a close working relationship with a central
headquarters whose function may consist of establishing policies and regulations, ordering
materials and supplies and providing training.
3.3.1.4 Archiving data and maintenance of records
Archiving upper-air data after completing an observation is very important to the World Weather
Watch and the World Climate Programme. A complete set of records containing pertinent
observational data must be kept by the station or the central headquarters, or in another
location. In addition, the data may be stored on a medium such as magnetic tape or disk.
Countries that can do so are advised to make provisions to supply the data upon request.
Further details can be found in 12.10.2, Part I, Chapter 12 of the Guide to Meteorological Instruments
and Methods of Observation (WMO‑No. 8).
In addition to the data archives mentioned above, the upper-air unit should keep the following
records to monitor operations:
(a) Various types of information relating to the observation, for example, bursting height,
reason for termination, problems encountered during the flight and type of radiosonde
used;
(b) A complete list of instruments and other office equipment used to obtain and transmit the
data.
3.3.1.5 Communications
The unit responsible for transferring the data from the upper-air unit to national communications
circuits and the Global Telecommunication System may vary from Member to Member.
Some countries may employ communicators or communications specialists who ensure that the
data are disseminated over the Global Telecommunication System network in a timely fashion.
In other countries the observer, or even personnel not associated with the upper-air unit, may be
given this responsibility. Some Members use trained private-sector individuals under contract
with National Meteorological Services.
For the data to be of any value, they must be entered into the distribution network at the
required times. Alternative means to communicate information should be available when the
primary communications link has been disrupted.
The following communications equipment is needed to disseminate data, depending, inter alia,
on the quality of communications circuits, remoteness of the upper-air unit and availability of
satellite ground stations:
(a) Interface to public communication networks;

(b) Telegraphs;
(c) Telex typewriters;
(d) Radioteleprinters;
(e) Radio facsimile broadcasts;
(f) Radios;
(g) Satellite-based communication equipment.
The data may be transmitted to the central headquarters, which sends them to the Global
Telecommunication System network. In some cases, another weather office or agency may be
responsible for transmitting the data to the Global Telecommunication System network.
3.3.1.6 Personnel
The type and number of upper-air unit personnel depend on the equipment used, and the level
of expertise and number of observations required. The type and degree of training depend on
the role and responsibilities of the staff members.
Personnel requirements by category are outlined below; examples of the recommended number
of personnel are provided in Tables III.3 and III.4．
(a) Station supervisor (WMO classification: Meteorologist; designated as “S” in Tables III.3

and III.4
Table III.3. Example of recommended observational personnel requirements – number per observation
Observing
Pilot balloons Radiosondes Rawinsondes Radiowinds
method
S SS O C M Ta S SS O C M Ta S SS O C M Ta S SS O C M Ta
Automatic - - - - - - 1 - 1 1b 1b 2 1 - 1 1b 1b 2 1 - 1 1b 1b 2
Semi-auto. 1 - 1 1b - 2 1 - 1 1b 1b 2 1 - 1 1b 1b 2 1 - 1 1b 1b 2
Manual 1 1b 1 1b - 2 1 1 1 1b 1b 3 1 1 2 1b 1b 4 1 2 1 1b 1b 4
a Minimum number of personnel needed to carry out observation; optional staff not included
b Optional positions for observing programme; supervisors considered part of observing programme
Table III.4. Example of recommended observational personnel requirements – number per week
Obser-vations/ Observing
Pilot balloons Radiosond Rawinsondes Radiowinds
day method
S SS O C M Ta S SS O C M Ta S SS O C M Ta S SS O C M Ta
1 Semi-auto. 1 - 2 2b - 3 1 - 2 2b 1b 3 1 - 2 2b 1b 3 1 - 2 3b 1b 3
Manual 1 2b 4 2b - 5 1 2 2 2b 1b 5 1 2 4 2b 1b 7 1 2 4 3b 1b 7
2 or 3 Semi-auto. 1 - 3 3b - 4 1 - 3 3b 1b 4 1 - 3 3b 1b 4 1 - 3 3b 1b 4
Manual 1 3b 6 3b - 7 1 3 3 3b 1b 7 1 3 6 3b 1b 10 1 3 6 3b 1b 10
4 Semi-auto. 1 - 4 4b - 5 1 - 4 4b 1b 5 1 - 4 4b 1b 5 1 - 4 4b 1b 5
Manual 1 4b 8 4b - 9 1 4 4 4b 1b 9 1 4 8 4b 1b 13 1 4 8 4b 1b 13
a Minimum number of personnel required to carry out observing programme; optional staff not included
b Optional positions for observing programme; supervisors considered part of observing programme
When more than one person is employed in the upper-air unit, a supervisor should be
designated. The relationship between the supervisor and the other staff is crucial to a well-
run office. The supervisor should be one of the most experienced people in the upper-air unit,
preferably with expertise in areas other than upper-air observations, such as hydrogen safety
and other upper-air instruments and equipment. Good communications and management skills
are also important. A supervisor’s main duty is to manage the upper-air unit so that it functions
efficiently. Another is to serve as the unit’s spokesman when communication with other weather
offices or agencies is required. In particular, this responsibility should include the following
duties:
(i) Seeking guidance from the central headquarters when a higher authority is required;
(ii) Establishing work schedules for station personnel;
(iii) Keeping inventories of all supplies and expendables and ordering them in time;
(iv) Ensuring that all relevant policies and regulations are carried out by the station
personnel and that the WMO Technical Regulations, Manuals and Guides and other
such documentation are kept up to date and available to the station personnel;
(v) Ensuring that all safety precautions are observed with respect to hydrogen gas, upper-
air instruments and equipment, power installations and other equipment.
(b) Shift supervisor (WMO classification: Meteorological Technician) (SS)
The designation of shift supervisors is desirable at a station performing manual upper-air

observations and is optional for all other methods of observation. Shift supervisors, who are
amongst the most experienced observers assigned to each shift, should have a broad operational
background. The minimum requirements can be fulfilled with on-the-job training.
(c) Observer (WMO classification – Meteorological Technician) (O)
The number of observers required to carry out an observation depends on the methods used
(automatic, semi-automatic, manual) and the observers’ level of experience. Although previous
experience in upper-air observations is not required, formal and on-the-job training are
necessary.
(d) Maintenance staff/technicians (WMO classification: Meteorological Technician) (M)
Minimum requirements for maintenance personnel/technicians are a high school or technical

school diploma. They must also have specialized training in the maintenance and operation of
some types of equipment, some knowledge of station equipment, a basic understanding of the
physical atmosphere and a minimum of two years’ recent experience.
(e) Communicator (C)
The role of communicators will depend on the volume of information transmitted and the
diversity of responsibility. Their basic training should include course work; further experience will
be gained through on-the-job training. In some cases, a licence for operating the appropriate
communications equipment may be required.
Notes:
1. The tables should be used as a staffing guide, not to establish minimum requirements.
2. For a description of the classification of meteorological personnel and their duties, see Guide to the Implementation
of Education and Training Standards in Meteorology and Hydrology (WMO‑No. 1083), Volume I – Meteorology.
3.3.1.7 Training
The purpose of a training programme is to ensure that the personnel in the upper-air facility
are able to meet all the demands made of them. These include the administration and
management of the station and the effective operation of the observing programme as well as
the implementation of new requirements or modifications to operating procedures that may be
requested. Training of a recurring nature is therefore important.
Technical training should cover operational and maintenance aspects. Operational training
is required for the meteorological technician who selects the pertinent meteorological data
from the equipment. Observers are key members of the team in this category since they are
responsible for data acquisition, sounding data reduction and preparation for local use and
transmission over the telecommunications systems. Training is essential, whether acquired on-
the-job or through formal course work.
Maintenance training is required for personnel responsible for preventive and corrective
maintenance of the system. In order to understand adequately the operation of electronic and
electromechanical devices, and to maintain them, it is necessary to understand the theory on
which they are based. The theory provides the foundation for understanding the operation
of present-day and proposed operational meteorological equipment. Instructions in theory
should therefore have priority in the training of a maintenance technician. Appropriate practical
training should also be provided to the staff before they attempt to maintain sophisticated and
equipment at their facilities.
A variety of training opportunities are often available at local or regional universities, technical
schools, or factories producing specialized meteorological equipment. On-the-job training
should be arranged either at the local facility or at another which performs the same function.
On-the-job-training in the use of complex equipment is sometimes provided upon the successful
conclusion of formal schooling instead of classroom study. Non-technical training is as important
as technical training as a basis for achieving operational efficiency of the facilities.
3.3.1.8 Quality standards
Reference should be made to:
(a) Manual on the Global Data-processing and Forecasting System (WMO‑No. 485), Part II, 2.1.3,
Minimum standards;
(b) Guide on the Global Data-processing System (WMO‑No. 305), Chapter 6.
3.3.2 Observations/measurements
3.3.2.1 General
The basic regulations are contained in 2.4, Part III, of the Manual on the Global Observing System
(WMO‑No. 544).
Upper-air soundings are taken with a number of different types of instruments on land and at
sea, at permanently established stations and on moving platforms, including research vessels.
For further information on applicable measurements, see Chapters 12 and 13, Part I, of the Guide
to Meteorological Instruments and Methods of Observation (WMO‑No. 8).
3.3.2.2 Pilot-balloon observations
Pilot-balloon observations are one of the oldest and simplest methods of upper-air observation
in use today. They involve the visual tracking of a pilot balloon as it rises by means of an optical
theodolite. An assumed ascension rate according to the balloon weight and the lifting gas
provides the height needed to calculate the wind speeds and direction. Observers read the
elevation and azimuth angles for a designated time period for as long as the balloon can be
tracked visually. The data may be plotted to derive the necessary wind data or entered into a
calculator or computer for semi-automatic data reduction.
Although this type of upper-air observation is primitive in comparison to other methods, it is

still used by some Members. It can be an inexpensive, simple-to-operate sounding procedure,
especially in climates which enjoy many cloud-free days. The major disadvantages of this
technique are that the observations are limited, even by small amounts of cloud cover. Moreover,
measurement uncertainty is directly proportional to the validity of the assumed ascension rate.
Further information can be found in 13.3.2, Chapter 13, Part I, of the Guide to Meteorological
3.3.2.3 Radiosonde observations
Of all the upper-air observations which use telemetry signals to collect data, radiosonde
observations continue to be the basic observation. Generally speaking, most radiosondes in use
today measure the basic variables of temperature, pressure and relative humidity (or dew point).
These measurements are carried out by sensors mounted in an instrument package which also
contains a radio-frequency transmitter. The transmitter communicates these data to the ground-
receiving equipment which may be converted into a strip chart recording or go directly into a
computer for further analysis. Regardless of the method employed, these data must be converted
into an easily recognizable, standardized form in accordance with the Technical Regulations
(WMO‑No. 49).
The radiosonde design and the exposure of its sensors should be such as to minimize adverse
effects of solar and terrestrial radiation, precipitation, evaporation or freezing on a sensor.
Suitable radiation corrections should be made if necessary. A control or check reading for each
sensor should be performed a few minutes before the radiosonde is released.
At synoptic upper-air stations, the vertical distances of an ascending radiosonde shall be

determined by means of hydrostatic computation or precision radar tracking. Variables measured
by a radiosonde and the desirable ranges and uncertainty requirements are given in Annex 12.A,
Chapter 12, Part I, of the Guide to Meteorological Instruments and Methods of Observation
(WMO‑No. 8).
A properly made radiosonde observation gives a two-dimensional picture of the atmosphere and
a three-dimensional picture when used in an upper-air network. For those Members not having
the use of a wind-finding apparatus, pilot-balloon observations can be made in conjunction with
radiosonde observations. The heights can be accurately derived from the radiosonde data which,
in turn, will produce accurate wind data.
Further information can be found in Chapter 12, Part I, of the Guide to Meteorological Instruments
3.3.2.4 Radiowind observations
A common radiowind observing method uses a windfinding radar that tracks a reflective surface
attached below the balloon. In practice, most operational windfinding radars have difficulty in
measuring height with sufficient accuracy to satisfy user requirements for pressure and height
measurements in the troposphere.
The major advantage of this observing method is that the necessary equipment for this type
of upper-air observation is generally small and can be mounted almost anywhere. This system
works best in climates which are not influenced by upper-air jets since the radar range is usually
limited to under 100 km. The disadvantage is that it may be influenced by passing high-altitude
aeroplanes causing the radar to lose track of its target.
Further information can be found in 13.2.2, Chapter 13, Part I, of the Guide to Meteorological
3.3.2.5 Rawinsonde observations
The most widely used type of upper-air observation in the world today is that of the rawinsonde,
which stands for radio-sounding wind. Rawinsonde and radiowind observations differ in their
observing method. Rawinsonde observations keep track of the radiosonde position and use
that information to make wind calculations. It uses the radiosonde as the active target. Another
difference is that the methods of collecting the positional data may vary. Two types of wind-
finding methods are used in most of today’s rawinsonde observations and include the use of
radio direction finding (RDF) systems and navigation aid (NAVAID) signals such as GPS and long-
range navigation (LORAN-C).
These systems vary in complexity, ranging from the use of strip chart recorders to highly
sophisticated computers which automatically analyse the data. In general, RDF involves the use
of large wind-finding equipment and NAVAID, relatively small wind-finding equipment. The pros
and cons of the various systems are discussed in 3.3.2.8.
3.3.2.6 Combined radiosonde and radiowind observations
Upper-air observations involve the use of the radiosondes in conjunction with radar. The
radiosondes are equipped with sensors for measuring meteorological variables. They have a
transmitter which communicates meteorological data and at the same time serves as an active
target for radar determination of the radiosonde position. This type of upper-air observation
provides the highest number of measured variables, such as temperature, humidity, pressure,
height of ascent and wind speed and direction.
The variables listed below can be measured or derived from

the basic measurements described in the previous sections:
• Wind speed and direction
• Constant pressure/height levels
• Tropopause data
• Dew point
• Stability index (optional)
• Mean winds (two levels)
• Wind shear
• Cloud parameters (optional)
• Maximum wind
• Freezing level (optional)
• Minimum/maximum temperature and relative humidity
for the observation
• Superadiabats and inversions (climate)
• Other data
3.3.2.7 Aerological soundings using automated shipboard or land-based upper-air

systems
The development and successful trials of a highly automated system known as the Automated
Shipboard Aerological Programme (ASAP) offers additional possibilities for obtaining upper-air
observations from ocean areas as well as isolated land areas.
ASAP involves the generation of upper-air profile data from data-sparse ocean areas using
automated sounding systems carried on board merchant ships plying regular ocean routes.
The profile data are available in real time on the Global Tele-communication System, for use
by operational centres. ASAP is of vital importance to both the World Weather Watch and the
Global Climate Observing System. Several National Meteorological Services operate ASAP units,
Balloon NCAR GOES ground station

300 g
GOES-West
Global positioning system

GOES
receiver
403 MHz
Launcher GOES uplink Apple

antenna computer
Global
positioning
system
Modem
Rotor GOES
Preamp Preamp
control transmitter
NCAR Telephone line
AES Telephone line
Automatic
upper-air HP B5
computer Shiptemp message Apple
system to Global Modem
computer
Telecommunication
System
Shipboard launch and tracking module
NCAR = National Center for Atmospheric Research, USA
Figure III.11. Example of an automated shipboard aerological programme system
and the programme is coordinated through the ASAP Panel, a component of the Joint WMO/
IOC Technical Commission for Oceanography and Marine Meteorology Ship Observations
Team. Most of the soundings are presently made from the North Atlantic and North–West Pacific
Oceans. The Ship Observations Team publishes an annual report, giving ASAP programme status
and statistics on data return and data quality.
The main components of the ASAP system are the launcher, automatic upper-air and
communication systems and the land-based ground station, which receives the data via satellite
and inserts them into the Global Telecommunication System. The release of the balloon is
automatic and its position as it ascends is determined by using the global positioning system to
calculate the winds aloft. All of the data processing is done automatically by computer, which
converts the sounding data to standard coded message form for relay via a geostationary
meteorological satellite using its data-collection system. The high degree of automation allows
the ASAP system to be operated by one person.
Figure III.11 provides an example of the system. The example is taken from a ship crossing the
Pacific between Japan and Canada. A sketch of the container is given in Figure III.12.
3.3.2.8 Upper-air systems
An upper-air sounding system consists of two primary components that are needed to make
one or more of the upper-air observations mentioned in 3.3.2.2 to 3.3.2.6: a radiosonde that
measures and transmits meteorological data and a ground station that receives the telemetry
and processes it into meteorological data products. Those systems in turn are composed of five
main parts:
(a) Radiosonde/transmitter;
(b) Antenna(s)/receiver(s);
(c) Signal processing system (decoder);
(d) System computer;
(e) Meteorological operating system (software).
An upper-air system can also contain peripheral equipment specific to certain manufactures such
as radiosonde ground check devices.
Figure III.13 contains a simplified system diagram of the RDF (a) and GPS (b) type system.
An important difference between the two formats is that the 1680 MHz receiver is located in the
antenna for RDF systems. The 403 MHz global positioning systems require two receivers (UHF
and differential GPS), both of which are located in the meteorological processor. This makes it
a much more complex and costly device than the corresponding signal processor used in RDF
systems.
ASAP
Figure III.12. Marine module for the shipboard installation of an ASAP launching and tracking
module
(a) Radiosonde
Signal
System computer processor
Antenna
(b) Radiosonde
Meteorological Antennas
System computer processor
Figure III.13. RDF (a) and GPS (b) upper-air sounding systems
3.3.2.8.1 Plug-and-play systems or interoperability
There are several reasons why upper-air systems have become closed rather than open, plug-
and-play type systems:
(a) Manufacturers use proprietary methods to decode, correct and process the pressure,
temperature and humidity data collected by their radiosondes. These methods cannot be
disclosed without putting trade secrets at risk;
(b) Plug-and-play compatibility is expensive to develop and there is no incentive for

manufacturers to provide it;
(c) Manufacturers have an incentive to control all parts of the system in order to maintain
quality and provide seamless integration. If one manufacturer does not control the
complete system, it becomes difficult to determine who is responsible if there is a system
malfunction;
(d) Users generally do not require it.
3.3.2.8.2 RDF systems interoperability
The 1680 MHz RDF systems have demonstrated the technical feasibility of interoperability. For an
RDF system to use a new radiosonde, two conditions must be met by the sonde manufacturer:
(a) A sonde-specific signal processing system must be supplied that is compatible with the
antenna and system computer;
(b) Certain algorithms have to be provided to the antenna supplier so the Meteorological
Operating System can apply sonde-specific calibration and data corrections.
After a new sonde has been integrated into the operating system, it should be possible to change
over from one sonde to another in a matter of minutes.
3.3.2.8.3 GPS systems interoperability
Although it is theoretically possible, operational interoperability has not been demonstrated in

403 MHz GPS systems. There are three main reasons for this:
(a) The signal processing system that is swapped out for compatibility in RDF systems
(Figure III.13 (a)) is a relatively simple and low-cost device. The meteorological processor
used in GPS systems (Figure III.13 (b)) is significantly more expensive since it includes the
system’s receivers as well as the sonde decoder;
(b) The UHF antennas and low-noise amplifiers included with 403 MHz GPS systems are not
standardized and need to be carefully integrated with the corresponding receivers in the
meteorological processor;
(c) Algorithms required for the meteorological operating system are more extensive than
merely calibration and solar correction. Since most sonde manufacturers use proprietary
GPS wind-finding schemes, this code would also have to be integrated.
For further details on present systems, see Part I, Chapters 12 and 13, of the Guide to
3.3.2.8.4 Optical theodolites
Optical theodolites, derivatives of the surveyor’s instrument, were one of the first upper-air
devices developed. They use a telescope-type apparatus which the observer uses to track a
balloon. For a specific time interval selected—usually one minute—the elevation and azimuth
angles are recorded according to the height, which has been estimated. Section 13.2.1,
Chapter 13, Part I, of the Guide to Meteorological Instruments and Methods of Observation
(WMO‑No. 8) gives a more complete description of this technique.
3.3.2.8.5 Radio direction-finding (RDF)
One of the most widely used methods of deriving wind information is through the use of RDF.
The basic design of an RDF system consists of a parabolic disk antenna, a radio receiver and either
a strip chart recorder or a direct link to a computer. To obtain wind information, the elevation
and azimuth angles and slant ranges of the antenna are sampled with respect to a given time
increment, usually at one-minute intervals. The antenna range for receiving the radiosonde
signal depends on the power and antenna gain.
RDF offers Members the ability to track accurately the radiosondes to an uncertainty of ± 0.5°
for the elevation and azimuth angles and ± 20 m for slant ranges. Wind calculations are derived
from spherical geometry techniques which make them easily accessible to computer reduction
algorithms.
The antennas may measure from 2 to 3 metres up to 5 to 6 metres in diameter. They usually
require shelter from the elements, and older versions require a fair degree of maintenance due
to the myriad moving parts. The antenna’s ability to collect accurate angular/slant-range data
may be affected by obstacles such as buildings and trees lying between the antenna and the
radiosonde.
3.3.2.8.6 Wind-finding radar
The derivation of winds is achievable through translation with the use of a tracking station.
Wind-finding radars, as the name implies, can derive wind information without having to use a
radiosonde to calculate heights. Although similar to radio theodolites in many ways, they acquire
data somewhat differently. Instead of homing in on a radio signal like a radio theodolite, the
radar emits pulses that are reflected from a target suspended below the balloon. These reflected
pulses measure the distance between the station and balloon which, when combined with the
elevation and azimuth angles, produce highly accurate wind data. Section 13.2.4, Chapter 13,
Part I, of the Guide to Meteorological Instruments and Methods of Observation (WMO‑No. 8) provides
a more complete description of this technique.
3.3.2.8.7 NAVAID systems
The principle of wind finding with the use of NAVAIDs is simple. A balloon or parachute,
equipped with a NAVAID receiver, retransmits the NAVAID signals to a base station. They
are transmitted from a number of fixed stations through the sonde to the base station. The
difference in the signals’ time of arrival is used to determine the range difference between pairs
of stations. Since the path from the sonde to the base station is identical for each transmitter,
the measurement of range differences eliminates the common path from the sonde to the base
station. The base station can, therefore, be in motion without introducing error to the wind
computation. The technique is ideally suited for measuring winds from a moving ship with
a balloon sonde. See 13.2.5, Chapter 13, Part I, of the Guide to Meteorological Instruments and
Methods of Observation (WMO‑No. 8) for a more complete description of this technique.
LORAN-C is a navigation system used to provide accurate ship navigation. A number of chains
cover the Pacific, Atlantic and Gulf Coasts, and the Aleutians. Global coverage, however, is not
available. Only a few LORAN-C NAVAID systems have been built, but they provide excellent wind
accuracy if located in areas of good coverage.
3.3.2.8.8 Other upper-air systems
3.3.2.8.8.1 Safesonde system
The Safesonde system consists of a base station, a reference transmitter and three translator
stations located 3 to 5 km from the base station. Signals are retransmitted from a radiosonde at
403 MHz and retransmitted at 1 680 MHz from the translator stations to the base station. Phase
comparison of the received signals permits the computation of the radiosonde location in three
dimensions. Temperature, humidity and pressure data are transmitted to the base station. These
data are then used to compute altitude independently as a check on the measured altitude. The
computations for all parameters are performed by a small computer with no need for an operator
after the balloon has been released.
Within the network, balloon motion is measured to an uncertainty of a few centimetres per
second. At large distances from the network, errors increase markedly. Network dimensions
can be increased to provide precise data at extended ranges. Typical errors for the system are
0.5‑m s–1 for 10-s averages up to 5 km altitude. At higher altitudes, uncertainty depends on
system baseline dimensions and balloon distance from the network. Uncertainty better than
1‑m s–1 should be achieved for 1-m averages over all altitudes.
3.3.2.8.8.2 Aircraft dropwindsondes
Dropwindsondes operate in the same fashion as radiosondes, telemetering data on pressure,

temperature and humidity. A parachute is used rather than a balloon and the sondes must be
designed to take high shock loadings on release. Current dropwindsondes may pose a danger to
populated areas because of their rugged design.
Until the development of the navigation-aid sonde, a number of expensive and abortive attempts
were made to develop a dropsonde which could measure winds. With the NAVAID sonde, the
problem was resolved. Measurements are made of the difference in phase of the signals received
by the sonde from the several transmitting stations.
3.3.2.9 Observational requirements
3.3.2.9.1 Time and frequency of observations
According to 2.4.2, Volume I, Part III, of the Manual on the Global Observing System
(WMO‑No. 544), the standard times of upper-air synoptic observations shall be 0000,
0600, 1200 and 1800 UTC. The actual time of observation in relation to the standard time of
observation is outlined in 2.4.10, Volume I, Part III, of the Manual on the Global Observing System
(WMO‑No. 544). The number and time of observations are specified in 2.4.8, 2.4.9 and 2.4.11 of
the aforementioned reference.
3.3.2.9.2 Type of observation
The central headquarters will decide whether upper-air synoptic observations, low-level
observations, or a combination of the two should be made, meeting the requirements set out in
2.4.6, Part III, Volume I, of the Manual on the Global Observing System (WMO‑No. 544).
3.3.2.9.3 Observer functions
Observers should follow the pre-release, data evaluation and verification procedures in
accordance with the standard operating procedures and other instructions given to the station.
The pre-release procedures include checking the radiosonde and ground equipment to ensure
that they are working properly, inflation of the balloon and preparations for entering the
observational records.
Data evaluation procedures may consist of automatic, semi-automatic or manual computations.

Some pilot-balloon calculations are now semi-automatic or automatic.
Data validation procedures are limited to some extent in automatic upper-air systems. Data
validation procedures for semi-automatic systems are carried out partly by the computer and
partly by the observer.
Observers may be required to carry out periodic equipment checks separately from the actual
observation and to regulate or adjust the equipment in accordance with the standard procedures
for the equipment in use. (Certain types of equipment such as radiotheodolites and barometers
require comparison with standards to verify accuracy of the data). When the equipment is
inoperative or malfunctioning, observers are advised to make an entry to this effect in an
equipment log. The upper-air unit must have back-up procedures or back-up equipment when
the primary equipment is inoperative. Manual computation may be necessary when a computer
becomes inoperative.
3.3.3 Special management considerations
3.3.3.1 General
Upper-air observation is a complicated and costly activity which is undertaken to obtain data
for a three-dimensional analysis of the atmosphere. There is, therefore, a need for rigorous
working standards at each station which should be ensured by proper arrangements for station
management and operation.
A Member operating a network of upper-air stations should establish an appropriate

organizational unit within the National Meteorological Service having responsibility for all
aspects of network management, such as station operation, maintenance and supervision;
logistics; procurement and supply of equipment and other necessary material to ensure the
efficient and uninterrupted functioning of the stations.
The basic principles to be followed in organizing management unit activities for an upper-air
station network are the same as those for a similar unit responsible for a surface synoptic network
(see 3.1.3 and 3.2.1). Therefore, only those aspects which are applicable to upper-air stations are
discussed in the present section.
3.3.3.2 Procurement of instruments and equipment
For further information on instruments and equipment, please refer to Chapters 12 and 13,
Part I, and Chapter 10, Part II, of the Guide to Meteorological Instruments and Methods of Observation
(WMO‑No. 8). The WMO Secretariat can also provide additional advice.
Useful information on currently used radiosondes and systems can be found in the WMO
Catalogue of Radiosondes and Upper-air Wind Systems in Use by Members in 2002 and Compatibility
of Radiosonde Geopotential Measurements for Period from 1998 to 2001, Instruments and Observing
Methods Report No. 80 (WMO/TD-No. 1197).
3.3.3.3 Maintenance
The purpose of a maintenance programme is to keep equipment in good operating condition

and obtain desirable performance from the system. The maintenance programme should include
preventive maintenance, equipment calibration, periodic cleaning and lubrication, performance
testing, corrective and adaptive maintenance and equipment modification, as necessary. Section
12.9, Chapter 12, Part I, of the Guide to Meteorological Instruments and Methods of Observation
(WMO‑No. 8) discusses this topic.
Preventive maintenance is important and must be conducted extensively on all equipment. It is

more efficient and effective to keep equipment running than to correct a breakdown. Scheduling
preventive maintenance is an absolute necessity for the successful continued operation of an
upper-air system. Original equipment manufacturers, usually through testing and evaluation
programmes, have determined a preventive maintenance programme to be followed by users,
which must be followed carefully during the entire period of use of the equipment to ensure
correct functioning. Where local guidelines regarding the maintenance programme do not
conflict with the manufacturer’s standards, they may be followed. Where conflict in guidelines
exists, clarification should be sought from the manufacturer.
The purpose of periodic equipment inspections and/or calibration is to ensure the continuous
operation of the equipment with a minimum of outage time. The inspection should include a
detailed visual examination to detect any physical deterioration, corrective action as necessary,
checking of the mechanical functions of the equipment to ensure their operation within
specifications and applicable tolerances and checking of all electrical functions to ensure that the
inputs and outputs meet the manufacturer’s specifications.
Periodic performance testing provides information on what can be expected when the
equipment is under operating conditions. It is also an effective way to discover and correct
malfunctioning equipment prior to use. Regularly scheduled performance testing is advised
to maintain the equipment in a satisfactory operating mode. Simulated operations should
be performed to check the equipment and ensure that all facets of the operation are within
specifications and provide the required data.
An effective corrective maintenance programme involves ensuring the availability of adequate

supplies, spares, and trained electronic and other maintenance personnel.
Original equipment manufacturers usually prescribe procedures and techniques to be followed

in determining and correcting equipment malfunctioning. These techniques and procedures are
based on laboratory testing and experience gained from field operations and must be followed
first in attempting to correct equipment failures and maintain the quality standards of the
operation. There are times when a disruption in the operation of a facility can arise from a local
and unusual occurrence which may not have occurred elsewhere. Such malfunctions should be
documented for future reference and forwarded to Members having similar equipment.
Trouble-shooting procedures are closely integrated with corrective maintenance and can be
viewed as a combined technique for correcting any malfunctions.
The design of equipment may feature one or more components whose mean time between
failures is below what is expected. These components must be given special attention in a
maintenance programme and if they deteriorate rapidly, the original equipment manufacturer
must be informed for possible redesign. Care must be exercised in making locally initiated
equipment modifications to ensure that the change is within the manufacturer’s specifications
and that no change occurs in the uncertainty and temporal resolution of the data.
3.3.3.4 Budgetary requirements
The purpose of determining budgetary needs is to ensure that resources are available for the
efficient and effective operation of the upper-air facility. Standards should be developed to
determine the number of people needed at a facility for the type of operation contemplated.
(See 3.3.1.6 and Tables III.3 and III.4.) Budgetary needs should be established on that basis.
Budgetary requirements for maintenance, supplies and other support activities should also be
prepared in a similar fashion. Resources must be made available for all personnel involved in the
station operations.
3.4 AIRCRAFT METEOROLOGICAL STATIONS
An aircraft meteorological station is defined in the Technical Regulations (WMO-No. 49), Volume I,

as “a meteorological station situated on an aircraft”, whereas a meteorological observing station
(meteorological station) is a “place where meteorological observations are made with the
approval of the WMO Member or Members concerned”. Meteorological observations from an
aircraft meteorological station (aircraft-based observations) are to be made for both aviation
and meteorological purposes by aircraft operating on national and international air routes. The
making of such observations is regulated by both WMO and the International Civil Aviation
Organization (ICAO) within the Technical Regulations (WMO-No. 49), Volumes I and II. More
specifically, detailed provisions for the making of such observations and requirements for making
them available in the WMO Information System (WIS) can be found in the Manual on the Global
Observing System (WMO-No. 544), Volume I, Part III, section 2.5.
Further guidance on aircraft-based observations can be found in the Guide to Aircraft-based

Observations (WMO-No. 1200). Members should use this guide as a source of information on
best practice in the implementation and operation of aircraft meteorological stations and the
provision of aircraft-based observations on the WIS within three categories:
1. WMO aircraft-based observations;

2. ICAO aircraft-based observations;
3. Other aircraft-based observations.
WMO aircraft-based observations are derived from aircraft-based observing systems operated
by WMO Members in collaboration with their national or other partner airlines, in which
requirements for aircraft-based observations are specified by WMO and its Members so as to
meet meteorological needs.
ICAO aircraft-based observations are derived from ICAO regulated aircraft observations, which
are made available to WMO and its Members under the provisions of ICAO as set out within the
Technical Regulations (WMO-No. 49), Volume II.
Other aircraft-based observations are those derived from aircraft-based observing systems
operated by other entities. In this case, while Members do not define specifications for the
operation of the observing system, they are urged to ensure that the observations are fit for
purpose.
As at 2016, the WMO Aircraft Meteorological Data Relay (AMDAR) observing system is the main
source of aircraft-based observations in the WIS. A detailed description of the AMDAR system
and recommendations on its implementation and operation by Members are contained in the
Guide to Aircraft-based Observations (WMO-No. 1200).
3.5 AERONAUTICAL METEOROLOGICAL STATIONS
3.5.1 General
Although the aim of commercial aviation is to be independent of meteorological conditions

and modern aviation has made considerable progress in all-weather operations, flight safety is
still weather related and atmospheric conditions continue to have a significant influence on the
economy and regularity of commercial aviation. Moreover, the introduction of lower aircraft-
operating minima and the increasing scale of operations have enhanced the requirement for
reliable and complete aerodrome information. It is the task of aeronautical meteorological

stations established at aerodromes and other points of significance to air navigation to provide
this information. The observations and reports made by aeronautical meteorological stations
are disseminated locally and to other aerodromes in accordance with regional air navigation
agreements. The procedures for observation and reporting services are developed and
promulgated jointly by WMO and the International Civil Aviation Organization (ICAO) on the
basis of operational requirements stated by the latter. The responsibility for providing the means
to satisfy these requirements rests with WMO (see the Guide to Practices for Meteorological Offices
Serving Aviation (WMO‑No. 732)).
The basic reference document to be used in making meteorological observations and reports
at aeronautical meteorological stations is the Technical Regulations (WMO‑No. 49), Volume II –
Meteorological Service for International Air Navigation, Part I, [C.3.1.],1 section 4.
The day-to-day activities for the provision of meteorological information for aeronautical
purposes call for close cooperation between the meteorological staff on the one hand and the
users, such as air traffic services and airport management units, airline flight planning centres
and aircrew, on the other. In particular, the type and accuracy of the data provided, the form and
speed of their transmission to the users, the methods and duration of their documentation and
the cost-effectiveness of the system should be regularly reviewed.
3.5.2 Instrumentation
The types of instruments used at an aeronautical meteorological station are, in general, the same
as those of a synoptic station. Certain instruments such as ceilometers and transmissometers are,
however, commonly used at aeronautical meteorological stations.
To be able to meet the demands for specific information for the approach and take-off areas,
the touchdown zone or portions of the runways, particularly at aerodromes with all-weather
operations, it is necessary to install multiple instruments. In these cases, it must be decided
which of the measurements is to be used on a routine basis for reports disseminated beyond the
aerodrome or in appropriate broadcasts for aviation.
Where only one instrument is used to measure a variable essential for take-off or landing, such as
surface wind, cloud base and atmospheric pressure, a back-up instrument should be available in
case of failure.
Instruments requiring electric power should be connected to an emergency power supply

available at the aerodrome. In view of the importance of individual weather variables for the
safety of take-off and landing operations and of the technical specifications of the instruments
used, it must be determined whether uninterrupted availability of electric power is required or
whether switching-over periods are acceptable and how they are to be limited.
At some aerodromes the installation of anemometers at distant sites or of other remote-sensing

devices suitable for the measurement of vertical wind shear or of gusts at the surface may be
advisable.
3.5.3 Location of meteorological stations and instruments
Special care is necessary in selecting sites for making observations or for the installation
of instruments to ensure that the values are representative of the conditions at or near the
aerodrome. It is particularly important that, while the siting and exposure of the instruments
meet operational requirements, the instruments or their operation do not pose hazards to air
navigation, and that the jet-blast or movement of aircraft on the aerodrome, such as taxiing,
take-off, runs, landing and parking, and the various aerodrome installations, do not unduly
1
See Guide to Meteorological Instruments and Methods of Observation (WMO-No. 8), Part I, Chapter 1, 1.3.5, for the main
purpose of inspections.
influence the measured values. Relevant instructions are contained in the Guide to Meteorological
Instruments and Methods of Observation (WMO‑No. 8), Part II, Chapter 2, and in the Manual of
Aeronautical Meteorological Practice (ICAO Doc 8896), Appendix C.
An important distinction should be made between observations at aeronautical meteorological

stations and synoptic observations. The latter aim at determining at one location a value of
a meteorological variable representative of a large area. Meteorological observations for
aeronautical purposes are often made at several locations and need to be representative
of more limited areas and times. As conditions vary from aerodrome to aerodrome and the
exact locations for which the data are needed are not always available for the installation
of instruments, owing to obstruction limits, universally applicable guidance cannot be
given. Detailed information on the representativeness of measurements and observations at
aeronautical meteorological stations can be found in the Guide to Meteorological Instruments and
Methods of Observation (WMO‑No. 8), Part II, Chapter 2, in the following sections: 2.2 Surface
wind; 2.3 Visibility; 2.4 Runway visual range; 2.5 Present weather; 2.6 Cloud; 2.7 Air temperature;
2.8 Dew point; and 2.9 Atmospheric pressure.
3.5.4 Observing and reporting programme
There are several types of observations as follows:
(a) Routine observations
At aerodromes, routine observations are normally made at hourly or half-hourly intervals

depending on regional air navigation agreements. At other aeronautical meteorological stations,
observations are made as required by air traffic services units and aircraft operations.
(b) Special and other non-routine observations
At aerodromes, routine observations are supplemented by special observations which are

made during the intervals between routine observations. Special observations refer to specified
conditions involving the deterioration or improvement of one or more meteorological variables.
Other non-routine observations, such as observations for take-off and landing, are made as
agreed between the meteorological authority and the appropriate air traffic services authorities.
(c) Continuous real-time observations
Information about certain meteorological parameters is required by air traffic services units
and operators almost continuously and in real time. These include information on surface wind
for take-off and landing and on cloud base or vertical visibility and runway visual range for all
weather operations.
As these requirements can generally not be met by a human observer, it is preferable to use, as
far as possible, integrated automatic systems for data acquisition, processing and dissemination/
display.
(d) Synoptic observations
The regulations for surface synoptic stations in 3.2 generally apply to aeronautical observations.
However, as aeronautical meteorological observations have high priority, they must be
completed first, in the event of competing interests. The observations needed for the preparation
of meteorological reports by aeronautical meteorological stations are specified in 2.6.6, Volume I,
Part III, of the Manual on the Global Observing System (WMO‑No. 544).
Some of the variables will require different procedures for reports disseminated within and
beyond the aerodrome.
Detailed instructions on observing and reporting surface wind, visibility, runway visual range,
present weather, clouds, air temperature, dew-point temperature and atmospheric pressure, and
on the inclusion of supplementary information are contained in [C.3.1.] 4.6, Volume II, of the
Technical Regulations (WMO‑No. 49).
Observing methods and commonly used instruments are discussed in Chapter 2, Part II, of the
3.5.5 Communications
The aim of communications with respect to aeronautical meteorological data and reports should
be to achieve the best possible response time of the system observer-forecaster-traffic controller-
pilot.
Plain language meteorological reports required for take-off and landing should be disseminated
by the quickest means available to the air traffic controller, the relevant airline operators and
the meteorological forecaster, if not co-located with the observer. The same applies to the
dissemination of meteorological reports which are to be included in the local automatic terminal
information service or the routine broadcast of meteorological information for aircraft in flight
broadcasts. An automatic dissemination/display system should be employed for measured data
required on a real-time basis.
Standard meteorological or aeronautical telecommunication systems, such as the GTS,

MOTNE, which stands for Meteorological Telecommunications Network Europe, or AFTN, the
Aeronautical Fixed Telecommunication Network, are generally sufficient for the dissemination of
coded meteorological reports.
3.5.6 Personnel and training
In addition to being fully trained in making meteorological observations, personnel at

aeronautical meteorological stations should be thoroughly familiar with the relevant
regulatory material, in particular the Technical Regulations (WMO‑No. 49, Volume II, [C.3.1.] 4)
and the Manual on Codes (WMO‑No. 306, Volume I.1, Part A). Information on the training of
meteorological technicians in the field of aeronautical meteorology is provided in the Guide to the
Implementation of Education and Training Standards in Meteorology and Hydrology (WMO‑No. 1083),
Volume I – Meteorology.
In view of the requirements of making observations at hourly and even half-hourly intervals,
a full understanding of observing, coding and reporting procedures is needed if reports are
to be prepared promptly. Additional observations may be needed at any time if the weather
deteriorates or improves with reference to established criteria for this purpose, or at the request
of the air traffic services units. Observers must be alert to such circumstances and prepared to
make the observations called for. The importance to being alert to aviation safety should be
reflected in appropriate duty rosters and the application of standards governing the maximum
acceptable number of uninterrupted working hours for the observers.
3.5.7 Quality standards
As aviation safety is involved, quality standards must be set very high and applied to essential
variables on a real-time basis. In order to initiate prompt corrective action, observers should
monitor all data whose acquisition, processing and display are carried out automatically. A
continuous record of the data should be provided to the users during the periods specified by
local agreements or requirements.
Instrument readings should be frequently checked, instruments recalibrated and, where

necessary, duplicated or provided with an emergency power supply. Contacts with pilots
after landing should be maintained in order to provide for a non-real-time feedback on the
representativeness of surface wind, visibility, runway visual range and cloud observations.
Further details can be found in the Guide to Practices for Meteorological Offices Serving Aviation
(WMO‑No. 732).
3.6 RESEARCH AND SPECIAL-PURPOSE VESSEL STATIONS
Many research and special-purpose vessels carry out a variety of activities during oceanic
expeditions but they are not always included in the VOS Scheme. Members with such vessels
should do their utmost to ensure that all such vessels make surface and upper-air meteorological
observations in accordance with the observing programme for sea stations (see 3.2.2.3). Upper-
air wind observations in particular are extremely important in the tropics and in data-sparse
areas.
Research and special-purpose vessels may also be equipped to make bathythermograph

observations during ocean crossings. The use of an expendable bathythermograph require
does not the ship to reduce speed or make course alterations. All arrangements for this type of
observation are made within the framework of the joint WMO/IOC Integrated Global Ocean
Services System. Procedures for the collection and exchange of BATHY and TESAC observations
are specified in the Guide to Operational Procedures for the Collection and Exchange of
Oceanographic Data.2 The preferred times for BATHY and TESAC observations are 0000,
0600, 1200 and 1800 UTC. However, observations made at any time are useful and should be
transmitted.
3.7 CLIMATOLOGICAL STATIONS
3.7.1 Organization
Each Member shall establish a network of climatological stations in the national territory.
A climatological station network should give a satisfactory representation of the climate
characteristics of all types of land in the territory of the Member concerned, for example, plains,
hills, mountainous regions, plateaux, coasts, inland areas and valleys.
Each Member should maintain an up-to-date directory of the climatological stations in its
territory similar to the synoptic station directory, as described in 3.2.1.2.6.
The Guide to Climatological Practices (WMO‑No. 100) and the Guide to Meteorological Instruments
and Methods of Observation (WMO‑No. 8) provide complementary information on the subject.
Chapter 2 of the former, covering climatological observations, is particularly relevant.
3.7.2 Climatological station networks
Spacing of the climatological stations should not exceed 100 km where feasible—the manned
station network should be supplemented when necessary by automatic stations—if possible,
maintaining this density in deserts and other sparsely populated areas. Observations from more
widely separated stations are also very valuable but stations of the network should not be further
apart than 500 km.
2
IOC Manuals and Guide No. 3, 1999, UNESCO, or the Manual on the Global Telecommunication System (WMO-No. 386),
Volume I, Part .
Climatological stations for measuring precipitation in the network should be more closely spaced
but the density depends on geographical features and economic considerations.
3.7.3 Station classification
According to the Appendix in Volume I of the Manual on the Global Observing System
(WMO‑No. 544), a climatological station network is composed of the following types of station:
(a) Reference climatological stations;

(b) Principal climatological stations;
(c) Ordinary climatological stations;
(d) Climatological stations for specific purposes.
Observing programmes of individual categories of climatological stations are described in 2.8,

Part III, of the Manual on the Global Observing System (WMO‑No. 544).
3.7.3.1 Reference climatological stations
Each Member should maintain at least one reference climatological station for different climatic
regions. A reference climatological station should be sited with an adequate and unchanged
exposure where the observations can be made in representative conditions. The station’s
surroundings should not be altered over time to such an extent as to affect the homogeneity of
the series of observations.
3.7.3.2 Principal climatological stations
Each Member should arrange for its principal climatological stations to be inspected at least
once a year and preferably twice a year, in summer and in winter. Special attention should be
paid to noting any possible changes in the siting of the station. To this effect, it is recommended
that every five years four pictures be taken from the thermometer screen, in the main compass
directions: north, east, south and west.
Each principal station should be located at a place and under an arrangement that will provide
for the continued operation of the station for at least 10 years, and for the exposure to remain
unchanged over a long period.
3.7.3.3 Ordinary climatological stations
The considerations for establishing ordinary climatological stations are similar to those for
principal climatological stations.
The operation of this type of station may be restricted to a much shorter period, but not less than
three years. Inspection should be carried out occasionally, preferably during the winter season,
to ensure a high standard of observation and the correct functioning of instruments.
3.7.3.4 Special-purpose stations
These stations are established by the Member for special observing programmes that are limited
by the number of variables involved and require appropriate instrumentation. The special-
purpose observing programme will determine its own frequency, spacing and timeliness on an
irregular basis.
3.7.3.4.1 Precipitation stations
The operation and observing programme of these special-purpose stations are confined to the
precipitation element only. The instrumentation consists of a standard rain gauge operated by
the Member or, in desert areas, of a mechanical or automatic storage rain gauge. These may be
supplemented by recording rain gauges. In the winter, a snow gauge and the measurement of
snow depth are required in many areas.
Each Member should arrange for its precipitation stations to be inspected at least every three
years, or more frequently if necessary, to ensure the maintenance of high observation standards
and the correct setting and operation of instruments. It is important that any changes in
the surroundings of the site be noted. Proper action should be taken to ensure the correct
functioning of the station.
3.7.4 Operation of stations
The observational requirements of a typical climatological station are described in 1.3, Chapter 1,
Each Member should arrange for observations at climatological stations to be made at fixed
hours, according to a standard time, UTC or other. These times should remain unchanged
throughout the year, regardless of the introduction of daylight-saving time by the authorities,
namely in summer and winter.
When two or more observations are made at a climatological station, they should be arranged at
times that reflect significant diurnal variations of climatic variables.
When changes are made at a station, including the replacement of a station or any changes
in surrounding observing sites, simultaneous observations should be carried out for an
overlapping period of at least one year to determine the effects of changed instruments or sites
on climatological data and to ensure the correct observational values.
When climatological observation times have been changed, simultaneous observations should
also be carried out on a skeleton network of representative stations for a period covering the
main climatic seasons of the area at former and new observation times.
3.7.5 Quality standards
Reference should be made to the following publications:
(a) Guide to Climatological Practices (WMO‑No. 100), Chapter II, 2.6, Station and network
operations, and Chapter III, 3.4, Quality control, (draft third edition, 2007);
(b) Guidelines on Climate Metadata and Homogenization (WMO/TD-No. 1186), 2.5, Data

processing.
3.7.6 Archiving
Efforts must be made to safeguard and preserve climatological data for future use.
3.8 AGRICULTURAL METEOROLOGICAL STATIONS
3.8.1 Organization
Each Member should establish within its territory a network of agricultural meteorological
stations. Such a network should give a true representation of the existing agricultural areas,
which are defined by biological and meteorological factors, to provide the required data.
Therefore, each station network should be sufficiently dense to delineate weather parameters
on the scale and magnitude required for agricultural meteorological planning, operation and
research, taking into account the agricultural potential and features of the country in question.
Each Member should maintain an up-to-date directory of agricultural meteorological networks,

as required for synoptic stations. In addition, the following information about each station
should be provided:
(a) Natural biomass, main agrosystem and crops of the area;

(b) Types of soil, physical constants and types of soil profiles.
The Guide to Agricultural Meteorological Practices (WMO‑No. 134) describes the basic requirements
of agricultural meteorology in detail. Chapter 2, Agricultural elements and their observation, is
particularly relevant.
The Guide to Meteorological Instruments and Methods of Observation (WMO‑No. 8) also provides
essential information on the subject. Reference is made mainly to sections dealing with
measurements of meteorological variables in relation to agricultural meteorological stations:
Chapters 1, 2, 4, 5, 7, 10 and 11, Part I; and Chapter 1, Part II.
3.8.2 Station classification
According to the Appendix in Volume I of the Manual on the Global Observing System
(WMO‑No. 544), agricultural meteorological stations are classified as follows:
(a) Principal agricultural meteorological station;

(b) Ordinary agricultural meteorological station;
(c) Auxiliary agricultural meteorological station;
(d) Agricultural meteorological station for specific purposes.
Observing programmes of agricultural meteorological stations are described in 2.11.5 and 2.11.6,
Volume I, Part III, of the Manual on the Global Observing System (WMO‑No. 544).
An example of the layout of instruments at an agricultural meteorological station is given in

Figure III.14.
According to the Manual on the Global Observing System (WMO‑No. 544), Volume I, Part III, 3.1.8,
agricultural meteorological stations should be inspected at least once a year.
3.8.3 Operation of stations
The information given under 3.7.4 applies in a general sense to agrometeorological stations.
3.9 SPECIAL STATIONS
3.9.1 General tasks and purposes of special stations
A wide range of special stations is used to measure or record meteorological variables of special
interest. These stations provide specialized information which is important to the overall purpose
of the World Weather Watch, although their main purpose is to meet national requirements
relating to toposcale and mesoscale of meteorological phenomena.
Some types of special stations, such as radars and reconnaissance aircraft, can cover large areas in
cost-effective ways. This may create some redundancy, which is unavoidable if routinely available
data are to be verified or reinforced efforts are made to ensure protection against system failure.
3.9.2 Types of special stations
3.9.2.1 Weather radar stations
3.9.2.1.1 General
Weather radar stations are in many cases co-located with surface or upper-air stations of
the basic synoptic network. Such stations should be established and equipped to carry out
radar observations in order to obtain information about areas of precipitation and associated
phenomena, and the vertical structure of cloud systems. The information obtained from radar
stations is used for operational purposes in synoptic meteorology—forecasting and warning of
dangerous weather phenomena such as tropical cyclones, the generation of numerical analyses
and guidance, aeronautical meteorology and hydrology, and research.
10 m
Rain-gauge
Screen
Rain recorder
Soil temperature
Grass minimum temperature

Evaporation pan
Sunshine recorder
Anemometer
Figure III.14. Instrument layout at an agricultural meteorological station in the northern

hemisphere
WMO Technical Note No. 181, Use of Radar in Meteorology (WMO‑No. 625), contains useful
guidance on the types of radar available, their possible usage, methods of operation and the
practical aspects of siting and maintenance.
Chapter 9, Part II, of the Guide to Meteorological Instruments and Methods of Observation
(WMO‑No. 8) provides further information.
3.9.2.1.2 Site selection
Several principles to be considered when selecting a site for a radar station are as follows:
(a) The location should be free of natural or man-made obstructions interfering with the
radar beam. Local construction plans should be examined to identify future potential
interference. Fixed targets should be as few as possible or at least not higher than 0.5°
above the level of the radar aerial;
(b) Many national regulations require a survey to ensure that people living in the area
surrounding the station site are not influenced by the microwave energy emitted;
(c) A licence for operating the radar at the planned site must be obtained from the radio-
telecommunication authorities concerned in order to avoid interference with any other
installation.
See 9.7.1, Chapter 9, Part II, of the Guide to Meteorological Instruments and Methods of Observation
(WMO‑No. 8) for more details.
Radar observations have been found most useful for the following tasks:
(a) Severe weather detection, tracking and warning;

(b) Surveillance of synoptic and mesoscale weather systems;
(c) Estimation of precipitation amounts;
(d) Wind shear detection.
Further information can be found in 9.1.3, Chapter 9, Part II, of the Guide to Meteorological
3.9.2.1.4 Organization
A radar meteorological observation is a manual or automated “evaluation” of the radar

echoes received from meteorological targets, coded as a message and transmitted to various
meteorological centres and other users at regular intervals.
The distance between two stations in an operational weather radar network should be a function
of the effective radar range. In the case of a radar network intended primarily for synoptic
applications, radars in mid-latitudes should be located at a distance of approximately 150 to
200 km from each another. The distance may be increased in latitudes closer to the Equator, if
the radar echoes of interest frequently reach high altitudes. Narrow-beam radars yield the best
accuracy for precipitation measurements.
Radar networks have a routine observing schedule. Each radar station may, however, increase its
observation times or take continuous observations according to the current weather situation.
A list of measurements and products can be found in 9.1.4, Chapter 9, Part II, of the Guide to
There should be at least one principal weather radar station or a national weather radar centre
which is responsible for receiving radar observational data from local stations and synthesizing
this data into a large-scale echo pattern for the entire network. The national weather radar centre
should also be responsible for regular inspection and quality control of network data.
An up-to-date directory of weather radar stations should be maintained by each Member within
its territory, giving the following information for each station:
(a) Name, geographical coordinates and elevation;
(b) Type of radar and some characteristics of the equipment used, such as wave length or
maximum transmitting power;
(c) Routine observing schedule.
A minimum radar network should consist of at least two radars together covering most of
the service area. Where necessary, individual radar can operate in conjunction with others in
neighbouring countries to form a network. Ground-level precipitation estimates from typical
radar systems are made for areas of typically 2 km2, successively for 5–10 minute periods.
A growing number of meteorological offices, governmental agencies, commercial users and

water authorities receive either the composite images or graphics produced at the weather radar
centre or single radar images directly from the radar sites.
3.9.2.1.6 Communications
Regular radar data are coded in code forms FM 20-VIII RADOB, found in the Manual on Codes
(WMO‑No. 306, Part A, Volume I.1) or FM 94 BUFR, in the Manual on Codes (WMO‑No. 306,
Parts B and C, Volume I.2) and disseminated in a timely fashion through the national or regional
telecommunication network. The type of communications equipment needed for disseminating
data depends on the temporal resolution of the data, the processing level and the quality of
communications available (telephone lines and the like).
3.9.2.1.7 Personnel
Weather radar personnel requirements with regard to category and number depend on the type
of equipment used, the level of automation and the number of observations required.
Maintenance and technical personnel responsible for the weather radar station or the entire
network must have specialized training in the maintenance and operation of equipment used
and a basic understanding of electronics and radar techniques.
A station supervisor is needed to carry out periodic checks of the calibration and the
interpretation methods used in manual or semi-automatic observations.
The relationship between surface rainfall and radar echo strength is unfortunately not fixed
or geographically universal. In addition, there are often significant echoes caused by ground
clutter and anomalous propagation that are not due to rainfall. The difficulty of correcting the
calculation of surface rainfall estimates objectively in real time is one factor that should be taken
into account when designing an interactive display system and interpreting radar images.
In addition to the quality control of radar observations, a combined digital satellite and radar
interactive system may enable its operators to use geostationary satellite data to extend the
surface rainfall analyses beyond the radar coverage area. This involves subjective judgement
and the use of algorithms that relate surface rainfall to cloud brightness and temperature.
Alternatively, real-time calibration of radar echoes with rainfall data from rain-gauges can also be
carried out when analysing rainfall data and estimating rainfall from radar echoes.
3.9.2.2 Radiation stations
3.9.2.2.1 General
It is recommended that Members should establish at least one principal radiation station in each
climatic zone of their territory and maintain a network of stations of sufficient density to permit
the study of radiation climatology. (See 2.12.3, Volume I, Part III, of the Manual on the Global
Observing System (WMO‑No. 544))
The terminology of radiation qualities and measuring instruments, and the classification
and calibration of pyranometers are given in Chapter 7, Part I, of the Guide to Meteorological
Every effort should be made at site radiation stations so as to allow adequate instrument
exposure and permit representative observations to be made. The location should have a free
horizon, undisturbed by obstructions. The station’s exposure and surroundings should not alter
over time to such an extent as to affect the homogeneity of the observation series.
3.9.2.2.3 Instrument selection
For details concerning radiation instruments and measurements, reference should be made
to Chapters 7 and 8, Part I, of Guide to Meteorological Instruments and Methods of Observation
(WMO‑No. 8).
The various observing programmes at principal and ordinary radiation stations are set
out in 2.12.3.5 and 2.12.3.6, Volume I, Part III, of the Manual on the Global Observing System
(WMO‑No. 544).
It is important that the data generated in a worldwide radiation measurement network be

homogeneous with respect not only to calibration, but to observation times as well.
The special requirements of all potential users should be considered in planning a radiation
station network. The answers to the following questions are therefore needed:
(a) How many stations are necessary to satisfy the requirements with respect to the spatial
resolution of the different kinds of meteorological radiation quantities?
(b) Which observing programme for each of the radiation quantities has to be set up for real-
time and non-real-time purposes?
The principal radiation station should be closely connected to or co-located with the National
Radiation Centre which is responsible for the calibration and checking of all radiometric
instruments used within the whole national radiation station network.
Detailed specifications for a National Radiation Centre are provided in Annex 7 C, Chapter 7,
The comprehensive fulfilment of the tasks of a National Radiation Centre is a prerequisite of an

adequately equipped and smoothly operating national radiation network.
The radiation measurements as specified in Chapter 7, Part I, of the Guide to Meteorological

Instruments and Methods of Observation (WMO‑No. 8) can be organized within the framework
of meteorological stations. An up-to-date directory of radiation stations should be maintained
by each Member in its territory, giving the information for each station, as well as the
information requested in 2.12.3.3, Part III, Volume I, of the Manual on the Global Observing System
(WMO‑No. 544).
The National Radiation Centre should be responsible for preparing and keeping up to date all
necessary technical information for the operation and maintenance of the national network of
radiation stations.
The results of all radiation measurements made at a radiation station should be collected and/
or transmitted to a designated centre according to arrangements which guarantee a timely
utilization of the data for operational and scientific research purposes. The collection of data may
be accomplished either through telecommunication channels or by mail.
Some of the regularly measured radiation data, such as diffuse solar radiation, sky radiation
and sunshine duration, are coded and then disseminated in a timely fashion to the National
Meteorological Centre for further data processing.
Data on sunshine duration is coded in tenths of hours and included once a day in section 3 of
code form FM 12-XIV SYNOP (see the Manual on Codes (WMO‑No. 306), Volume I.1, Part A, for
regional exchange of meteorological data). In contrast, data on global and sky radiation data
may be coded and distributed nationally in connection with other synoptic observations using
the same data collection procedures and telecommunication channels.
3.9.2.2.8 Personnel
National Radiation Centre staff should ensure continuity and include at least one qualified
scientist with experience in radiation. The staff is also responsible for giving instructions to the
staff of any other stations in the network and for maintaining close liaisons with them.
Observers at radiation stations must be trained in order to ensure accurate and reliable radiation
data. They may require special training in the use of sophisticated equipment and instruments.
All radiation data intended for permanent storage or non-real-time investigations should
be subjected to either manual or automatic quality control. Errors and obscurities should be
resolved as soon as possible.
3.9.2.3 Atmospherics detection stations
3.9.2.3.1 General
Atmospherics or sferics may be defined as electromagnetic waves resulting from electric

discharges, for example, from lightning in the atmosphere.
The main purposes of this type of special station are to deduce the presence of atmospherics
from observations and classify their activities. Technical progress now offers the prospect of
locating distant thunderstorms by means of automated atmospherics detection systems.
Certain characteristics of atmospherics, when determined by special techniques, can usefully be

employed in combination with other observations, especially for mesometeorological purposes,
to analyse severe storms in order to determine their characteristics, forecast their severity and
improve early warning to the community. In particular, lightning detection networks have
proven useful in augmenting radar detection of storms, especially in mountainous terrain where
radar interference may occur.
For details on how to locate the sources of atmospherics, reference should be made to Chapter 7,
Part II, of the Guide to Meteorological Instruments and Methods of Observation (WMO‑No. 8).
For reasons of cost-effectiveness, lightning location systems are generally installed at the site
of manned or automatic synoptic stations or at the site of weather radar stations. If a lightning
or thunderstorm sensor is to be used for automatic detection and reporting of the presence or
absence, direction, range, intensity and movement of such phenomena within an operational
network of atmospherics detection stations, the distance between two stations should be not
more than 150 to 250 km.
The area covered by a system of at least three atmospherics detection stations can be extended
to several tens of kilometres in a local warning system and to 200 to 400 km in a regional warning
system.
Before making significant installation investments, it is necessary to study site facilities, especially
the availability of electrical power, telecommunication facilities and staff. Considerations for the
site selection of automatic weather stations largely apply in this case as well.
3.9.2.3.3 Capital equipment
The type of equipment to be used depends on the purpose for which the observations are
planned and the technology or technique to be used.
Real-time collection, transmission and processing are necessary to derive the full use of lightning
location data. The observing programme should take into account the requirements of different
users and should be employed in conjunction with other observing systems.
The observing programme depends on the following factors:
(a) type of equipment used at the observing site, for example:
(i) Direction finders (an optimum distance of 500 to 1 000 km between the stations is
required);
(ii) Time-of-arrival receivers (the number of stations for an effective service is five);
(iii) Local lightning detectors (effective lightning counters are useful only within a radius
of 20 to 50 km);
(b) type of measuring system employed, for example:
(i) Manual systems (for instance, only for sampling periods H-10 to H; continuous
observation is not practicable);
(ii) Semi-automatic systems (computational facilities are needed);
(iii) Automatic systems (a sampling process with time provided for communications and
data processing).
Strong organizational principles must be followed in an operational network of atmospherics

detection stations, and a control station is necessary. Automatic systems are preferred where the
prerequisites for a fully automatic network are given.
Lightning location systems are used not only for operational purposes, often along with weather
radar observations, but also for non-real-time or research activities.
In general, the events must be plotted manually or automatically onto charts during one-day or
one-month periods, depending on requirements. The events should be recorded in a cumulative
manner only, for example, for making decisions on planning electric power lines.
Relevant information can be found in Chapter 7, Part II, of the Guide to Meteorological Instruments
3.9.2.3.8 Personnel
At least one observer per station is needed for the operation of a network of manned atmospheric
detection stations. Observers must be qualified to carry out this job effectively, including
equipment calibration and testing and careful reading of the different measuring scales. In some
countries, lightning information may be purchased from companies that maintain their own
networks.
In an automatic system, the task of supervising the error-free operation of a lightning location
sensor may be carried out by an ordinary observer with special training.
In modern equipment, a built-in microprocessor controls the data collection, derives an estimate
of thunderstorm movement and intensity and formats the processed thunderstorm data for
transmission to the automatic weather station and/or to the meteorological office concerned. In
this case, an electronics expert should be available for routine maintenance and repair.
Relevant information can be found in Chapter 7, Part II, of the Guide to Meteorological Instruments
3.9.2.4 Meteorological reconnaissance aircraft stations
3.9.2.4.1 General
A meteorological reconnaissance aircraft station is defined as a station situated in a

meteorological reconnaissance aircraft. Observations by aircraft can provide a valuable addition
to the meteorological information acquired by more conventional methods. As a result of
the latest developments in the techniques and instruments for automatic meteorological
observations and reporting by aircraft using space-based telecommunication facilities, modern
equipment installed in wide-bodied commercial aircraft on long-distance routes can provide
valuable upper-air temperature, humidity and wind data. The information obtained in this
way, particularly from remote and inaccessible areas of the world where routine surface-based
observations are sparse or non-existent, is of great value.
Since commercial aircraft are bound to their flight routes and timetables, there is a continuing
need to organize routine or special aircraft weather reconnaissance flights, for example, in the
event of hurricanes. Such meteorological reconnaissance aircraft should be devoted exclusively
to the task of meteorological observing, and therefore adequately equipped with meteorological
instrumentation, conducting the required flight patterns without regard to any other
commitment.
The instructions in 2.12.6, Part III, Volume I, of the Manual on the Global Observing System
(WMO‑No. 544), should be followed.
The site selection for the air base of meteorological reconnaissance aircraft and the type of
reconnaissance flight plan vary according to the purpose of the mission and depend on other
conditions as well. In some cases, where modern instrumented aircraft with satellite data links
to the National Meteorological Centre or the relevant meteorological offices are available, the
aircraft range must be taken into account when planning the flight track as triangles or polygons
in order to cover as vast a synoptic area as possible.
If, for example, reconnaissance flights are planned for research activities in tropical cyclones
in order to determine the position of vortex centre, maximum wind and minimum sea-level
pressure, or isobaric height, a variety of meteorological variables within 150 km of the storm
centre are necessary to produce analysed fields and detailed storm tracks in real time.
Depending on the reconnaisance task, the aircraft should be equipped with remote-sensing
technology, a video-recording device and, if possible, meteorological instruments providing
pressure, temperature and humidity observations.
The observing programme may be determined in advance or varied on a flight-to-flight basis.

In general, a fixed programme in which the aircraft makes flights at the same time daily, along
identical tracks and with altitude variations at the same geographical points, is most suitable
for synoptic or area reconnaissance. The aircraft usually reports location, pressure, temperature,
winds and altitude, and some are specially fitted with meteorological radar.
The type of aircraft selected for meteorological reconnaissance flights is best determined by the
task to be performed.
Operationally, three types of flights are available to the National Meteorological Centre or the
meteorological office, differing according to the purpose for which the information is required,
and hence the type of observation to be obtained:
(a) Low-level flight, in which the aircraft simulates as closely as possible a series of normal
synoptic surface observations;
(b) Vertical flight, providing a vertical cross-section of the atmosphere at or near a fixed point;
(c) High-altitude flight, in which a horizontal cross-section of the observable parameters is

given at a chosen level.
In practice, a single flight may be devoted to one or a combination of these conditions. The flight
plan may consist solely of a vertical ascent over base, or of level flights at one or more altitudes,
with or without measurements taken as vertical soundings during the ascents and descents
between levels.
Adequate means of communication, related to the range of the meteorological reconnaissance

flight and the amount of data to be transmitted, are required.
If limited computing capacity prevents extensive data processing aboard the aircraft, the
unanalysed observations must be sampled at short intervals—a few minutes—and transmitted at
high speed to the National Meteorological Centre or the meteorological office concerned where
they can be processed with other available meteorological data.
3.9.2.4.8 Personnel
Personnel requirements depend on the type of aircraft, the quantity and nature of the special
instrumentation and the exact purpose of the meteorological reconnaissance aircraft station.
To derive full benefits from the flight, at least one crew member should be a meteorologist
who is specially trained in aircraft measurements and observations. Certain circumstances may
necessitate accepting a regular crew member for this task.
Ground personnel to support the meteorological reconnaissance flights should be highly

qualified to perform aircraft and instrument maintenance.
Accurate height and airspeed measurements are essential, and the necessary corrections to the
instruments should be readily available.
Special meteorological instrumentation is required and should be chosen and installed to

provide adequate accuracy for the purpose.
3.9.2.5 Meteorological rocket stations
3.9.2.5.1 General
Meteorological rocket sondes are used to obtain information on atmospheric variables from the
stratosphere and mesosphere, generally between 20 and 90 km above the Earth’s surface.
Data obtained by rocket sonde systems are mainly used for the calibration and verification of
vertical temperature profiles derived from satellite infra-red radiometers.
See Chapter 6, Part II, of the Guide to Meteorological Instruments and Methods of Observation
(WMO‑No. 8), for more information.
The main principles to be considered when selecting a site for a meteorological rocket station are
as follows:
(a) A survey should be conducted to ensure a high level of security for the people living in the
vicinity of the planned launching site;
(b) A licence for the launching site, which must not be located within air traffic zones, shall be
applied for and obtained from the authorities concerned, including those responsible for air
traffic control;
(c) The launching schedule must be checked and approved by the authorities.
Problems in connection with safety and high costs tend to limit the number of stations and
frequency of launching.
Two worldwide sounding cross-section networks located approximately along the meridians
60°E and 70°W have been formed with close international cooperation.
The variables observed or computed include temperature, wind direction and speed. By setting
dates and times for launching, it has been possible to prepare meridional cross-sections of the
upper atmosphere.
A central body, World Data Centre A, undertakes collection of the data and the various
exchanges of data between participating Members. These data are used to conduct systematic
studies on topics such as the general circulation, solar high-atmosphere relationships, the
correlation between geomagnetism and meteorological parameters, the composition of
standard atmospheres, verification of satellite data and stratospheric warming.
Where altitudes higher than 20 km are concerned, meteorological variables such as temperature,
wind and air density should be determined for mandatory and significant levels.
The launching programme should be based on international agreements. Many different types
of rockets and sensors are in use and various data reduction techniques are employed.
For each launching, a report known as FM 39-VI ROCOB is compiled and disseminated over the
Global Telecommunication System.
3.9.2.5.7 Personnel
Personnel requirements (categories and number) at a meteorological rocket station depend on

the equipment used, the level of automation and expertise and the number of rocket launches
per week.
One person should be responsible for all facets of station operation and should be one of the
most experienced members in this field. Qualified scientists and engineers are required for the
preparation, execution and interpretation of the rocket launch.
International intercomparisons have been conducted to ensure uniformity of the results obtained
by the existing systems.
Modifications to certain measuring systems and laboratory experiments should be carried out
after each intercomparison in order to make the various systems and the evaluation of corrections
more uniform.
3.9.2.6 Global Atmosphere Watch (GAW) stations
3.9.2.6.1 General
The WMO Global Atmosphere Watch is designed to meet the need for monitoring the chemical
composition and related characteristics of the atmosphere on global and regional scales. Such
information is required to improve understanding of the behaviour of the atmosphere and its
interactions with the oceans and the biosphere and to enable prediction of the future states
of the Earth system. The GAW integrates many monitoring and research activities involving
high-quality measurement of atmospheric composition. Some GAW components have been in
operation since the 1950s.
Atmospheric chemical observations should be carried out at GAW stations with the same
attention as is given to the measurements of other meteorological variables. National
meteorological and/or environmental protection services are encouraged to ensure that chemical
composition observations become an integral part of atmospheric observations in general.
The network of GAW stations comprises the following two main categories of stations:
(a) Global, or baseline stations are established to provide measurements needed to address
atmospheric environmental issues of global scale and importance, for example, climate
change and depletion of the ozone layer.
The required number of global stations should be such that a minimum of one per principal
climatic and ecological zone is achieved. Towards this goal, Members are being encouraged
to establish and/or cooperate in establishing some 30 stations of this category at selected
sites. At the time of writing there are 24 GAW global stations.
(b) Regional stations are established to provide measurements primarily to help assess regional
aspects of global atmospheric environmental issues and atmospheric problems in various
regions or countries, for example, acid rain, near-surface ozone, deterioration of ecosystems
and materials and air pollution in rural areas.
The number of regional stations should allow regional aspects of global environmental
issues and environmental problems of interest to the Region or country) concerned to be
adequately addressed. To this end, Members are being encouraged to establish at least 400
stations of this category.
For further information, see the Global Atmosphere Watch Measurement Guide (WMO/TD-No. 1073,
the Updated Guidelines for Atmospheric Trace Gas Data Management (WMO/TD-No. 1149), the
WMO/GAW Aerosol Measurement Procedure: Guidelines and Recommendations (WMO/TD-No. 1178),
the Manual for the GAW Precipitation Chemistry Programme: Guidelines, Data Quality Objectives and
Standard Operating Procedures (WMO/TD-No. 1251) and the Baseline Surface Radiation Network
(BSRN): Operations Manual (WMO/TD-No. 879). See http://www.wmo.int/pages/prog/arep/gaw/
gaw-reports.html for a complete list of GAW reports.
Global Atmosphere Watch stations should be established only at sites where direct pollution
effects can be avoided. Therefore, strict siting criteria have been established and listed in the
WMO Global Atmospheric Watch (GAW) Strategic Plan: 2008–2015, for the two main categories
of station. The following criteria should be met for regional stations:
(a) The station location should be regionally representative of the variables measured and free
of the influence of significant local pollution sources;
(b) Adequate power, air conditioning, communication and building facilities to sustain long-
term observations with greater than 90 per cent data capture (i.e. <10 per cent missing
data) should be available;
(c) Technical staff should be trained to operate the equipment;
(d) The agency in charge should be committed to long-term observations of at least one of the
GAW variables in the GAW focal areas;
(e) GAW observations should be of high quality and linked to the GAW Primary Standard;
(f) Data and associated metadata should be submitted to one of the GAW World Data Centres
no later than one year after the observations are made. Metadata changes, including
instrumentation, traceability and observation procedures, should be reported to the
relevant World Data Centre;
(g) If required, data should be submitted to a designated data distribution system in near-real
time;
(h) Standard meteorological in situ observations necessary for the accurate determination and
interpretation of GAW variables should be accurate and precise;
(i) The station’s characteristics and observational programme should be updated in the GAW
Station Information System on a regular basis;
(j) A station log-book, namely a record of observations made and activities that may affect
observations, should be maintained and used in the data validation process.
In addition to the above-mentioned requirements, GAW global stations should meet the
following criteria:
(k) Variables should be measured in at least three of the six GAW focal areas (see (d));
(l) Global stations should boast a strong scientific supporting programme with appropriate
data analysis and interpretation within the country and, if possible, the support of more
than one agency;
(m) Measurements of other atmospheric variables important to weather and climate, including
upper air radio sondes at the site or in the region, should be taken;
(n) A facility where intensive campaign research can augment long-term routine GAW
observations, and testing and development of new GAW methods can be undertaken
should be provided.
Siting requirements may also differ, depending on the measurement programmes and the
specialities of variables to be monitored.
The capital equipment needed for Global Atmosphere Watch stations depends on the purpose of
the station and differs mainly in terms of scientific objectives and technical feasibility with regard
to the monitoring programme concerned.
Further information on equipment requirements can be found in the Global Atmosphere Watch
Measurements Guide (WMO/TD-No. 1073) and the GAW publications mentioned in 3.9.2.6.1.
The observing programmes may differ according to the recommended priority to be given to
the measurement of atmospheric composition at global and regional stations as laid down in
2.12.8.4, 2.12.8.5 and 2.12.8.6, Part III, Volume I, of the Manual on the Global Observing System
(WMO‑No. 544) and in Chapter B.2, Volume I, of the Technical Regulations, (WMO‑No. 49). For
further information, see Chapter 17, Part I, of the Guide to Meteorological Instruments and Methods
of Observation (WMO‑No. 8).
3.9.2.6.5 Organization and operation
The basic responsibilities for monitoring the operation of the Global Atmospheric Watch shall rest
with participating Members.
All GAW stations are operated by Members in accordance with the Global Atmosphere Watch
Measurements Guide (WMO/TD‑No. 1073). Data collection and preparation for WMO data
publication are undertaken by the GAW Data Centres.
3.9.2.6.6 Central collection, publication and availability of data
Global Atmospheric Watch data are centralized and published as follows:
(a) The agency operating the station(s) in each participating country submits all data from the
station(s) on data forms or agreed formats to the following centres according to their areas
of specialization:
(i) Total ozone, profile ozone, solar ultraviolet radiation:

World Ozone and Ultraviolet Radiation Data Centre
c/o Experimental Studies Section
Environment Canada
4905 Dufferin Street
Toronto, Ontario
Canada M3H 5T4
http://www.woudc.org
(ii) Carbon dioxide, methane, nitrous oxide, chlorofluorocarbons,

hydrochlorofluorocarbons, hydrofluorocarbons, surface ozone:
World Data Centre for Greenhouse Gases
c/o Japan Meteorological Agency
1-3-4, Otemachi, Chiyoda-ku
Tokyo 100-8122, Japan

http://gaw.kishou.go.jp/wdcgg.html
(iii) Solar radiation, radiation balance, sunshine duration:

World Radiation Data Centre
Voeikov Main Geophysical Observatory
7, Karbyshev Street
194021 St Petersburg, Russian Federation
http://wrdc.mgo.rssi.ru/
(iv) Aerosol optical depth, aerosol light scattering and backscattering, aerosol chemistry,
aerosol light absorption, condensation nuclei concentration, aerosol size distributions:
World Data Centre for Aerosols
Joint Research Centre, European Commission
Ispra, Italy
http://www.gaw-wdca.org/
(v) Precipitation chemistry:

World Data Centre for Precipitation Chemistry
Atmospheric Sciences Research Centre,
State University of New York
Albany, New York, USA
http://www.wdcpc.org/
(vi) Atmospheric composition as measured from space with emphasis on ozone and
aerosols:
World Data Center for Remote Sensing of the Atmosphere
DLR-DFD-KA
Oberpfaffenhofen
D-82234 Wessling, Germany
http://wdc.dlr.de/
(b) The World Ozone and Ultraviolet Radiation Data Centre at Environment Canada processes
all ozone data (surface, total column, vertical distribution) and publishes the Ozone Data
for the World (ODW) annually. The Centre also issues an index which contains the list of
stations, a station catalogue with information on stations (names, organizations operating
the stations, types of observation, types of instrument used, observing programmes) and
a catalogue of available ozone data (stations, period, data type, ODW volume numbers in
which the data are contained);
(c) Every six months the WMO World Data Centre for Greenhouse Gases at the Japan
Meteorological Agency processes and publishes the data for which it is responsible. Every
year, in October/November the World Data Centre for Greenhouse Gases assists the WMO
Secretariat with the preparation of the WMO Greenhouse Gas Bulletin.
Details on GAW data availability, including conditions and procedures for ordering, can be
obtained directly from the Centres listed in (i) through (vi).
Each Global Atmosphere Watch station should retain its own complete original data record and
documentation of station operations. The operating agencies will make whatever use of these
data as suits their needs. However, accessibility to the entire data reservoir is essential, which
should be accomplished through centralized archiving. All data from GAW stations should be
sent every two months to the relevant GAW Data Centre (see 3.9.2.6.6).
3.9.2.6.8 GAW-ozone-observing stations
3.9.2.6.8.1 General
In view of the radiation properties of ozone as a minor atmospheric constituent, this gas is
a significant contributor to the atmosphere’s radiation energy. In addition to its radiation
properties, ozone reacts photochemically with many other trace species, some of which are
anthropogenic in origin. The meridional and vertical distributions of ozone in the atmosphere are
determined by a complex interaction of atmospheric dynamics and photochemistry.
Ozone-observing stations have been promoted by WMO since the mid-1950s and are now a part
of the Global Atmosphere Watch. These stations regularly measure total column ozone and the
vertical distribution of the ozone in the troposphere and the stratosphere. Surface ozone near the
ground has generally been measured at selected GAW background pollution stations.
The available observational data have documented the geographic and seasonal average
ozone distributions and have indicated the presence of variations on many time and space
scales. These variations are, in part, associated with meteorological processes and may also be
affected by anthropogenic and solar influences. An improved representation of the global ozone
distribution is essential to provide a definitive pattern of the space and time variations of ozone
over the globe and for periods up to a decade or longer. The GAW Ozone Network has played a
crucial role in the discovery of the Antarctic ozone hole and in the quantification of the effect of
chlorofluorocarbons and halons on the ozone layer all over the globe. Data from the total ozone
and ozonesonde stations, stored at the World Ozone and Ultraviolet Radiation Data Centre, have
played a central role in the quadrennial WMO/United Nations Environment Programme Scientific
Assessments of Ozone Depletion. These data will also be important in the quest for eventual
ozone recovery, which is expected to happen around 2050–2070, depending on the region.
The following three characteristics of atmospheric ozone should be routinely measured and
reported from ozone-observing stations:
(a) Surface ozone;

(b) Total ozone;
(c) Vertical profile of ozone.
Members who establish ozone-observing stations should follow the Global Atmosphere Watch
Measurements Guide (WMO/TD-No. 1073).
3.9.2.6.8.2 Network considerations
The site selection for an ozone-observing station depends mainly on the types of routine
atmospheric ozone observations intended to be made and available facilities. Observing sites
should be selected so as to minimize effects of pollution and cloudiness as much as possible.
Total ozone is determined from ground-based and satellite-borne instruments, and vertical
ozone profiles are derived from ground-based, balloon-sonde and satellite techniques.
The GAW ozone system reflects the concern for the quality control and cross-validation of all
components of the system. In addition, the observations must be part of a network sufficiently
representative in space and time to allow for the documentation of all geophysically significant
ozone variations. This requires guarantees of long-term network maintenance and observing
system stability. The derived data need to be made readily available to the user community in a
timely manner and in an appropriate form; this is achieved through the services offered by the
World Ozone and Ultraviolet Radiation Data Centre in Toronto under the auspices of WMO.
The existing network of surface-based observing stations for the total ozone field is very irregular,
with a high density of stations in Europe, North America, and parts of Asia, and low density
over the tropics and the oceans, and in the southern hemisphere in general. Results of field
examinations suggest that total ozone should be sampled with the spatial resolution of mid-scale
waves, or at intervals of 30° longitude or less, implying the need for about 100 well-distributed
total ozone stations throughout the world. Information about the spatial statistical structure of
the total ozone field may be derived for the optimal location of new stations by means of iterative
procedures based on optimization criteria and internal network redundancy. Through the efforts
of international and national organizations, substantial progress has been made in improving the
consistency of the Dobson ozone network and extending the spatial coverage. However, there is
a clear need for further progress in these directions.
Balloon-borne ozone soundings play an important role in furthering knowledge of the global
vertical ozone distribution. However, three quarters of this kind of sounding have been made in
the latitude belt between 35°N and 55°N.
The following requirements should be noted:
(a) Additional ozone-observing stations representative of “unpolluted” areas—preferably

connected with a meteorological observatory—should be established, especially in the
tropical region and in the southern hemisphere;
(b) The current ground-based ozone network should be expanded by the addition of about
10 suitably sited Dobson and/or Brewer stations;
(c) A subset of Dobson and Brewer stations should be maintained and calibrated in such a way
that they can assist in the validation of satellite measurements.
Adequate answers to ozone questions can be provided only if there is a continuous flow of
reliable total and vertical profile ozone data sufficient to form a coherent data set. This goal can
be achieved effectively only by coordinating the diverse existing and planned ozone-measuring
programmes by Members concerned. They are urged to cooperate closely with the World Ozone
and Ultraviolet Radiation Data Centre in Toronto by providing the data in a timely manner.
3.9.2.6.8.3 Capital equipment
The capital equipment needed for a surface-based ozone-observing station comprises at least
one of the following main components:
(a) Total ozone
Total ozone may be measured by one of the following types of equipment:
(i) Dobson ozone spectrophotometer;
(ii) Brewer ozone spectrophotometer;
(iii) Filter ozonometer (type M-83, M-124);
(iv) System of Analysis of Observations at Zenith (SAOZ) spectrophotometer;
(v) Other ultraviolet-visible spectrometers using the differential optical absorption

spectroscopy technique.
(b) Vertical ozone profile
Vertical ozone profile may be measured by the following techniques and equipment:
(i) Surface-based (Dobson/Brewer) stations where the vertical profile of ozone is

measured remotely from the ground by the Umkehr inversion method;
(ii) Balloon ozonesonde stations employing the electrochemical concentration cell

sondes;
(iii) Lidar, or optical radar systems, which are usually limited to operation at night and
when there are no appreciable clouds;
(iv) Microwave instruments;
(v) Ozonesonde and laser radar.
See Chapter 16, Part I, of the Guide to Meteorological Instruments and Methods of Observation
(WMO‑No. 8) for more details.
3.9.2.6.8.4 Observing programme
Total ozone should be observed daily, because the frequency of observations is as important
as the spatial density of observation. Data acquisition depends on the local weather conditions
encountered. Difficulties arising from cloudy sky and/or haze should be noted in order to reduce
their bias in the data acquisition.
A balloon-borne ozone measurement network has been built up within GAW. It features routine
soundings carried out according to an agreed launching schedule of at least one per week at
some 60 stations using one of five versions of electrochemical ozonesondes.
3.9.2.6.8.5 Organization
Findings of the WMO Global Ozone Research and Monitoring Project have shown the need for a
network of ozone soundings and total ozone stations as an integrated part of the GAW network.
Each Member with more than one station taking part in ozone measurements should select one
to become the national focal point or national ozone centre of the relevant country. This special
station or observatory should be responsible for regular inspection, quality control, sampling,
archiving and distributing ozone data. Common efforts should be directed towards instrument
calibration and data intercomparisons to be carried out within the Global Atmosphere Watch in
order to improve the ozone database. The data from calibrations and intercomparisons should be
routinely deposited at the WMO World Ozone and Ultraviolet Radiation Data Centre in Toronto.
Regional Associations may designate a Regional Ozone Centre that could take an active role in
implementing related activities, namely:
(a) To organize intercomparisons of instruments for measuring total ozone and/or the vertical
profile of ozone;
(b) To carry out instrument checks;
(c) To offer training courses for various ozone measurement methods;
(d) To assist in analysis and evaluation of ozone data in close cooperation with the WMO World
Ozone and Ultraviolet Radiation Data Centre.
The World Ozone and Ultraviolet Radiation Data Centre plays an important role in publishing
and archiving all available ozone data, relevant calibrations and documentation provided by
Members.
3.9.2.6.8.6 Operations
An ozone-observing station network should focus on the following points:
(a) Calibration of instruments used, documentation of major changes and the like;
(b) Cross-validation of the ozone-measurement system used in order to detect possible sources
of errors;
(c) Commitment to long-term continuous operation of ozone stations.
For many applications of ozone measurements, and particularly for the detection of long-term
trends, a long record is necessary. Thus, stations and systems with long histories should be given
priority in future programmes.
While the implementation of specialized techniques is encouraged, they should be carefully

assessed to determine their likely long-term operation in a network setting.
The operational validation of an ozone measurement system should consist of the following four
parts or steps:
(a) Complete calibration and a thorough understanding of the performance characteristics of

the measuring instruments, including their errors;
(b) Identification of error sources due to mathematical modelling and external inputs to the
evaluation algorithm;
(c) Thorough understanding of the evaluation algorithm and the propagation of the above-
mentioned errors by the algorithm;
(d) Intercomparison or cross-validation with independent observations.
Neglect of any of these four steps will lead to an incorrect assessment of the performance of the
measuring system and the validity of the derived data products.
For all measuring systems, the problem of long-term stability of the instruments must be
considered. Ozone measuring systems also have to take into account changes in atmospheric
composition such as aerosols that may affect the retrieval schemes. Lastly, in connection with
instruments operating at ultraviolet wavelengths, the stability of the primary source of their
radiation, the Sun, is a critical matter of concern. Solar ultraviolet variability as related to ozone
measurement should be assessed by means of continuous satellite monitoring.
3.9.2.6.8.7 Communications
Ozone-observing stations should maintain a complete original data record and documentation
of station operations. In most cases, the data will have to be scrutinized by ozone experts.
Special precautions are also necessary before the exchange of ozone data as agreed between the
relevant Members. In connection with the Antarctic and Arctic ozone decline, a need for data
transmission in near-real time has appeared. Data centres have been established to collect ozone,
ultraviolet and meteorological data in near-real time and make them available to the scientific
community. There is a need to standardize delivery routines, data formats and metadata
reporting. The use of the WMO Information System should be considered for the exchange of
ozone and other atmospheric composition data. Some stations already transmit their ozone data
to the GTS. The number of stations submitting data in this manner should be increased and
eventually involve all ozone stations in the GAW network. Satellite data exchange should take
place expeditiously.
To ensure an adequate inventory of ozone data, it is recommended that a summary regarding

data availability be included in the Ozone Data for the World and updated about once every
year. This should include information on satellite data, surface-based total ozone measurements
and ozonesondes. Graphs should show the availability of data from particular stations and
satellites.
The exchange of total ozone data and/or validated ozone profiles will normally take place by
electronic mail or ftp within two months of the measurement.
3.9.2.6.8.8 Personnel
The availability of specially trained observers and technicians is fundamentally important

in the implementation of the GAW ozone system. At least two or three persons should be
employed at an ozone-observing station. The station supervisor must be highly qualified, with
preference given to experts with a university-level education. Advanced training in ozone-
measuring techniques is indispensable for effective participation in the GAW, especially at ozone
observatories and laboratories.
3.9.2.6.8.9 Quality standards
The responsibility for observational data quality rests with the agency providing the data.
With regard to ozone-observing stations, strong emphasis should be placed on instrument
maintenance and calibration, strict adherence to proper observing techniques and careful
rechecking of all clerical steps.
Like all measuring instruments, the accuracy of ozonesondes is also limited by sources of error.
More detailed information on this point, in particular a discussion of comparison, calibration
and maintenance of diverse instruments, can be obtained in JOSIE-2000, Jülich Ozone Sonde
Intercomparison Experiment 2000 (WMO/TD-No. 1225).
3.9.2.6.8.10 Archiving
Ozone soundings must be scrutinized carefully by the agency in charge of the station before
results are submitted to the World Ozone and Ultraviolet Radiation Data Centre.
Regarding surface-based total ozone station networks, the raw data should be recorded on
standard forms which should be archived at the station with due care to include all relevant
information, or by the agency responsible for the station. Preservation of the raw data is
necessary if retrospective corrections are to be made. A directory of all archived raw data should
be published by the World Ozone and Ultraviolet Radiation Data Centre at regular intervals.
In addition, information required for reduction of the raw data, such as calibration tables,
standards, sensor corrections, atmospheric corrections, zenith charts and cloud corrections
should be archived at the national ozone centres.
3.9.2.7 Planetary boundary-layer stations
3.9.2.7.1 General
Some Members have operated planetary boundary-layer stations on a regular or experimental

basis. The following guidance based on their experience will be useful if the need for a fully
operational network is established.
Measurements of the planetary boundary-layer may be carried out on a regular basis at a

permanent site, at special observatories, for example, or on special occasions at particular sites by
mobile teams.
Wind-profiling and Doppler radars are proving to be extremely valuable in providing high-
resolution data in space and time for these measurements. Wind profilers are especially useful
in making observations at times between balloon-borne soundings, and have great potential
as a part of integrated networks. Doppler radars are used extensively as part of national
and, increasingly, regional networks, mainly for short-range forecasting of severe weather
phenomena. The Doppler radar capability of making wind measurements and estimates of
rainfall amounts is particularly useful.
Current ground-based systems that can be used for these purposes are described in more
detail in Chapter 5, Part II, of the Guide to Meteorological Instruments and Methods of Observation
(WMO‑No. 8).
The capital equipment varies depending on the measuring techniques used.
3.9.2.7.4 Operational observing programmes
Over the past few decades there have been a number of boundary-layer observing programmes.
In some cases these have been organized specifically for boundary-layer studies, while in other
cases the boundary-layer studies have formed a component of larger experiments, for example,
those established within the Global Atmospheric Research Programme. Each of these observing
programmes has increased the understanding of boundary-layer processes and provided data
sets invaluable to research in this field.
Most of the observing methods and techniques are described in Chapter 5, Part II, of the Guide
to Meteorological Instruments and Methods of Observation (WMO‑No. 8). Further background
information can be found in WMO Technical Note No. 165, The Planetary Boundary Layer
(WMO‑No. 530).
3.9.2.7.5 Personnel
The personnel engaged in supervising planetary boundary-layer stations should have a

university or equivalent degree in electronics or mechanical engineering combined with a good
knowledge of meteorology.
Procedures should be developed to meet a minimum standard level in quality control of data
derived from the sophisticated instruments or equipment used for special investigations—
sodars, lidars and wind profilers, for example. Owing to the very high resolution in time and
space for data acquisition at planetary boundary-layer stations, accuracy requirements can be
more stringent than in most other cases. For this reason, quality control becomes a difficult task
for variables with very short time constants.
3.9.2.8 Tide-gauge stations
3.9.2.8.1 General
Tide-gauge stations should be established along coasts subject to storm surges. They provide
sea-level measurements which must be filtered to remove high-frequency fluctuations such as
wind waves, in order to provide time-series data from which tides and tidal predictions can be
determined.
These stations provide the basic tidal datums for coastal and marine boundaries and for chart
datums as required by the tsunami, seiche, and storm-surge warning services. Global sea-level
measurements are necessary to monitor possible increases due to global warming. Coastal sea-
level measurements are vital for hydrographic surveys and give indications of ocean circulation
patterns and climate change. Likewise, the archived data of sea-level heights may be very
important for making decisions on vessel navigation, coastal processes and tectonic studies, and
for numerous other engineering and scientific purposes and investigations.
An international network of sea-level measuring stations including tide-gauge stations forms the
basis of the Global Sea Level Observing System (GLOSS), coordinated by the Intergovernmental
Oceanographic Commission of UNESCO. For further information, reference should be made
to the Manual on Sea-Level Measurement and Interpretation, IOC Manuals and Guides No. 14,
UNESCO.
Location of tide-gauge stations should be selected in such a way that the measurements
returned are representative of the open ocean conditions. In particular, influences on the
measurements from overflow, salinity, hydraulics, density, stratification, stability, and wave and
storm resistance effects should be minimized as much as possible, or, taking those influences
into account, it should be possible to effectively apply the necessary corrections to make the
measurements as representative as possible of the open ocean conditions.
Some locations may be selected to provide bay, estuary, or river information for marine boundary
datum determinations or similar studies. A selected set of tide-gauge benchmarks must be
accurately connected to a global geodetic reference system.
The following points should be considered when selecting and setting up a tide-gauge station:
(a) Stable support structure such as a pier, or bulkhead, for the installation of water-level
measurement sensors. The support structure must be above the highest expected water
level, and water depths must be below the lowest expected water level;
(b) Space for a small shelter (typically 1.5 m x 1.5 m) to house the instrumentation (or 2 m x 2 m
clear wall space to mount the equipment in an existing building);
(c) If satellite data transmission of an automatic tide-gauge station is conducted, the antenna
must have a clear line of sight to the satellite that serves as data collection platform;
(d) Locations near geodetic first- or second-order vertical control networks, if existent, are
highly desirable;
(e) Utility services are highly advisable but not critical. The site should have AC power available
nearby; however, many measuring systems can operate solely on solar panel power if
necessary. Telephone lines are desirable at tide-gauge sites to allow direct communication
with the instrumentation.
Each station consists of a stable structure from which the measurements can be made, of water-
level measurement equipment, and of a suite of fixed physical objects, or benchmarks, used to
reference the vertical datums.
If continuous data collection is of vital importance, the primary data collection platform (DCP)
(see 3.9.2.8.7) should have a backup DCP with another water-level measurement sensor. In
these cases, further auxiliary data, such as wind speed and direction, atmospheric pressure, air
temperature, relative humidity and water conductivity may also be accommodated by the DCPs.
A telephone line should be available to allow access to pre-quality-controlled data by selected

real-time users.
Visual readings by human observers should be made at the following times, listed in descending
order of preference:
(a) Hourly, particularly in coastal storm situations;

(b) Times of high and low water;
(c) Main synoptic hours of 0000, 0600, 1200 and 1800 UTC.
Where the installation of automatic sea-level measurement equipment combined with the
data collection platform is feasible, the system may be programmed to average a series of
measurements. Data should be stored temporarily at the station site in the DCP memory
and periodically transmitted via satellite or land-line to a central collection station for further
processing and long-term storage. For further information, reference should be made to the
Manual on Sea-Level Measurement and Interpretation, IOC Manuals and Guides No. 14, UNESCO.
Owing to the vital impact of the deviation of tidal heights which might be generated by tsunamis
or storm surges on coastal community activities, real-time information on water-level deviations
is sorely needed. Although many tide-gauge stations are still equipped with simple water-
level gauges where visual readings have to be made by human observers, there are already
fully automatic water-level observational networks in operation. Where feasible and necessary,
preference should be given to creating a network that utilizes automatic data-acquisition and
recording equipment to measure water levels along the coastline.
To protect life and property from flood situations resulting from storm surges, the meteorological
hydrological warning system should be closely linked with public alert and coastal defence
systems. Where the warning time required exceeds the capability of meteorological forecasting,
an alert system composed of several phases of increased alert should be introduced, resulting in
a higher frequency of observations at manned tide-gauge stations.
The tide-producing forces are distributed in a regular manner over the Earth, varying with
latitude. However, the response of the oceans and seas to these forces differs, depending on the
hydrographic features of each basin. As a result, the tides as they actually occur differ significantly
along the coast and in bays and estuaries. An attempt should be made to space the stations so
that the changes in the tidal characteristics are represented. Among the numerous tidal features
which vary at different places, those relating to the time of tide, range of tide, and type of tide are
the principal features reviewed to form the network.
An up-to-date directory of tide-gauge stations where water-level measurements are taken should
be maintained, providing the following information for each station:
(a) Name of station and geographical coordinates;

(b) List and description of equipment and measuring techniques;
(c) Description of support structure;
(d) Description of benchmarks;
(e) Dates of stability and calibration checks of water-level measurement equipment;
(f) Station or data collection platform access information, such as:
(i) Telephone number;

(ii) Satellite platform ID and channel;
(g) Corrections to reference data to chart datum or description of “zero” reference.
IOC Member States agreeing to participate in the Global Sea Level Observing System are
requested to perform the following tasks:
(a) Ensure that all operating GLOSS stations report monthly mean sea-level data values to the
International Council for Science Permanent Service for Mean Sea Level within one year of
acquisition;
(b) Make hourly values of sea-level data available for international exchange;
(c) Upgrade existing stations which are below GLOSS standards;
(d) Install new stations in consultation with the Intergovernmental Oceanographic

Commission.
In order to standardize the procedures for sea-level measurements, national instructions should
be in accordance with the Manual on Sea-Level Measurement and Interpretation, IOC Manuals and
Guides No. 14.
Tide-gauge stations must have access at least to a public telecommunication network with the
aim of making data available after timely quality control checks at manned stations. Data from
automated stations accommodated by data collection platforms can be transmitted via satellite
to the main computer of the relevant service centre for quality control and further analysis and
dissemination of water-level information. Even data which have not been subject to quality
control should be made available immediately from the downlink through a decode programme
on a personal computer for public information services.
Selected users can be authorized to access the water-level data directly from the measurement
equipment if tide-gauge stations have a telephone line.
3.9.2.8.8 Personnel
Tide-gauge station personnel should be familiar with national observing instructions and
guidance material. The operation and maintenance personnel responsible for these stations,
especially when automated water-level measuring systems are used, must have specialized
training in the following areas: structure maintenance, electronic equipment installation and
repair, scuba (self-contained underwater breathing apparatus) diving for inspecting and cleaning
the underwater components, and surveying for running differential levels to monitor stability of
the equipment and benchmarks.
Owing to the various locations of tide-gauge stations, there are no preset quantified confidence
or uncertainty limits that state datum variability in a generic sense, for example, +/– 0.1 feet. Such
accuracies are site-specific, relating to physical environment, vertical stability, signal-to-noise
ratios, gauge operation, length of series, closeness of control stations and the like. Users should
be provided with estimated confidence limits for data on a case-by-case basis.
The readings of individual sea levels should be made with a target resolution of 0.1 m.
Connections between the benchmark and the gauge zero should be made to an uncertainty of a
few millimetres every six months.
Members participating in the Global Sea Level Observing System should send their monthly
and annual sea-level mean values to the Permanent Service for Mean Sea Level at Bidston
Observatory, Merseyside, UK, together with details of gauge location, missing days and a
definition of the datum to which the measurements are referred. Received data are checked
for consistency. If possible, values are reduced to revised local reference; this involves the
identification of a stable, permanent benchmark close to the tide-gauge station and the
reduction of all data to a single datum level which is referred to this benchmark, ensuring
continuity with subsequent data.
References
Drifting Buoys in Support of Marine Meteorological Services, WMO Marine Meteorology and Related
Oceanographic Activities Report No. 11
Global Atmosphere Watch Guide (WMO/TD-No. 553)
Global Atmosphere Watch Measurements Guide (GAW No. 143, WMO/TD‑No. 1073)
Guide to Agricultural Meteorological Practices (WMO‑No. 134)
Guide to Data Collection and Location Services Using Service Argos, WMO Marine Meteorology and Related
Oceanographic Activities Report No. 10
Guide to Practices for Meteorological Offices Serving Aviation (WMO‑No. 732)
International List of Selected, Supplementary and Auxiliary Ships (WMO‑No. 47)
Location and Data Collection Satellite System, ARGOS User’s Guide
Manual on Sea-Level Measurement and Interpretation, International Oceanographic Commission Manuals and
Guides No. 14, UNESCO
Manual on the Global Observing System (WMO‑No. 544), Volume I
Manual on the Global Telecommunication System (WMO‑No. 386)
Marine Observer’s Guide, UK Meteorological Office
Meteorological Observations at Oil Fields Offshore, Norwegian Meteorological Institute
Meteorological Observations Using Navaid Methods, Technical Note No. 185 (WMO-No. 641)
Technical Regulations (WMO‑No. 49)
The Planetary Boundary Layer, Technical Note No. 165 (WMO‑No. 530)
The Planning of Meteorological Station Networks, Technical Note No. 111 (WMO‑No. 265) (out of print)
Use of Radar in Meteorology, Technical Note No. 181 (WMO‑No. 625)
Weather Reporting (WMO‑No. 9)
WMO/IOC Integrated Global Ocean Services System (IGOSS), General Plan and Implementation Programme,
1982–1988
APPENDIX III.1. FUNCTIONAL SPECIFICATIONS FOR AUTOMATIC
WEATHER STATIONS
Minimum
Maximum effective Mode of
Variablea reported BUFR/CREX e Statuse
range b observation d
resolution c
Atmospheric pressure
Atmospheric pressure 500–1080 hPa 10 Pa I, V 0 10 004 OP
Temperaturef
Ambient air temperature

–80 °C – +60 °C 0.1 K I, V 0 12 101 OP
(over specified surface)g
Dewpoint temperatureg –80 °C – +60 °C 0.1 K I, V 0 12 103 OP
Ground (surface)
temperature (over –80 °C – +80 °C 0.1 K I, V 0 12 120 VAL
specified surface)g
Soil temperatureg –50 °C – +50 °C 0.1 K I, V 0 12 130 OP
Snow temperatureg –80 °C – 0 °C 0.1 K I, V 0 12 131 VAL
0 13 082
Water temperature – river, OP
–2 °C – +100 °C 0.1 K I, V or
lake, sea, well OP
0 22 043
Humidity f
Relative humidity 0–100% 1% I, V 0 13 003 OP
Mass mixing ratio 0–100% 1% I, V 0 13 110 VAL
Soil moisture 0–103 g kg–1 1 g kg–1 I, V 0 13 111 VAL
Water vapour pressure 0–100 hPa 10 Pa I, V 0 13 004 OP
2 01 130
Evaporation/ 0.1 kg m–2,
0–0.25 m T 0 13 033 OP
evapotranspiration 0.0001 m
2 01 000
Object wetness duration 0–86 400 s 1s T 0 13 112 VAL
Wind
0; h,i 1–360
Direction 1 degree I, V 0 11 001 OP
degrees
Speed 0–75 m s–1 0.1 m s–1 I, V 0 11 002 OP
Gust speed 0–150 m s–1 0.1 m s–1 I, V 0 11 041 OP
X, Y component of wind
0 11 003
vector Z component of –150–150 m s–1
0.1 m s–1 I, V 0 11 004 OP
wind vector (horizontal –40–40 m s–1
0 11 006
and vertical profile)
BUFR table
Turbulence type (low
up to 15 types Not yet I, V – N
levels and wake vortex) j
specified
BUFR table
Turbulence intensityj up to 15 types Not yet I, V – N
specified
Radiation k
Sunshine duration 0–86 400 s 60 s T 0 14 031 OP

Minimum
resolution c
Background luminance 0–1·105 Cd m–2 1 Cd m–2 I, V 0 14 056 VAL
Global downward solar

0–1·10 8 J m–2 1·102 J m–2 I, T, V 0 14 028 OP
radiation
Global upward solar

–1·10 8 –0 J m–2 1·102 J m –2 I, T, V 0 14 052 VAL
radiation
Diffuse solar radiation 0–1·10 8 J m–2 1·102 J m –2 I, T, V 0 14 029 OP
Direct solar radiation 0–1·10 8 J m–2 1·102 J m –2 I, T, V 0 14 030 OP
Downward long-wave
0–6·107 J m–2 1·103 J m–2 I, T, V 0 14 002 OP
radiation
Upward long-wave
–6·107–0 J m–2 1·103 J m–2 I, T, V 0 14 002 OP
radiation
Net radiation –1·10 8 –1·10 8 J m–2 1·102 J m –2 I, T, V 0 14 053 VAL
UV-B radiation l 0–26·10 4 J m–2 1 J m–2 I, T, V 0 14 072 VAL
Photosynthetically active
0–6·107 J m–2 1·103 J m–2 I, T, V 0 14 054 VAL
radiationm
Surface albedo 0% – 100% 1% I, V 0 14 019 OP
–1·10 8 – 1·10 8 J
Soil heat flux 1·102 J m –2 I, T, V 0 14 057 VAL
m–2
Clouds
Cloud base height 0–30 km 10 m I, V 0 20 013 OP
Cloud top height 0–30 km 10 m I, V 0 20 014 OP
Cloud type, convective vs

up to 30 classes BUFR table I 0 20 012 OP
other types
1–700
Cloud hydrometeor 1 hydrometeor
hydrometeors I, V 0 20 130 VAL
concentration dm–3
dm–3
Effective radius of cloud

2·10 –5 –32·10 –5 m 2·10 –5 m I, V 0 20 131 VAL
hydrometeors
1·10 –5 –1.4·10 –2
Cloud liquid water content 1·10 –5 kg m–3 I, V 0 20 132 VAL
kg m–3
Optical depth within each Not yet

Not yet specified I, V – N
layer specified
Not yet
Optical fog depth Not yet specified I, V – N
specified
Height of inversion 0–1 000 m 10 m I, V 0 20 093 VAL
Cloud cover 0–100% 1% I, V 0 20 010 OP
Cloud amount 0–8/8 1/8 I, V 0 20 011 OP
Precipitation
0.1 kg m–2,
Accumulation n 0–1 600 mm T 0 13 011 OP
0.0001 m
Depth of fresh snowfall 0–1 000 cm 0.001 m T 0 13 118 VAL
Duration up to 86 400 s 60 s T 0 26 020 OP

Minimum
resolution c
Size of precipitating 0 13 058
1·10 –3 –0.25 m 1·10 –3 m I, V OP
elemento 0 20 066
0.1 kg m–2 s–1,

Intensity – quantitative 0–2 000 mm h–1 I, V 0 13 155 OP
0.1 mm h–1
Type up to 30 types BUFR table I, V 0 20 021 OP
1·10–3 kg
Rate of ice accretion 0–1 kg dm–2 h–1 I, V 0 13 114 VAL
dm–2 h–1
Obscurations
Obscuration type up to 30 types BUFR table I, V 0 20 025 OP
Hydrometeor type up to 30 types BUFR table I, V 0 20 025 OP
Lithometeor type up to 30 types BUFR table I, V 0 20 025 OP
Hydrometeor radius 2·10 –5 –32·10 –5 m 2·10 –5 m I, V 0 20 133 VAL
Extinction coefficient 0–1 m–1 0.00001 m–1 I, V 0 15 029 VAL
Meteorological optical
1–100 000 m 1m I, V 0 15 051 VAL
range p
Runway visual range 1–4 000 m 1m I, V 0 20 061 OP
Other weather type up to 18 types BUFR table I, V 0 20 023 OP
Lightning
Rate of lightning discharge 0–4 500 000 h–1 1 h–1 I, V 0 20 126 VAL
Lightning discharge type

(cloud to cloud, cloud to 3 types BUFR table I, V 0 20 023 OP
surface)
Lightning discharge
2 types BUFR table I, V 0 20 119 VAL
polarity
Lightning discharge Not yet

Not yet specified I, V – N
energy specified
Lightning – distance from

0–2·105 m 103 m I, V 0 20 127 VAL
station
Lightning – direction from

1–360 degrees 1 degree I, V 0 20 128 VAL
station
Hydrologic and marine observations
Flow discharge – rivers 0–2.5·105 m3 s–1 0.1 m3 s–1 I, V 0 23 040 VAL
Flow discharge – wells 0–50 m3 s–1 0.001 m3 s–1 I, V 0 23 041 VAL
Ground water level 0–1 800 m 0.01 m I, V 0 13 074 VAL
Ice surface temperatureg –80 °C – +0 °C 0.5 K I, V 0 12 132 VAL
Ice thickness – rivers, 0 08 029

0–50 m 0.01 m I, V VAL
lakesq 0 13 115
Ice thickness – glaciers, 0 08 029

0–4 270 m 1m I, V VAL
seasq 0 13 115
Minimum
resolution c
2 01 133
2 02 129
Ice thicknessr 0–3 m 0.015 m T 0 20 031 OP
2 02 000
2 01 000
0 13 071
Water level 0–100 m 0.01 m I, V OP
0 13 072
Wave height 0–50 m 0.1 m V 0 22 021 OP
2 01 129
Wave periodr 0–100 s 1s V 0 22 011 OP
2 01 000
Wave direction 0; i 1–360 degrees 1 degree V 0 22 001 OP
2 01 135
1D spectral wave energy
0–5·105 m2Hz–1 10 –3 m2Hz–1 V, T 0 22 069 OP
densityr
2 01 000
2 01 135
2D spectral wave energy
0–5·105 m2Hz–1 10 –3 m2Hz–1 V, T 0 22 069 OP
densityr
2 01 000
2 01 130
Water practical salinityr 0–400 psu s 10 –3 psu I, V 0 22 064 OP
2 01 000
2 01 132
Water conductivityr 0–600 S m–1 10 –6 S m–1 I, V 0 22 066 OP
2 01 000
2 07 001
Water pressurer,t 0–11·107 Pa 100 Pa I, V 0 22 065 OP
2 07 000
0.5 kg m–1 (on

Ice mass 0–50 kg m–1 T 0 20 135 VAL
32 mm rod)
Snow density (liquid water

100–700 kg m–3 1 kg m–3 T 0 13 117 VAL
content)
Tidal elevation with 2 01 129

respect to local chart –10 – +30 m 0.001 m I, V 0 22 038 OP
datumr 2 01 000
Tidal elevation with 2 01 129

respect to national land –10 – +30 m 0.001 m I, V 0 22 037 OP
datumr 2 01 000
Meteorological residual
tidal elevation (surge or –10 – +16 m 0.001 m I, V 0 22 040 OP
offset)r,u
0 22 004
Ocean current – direction 0; i 1–360 degrees 1 degree I, V or OP
0 22 005
0 22 031
Ocean current – speed 0–10 m s–1 0.01 m s–1 I, V or OP
0 22 032
Other surface variables
Runway conditions up to 10 types BUFR table I, V 0 20 085 OP

Minimum
resolution c
Braking action/friction
up to 7 types BUFR table I, V 0 20 089 OP
coefficient
State of ground up to 30 types BUFR table I, V 0 20 062 OP
Type of surface specified up to 15 types BUFR table I, V 0 08 010 OP
Snow depth 0–25 m 0.01 m T 0 13 013 OP
Other
Gamma radiation dose

1–103 nSv h–1 0.1 nSv h–1 I, T 0 24 014 VAL
ratev
Categories of stability 9 types BUFR table I, V 0 13 041 OP
a Name of variable, in line with WMO vocabulary and the Technical Regulations.
b Maximum effective range – maximum range of measuring capability; units traceable to SI.
c Minimum reported resolution – lower resolution of reporting is not permitted.
d Mode of observation – type of data being reported:
I Instantaneous – 1-minute value; instantaneous as defined in the Guide to Meteorological Instruments and
Methods of Observation (WMO-No. 8), Part II, Chapter 1, 1.3.2.4
V Variability – average (mean), standard deviation, maximum, minimum, range, median, etc. of samples; those
reported depend on meteorological variable
T Total – integrated value during defined period; maximum 24 hours for all parameters except radiation, which
requires a maximum of 1 hour (exception, see note (k) and precipitation accumulation (6 hours maximum)).
The relevant element descriptor shall be preceded by a time period descriptor 0 04 024 (in hours) or 0 04 025
(in minutes)
A Average (mean) value
e BUFR/CREX descriptors for representation of the listed variables;
OP: operational descriptors for BUFR/CREX Table B, Version Number 14 and subsequent versions
VAL: descriptors which became operation on 2 May 2012 (BUFR-CREX Table B, version Number 18)
N: not yet specified requirements
f Humidity-related variables (i.e., dew point temperature) expressed as temperature are collected under
temperature.
g Temperature data represented by 0 12 101, 0 12 103, 0 12 113, 0 12 120, 0 12 130, 0 12 131 and 0 12 132 shall be
reported with precision in hundredths of a degree even if they are measured with the accuracy in tenths of a
degree. This requirement is based on the fact that conversion from the Kelvin to the Celsius scale has often resulted
in distortion of the data values. Temperature t (in degrees Celsius) shall be converted into temperature T (in
degrees Kelvin) using equation T = t + 273.15.
h Direction to indicate 0 (zero) if speed = 0
i Calm
j If the UNIT is specified as “BUFR Table”, the BUFR descriptor cannot be proposed without the content of the table
being available.
k Radiation energy amounts given over a 24-hour period.
l Definition of UV-B according to the Guide to Meteorological Instruments and Methods of Observation (WMO-No. 8),
Part I, Chapter 7, 7.6 (b). Descriptor 0 14 072 (Global UV irradiation) was recommended for validation in September
2008, revised in July 2010.
m Photosynthetically active radiation (PAR). Various forms of the electromagnetic energy flux in the 400–700 nm
wavelength range, either as integrated spectra or using different weighting functions; for example, converted to
the photosynthetic photon flux (PPF) in quanta per second per square metre, or mole of quanta per second per
square metre or microeinsteins per second per square metre. Approximate conversion is 1 J m–2 s–1 equivalent to
5 µE m–2 s–1 based on a mean wavelength of 550 nm.
n Liquid water equivalent. Maximum interval: 6 hours.
o 0 13 058 (Size of precipitating element) is capable of expressing size of any precipitation element, apart from
hailstones. Size of hailstones shall be represented by 0 20 066.
p MOR uniquely related to “extinction coefficient”, σ, by MOR = –ln(5%)/σ
q Ice thickness 0 13 115 shall be preceded by 0 08 029 (Surface type) set to 11, 12, 13 or 14 to specify river, lake, sea or
glacier, respectively.
r These requirements being confirmed, it has to be noted that the selected descriptors are suitable for the normal
operating conditions and shall be combined with appropriate operator descriptors to allow representation of the
extreme values or the requested high precision.
s Absolute salinity (kg.kg-1) is now being used for ocean applications (IOC Resolution XXV-7). However, salinity
that is reported to national oceanographic data centres remains the Practical salinity (psu). Ocean water is about
35 psu. Lake Assal (Djibouti) is the most saline body of water on Earth with 348 psu salt concentration.
t Operator 2 07 Y is recommended to be used with Water pressure 0 22 065 (Pa, –3, 0, 17), if the data are produced
in BUFR, Edition 4. The same result, i.e., change to (Pa, –2, 0, 21) would be obtained by the combined use of the
less sophisticated operators 201Y and 202Y:
2 01 132
2 02 129
0 22 065 Reported value of “Water pressure”
2 02 000
2 01 000
u The following sequence is to be used to change data width and reference value of 0 22 040 (m, 3, –5000, 14) to
become (m, 3, –10000, 15):
2 01 129
2 03 015
0 22 040 New reference value = –10000
2 03 255
0 22 040 Reported value of “Meteorological residual tidal elevation”
2 01 000
2 03 000
v Gamma radiation dose rate 0 24 014 is intended to be used for reporting of this element under normal conditions,
nuclear accidents excluded.
APPENDIX III.2. BASIC SET OF VARIABLES TO BE REPORTED BY
STANDARD AUTOMATIC WEATHER STATIONS FOR MULTIPLE USERS
[Fixed]
SYNOP Ocean Aeronautical Principle
ocean
Variables land observing meteorological climatological Standard
weather
stations a platforms b station a station a
stations a
Atmospheric
MA MA MA Xc X A
pressure
Pressure
tendency and [M] M [M] [A] – – [A]
characteristics
Air temperature M d A MA M [A] X X e A
Humidity f MA M [M] [A] X g X A
Surface wind h MA MA M [A] X X A
Cloud amount
M M [M] X i X A i
and type
Extinction profile/
M [A] M – X X A j
cloud-base
Direction of
[M] – – – – –
cloud movement
Weather, present
M M M X X A j
and past
State of the not not

[M] – X k [A]
ground applicable applicable
Special
[M] [A] M [M] – – –
phenomena
Visibility M [A] M M X X A
Amount of
[M] [A] [A] [A] – X A
precipitation
Precipitation, yes/
A [A] [A] – X A
no
Intensity of
[A] – [A] – – –
precipitation
not not
Soil temperature – – X A
applicable applicable
Sunshine and/or
– – [A] – X A
solar radiation
Waves – M [A] [M] [A] – – A l
Sea temperature – MA [M] A – – A l
Sea ice and/or not

M M – – –
icing applicable
Ship’s course and not

– [M] [A] –m – [A] l
speed applicable
Sea level – –n [M] [A] not applicable – [A] l
a Source: Manual on the Global Observing System (WMO-No. 544)

b Proposed by the representative of the Joint WMO/IOC Technical Commission for Oceanography and Marine
Meteorology (JCOMM), to become valid for voluntary observing ships, drifting and moored buoys, rigs and
platforms, tide gauges, profiling floats (for review after consultation with JCOMM Expert Teams)
c Also QNH and QFE (see key, Appendix III.3)
d Optional: extreme temperatures
e Inclusive extreme temperatures
f Dew-point temperature and/or relative humidity and air temperature
g Dew-point temperature
h Wind speed and direction
i Cloud amount, towering cumulus and cumulonimbus only
j Restricted to what is feasible
k Snow cover
l Sea and coastal stations only
m Only for helidecks on ships
n Coastal stations and offshore platforms only
Explanation
M Required for manned stations

[M] Based on a regional resolution
A Required for automatic stations
[A] Optional for automatic stations
X Required
APPENDIX III.3. AUTOMATIC WEATHER STATION METADATA
Metadata—which refers to station and data history as applied to measurement and

observation— describe the location, instrument and method of observation, quality and other
data characteristics. Metadata are important for data users in understanding the origins of
meteorological values, and most importantly, in understanding elements that are particularly
sensitive to exposure, such as precipitation, wind and temperature.
Metadata are an extension of the station administrative record, containing all possible
information related to station siting, instrument installation, instrument type, maintenance
schedule and scheduled and unscheduled system changes that occur during the history of an
observing system. Expanded metadata information should also include digital images.
Metadata are dynamic. Station location, ground cover, instruments, observation measurement
and reporting practices, algorithm processing, data formats and the like change over time. As
computer data management systems gradually become an important component of the data
delivery systems, metadata should be available in near-real time since a computer database
allows for computerized composition, updating and delivery.
1. AUTOMATIC WEATHER STATION METADATA REQUIRED FOR OPERATIONAL

PURPOSES
1.1 Station information
Type of metadata Description Example

Station name Official station name Bratislava-Koliba
Station index number or Number used by the National 11813, A59172

identifier Meteorological Service to
identify a station
WMO block and station numbers BUFR descriptors 0 01 001 and 11 and 813
0 01 002a
Geographical coordinates Latitude and longitude of the 18.7697 degrees

station reference point with 18.5939 degrees
respect to WGS 84b
Reference time Actual time of observations in 0655

UTC
Elevation above mean sea level Vertical distance of a reference 260.25 m

point of the station measured
from mean sea level with respect
to EGM 96
Surface qualifier BUFR descriptor 0 08 010 Land grass cover
Roughness classification Davenport classification of 2

effective terrain roughness
Datum level to which station’s Datum levels to which the Pressure sensor: 123.45 m MSL;
atmospheric pressure data refer; atmospheric pressure is reduced Station: 125.67 m MSL;
elevation data used for QFE/QNH Aerodrome reference point:
124.56 m MSL
a The current limitation of WMO station number to 999, also limited by BUFR descriptor 0 01 002, which has a data
width of 10 bits, is a problem for a wide exchange of observations. More than 999 stations often exist in the area
covered by a given WMO block number. Not all the observations available are currently disseminated on the Global
Telecommunications System. To disseminate the observations from all potentially available stations, the WMO
station number should be expanded, that is, a new descriptor is defined and used.
b To add a note for the relevant descriptor denoting latitude and longitude with reference to WGS 84.
1.2 Individual instrument information

Principle of operation Description of method or system
used
• Method of measurement/ Type of operation principle Direct current principal, polymer

observationa describing measurement/ capacitance
observing method used BUFR
descriptors 0 02 175–0 02 189
• Type of detection system Complete set of measuring Optical scatter system combined
instruments and other with precipitation occurrence
equipment assembled to carry sensing system
out specified measurements
BUFR descriptors 0 02 175–0 02
189
Siting and exposure Siting classificationb
• Height above ground (or 1.75 m, -0.1 m

level of depth)
• Representative height of Standard height for the 1.25 m

sensor above ground measurement
Expected instrument A performance classification Class A: for an instrument

performance (to be defined) should include following WMO instrument and
instrument uncertainty and maintenance recommendations
periodicity of preventive Class D: for an instrument with
maintenance and calibration. unknown characteristics and/or
All these elements determine unknown maintenance
the expected instrument Class B and C: intermediate;
performance
Adjustment procedures in Adjustment applied to the data BUFR descriptor 0 08 083

relation to the nominal value
a To add a note for the existence of shielding and type of shielding applied and whether artificially ventilated or not.
b To be standardized. France has defined a classification using values from 1 to 5. The National Oceanic and
Atmospheric Administration (NOAA)/National Climatic Data Center (NCDC) uses a similar classification system.
It is recommended that the Commission for Instruments and Methods of Observation develop guidelines for such
classification, possibly in collaboration with ISO/TC146/SC5 meteorology.
1.3 Data-processing information

Measuring/observing
programme
• Data output Quantity that is delivered by an 2-m average value

instrument or system
• Processing interval Time interval from which the 2, 10 m (wind)

samples are taken
1.4 Data-handling information

Quality control flag for each Description of quality control 1 = good, 2 = inconsistent,
parameter flags 3 = doubtful, 4 = erroneous,
5 = not checked, 6 = changed
2. AUTOMATIC WEATHER STATION METADATA REQUIRED FOR NEAR-REAL

TIME AND NON-REAL TIME PURPOSES
2.1 Station information
There is a great deal of information relating to a station’s location, local topography and others.
Basic station metadata include the following:

Station name Official station name Bratislava–Koliba
Station index number or Number used by National 11813, A59172

identifier Meteorological Service to
identify station
WMO block and station numbers BUFR descriptors 0 01 001 and 0 11 and 813
01 002
Geographical coordinates Latitude and longitude station 18.7697 degrees

reference point with respect to 18.5939 degrees
WGS 84
Reference time Actual time of observations in 0655

UTC
Elevation above mean sea level Vertical distance of station 260.25 m

reference point of the station
measured from mean sea level
with respect to EGM 96
Surface qualifier BUFR descriptor 0 08 010 Land grass cover
Types of soil, physical constants Description of soil type below Clay

and soil profile station, soil characteristics
Types of vegetation and Description of station’s Natural; grass, 7 December 2004

condition, date of entry environment, land
Local topography description Description of station’s Valley station

surroundings, with emphasis on
topographic features that may
influence weather at station
Roughness classification Davenport classification of 2

effective terrain roughness
Type of automatic weather Basic information on the Automatic Weather Stations:

station, manufacturer, hardware automatic weather station Model Vaisala MILOS 500
and software versions, model installation Hardware v1.2
details, (model, serial number, Operating system v1.2.3,
software version) application program v1.0.2
Modem model ABCD, hardware
v2.3, software v3.4.5, power
supply model XYZ, hardware v4.5

Station’s observing programme Information on types of 1-hour synoptic observations
observation made, variables
measured
• Parameters measured List of variables measured Temperature, pressure, humidity,

wind speed and direction
• Reference time Reference time of observations UTC
• Message codes and Actual time of observations METAR: Start 0000 UTC
reporting times (offset and 1-hour intervals; SYNOP: Start
interval) 0000 UTC
3-hour intervals
Automatic weather stations: Start
0000, 1-minute intervals
Datum level to which Datum levels to which the Pressure sensor: 123.45 m
atmospheric pressure data of the atmospheric pressure is reduced MSL; station: 125.67 m MSL;
station refer; elevation data used aerodrome reference point:
for QFE/QNH 124.56 m MSL
2.2 Individual instrument information
Relevant metadata should include the following:

Sensor type Technical information on the Temperature, humidity, pressure
sensor used to measure the
variable
• Manufacturer Vaisala, Campbell
• Model HMP45C, pressure, temperature,

humidity-2000
• Serial number 12345…
• Hardware version V1.2.3
• Software version V2.3.4
Principle of operation Description of method or system

used
• Measurement/observing Type of operation principle Direct current principle, polymer

method describing method of capacitance
measurement/observation used

189
• Type of detection system Complete set of measuring Optical scatter system combined
instruments and other with precipitation occurrence
equipment assembled to carry sensing system
out specified measurements

189
Performance characteristics Operating range of sensors -50 – +60 ºC, 0–100%
Unit of measurement SI unit in which the variable is K, Pa, m s–1

measured

Measuring range Interval between upper and -50 – +60 ºC, 0–75 m s–1
lower value limits for which
variable reported
Resolution Smallest change in a physical 0.01 K

variable which will cause
variation in measurement system
response
Uncertainty Variable associated with result of ±0.1 K

measurement that characterizes
the dispersion of values that
could be reasonably attributed to
measurement; interval in which
the “value” of the variable at the
time of measurement is expected
Instrument time constant Time required for instrument 20 s

to indicate given percentage
(63.2%) of final reading resulting
from input signal
Interface time constant Time required for interface 5s

electronics to indicate given
percentage (63.2%) of final
reading resulting from input
signal
Time resolution Sampling frequency 3 s, 10 s
Output averaging time Time period to determine 1 m; 2 m; 10 m

reported value
Siting and exposure Siting classification
• Location Screen, mast, tower
• Degree of interference from

other instruments or objects
• Shielding Screen, naturally aspirated
• Shielding time constant Time required for instrument 10 seconds

exposure method (solar radiation
screen, or vent) to indicate a
given percentage (63.2%) of
final reading resulting from input
signal
• Height above ground (or 1.75 m, -0.1 m

level of depth)
• Representative height of Standard height for measurement 1.25 m

sensor above ground
Expected instrument Performance classification (to Class A: instrument following

performance be defined) should include: WMO instrument and
uncertainty of instrument maintenance recommendations;
and periodicity of preventive Class D: instrument with
maintenance and calibration; unknown characteristics and/or
these factors determine expected unknown maintenance; Classes B
performance of instrument and C: intermediate
Data acquisition
• Sampling interval Time between successive 3 s, 10 s, 30 s

observations

• Averaging interval Time interval from which samples 1, 2, 10, 30 m
are used
• Type of averaging Method used to calculate Arithmetic; exponential;

average harmonic
Adjustment procedures for Adjustment applied to data BUFR descriptor 0 08 083

nominal value
Calibration data
• Correction Value to be added to or C = R (1+0.6R)

subtracted from instrument
reading to obtain correct value
• Calibration time Date of latest calibration 12 December 2003
Preventive and corrective maintenance
• Recommended/scheduled Frequency of preventive Once every 3 months

maintenance maintenance
• Calibration procedures Type of method/procedure used Static/dynamic calibration
• Calibration frequency Recommended frequency 12 months
• Procedure description
Results of comparison with Sensor field test results 98%

travelling standard immediately after installation
Results of comparison with Sensor field test results 103%

travelling standard immediately prior to removal
SI International System of Units
2.3 Data processing information
Metadata related to processing procedures should include the following information for each
individual meteorological parameter:

Measuring/observing
programme
• Observation time 10th, …, 60th m
• Reporting frequency 10 m
• Data output Quantity delivered by instrument 2-m average value

or system
• Processing interval Time interval from which samples 2, 10 m (wind)

taken
• Reported resolution Resolution of variable reported 0.1 m s–1
Data-processing method, Method used A running 10-m average

procedure, algorithm
Formula used to calculate the VIS=N/(1/V1+1/V2+ … +1/Vn)

element
Observing/measuring method Type of data being reported Instantaneous, total, mean value,
variability
Input source (instrument, Measured or derived variable WAA 151

element and the like)

Constants and parameter values Constants, parameters used to g = 9.806 65ms–2
compute derived parameter
2.4 Data-handling information
Important metadata components include the following:

Quality control procedures, Type of quality control Plausible value check; time
algorithms procedures consistency check, internal
consistency check
Quality control flag for each Description of quality control 1 = good, 2 = inconsistent,
parameter flags 3 = doubtful, 4 = erroneous,
5 = not checked, 6 = changed
Processing and storage Different procedures used in data A computation of visibility

procedures reduction and data conversion derived from extinction
coefficient
Constants and parameter values
2.5 Data transmission information
The transmission-related metadata of interest are as follows:

Transmission method Means of transmission Global System for Mobile
Communication/General Packet
Radio Service, OrbComm; radio
Data format Type of message used for data BUFR; SYNOP

transmission
Time of regular transmission of

Transmission time 11th m; 60th m
data
Transmission frequency Frequency of data transmission 10 m; 1 h

APPENDIX III.4. THE WMO VOLUNTARY OBSERVING SHIP SCHEME
1. INTRODUCTION
Note: This appendix was originally structured as Chapter 6 of the Guide to Marine Meteorological
Services (WMO‑No. 471).
The international scheme under which ships plying the various oceans and seas of the world
are recruited for taking and transmitting meteorological observations is known as the WMO
Voluntary Observing Ship (VOS) Scheme. The forerunner of the scheme dates back to 1853, the
year in which delegates of ten maritime countries came together at a conference in Brussels,
on the initiative of Lieutenant Matthew F. Maury, then director of the US Navy Hydrographic
Office, to discuss the establishment of a uniform system for the collection of meteorological and
oceanographic data from the oceans and their use for the benefit of shipping. In the twentieth
century, the system was recognized in the International Convention for the Safety of Life at Sea
(SOLAS) as amended in 1974, which specifies in Chapter V, Regulation 5, that “the Contracting
Governments undertake to encourage the collection of meteorological data by ships at sea and
to arrange for their examination, dissemination and exchange in the manner most suitable for
the purpose of aiding navigation”.
Voluntary observing ships make a highly important contribution to the Global Observing
System of the World Weather Watch. They also contribute substantially to the Global Climate
Observing System (GCOS) and the Global Ocean Observing System (GOOS) jointly established
by the Intergovernmental Oceanographic Commission (IOC) of the United Nations Educational,
Scientific and Cultural Organization (UNESCO), WMO, the International Council for Science
(ICSU) and the United Nations Environment Programme (UNEP). Relevant standard and
recommended practices and procedures are contained in the Manual on the Global Observing
System (WMO-No. 544), Volume I, Part III, Section 2.3.3. Although new technological means,
such as satellites and automated buoys, are used to gather data from the oceans, voluntary
observing ships continue to be the main source of oceanic meteorological information.
From the beginning of such activity, shipping has assisted in the scientific exploration of the
oceans, as well as in the development of suitable measuring techniques for use by shipborne
observers. Nowadays, the cooperation of voluntary observing ships is sought in each of the
large-scale scientific experiments conducted by special research vessels in order to gather
the additional data needed for complete analyses of environmental conditions. In addition,
the participation of these ships is regularly requested in technical studies and investigations
concerning observing methods, such as the measurement of sea-surface temperature,
precipitation and wind.
2. CLASSIFICATION OF VOLUNTARY OBSERVING SHIPS
2.1 Types of surface synoptic sea stations
Meteorological observing stations include different types of surface synoptic sea stations.
Since this appendix of the Guide emphasizes the mutual collaboration between marine users
and meteorologists regarding the making of marine meteorological observations by voluntary
observing ships, only the activities of National Meteorological Services with regard to voluntary
observing ship stations are described in the following paragraphs. There are eight types of
mobile ship stations engaged in the WMO VOS Scheme, namely:
(a) Selected ship stations;

(b) Selected ship stations equipped with automatic weather stations (AWSs);
(c) VOSClim (VOS Climate) ship stations;
(d) VOSClim ship stations equipped with AWSs;

(e) Supplementary ship stations;
(f) Supplementary ship stations equipped with AWSs;
(g) Auxiliary ship stations;
(h) Auxiliary ship stations equipped with AWSs.
The types of observation normally made by each of these types of ship station is shown in Table 1
below.
Table 1. Observations made by mobile ship stations
Selected Selected VOSClim VOSClim Supple- Supple- Auxiliary Auxiliary

AWS AWS mentary mentary AWS
AWS
Present and
x x x x
past weather
Wind
direction and x x x x
speed
Cloud
x x x x
amount
Cloud type
and height of x x x
base
Visibility x x x x
Temperature x x x x x x
Humidity
x x x x
(dew point)
Atmospheric
x x x x x x x
pressure
Pressure
x x x x
tendency
Ship’s course
x x x x
and speed
Sea
x x
temperature
Period and
height of x x
wind waves
Direction,
period and
x x
height of
swell
Sea ice and/

or icing (if x x x x
appropriate)
Special
phenomena
x x
(if
appropriate)
Selected Selected VOSClim VOSClim Supple- Supple- Auxiliary Auxiliary

AWS AWS mentary mentary AWS
AWS
Maximum
height of deck
cargo above
- - x x - - - -
the Summer
Load Line
(SLL)
Height
difference
from the SLL - - x x - - - -
to the water
line
Course of ship
- - x x - - - -
over ground
Ship’s ground
- - x x - - - -
speed
Ship’s
- - x x - - - -
heading
x = mandatory
2.2 Selected ship station
A mobile ship station equipped with sufficient certified meteorological instruments for
making observations. It transmits regular weather reports and enters the observations in a
meteorological logbook. A selected ship should have at least a barometer, a thermometer
to measure sea-surface temperature, a psychrometer (for air temperature and humidity), a
barograph and possibly an anemometer. Selected ships constitute the large majority of voluntary
observing ships.
2.3 Selected ship station equipped with an automatic weather station
A mobile ship station equipped with an AWS system comprising certified meteorological
instruments to measure at least air pressure, pressure change, temperature and humidity.
Optional sensors would provide wind speed and direction, and sea temperature measurements.
The AWS may or may not have the facility for manual input of the visual elements, and for
transmitting reports at least three hourly or more frequently. The AWS should have the facility for
logging the data.
2.4 VOSClim ship station
A mobile ship station equipped with sufficient certified meteorological instruments for making
observations. It transmits regular and timely weather reports, enters the observations in an
electronic logbook compliant with the International Maritime Meteorological Tape (IMMT)
and has a proven record of providing high-quality observations. A VOSClim ship should have
at least a barometer, a thermometer to measure sea-surface temperature, a psychrometer (for
air temperature and humidity), a barograph and possibly an anemometer. The full range of
metadata must be maintained in the International List of Selected, Supplementary and Auxiliary Ships
(WMO-No. 47); the full suite of digital images, sketches and drawings must be available; and
the delayed-mode (IMMT) data must be submitted to the Global Collecting Centres (GCCs)
in accordance with the procedures described in the Guide to Marine Meteorological Services
(WMO-No. 471), Chapter 3. It is highly desirable for a VOSClim ship to be inspected at intervals
of less than six months.
2.5 VOSClim ship station equipped with an automatic weather station
A mobile ship station equipped with an AWS system comprising certified meteorological
instruments for measuring at least air pressure, pressure change, temperature and humidity.
Optional sensors would provide wind speed and direction, and sea temperature measurement.
The AWS may have a facility for manual input of the visual elements, and may transmit reports
at least three hourly or more frequently. The AWS must have the facility for logging the data
including the additional IMMT delayed-mode VOSClim groups. The full range of metadata must
be maintained in the International List of Selected, Supplementary and Auxiliary Ships (WMO-No. 47);
the full suite of digital images, sketches and drawings must be available; and the delayed-mode
(IMMT) data must be submitted to the GCCs in accordance with the procedures described in
the Guide to Marine Meteorological Services (WMO-No. 471), Chapter 3. It is highly desirable for a
VOSClim ship to be inspected at intervals of less than six months.
2.6 Supplementary ship station
A mobile ship station equipped with a limited number of certified meteorological instruments
for making observations. It transmits regular weather reports and enters the observations in a
meteorological logbook.
2.7 Supplementary ship station equipped with an automatic weather station
A mobile ship station equipped with an AWS system comprising a limited number of certified
meteorological instruments and reporting regularly.
2.8 Auxiliary ship station
A mobile ship station normally without certified meteorological instruments, which transmits in
a reduced code form or in plain language, either on a routine basis or on request, in certain data-
sparse areas and under certain conditions.
2.9 Auxiliary ship station equipped with an automatic weather station
A mobile ship station equipped with an AWS system comprising non-certified meteorological
instruments and reporting regularly.
2.10 International list of selected, VOSClim, supplementary and auxiliary ships
The various ship stations described above constitute an important source of marine data.
In analysing these data, National Meteorological Services should be aware of the type of
instrumentation on board a given ship, or of the particular method of observation when
several methods are in use. For this purpose, WMO issues the annual International List of
Selected, Supplementary and Auxiliary Ships (WMO-No. 47) on the basis of information supplied
by Members in accordance with the Manual on the Global Observing System (WMO-No. 544),
Volume I, Part III, 2.3.3.3 and 2.3.3.4. The information therein covers such particulars as:
(a) Recruiting country;

(b) Metadata format version;
(c) Date of report preparation;
(d) Name of ship;
(e) Call sign;
(f) Vessel type;
(g) Vessel dimensions;
(h) Area or routes the ship normally plies;
(i) Type of barometer;
(j) Type of thermometer;

(k) Exposure of thermometer;
(l) Type of hygrometer or psychrometer;
(m) Exposure of hygrometer or psychrometer;
(n) Method of obtaining sea-surface temperature;
(o) Type of barograph;
(p) Various other meteorological instruments used aboard the ship;
(q) Types of radio equipment, including the International Maritime Satellite (INMARSAT)
system;
(r) Height of barometer, in metres, measured from maximum load line;
(s) Height of anemometer, in metres, measured from maximum load line;
(t) Depth of sea temperature measurement;
(u) Satellite transmission system;
(v) Make and model of AWS system;
(w) Name and version of electronic logbook software.
The International List of Selected, Supplementary and Auxiliary Ships (WMO-No. 47) needs to be
regularly updated (see the Manual on the Global Observing System (WMO-N. 544), Volume I,
Part III, 2.3.3.3) because of frequent changes in the international merchant fleet and changes
in the recruitment of observing ships. Members are asked to provide the WMO Secretariat at
least every quarter, but preferably every month, with updates of their list of selected, VOSClim,
supplementary and auxiliary ships, as an email attachment in approved format. This is the most
efficient means of keeping the master list updated, as no retyping is required. The Secretariat
makes available the master list at http://www.wmo.int/pages/prog/www/ois/pub47/pub47-
home.htm.
Furthermore, a new overarching standard for collecting and making available metadata is being
implemented for all WMO observing systems. More information can be found in the Manual on
the WMO Integrated Global Observing System (WMO-No. 1160) and the Guide to the WMO Integrated
Global Observing System (WMO-No. 1165).
3. RECRUITMENT OF VOLUNTARY OBSERVING SHIPS
3.1 REQUIREMENT TO RECRUIT SHIPS
The recruitment of voluntary observing ships is the responsibility of each Member participating
in the scheme.
Each Member shall recruit as mobile ship stations as many ships as possible that traverse data-
sparse areas and regularly follow routes through areas of particular interest. In fulfilling this
obligation, each Member contributes to the common objective of obtaining sufficient coverage
of meteorological observations over the sea. While a uniform coverage of the oceans is desirable,
this is difficult to achieve in view of the large differences in the density of shipping traffic.
This traffic is comparatively dense in the northern hemisphere, but this is not the case in the
tropics or in the southern hemisphere. Consequently, greater attention should be given to the
recruitment of voluntary observing ships in these areas. Monthly maps showing the density of
observations received from ships are available from the Joint WMO/IOC Technical Commission
for Oceanography and Marine Meteorology (JCOMM) in situ Observation Programme Support
Centre (JCOMMOPS) at http://www.jcommops.org.
Meteorological services in many countries are required to provide more detailed information
on the weather and sea conditions in coastal areas. Some services recruit ships of local shipping
companies to make and transmit observations during their voyage from harbour to harbour
along the coast. Their observations have been widely recognized as being of great value.
3.2 Criteria for recruitment
Several criteria can be used in deciding whether a particular ship should be recruited as a
selected, selected AWS, VOSClim, VOSClim AWS, supplementary, supplementary AWS, auxiliary
or auxiliary AWS ship, to satisfy national and international needs. Questions that should be
examined are whether all the necessary instruments can be installed with adequate exposure,
whether the ship's officers will have the time to record and transmit the observations and
whether the necessary regular contact can be established for training the observers and for
the receipt of electronic or hard copy logbook data. Ship owners and masters are generally
very cooperative in these matters; however, it is advisable that these questions be thoroughly
discussed at the recruiting stage. In all cases, observations should never be undertaken if they
will impair the safe navigation of the recruited ship.
Contrary to the early days of the VOS Scheme, ships are now registered in a variety of countries.
Ships registered in ports outside those of the recruiting country can therefore be recruited.
However, it is advisable to contact the Meteorological Service of the flag state beforehand and
to check that the ship hasn’t already been recruited according to the International List of Selected,
Supplementary and Auxiliary Ships (WMO-No. 47). Care should be taken to ensure that duplicate
recruitment is avoided. Selected or supplementary ships thus recruited should, however, visit the
ports of the recruiting country often enough to permit regular contact.
Members should establish a suitable organizational structure for the maintenance of their marine
networks and for the recruitment of voluntary observing ships. It will often be necessary to
contact shipping companies, managers of fleets and shipping agencies to enlist their cooperation
for arranging visits to ships and the provision of instruments. Port Meteorological Officers
(PMOs) play a major role in the recruitment of ships. Appropriate measures should also be taken
for the provision of instruments, instruction material and other necessary documents to ships, for
the collection and examination of ships’ logbooks, for visits to ships and for the various financial
questions involved. A special officer within the national PMO unit should be made responsible for
ship recruitment.
Complaints about meteorological observations from a particular observing ship should be

directed to the Member with which the ship is registered. If the ship was recruited by another
Member, the Member receiving the complaint should forward it to the Member concerned.
No prior arrangements are required with the Meteorological Service of the country of registry for
the recruitment of an auxiliary ship or auxiliary AWS ship.
3.3 Personnel and training
An essential step in recruiting voluntary observers for ships’ observations is to obtain the
permission of the owner and the master of the ship. When this has been done and the
observer(s) identified, the PMO provides instructions in the following areas:
(a) General care of instruments;

(b) Exposure and reading of hygrometer and psychrometer;
(c) Obtaining sea-water temperatures;
(d) Use of WMO codes;
(e) Coding and transmission of observations.
Once a ship has been recruited, the PMO should endeavour to visit it at least every three months
to check instrument accuracy and renew supplies of forms and documents, for example, codes
and regulations. The officer should take the opportunity to foster an interest in meteorology
in the crew members concerned and explain to them the mutual value of accurate weather
information to seafarers and meteorologists.
4. METEOROLOGICAL OBSERVATIONS FROM SHIPS
4.1 Danger messages
The International Convention for the Safety of Life at Sea (1974), in its Chapter V, Regulation 31,
specifies that ship masters are obliged to issue a danger message when a ship meets with objects
or conditions that are of direct danger to navigation. As far as meteorological phenomena are
concerned, danger messages should contain information on dangerous ice, tropical storms,
sub-freezing air temperatures associated with gale force winds causing severe ice accretion on
superstructures, or winds of force 10 or above on the Beaufort scale for which no storm warning
has been received.
Details concerning the contents of danger messages and their transmission are provided in
SOLAS, Chapter V, Regulation 32. The information given in these messages directly serves the
safety of navigation. Those containing meteorological information are of vital importance to
National Meteorological Services for the preparation of weather and sea bulletins.
4.2 Surface observations
4.2.1 Content of surface observations from ships
The elements observed by the various types of voluntary observing ship are shown in Table 1.
4.2.2 Programme for surface observations on board ships
The basic programme for making surface observations on board ships consists of the following
procedures:
(a) Synoptic observations should be made at the main standard times: 0000, 0600, 1200 and
1800 UTC. When additional observations are required, they should be made at one or more
of the intermediate standard times: 0300, 0900, 1500 and 2100 UTC;
(b) While making observations, atmospheric pressure should be read at the exact standard
time, the observation of other elements being made within the ten minutes preceding the
standard time;
(c) When operational difficulties on board ship make synoptic observations impracticable at
a main standard time, the actual time of observation should be as near as possible to the
main standard time. In special cases, the observations may even be taken one full hour
earlier than the main standard time, for example at 2300, 0500, 1100 and 1700 UTC. In
these cases the actual time of observation should be indicated; however, these departures
should be regarded only as exceptions;
(d) When sudden or dangerous weather developments are encountered, observations should
be made for immediate transmission without regard to the standard times of observation
(see paragraph 4.1 above for obligations under SOLAS);
(e) In accordance with SOLAS, Chapter V, Regulation 32, when a master has reported a
tropical cyclone or other dangerous storm, it is desirable, but not obligatory, that further
observations be made and transmitted hourly, if practicable, but in any case at intervals
of not more than 3 hours. So long as the ship remains under the influence of the storm,
National Meteorological Services may also request more frequent observations for storm
warnings, particularly for tropical cyclones, and special observations may also be requested
for search and rescue operations or other safety reasons;
(f) When required for scientific studies, supplementary observations should be made at
intermediate standard times, subject to non-interference with navigation duties;
(g) Ships’ officers should be encouraged to continue taking and reporting observations while
the ships are in coastal waters, provided it does not interfere with their duties for the safety
of navigation;
4.2.3 Observation of sea and swell
The distinction between two separate wave trains, particularly between sea and swell waves, can
be difficult for an inexperienced observer. Sea waves1 are systems of waves observed within the
wind field producing them. Swell waves are systems of waves observed at a point remote from
the wind field that produced them, or observed when the wind field which generated the waves
no longer exists.
The distinction between sea and swell waves can be made on the basis of one of the following
criteria:
(a) Wave direction: If the mean direction of all waves with more or less similar characteristics
differs by 30° or more from the mean direction of waves with different appearance, the two
sets of waves should be considered as belonging to separate wave systems;
(b) Appearance and period: When typical swell waves characterized by their regular
appearance and long-crestedness arrive approximately, i.e. within 20°, from the direction of
the wind, they should be considered as a separate wave system if their period is at least 4 s
greater than the period of the larger waves of the existing sea waves.
More guidance on the observation of sea and swell waves, as well as the observation of sea ice,
can be found in the Guide to Meteorological Instruments and Methods of Observation (WMO-No. 8),
Part II, Chapter 4.
4.3 Upper-air observations
In the past, very few mobile ship stations were equipped for making upper-air synoptic
observations. An automated means of making upper-air soundings from a merchant ship
has now been developed under the Automated Shipboard Aerological Programme (ASAP).
A balloon is filled with helium and released by a ship’s officer. After launch, the observations
are automatically received, encoded and transmitted to the National Meteorological Service.
However, the number of ships making upper-air observations is still small and mostly
concentrated in the North Atlantic.
An upper-air synoptic observation consists of one or more of the following:
(a) Atmospheric pressure;

(b) Air temperature;
(c) Humidity;
(d) Wind speed and direction.
The standard times of upper-air synoptic observations are 0000, 0600, 1200 and 1800 UTC,
although most ASAP ships report only two times per day. The actual launch time of regular
upper-air synoptic observations is about 60 minutes before these standard times to allow for
relaunches as well as delayed satellite transmissions. The actual time of a balloon observation
may deviate from this time range if wind observations at considerably greater heights can be
achieved.
1
The term “wind wave” is often used as a synonym of sea wave.
In the basic programme of upper-air soundings from mobile ships, the general objective is to
obtain reports from positions that are not more than 1000 km apart and the observations are
typically required at 0000 and 1200 UTC. These observations are to be coordinated within the
framework of an international programme to ensure that data are obtained from those parts of
the oceans where upper-air data are most needed. Members establishing a programme of upper-
air observation on board voluntary observing ships are required to complete the ASAP section of
the Ship Observations Team national annual report.
4.4 Sub-surface observations
Selected ships may also be equipped to make bathythermograph observations during ocean
crossings. The use of an expendable bathythermograph (XBT) does not oblige the ship to reduce
speed or make course alterations. All arrangements for this type of observation are made within
the framework of the Ship Observations Team of the Joint WMO/IOC Technical Commission for
Oceanography and Marine Meteorology (JCOMM) and its Ship of Opportunity Programme
(SOOP).
Procedures for the collection and exchange of BATHY and TESAC (temperature, salinity and
current) observations are specified in the Guide to Operational Procedures for the Collection and
Exchange of JCOMM Oceanographic Data (IOC Manuals and Guides No. 3) and the Manual on the
Global Telecommunications System (WMO-No. 386), Part I, Attachment I-1. The preferred times
for BATHY and TESAC observations are 0000, 0600, 1200 and 1800 UTC. However, observations
taken at any time are useful and should be transmitted.
4.5 Special observations
In relation to international programmes of scientific or economic significance, observations of

a special nature are needed from ships at sea, and WMO is requested to assist through its VOS
Scheme. One such example is the request for observations on locust swarms in the seas around
Africa, the Middle East and the Arabian Sea. This programme, which is of great importance to the
agricultural economy in the countries concerned, is described in Annex A to this chapter.
Another example is the report of freak waves. A freak wave is defined as a wave of considerable
height preceded by a deep trough. It is the unusual steepness of the wave that makes it
dangerous to shipping. Favourable conditions for the development of freak waves seem to be
strong current flows in the opposite direction to a heavy sea, especially when this occurs near
the edge of the continental shelf. The reports may contribute to a mapping of these particularly
dangerous areas and to a better understanding of the phenomenon. Guidelines covering the
content and form of the report and the forwarding arrangements are described in Annex B to this
chapter (see also Guide to Marine Meteorological Services (WMO-No. 471), Chapter 3, 3.3.1).
Sea-surface currents are also subject to special observation. The relevant data are derived from
measurement of ships’ set and drift; they form the basis for consideration of the ocean-surface
current circulation. They are of value in research and climatic studies and are collated by the
International Surface Current Data Centre (ISCDC) in the United Kingdom, which sends a
copy of the stored data to the World Data Service for Oceanography, in the United States. In
order to improve the database of sea-surface currents, all vessels are encouraged to obtain and
supply such data on a voluntary basis. Details of the form of the report and the forwarding
arrangements are given in Annex C to this chapter (see also Guide to Marine Meteorological Services
(WMO-No. 471), Chapter 3, 3.3.2).
4.6 Coding of observations
Ships’ observations are coded in the international meteorological codes published in the Manual
on Codes (WMO-No. 306), Volume I. The various code forms are given code names which
are sometimes included in the heading of the ship's report. In all cases, however, a four-letter
identification group is used (see code 2582 in the Manual on Codes). The identification groups
normally used by ships are shown in Table 2.
Table 2. Identification groups of codes reported by ships
Code name Identification group(s) Content of the code

SHIP BBXX Surface report from a sea station
PILOT SHIP QQAA, QQBB, Upper-wind report from a sea station; Parts A, B, C and D
QQCC, QQDD respectively
TEMP SHIP UUAA, UUBB, UUCC, Report of upper-level pressure, temperature, humidity and wind
UUDD from a sea station; Parts A, B, C and D respectively
BATHY JJVV Bathythermal observation
TESAC KKYY Observation of temperature, salinity and current from a sea station
TRACKOB NNXX Report of a marine surface observation along a ship’s track
BUFR BUFR Binary Universal Form for the Representation of meteorological

data (specific sequences and/or templates should be used for
specific ship reports)
CREX CREX Character form for the Representation and EXchange of data
(specific sequences and/or templates should be used for specific
ship reports)
4.7 Electronic meteorological logbooks
The manual coding of shipboard observations has been greatly aided by the use of electronic
logbook software and by the increased availability of satellite communications on merchant
ships. Observations are taken manually in the traditional way and then entered into a dedicated
software program loaded onto a personal computer. This may be in the form of a laptop
provided by a National Meteorological Service, or the ship’s computer (with the permission of
the ship owner). The computer program then:
(a) Provides screen prompts to assist with data entry;
(b) Calculates the true wind, mean sea-level (MSL) pressure and dew point;
(c) Checks the validity of some data, for example, month in range 1–12 and observations near
climatological extremes;
(d) Allows the real-time observation in SHIP code to be downloaded to electronic media
such as a USB device so that it can then be transferred to the ship’s INMARSAT system for
transmission to the National Meteorological Service. Because most ocean going ships are
required to carry INMARSAT-C equipment, electronic devices can usually be placed in the
INMARSAT terminal and the observation can be transmitted without rekeying. However,
some ship INMARSAT equipment may not have this facility, in which case the data will need
to be transcribed;
(e) Automatically formats and stores the observation in IMMT format (referred to in the
Guide to Marine Meteorological Service (WMO-No. 471), Chapter 3, 3.2.7), which can be
subsequently downloaded to electronic media. The data are usually collected by a PMO at
the time of inspection, or are emailed directly from the ship to the National Meteorological
Service when email is available.
5. ON-BOARD METEOROLOGICAL INSTRUMENTATION
5.1 General
Full guidance on the basic meteorological instruments suitable for use on board ships making
observations under the VOS Scheme, together with advice on methods of observations, is
provided in the Guide to Meteorological Instruments and Methods of Observation (WMO-No. 8),
Part II, Chapter 4.
Experience shows that certain features of these meteorological instruments on board ships
require constant attention. The following comments emphasize where special care should be
taken and are fully complementary to the general guidance in the above-mentioned Guide.
5.2 Instruments measuring atmospheric pressure
Note: It is mandatory for Members to avoid the use of mercury in their instruments or, where mercury is still in use,
to obey safety precautions. See the Manual on the WMO Integrated Global Observing System (WMO-No. 1160), 3.3.2.1, and
the Guide to Meteorological Instruments and Methods of Observation (WMO-No. 8), Part I, Chapter 3, 3.2.7.
Aneroid barometers, precision aneroid barometers and digital barometers are commonly used
on voluntary observing ships to measure atmospheric pressure. These instruments are subject
to drift and require regular checking by a PMO using a transfer standard barometer, preferably
at intervals not exceeding three months. A permanent record of all such checks should be
maintained by the PMO, with a copy attached to the barometer showing the date of the check
and the ambient temperature and pressure.
Some aneroid (dial type) barometers are set to indicate mean sea-level pressure when they
are installed on the ship. Other aneroid barometers, precision aneroid barometers and digital
barometers require correction to mean sea level. The barometer height can vary significantly
with the loading of the ship, so the barometer correction table for height needs to provide a
range of height reduction constants. The draught of very large tankers can vary by as much as
10 metres between a sea-going ballast condition and a fully-loaded condition. If the barometer
elevation is great, air temperature may also have to be taken into consideration when preparing
reduction tables. At all times, the limit of accuracy of the applied reduction should be kept within
0.2 hPa.
The correction of the barometer to mean sea level may be made manually by use of correction
tables or, in the case of ships using electronic logbook software, may be computed by the
software.
Barographs used on board ships should be supplied with an efficient built-in damping device,
and the instrument should be mounted on shock-absorbing material in a position where it is
least likely to be affected by concussion, vibration or movement of the ship. The best results are
generally obtained from a position as close as possible to the centre of flotation. The barograph
should be installed with the pen arm oriented athwart ship to minimize the risk of its swinging
off the chart.
5.3 Instruments measuring wind speed and direction
In order to ensure that wind reports from ships equipped with instruments are comparable
with estimated winds and wind reports from land stations, anemometer readings should be
averaged over 10 minutes. It is difficult to estimate 10-minute means by watching the dial of an
anemometer. An overestimation of more than 10% is not uncommon. It is therefore preferable
that the instrument read-out used for reporting wind velocities be automatically averaged over
10 minutes. If such read-outs are not available, careful instructions should be given in order to
avoid overestimation.
Due to the flow distortion caused by superstructure, masts and spars, the site of the
anemometer’s sensor has to be carefully selected, preferably as far forward and as high as
possible, ideally on the foremast, if this is possible. Wind speed also needs to be corrected for
effective height (for further information see Wind Measurements Reduction to a Standard Level, by
R.J. Shearman and A.A. Zelenko (MMROA Report No. 22, WMO/TD-No. 311).
Any anemometer mounted on a ship measures the movement of air relative to the ship. It is
essential that the true wind be computed from the relative wind and the ship’s velocity. A simple
vector diagram may be used, although in practice this can be a frequent source of error. Special
slide rules and handheld computers are available, and programs can be installed on small digital
computers.
5.4 Instruments measuring temperature and humidity
Temperature and humidity observations should be made by means of a psychrometer with

good ventilation and exposed in the fresh airstream on the windward side of the bridge. Many
countries use a louvred screen and secure it on each side of the vessel, so that the observation
can be made on the windward side. The muslin and wick fitted to a wet-bulb thermometer in a
louvred screen should be changed at least once a week, more often in stormy weather, and the
water bottle filled.
Automated or distant-reading thermometers and hygrometers should be sited in a well-

ventilated and exposed screen with good radiation protection, and placed as far as possible from
any artificial source of heat. It is advisable to compare the readings with standard psychrometer
observations at the windward side of the bridge at regular intervals, particularly when new types
of equipment are introduced.
5.5 Instruments measuring sea temperature
It is important that the temperature of the uppermost thin film of water (measured by infrared
radiometers) be distinguished from the temperature of the underlying mixed layer. It is the
representative temperature of the mixed layer that should be reported by voluntary observing
ships.
The bucket method is the simplest and probably the most effective method of sampling
this mixed layer, but unfortunately this method can only be used on board vessels with low
freeboards and moving slowly. Other methods are:
(a) Intake and tank thermometers, preferably with distant reading display and used only when
the ship is moving;
(b) Hull-attached thermometers located forward of all discharges;
(c) Trailing thermometers;
(d) Infrared radiometers.
These instruments are described in the Guide to Meteorological Instruments and Methods of
Observation (WMO-No. 8), Part II, Chapter 4.
6. TRANSMISSION OF SHIP’S OBSERVATIONS TO THE SHORE
6.1 General guidance
Weather reports from mobile ship stations should be transmitted to a coastal radio station as
soon as possible after the time of observation; hence, the meteorological report, as soon as it is
made on board ship, should be handed to the ship’s radio officer without delay so that it can be
cleared to shore as rapidly as possible. Regulations for the transmission of weather reports from
mobile ship stations to designated coastal radio stations are given in the Manual on the Global
Telecommunication System (WMO-No. 386), Part I, Attachment I–1. The relevant procedures are
reproduced below for ready reference. Weather reports from mobile ship stations should be
transmitted without special request from the ship to the nearest coastal radio station situated
in the zone in which the ship is navigating. If it is difficult to contact promptly the nearest radio
station in the zone in which the ship is navigating, owing to radio propagation conditions
or other circumstances, the weather messages should be cleared following the established
procedures in the order given below:
(a) Transmission to any other coastal radio station in the zone in which the ship is navigating;
(b) Transmission to any coastal radio station in an adjacent zone within the same region;
(c) Transmission to any coastal radio station in any other zone within the same region;
(d) Transmission to a coastal radio station in an adjacent zone in a neighbouring region or,
failing that, to any other station in a neighbouring region;
(e) Transmission to another ship or an ocean weather station with the function of or willing to
act as a relay station.
Maritime mobile radio systems used for ship-to-shore communications as above can cause
problems, for various technical reasons, in the collection of ships’ weather reports for subsequent
distribution over the Global Telecommunication System. The use of new communication
techniques, especially through satellites, offers a promising solution to these problems. Special
mention may be made of the system known as INMARSAT, designed for full communication
capability for public ship-to-shore communication. The use of this system has, however,
important technical and financial implications for National Meteorological Services, and WMO
has been studying them. Other satellite data telecommunication systems are now also being
used in a cost-effective way.
6.2 The International Maritime Satellite system
Ship reports can be readily transmitted to an INMARSAT Land Earth Station (LES) which has been
authorized to accept these reports. Such reports should always be sent via Special Access Code
41 to ensure that they are automatically routed to the National Meteorological Service and that
no cost is incurred by the ship. The National Meteorological Service of the country operating the
LES pays the cost. There are a number of such LESs in each satellite footprint and they are listed,
together with the area from which they will accept reports, on the WMO website at http://www.
wmo.int/pages/prog/amp/mmop/inmarsat_les.html. To place a limit on the costs incurred by
a National Meteorological Service, a LES may be authorized to accept reports only from ships
within a designated area of ocean. These limits should be drawn to the attention of the relevant
ship’s officers when recruiting a ship under the VOS Scheme.
An increasing number of ships are now willing to use their INMARSAT system to send their
weather reports by email directly to National Meteorological Services. In such cases, however,
the cost of the transmission will be incurred by the shipowner, so it must be ensured that they are
willing to accept such costs. In addition, National Meteorological Services will need to establish
a secure system for the receipt and routing of the reports through their message switching
systems.
6.3 Service Argos
Service Argos is a system for receipt of data from automatic weather stations through orbiting
satellites and has been used for many years to collect data from drifting buoys and profiling
floats. The data are sent from the satellite to ground stations for processing and distribution on
the Global Telecommunication System (GTS).
6.4 Other providers of satellite data telecommunication services
There are now private satellite data telecommunication service providers that allow ship
observations to be collected via specific satellite systems such as Iridium. The data can be
transmitted in free format to shore, and the Member recruiting the ship should be responsible
for converting the raw data to geophysical units and applying the necessary quality control
procedures before the dissemination of the data over the GTS.
7. DISTRIBUTION OF SHIPS’ WEATHER REPORTS OVER THE GLOBAL

TELECOMMUNICATION SYSTEM
Ship weather reports received at a National Meteorological Centre (NMC) from INMARSAT Land
Earth Stations and coastal radio stations should be assembled into meteorological bulletins and
transmitted over the GTS with minimum delay. Some Centres transmit a bulletin of available ship
weather reports every 15 minutes. Because ship weather reports are a vital input to a variety of
forecast models runs, it is important that data from different parts of the world are received with
minimum delay.
8. METEOROLOGICAL LOGBOOKS FOR SHIPS
8.1 Layout
The recording of observations in permanent form is obligatory for selected, VOSClim and
supplementary ships and recommended for auxiliary ships. Although most ships now use
electronic logbooks for compiling their observations, a small number of ships still record their
observations in a hard copy meteorological logbook. The layout of logbooks is a national
responsibility. Generally, the order of parameters recorded in the logbook follows the order
of elements in the WMO SHIP code format. Thus the logbook can be used both for recording
the synoptic weather report that is to be transmitted and to include additional information
required for climatological purposes. For the latter use, the entries are subsequently converted to
IMMT format (see Guide to Marine Meteorological Services (WMO-No. 471), Chapter 3, 3.2.7, and
Annex 3.C).
Logbooks should contain clear instructions for entering observations. Code books or code cards
should also be provided, along with logbooks, for ready reference and to help correct wrong
entries as necessary. It is useful to mark in the logbook those columns that are earmarked for
entries to be transmitted as part of the weather report. In some national logbooks, these columns
are lightly shaded or coloured whereas in others they are inserted in a special frame. Space is
often also provided in logbooks to enter the various readings used to compute a meteorological
element such as air pressure reduced to sea-level or actual wind derived from a measured
apparent wind and the ship’s movement. This will enable a check of the computations carried
out on board ship for subsequent quality control of the data during processing for climatological
purposes.
To facilitate the supply of meteorological logbooks to ships that do not regularly call on their
home ports, PMOs in various ports keep a stock of logbooks of different National Meteorological
Services. In addition, they may keep stocks of observing and coding instructions in different
languages.
Ships should be requested to return a completed logbook to the National Meteorological Service
or PMO that has recruited the ship. The period covered by a logbook should ideally not be more
than three months, so that the delay in entering the observations in the climatological system is
not too great.
Logbooks should be returned with information regarding the ship, the instruments used and
other details of a general nature, and space should accordingly be provided for these entries.
The names of the master, the observers and the radio officer (if carried) should also be included,
particularly if an award system exists in the country where the ship has been recruited.
8.2 Supply and return
The observations made by voluntary observing ships using electronic logbook software are
archived by the programme and need to be downloaded by PMOs at regular intervals. Some
voluntary observing ships still use hard copy logbooks, so PMOs need to issue these ships with
the required stationery and collect the completed logbooks. The completed paper logbooks and
the electronic data are generally considered to be the property of the National Meteorological
Service that has recruited the ship.
The National Meteorological Service should archive the paper and electronic logbook data and
submit them to the GCCs under the Marine Climatological Summary Scheme (MCSS).
In order to avoid the entry of duplicate data into the international archiving system,
meteorological logbooks from ships registered in a foreign country should be produced and
stored through appropriate arrangements with the National Meteorological Service of the
country of registry.
8.3 Scrutiny of entries
However clear the instructions relating to entering observations in a logbook, there is always
the possibility of errors occurring in entering data into a logbook. Completed logbooks must
therefore be scrutinized upon receipt and obvious errors corrected. It is of great importance
that recurrent types of error be brought to the attention of the observers concerned so that any
misinterpretation of the instructions or erroneous practices in reading instruments or making
entries can be corrected. When the logbooks are received by the PMO, a first check should be
made as soon as possible to permit a personal conversation with the appropriate ship’s officers.
Such conversations or written responses commenting on the logbooks that have been received
constitute an important element of the continuous training of shipborne observers. Without
this feedback ship officers would soon become uncertain as to the quality of their work or the
implementation of certain observing or coding procedures and, with the inevitable waning of
interest, the quality of their observations may deteriorate.
Ships’ officers often include questions on coding matters or on any special phenomena observed
by them in the remarks column of the logbook. Response to these questions is important, as
this helps maintain interest in meteorological work. Some countries have instituted special
periodicals for meteorological observers on board their ships in which these questions are
discussed and explained (see section 11 below).
9. PORT METEOROLOGICAL OFFICERS
In recruiting voluntary observing ships and assisting them in their meteorological work, direct
contact with ships’ officers is often needed to provide them with instruction material and other
documents, to inspect meteorological instruments on board ships, to collect completed hard
copy logbooks or to download log files from electronic logbooks, and to provide feedback on the
quality of their observations. For this purpose, PMOs, ideally with seagoing experience, should
be appointed at the main ports routinely visited by observing ships.
Port Meteorological Officers are representatives of the National Meteorological Service of

their country, and are, therefore, the point of contact with local maritime authorities. Port
Meteorological Officers play a very important role and the efficiency of the system of voluntary
ship observations often depends on the initiative displayed by these officers. They are in a
good position to discuss with ships’ officers any problems they have encountered and offer
suggestions, bring to their attention any changes in procedures that may have taken place and
give them the latest information that they may wish for. Opportunity should also be taken to
explain various meteorological and/or oceanographic programmes whenever observations are
specially needed from ships. Meteorological instruments on board ships should be checked and
other advice or assistance in meteorological matters should be given by PMOs upon request by
the master of any ship.
Port Meteorological Officers should also report to the meteorological authorities in their country
if the meteorological work carried out on board a ship has not been entirely satisfactory.
Members should immediately respond to these reports; when they concern the work carried out
under the authority of another Member, the latter should be informed. If action has to be taken
following complaints, this can best be done through the PMOs who can play a very important
role using a tactful approach with the masters. If constructive criticism is expressed in positive
terms, goodwill can be maintained all round.
The scope of the work of PMOs depends largely on the importance of marine traffic in the
particular area served. Before deciding to establish a PMO in a given port, a study must be made
of the various services that should be provided. As marine activities develop, a review should
be made from time to time to see whether new services should be provided. Guidelines for
organizing PMO activities are given in Annex D to this chapter, and are also available on the VOS
website. A list of PMOs with their addresses and telephone numbers is available on the JCOMM
website.
10. INCENTIVE PROGRAMME FOR VOLUNTARY OBSERVING SHIPS
In recognition of the valuable work done by ships’ officers in taking and transmitting
meteorological observations and as an incentive to maintaining a high standard of observation,
many maritime countries have established a national award or certificate system. These systems
vary greatly from country to country: in some countries the ships receive the awards, while
in others awards are made to the individual masters or officers. Sometimes recognition for
the meteorological work done on board ships is given in the form of books, charts and other
documents presented to the ship.
Members are encouraged to continue the practice of issuing national awards or certificates
to selected, VOSClim, supplementary and auxiliary ships recruited by them, or to the ships’
personnel, as a sign of their participation in the WMO VOS Scheme.
In addition to national award schemes, the JCOMM Ship Observations Team has produced
a Certificate of Appreciation that can be issued by National Meteorological Services to
participating observing ships.
11. MARINE METEOROLOGICAL PUBLICATIONS PRODUCED BY NATIONAL

SERVICES FOR SEAFARERS AND MARINE OBSERVERS
A number of National Meteorological Services in maritime countries publish magazines

for masters and officers of ships participating in the WMO VOS Scheme. Although content
and format differ widely, all these periodicals have two goals in common: first, to stress the
importance of ships’ participation in the marine observing programme and second, to offer
timely marine meteorological information of interest. A list of these periodicals is given in
Annex E to this chapter.
The periodicals cover:
(a) Incidents where ships’ observations proved particularly useful;

(b) Commendations on active participation in the WMO VOS Scheme;
(c) Hints on observing practices;
(d) Changes in broadcast schedules of weather and sea bulletins or radiofacsimile broadcasts;
(e) Articles on important weather features of particular ocean areas.
Members are encouraged to produce such periodicals and supply them to voluntary marine
observers.
ANNEX A. LOCUST REPORTS FROM SHIPS
(See Appendix III.4, section 4, subsection 4.5)
Members concerned should instruct reporting ships, regardless of their nationality, operating in
the seas around Africa and the Middle East and in the Arabian Sea, to send by radio and in plain
language reports on any locusts seen to the Food and Agriculture Organization of the United
Nations (FAO) in Rome (e-mail：[email protected]). Costs are paid in principle by FAO.
Each locust report should contain the following:
(a) Date and time (specifying UTC or zone time) when locusts were first seen;
(b) Latitude and longitude, if possible to nearest minute, where locusts were first seen;
(c) Time and position at which locusts were last seen;
(d) Whether isolated locusts (seen in flight singly), locust groups (flying locusts seen
intermittently in numbers), swarms (flying locusts seen continuously in numbers over
a period of at least a minute), dense swarms (obscuring part of the horizon or other
background), locusts appearing on board or floating dead locusts (isolated, groups or
swarms) were observed;
(e) Colour of locusts (yellow, pink, grey);
(f) Wind direction and speed.
Details of such reports should be entered in the ship’s meteorological logbook or recorded in the
ship’s electronic logbook, even when it has not been possible to send a radio report.
ANNEX B. GUIDELINES FOR REPORTING ON FREAK WAVES AND FOR
RECORDING INFORMATION IN METEOROLOGICAL LOGBOOKS, WITH AN
EXAMPLE OF A SPECIAL LOG SHEET
GUIDELINES
It is recommended that the following information be recorded in meteorological logbooks:
(1) Information on freak waves:
Date: ................... Time: ................... Ship’s position: ...................
Full description of freak wave (including height and horizontal distance between crest and
trough, if possible)
Weather conditions:......................................................................................................
State of sea:....................................................................................................................
Any other factors that may have influenced the state of the sea:...............................
Any damage sustained by ............................................................................................
(2) Information to be attached to freak wave reports by National Meteorological Centres:
Ship’s name: ........................................................................................................
Gross registered tonnage: ..................................................................................
Ship’s radio-call sign: ..........................................................................................
EXAMPLE OF LOG SHEET
FREAK WEATHER REPORT
A freak wave may be defined as a wave of considerable height, ahead of which there is a deep
trough. Thus it is the unusual steepness of the wave that is its outstanding feature and that makes
it dangerous to shipping.
ss/mv:..........................................................................................................
Call sign:.....................................................................................................
Gross tons:..................................................................................................
Date:................................Time:................................ GMT
Ship’s position:...............................................................................................
Description of freak waves
Height:...............................m Direction if known:...............................

Horizontal distance between crest and trough:........................m
Depth of water:..............m (either by sounding or from chart)
Remarks:.........................................................................................................................
........................................................................................................................................
Weather conditions
Wind direction:................................. Wind speed:..........knots
Any other weather factor applicable:...........................................................................
........................................................................................................................................
State of the sea
Sea waves: Height ..................m Period ............sec
Swell waves: Direction ................... Period ............sec Height ...............m
Any other factor that may have influenced the state of the sea (tide, currents, etc.)
……………………….............................................................................................................
……………………….............................................................................................................
Damage to ship (if any): .................................................................................................
Signature of observer: .......................................................................................................
Signature of master: ...........................................................................................................

ANNEX C. GUIDELINES FOR THE OBSERVATION AND RECORDING OF SEA
CURRENT DATA ON BOARD SHIP
1. INTRODUCTION
The knowledge we now possess regarding surface currents in the world seas is, for the most part,
based on information from current observations taken on board ships.
The systematic collection of surface current information begun in the middle of the nineteenth
century. The famous Lieutenant Matthew F. Maury of the US Navy was one of the first who saw
the importance of gathering wind and current data from ship logbooks. In 1845, he published
the first of a series of wind and current charts.
For constructing current charts, as many observations as possible are required, covering many
years. As the variability of local currents can be examined only on the basis of a large number of
observations, and as the number needed has not been reached for any place at sea, there is still
a great need for current observations, especially from areas less frequented by ships outside the
major shipping lanes. More observations are also needed to establish, year to year, variations in
currents, as some of these are of great significance for marine science, for example, El Niño. The
only way of obtaining enough observations is through the cooperation of voluntary observers.
By making and reporting observations of currents experienced, the seaman not only gains
practical knowledge himself, but benefits shipping generally by adding to our statistical
knowledge, so that up-to-date information can be published.
2. METHODS OF OCEAN-CURRENT OBSERVATIONS AND DEFINITIONS
Current observations are made by calculating the difference between the dead reckoning
(DR) position of the ship, after allowing for leeway, and the position according to a reliable
astronomical, land, radio, radar, electronic or satellite fix. The result is the set and drift over
the ocean floor experienced by the ship during the interval since the previous reliable fix was
obtained, and applies to a mean depth of about half the ship’s draught.
The sea of current is the direction in which it acts; that is the direction toward which it flows. So,
the current set is from the DR position to the fix.
The drift of a current is the distance measured in nautical miles from the DR position to the fix.
The leeway is the angular difference between the ship’s course and the ship’s direction of
movement through the water (i.e., the direction shown by the wake). Leeway occurs when a
ship is subjected by the wind to a pressure from a beam. The angle is rarely more than a few
degrees, but there is a considerable loss of accuracy in the observation of the current if a realistic
allowance is not made for leeway.
The FROM position is the true position at the beginning of the stretch over which the current is
calculated.
The TO position is the true position at the end of the stretch over which the current is calculated.
The DR position is the position of the ship determined by applying to the last well-determined
position (the FROM position) the run that has since been made, using only the true courses
steered (corrected for leeway, if necessary) and the distance run, as determined by log or engine
revolutions, without considering current. It is important that the true course is corrected for the
influence of the wind, so that the difference between the DR position and the true fix is caused
only by the current.
3. THE CALCULATION
The calculation is made in two steps and is based on the following data:
(1) First step – Calculation of the DR position
Data:
(a) Position FROM;
(b) Course(s) steered, corrected for possible wind influence without considering current;
(c) The distance, calculated from speed and time, run along each of the course lines
without considering current.
(2) Second step – Calculation of the current
Data:
(a) DR position;
(b) Position TO.
It is possible to make both calculations by computer. In this case, it is necessary that all three data
fields for the first step and also the position TO are entered in the logbook by the observer.
The advantages of making the calculation by computer are that the extra work involved for the
observers on board is avoided and that errors in the calculation are practically eliminated. A
disadvantage, however, is that errors in the basic data cannot be discovered and this inevitably
leads to faulty results. On the other hand, the observer is in a position to check the basic data for
possible mistakes; also, he/she can check if the data are reliable enough for current calculation.
Calculation by computer, therefore, means that the observer has greater responsibility for
entering the basic data correctly and for their reliability. For this reason, it is advisable to always
enter the data carefully and to check them.
However, in many cases, the officer will wish to calculate the current for his own interest and use,
and this is to be encouraged. When the current is calculated on board, it should be entered in the
logbook along with the data from which it was calculated.
4. THE OBSERVATION
The following notes are intended to give practical guidance on the ways in which the most useful
observations of currents can be made. The usefulness of an observed current depends largely
on its representativeness and its accuracy. Nevertheless, an observation that might normally
be rejected as being unlikely to have the desired accuracy might still be of value if it came
from an area of sparse shipping, that is to say, one about whose currents little was known. The
observation of currents is particularly desirable in such areas.
The representativeness and accuracy of current observations are discussed below in more detail:
(a) Representativeness of observed currents

Ideally, each observation would represent a single current. In practice though, an observation
is made over a distance in which there is likely to be some variation in current. An observation
is not required if it is likely to incorporate currents from two different systems. In particular, it is
desirable to interrupt an observation when passing a cape, a strait or a current rip, as they are
likely to form boundaries between different current systems. Also, observations should not be
made with the distance between FROM and TO positions in excess of about 500 nautical miles or
with the time interval between these positions in excess of about 24 hours. Observations should
not be made where there are tidal influences, for example, on coastal passages;
(b) Accuracy of fixes
The accuracy of current observations depends largely on the accuracy of at least two fixes. In
general, fixes accurate to within two nautical miles are required. Observations based on noon
(sun) positions, derived by running fix, usually have less than the desired accuracy; the accuracy
of such fixes depends on a due appreciation of the currents experienced — the very element
we are trying to determine. On the other hand, the fixes derived from observing two or more
planets or stars at twilight, are likely to be very suitable for calculating currents. When suitable
equipment is available, fixes by such accurate means as the Global Navigation Satellite System
give especially useful current observations;
(c) Course
The true course, corrected for compass error, must be used. An error in the DR position due to
an incorrect course has a direct influence on the current calculation. Therefore, the course must
be corrected for leeway, whenever necessary. Estimating the correction for wind is not simple
and can only be made with experience. However, at a National Meteorological Service receiving
current observations, it is hardly possible to make such corrections, because they are so very
dependent on the type of ship and on its draught. If estimation of the leeway is impossible, for
example, because of stormy weather, no current observation should be made. When, for some
reason, the ship is stopped, it is also better to make no current observation if the wind is more
than Beaufort force 3;
(d) Speed
It is of great importance that the speed of the ship through the water is known as accurately
as possible. An electronic log is especially useful. With other, more common, types of log, the
speed cannot be determined so precisely and a compromise between log distance and distance
by engine revolutions, making due allowance for slip, possibly gives the best results. The slip
depends on several factors (such as draught, loading conditions, sea and swell and the time
elapsed since the ship was in dry dock), but some of their effects are often hard to determine;
(e) Changes in course and speed between the FROM and TO positions
Between the FROM and TO positions, it is possible for the course to have been changed one
or more times; also, it can happen that different corrections for leeway must be applied over a
distance sailed with a constant course. In such circumstances, the distance is divided into parts,
each with a constant course and speed through the water. If the current is not calculated on
board, but later by computer for each part, each distance must be determined from speed and
time noted in the logbook. More than three parts are not acceptable;
(f) Period between FROM and TO positions
The main consideration is that the period should be long enough for the current to have a
measurable effect, yet short enough to make it unlikely that any large variation in current would
have occurred over the distance covered. Thus, the desirable period depends on the accuracy of
available navigational data. Exceptionally, with very accurate data, for example, satellite fixes and
speed through the water measured by electronic log, the current might justifiably be measured
over a period as short as one or two hours. Also, when coasting, a period of a few hours between
two shore fixes may be taken. Usually, however, a longer period is desirable and a period of
about 12 hours between stellar fixes, determined at dusk and at dawn, for instance, would be
very suitable. A period of about 24 hours is necessary when the only positions determined
have been by running fix, for example noon (sun) positions, but such observations are barely
acceptable. Observations from still longer periods are not acceptable. Since observations of
current should be independent, periods of observation should not overlap.
ANNEX D. GUIDELINES FOR ORGANIZING THE ACTIVITIES OF A PORT
METEOROLOGICAL OFFICER
(See Appendix III.4, section 9)
1. INTRODUCTION
The functions of a Port Meteorological Officer (PMO) cover seven broad areas:
(a) Recruitment of ships to take part in the Voluntary Observing Ship (VOS) Scheme;
(b) Regular liaison with recruited ships to ensure the highest standard of observations;
(c) Collection of completed ships’ meteorological logbooks and data from electronic logbooks;
(d) Acting as an interface between the National Meteorological Service and the marine
community;
(e) In large ports, acting as a focus for the provision of meteorological services in the port;
(f) Assisting with arranging deployment of drifting buoys and profiling floats;
(g) Inspection of ships fitted with upper-air radiosonde equipment, an automatic weather
station (AWS) system or expendable bathythermograph (XBT) equipment.
1.1 Personnel requirements
Each maritime Member of WMO should endeavour to appoint PMOs with maritime experience
at its main ports. Their maritime experience enables them to communicate effectively with
the ship’s master and other officers. They should also have experience in, and knowledge of,
meteorology, theoretical as well as practical. Knowledge of the English language would be
an advantage, as most ships' officers whose mother tongue is not English are able to express
themselves in this language. The necessary training of PMOs is described in the Manual on Marine
Meteorological Services (WMO‑No. 558), Part IV, Section 3.
1.2 Location of the office of a Port Meteorological Officer
The office of the PMO should preferably be situated in the centre of the harbour area. This allows
the maximum of ships to be visited and facilitates visits by observers from voluntary ships to the
PMO’s office, giving them access to meteorological information. The PMO will need appropriate
transport for instruments and supplies to ships as required.
2. DUTIES OF A PORT METEOROLOGICAL OFFICER
2.1 Recruitment of observing ships
2.1.1 Merchant shipping
Recruitment of observing ships should be in the hands of the PMOs but subject to overall
guidance from the relevant section of the National Meteorological Service. A worldwide
distribution of observing ships is the objective to attain and every effort should be made
to recruit ships that operate in data-sparse areas, for example, the oceans of the southern
hemisphere.
Port Meteorological Officers often prioritize the recruitment of ships that are registered in their
own country, but ships of other registry are commonly considered for recruitment if they are
regular callers and if the PMO considers that they would make a useful addition to the voluntary
observing fleet.
Points to be considered when recruiting ships are:
(a) Willingness of masters and officers to carry out voluntary weather observations and to
submit reports throughout the voyage;
(b) Suitability of the ship to carry and care for the instruments.
Permission to recruit a ship should, whenever possible, be obtained from the ship owners or
managers, usually through the marine superintendent of the company, and from the master. It
is recommended that only a verbal undertaking by a ship’s master to carry out the work of an
observing ship should be obtained. This service is voluntary, and it is therefore not desirable to
create the impression that a formal binding contract will be imposed.
When a ship agrees to participate in the scheme or volunteers, the PMO equips the ship with the
necessary instruments and stationery. This needs to be done quickly as many ships do not spend
much time in port. A list of the instruments issued to the ship should be recorded along with the
metadata required by the PMO for the International List of Selected, Supplementary and Auxiliary
Ships (WMO-No. 47).
If calibrated National Meteorological Service instruments are available, the ship should be
recruited as a selected or a VOSClim ship. If available, electronic logbook software should be
installed and training given on how to prepare observations.
Suggested lists of instruments and stationery for the various types of observing ship are as
follows:
Selected and VOSClim ships
- One suitably certified precision or digital barometer;

- One barograph (unless the digital barometer includes a tendency display);
- One whirling psychrometer or two screens and two sheathed thermometers (1 air, 1 wet
bulb) for each screen, plus two spares, or a suitable digital electronic device to measure
temperature and humidity;
- Two sea thermometers and suitable sea buckets (if the bucket method is to be used for
measuring sea-surface temperature);
- Electronic logbook software (or hard copy meteorological logbooks);
- Barograph charts;
- Plotting charts;
- Code and decode information (usually in the form of a code card);
- State of sea card or booklet;
- Booklets for observers of cloud types;
- Reduction to mean sea-level card (for ships where the pressure height correction is not
automatically applied by the electronic logbook software);
- Dew-point tables (for ships that are not equipped with electronic logbook software).
Supplementary ships
- One suitably certified precision or digital barometer;

- One whirling psychrometer or two screens and two sheathed thermometers (1 air, 1 wet
bulb) for each screen, plus two spares, or a suitable digital electronic device to measure
temperature and humidity;

- Booklets for observers of cloud types;
- Reduction to mean sea-level card (for ships where the pressure height correction is not
automatically applied by the electronic logbook software.
Auxiliary ships
- Aneroid barometer correction card;

- Booklets for observers of cloud types.
Ships’ officers should be asked to keep the instruments provided by the National Meteorological
Service in good and clean condition. The position for the barometer in a ship’s chart room
should be chosen with care in consultation with the master. It should be as safe as possible from
accidental damage, in a good light and clear of artificial heating. Advice should be given as to the
best exposure for the thermometer screen if issued under differing conditions. The screen should
be kept white. Special attention should be drawn to the care required in ensuring accurate sea
temperature observations.
Port Meteorological Officers should ensure that observing officers understand the importance of
reading wet and dry bulb temperatures in any one observation to the same degree of precision.
All temperatures are required to be read to the nearest tenth of a degree. When this is not
possible and the temperatures are read to the nearest whole degree, the tenth figure is reported
as a solidus and not by a zero.
Subject to financial constraints, ships under construction may be supplied with distant reading
equipment. Port Meteorological Officers should inform their headquarters of any ships being
built in their area that would be suitable. The owners and marine superintendents of those ships
could then be approached by headquarters with a view to installing the necessary cabling and
equipment during the construction. When the necessary agreements and financial approval
have been obtained from the ship owners or managers, the PMO should be informed. He/she
should then arrange to visit the ship with a technician, if necessary, to discuss the siting and
installation of the instruments.
It is of the greatest importance that the PMO’s initial guidance and instruction to newly-recruited
ship officers be as thorough and complete as possible. This will immediately ensure uniformity in
observing technique.
2.1.2 Fishing vessels and small craft
To help extend the collection of marine meteorological data, small craft fitted with good
communication equipment may be supplied with instruments or they may be recruited as non-
instrumental observing ships and requested to report surface weather conditions, whenever
possible. They become auxiliary ships under the VOS Scheme.
Large fishing vessels and yachts can supply most valuable meteorological information from
important areas from which there are normally very few ships’ weather reports.
In ports from which fishing vessels and large yachts sail, the PMOs should do all that is possible
to encourage and interest owners and captains in marine meteorology. Captains should be
assured of the usefulness to forecasting centres of their voluntary weather reports.
2.2 Visits to ships
Visits and inspections are primarily intended to be occasions for giving encouragement and
guidance to marine observers and for thanking them for their work, but they are also the
occasion for checking on the continued accuracy of the instruments. Observing ships should, if
possible, be visited at intervals of no more than three months and a report should be made on
their instruments. A point to remember when visiting ships is that all the facilities being made
available to the visitor are at the discretion and invitation of the ship’s staff.
At each inspection, any defective instruments belonging to the National Meteorological Service
should be replaced and a receipt should, if possible, be obtained from the master or his senior
officer for all instruments issued.
The barometer is probably the most important instrument for weather observation. The reading
should be checked by comparison with a PMO’s transfer standard barometer, such as a Vaisala
digital barometer.
The barometer should be withdrawn from a ship if the difference from the transfer standard
barometer exceeds 0.3 hPa.
It is recommended that a record card be kept for each barometer issued to a ship. The difference
between the ship’s barometer and the transfer standard barometer is to be registered on the
card. The difference, however small, should always be entered in a form, so that an accurate
record can be kept of the behaviour of each barometer. Plus or minus signs should be used to
indicate high or low differences: the plus sign when the ship’s barometer reading is higher than
the transfer standard and the minus sign when the barometer is lower than the standard.
Distant reading equipment, if fitted aboard ships, should be checked at each inspection.
Handheld anemometers, if issued to ships, should be returned to the National Meteorological

Service once a year for recalibration and a replacement should be issued.
In making out reports on instruments, care should be taken to distinguish between instruments
belonging to the National Meteorological Service and the ship’s own instruments. Where the
ship’s own instruments are used for observation, the PMO should record this on the visit form.
This is necessary to avoid confusion between the property of the National Meteorological Service
and that of the ship owners or officers.
A standard inspection form should be used for each visit. Space should be available on this form
for recording information such as:
(a) Any replacement of instruments;
(b) Any instruments that are the property of the ship owners or officers;
(c) Any instruments supplied by other authorities, such as XBTs or plankton recorders, which
affect the appropriate entry in the International List of Selected, Supplementary and Auxiliary
Ships (WMO-No. 47);
(d) Any metadata required by the International List of Selected, Supplementary and Auxiliary Ships
(unless these data are collected using the ship’s electronic logbook).
The inspection report should be forwarded to the relevant section of the National Meteorological
Service as soon as possible after the inspection.
On visiting an observing ship, the PMO should ascertain that the necessary hard copy logbooks
(if applicable) and stationery are on board and are up to date. The ship’s officers should be
encouraged to understand the international meteorological codes and be familiar with the
procedures for transmitting weather messages to the meteorological centres ashore.
Courtesy visits should, if possible, be made to voluntary observing ships of other countries when
they are in local ports, and advice and assistance given as necessary.
2.2.1 Withdrawal of instruments
It should be the duty of the PMO to recover instruments from ships that cease to observe. When
ships cease observing for any reason, the fact should be recorded. Port Meteorological Officers
should check the shipping papers and journals to ascertain, among others, ship sales and change
of registry and, if these take place abroad, they should consider requesting the assistance of the
PMO from the relevant country and port.
On receipt of this information, the ship’s name will be removed from the national fleet list in the
relevant National Meteorological Service.
When withdrawing instruments, care should be taken that those not belonging to the National
Meteorological Service are not included.
2.3 Collection of ships’ hard copy meteorological logbooks
When the hard copy meteorological logbooks are completed, ships normally return them to the
National Meteorological Service, but some may prefer to hand them to a PMO. The latter should
check the meteorological logbook of all visiting ships and, if it is full or nearly full, they should
forward it to the relevant section of their National Meteorological Service as soon as possible
after collection.
It is important to return completed logbooks from observing ships. When visiting observing
ships, a PMO should therefore ascertain that the logbooks have been returned. If the book in
current use was started more than six months previously, it should be withdrawn and the officers
asked to start a new one. Port Meteorological Officers should take the opportunity, whenever
possible, to give any advice as to the method of writing up the logbooks.
Port Meteorological Officers should make a point of visiting observing ships’ crews who appear
to have difficulties in completing their logbooks and ascertain the reason.
2.4 General liaison with ships
The first duty of PMOs is the care and supervision of the work of voluntary marine observers;
they should encourage the merchant navy to apply marine meteorology for safe and efficient
navigation, comfort of passengers and care of cargo.
A PMO is the channel used to communicate to marine observers advice, instruction and
correction but also the gratitude of the meteorological departments responsible for coordinating
the work. Thus a courtesy call by a PMO on the master and officers of a ship should be regarded
as more valuable than a letter or e-mail; a complimentary card should be left if it was not possible
to see the master.
Port Meteorological Officers should become familiar with the current international
meteorological codes for ships so that they are able to explain them to the masters and officers of
the voluntary observing fleet.
Advice and encouragement to voluntary observing officers should be given at every opportunity
during visits and, for example, through any national marine meteorological publications written
for the voluntary observing ships.
Marine observers and others interested in marine meteorology should be encouraged to

contribute papers or remarks on pertinent subjects for publication in meteorological journals.
Special attention should be drawn to the pages, where provided, for additional remarks in
the meteorological logbooks. Masters and officers should be encouraged to describe their
experiences not only as regards weather, but all subjects of scientific interest. It is important that
PMOs maintain contact with their national navigation schools and colleges and give them any
advice and assistance they may require.
Port Meteorological Officers of a National Meteorological Service should remember that it is their
duty to secure, through the voluntary service of ships’ officers, the best possible information on
meteorological conditions at sea; it is, however, advisable to avoid imposing a workload that may
interfere with or adversely affect the main duties of a ship’s officer since an excessive workload
may be detrimental to his or her main duties.
Port Meteorological Officers should make themselves thoroughly familiar with the scheme
of communication for observing ships’ routine weather reporting. They should give every
encouragement and all necessary advice and instruction to observing ships.
Attention should be drawn to the procedures regarding the Special Access Code 41 for ships
fitted with the International Maritime Satellite (INMARSAT) system. Messages addressed to
National Meteorological Services without the code 41 are chargeable to the ship.
Port Meteorological Officers should explain the use of radio weather bulletins and gale, storm
and tropical cyclone warnings issued specially for shipping, and which radio weather bulletins,
including facsimile broadcasts, are the most suitable for masters and officers. They should be
familiar with maritime safety information broadcasts such as SafetyNet and Navtex forecasts and
warnings. Information on this and other meteorological services available to mariners should
also be given to navigation schools.
Port Meteorological Officers should try to keep in touch with the management and marine
superintendent of shipping companies with offices in their area and pay them regular visits.
2.5 Provision of port meteorological services
Shipping, fishing and other marine interests should be informed on how weather forecasts can
readily be obtained in port. They should also be kept informed of all meteorological services
available to mariners.
Weather information useful to shipping, fishing or small craft should, if possible, be available
at the Port Meteorological Office and details of marine forecast products obtainable on the
Internet should also be provided. In large ports with a network of automatic weather stations the
latest observations may be displayed electronically at the office of the PMO (see Guide to Marine
Meteorological Services (WMO-No. 471), paragraphs 1.7.5 and 7 for more information on services
in ports).
As the first point of contact for ships’ officers on meteorological matters, the PMO may be asked
for more specific technical information, for example, on cargo ventilation. If the PMO is unable
to answer the query himself, he should transmit it to the appropriate section of the National
Meteorological Service and ensure that a prompt reply is made.
ANNEX E. MARINE METEOROLOGICAL PUBLICATIONS PRODUCED
BY NATIONAL SERVICES AND INTERNATIONAL ORGANIZATIONS OF
INTEREST TO SEAFARERS AND MARINE OBSERVERS
(See Appendix III.4, section 11)
Title of publication Issues per year Country of origin Language

Boletín Climático 3 Cuba Spanish
Marino
Météo – Le magazine 4 France French
Guide de l’observateur 1 France French
météorologiste en mer
Der Wetterlotse 6 Germany German
Newsletter V.O.S. from 2 Hong Kong, China English
Hong Kong, China
Ship and Maritime 3 Japan Japanese
Meteorology (Fune to
Kaijou Kishou)
Meteorological 4 Netherlands Dutch and English
Information Bulletin
Maritime
Monthly Weather 12 Qatar English
Summary
IMO News 4 United Kingdom English
Mariners Weather Log 4 United States English
Storm Data 12 United States English
WMO Bulletin Switzerland English, French,
4
Russian, Spanish
PART IV. THE SPACE-BASED SYSTEM
4.1 GENERAL
4.1.1 History of the space-based subsystem
The first experimental meteorological satellite was launched by the United States of America
on 1 April 1960. It provided basic, albeit very useful, cloud imagery and was such an effective
proof of concept that the decision was rapidly made to implement a series of operational polar
satellites. The automatic picture transmission (APT) system was flown for the first time in 1963
and improved access to image data. APT systems have been flown on many satellites since then
to provide images several times a day to relatively inexpensive user stations around the world.
By 1966, the United States had launched its first experimental geostationary meteorological
satellite demonstrating the value of a fixed viewing point relative to the Earth, enabling
frequent images to be taken and used to generate animated views of the world’s weather. In
1969 the then Union of Soviet Socialist Republics launched the first of a series of polar-orbiting
satellites. In 1974, the United States launched the first operational geostationary satellite. Similar
geostationary meteorological satellites were launched and operated by Japan and by the
European Space Agency (ESA) in 1977. Thus, within 18 years of a first practical demonstration,
a fully operational meteorological satellite system was in place, giving routine data coverage of
most of the planet. The network stabilized in terms of sensor data and services during the 1980s
but there was a resurgence of experimental environmental satellites and an upgrading of the
operational subsystem during the 1990s with the contribution of new operators such as India
and China. Further significant improvements in sensor performance and additional spacecraft
are being implemented during the first decade of the new century, contributing to a more robust
constellation of operational and research and development (R&D) satellites around the Earth.
The rapid development of a new inter-national observing system, entailing major investments,
has been unprecedented and indicates the enormous value of these satellites to meteorology and
society when allied to massive improvements in the ability to communicate process and display
information.
4.1.2 Relation to the surface-based subsystem
Several factors make meteorological satellite data unique compared with those from other
sources; it is worthwhile noting a few of the most significant:
(a) Because of its high vantage point and broad field of view, a satellite can provide a regular
supply of data from those areas of the globe yielding very few observations from the
surface-based subsystem.
(b) The atmosphere is broadly scanned from satellite altitude which enables large-scale
weather systems to be seen in a single view.
(c) The ability of geostationary satellites to view a major portion of the atmosphere continually
from space makes them particularly well suited for the monitoring and warning of short-
lived storms.
(d) Advanced communications systems initially developed as an integral part of satellite

technology permit rapid data transmission from the satellite, or their relay from automatic
stations on Earth and in the atmosphere, to operational users. This is still the case after
four decades, although the current trend is towards carrying broadcasting and observing
functions on separate spacecraft that are optimized for telecommunications or Earth
observation.
(e) Information about the atmosphere or surface is obtained indirectly by measuring properties
of electromagnetic radiation reaching a satellite-borne sensor. The use of such data poses
PART IV. THE SPACE-BASED SYSTEM 163
particular problems. These generally manifest themselves in the difficulty of obtaining the
required vertical resolution in some measurements and long-term stability. Furthermore,
errors tend to be spatially correlated, which makes it challenging to use the measurements
in defining differential field properties. The underlying surface, whether it be the actual
surface or cloud top, can make contributions to the upwelling radiation by inferring the
properties of the overlying atmosphere.
On the other hand, in situ measurements can be directly interpreted but may be influenced by
local factors, which raises an issue of representativeness. Furthermore, the density of observing
networks is not homogeneous.
The density of conventional surface observations and upper-air soundings in some portions
of the globe, such as North America or Western Europe, far exceeds the density of similar
information over oceanic and less-developed regions. Reports from ships, aircraft, island stations
and some buoys are often the only surface-based observations available from the oceans,
and many of the data are concentrated in limited geographical areas covered by commercial
transportation routes. Polar-orbiting and geostationary satellites provide the only other source
of environmental data over these and other data-sparse areas. Winds inferred from cloud and
atmospheric pattern motions monitored by the geostationary satellites, particularly over the
otherwise data-sparse low latitudes, occupy a further important niche. The quality of forecasts
and services is directly correlated to the information available on atmospheric conditions,
regardless of what scale of motion is examined, although the effects of any deficiencies may not
be co-located with them.
Forecasts from numerical models are the foundation for routine regional and local weather
forecasts. The global temperature-sounding data from polar-orbiting weather satellites, first
available in the late 1960s and used operationally since the mid-1970s, have encouraged
numerical modelling activities. Computer power and model improvements have made it
both necessary and possible to develop increasingly sophisticated methods for extracting the
temperature and humidity profiles from the satellite radiances. Sounding data assimilation has
a significantly positive impact for both hemispheres thanks to the improved vertical resolution
allowed by advanced hyperspectral sounders.
Strengths and weaknesses from surface and space-based measurements are complementary,
which why the Global Observing System should be considered a composite that builds upon the
strengths of both components. Satellite observations are critical to the generation of warnings
and forecasts of hazardous conditions such as storms, tropical cyclones, polar lows, high winds
and waves. More than 90 per cent of the volume of data assimilated in global numerical weather
prediction models comes from space systems. Nevertheless, direct measurements from surface,
radiosondes and aircraft remain indispensable to provide geophysical variables that are not easily
derived from remote sensing, to monitor small-scale phenomena and to provide independent
validation and calibration data.
4.1.3 Coordination
The combination of operational environmental satellites making the space-based subsystem is

a series of independent national or regional systems coordinated by mutual agreement among
satellite operators and WMO through the Coordination Group for Meteorological Satellites
(CGMS). The Group’s membership involves the operators of meteorological satellites, including
operational and R&D satellites, and WMO in its capacity as a major user organization. It
currently includes meteorological and/or space agencies from China (CMA and CNSA), Europe
(EUMETSAT and ESA), France (CNES), India (IMD), Japan (JMA and JAXA), the Republic of Korea
(KMA), the Russian Federation (ROSHYDROMET and ROSKOSMOS) and the United States
(NOAA and NASA). The Group had its first meeting in September 1972, when it was known as
the Coordination of Geostationary Meteorological Satellites, and has met on an annual basis ever
since. It deals with the coordination of a wide-ranging list of operational aspects of the satellite
systems, such as contingency planning, optimization of the mutual locations of geostationary
and polar-orbiting spacecraft and telecommunications standards. It has helped to ensure the
standardization of key facilities across the entire global system and fosters cooperation on sensor
calibration, product derivation and user training. More information on the Coordination Group
for Meteorological Satellites can be found at http://www.cgms-info.org/.
As regards contingency planning, satellite operators within the Coordination Group for
Meteorological Satellites have established a practice of mutual support among geostationary
satellite systems, whenever necessary and feasible. Troubles experienced by a spacecraft
generate a contingency situation over one region when continuity of critical missions can no
longer be maintained and a replacement satellite cannot be launched before a long period of
time. In such cases, if another satellite operator has a spare satellite in orbit with sufficient fuel
capacity to allow for additional manoeuvres, it is possible to relocate this spare satellite from
its original location to the part of the globe that needs temporary coverage support. Although
all geostationary meteorological satellites have some common mission objectives and a core
set of similar capabilities, they are not interchangeable. Owing to diverse regional and national
standards and to the emergence of different technologies from the independent phasing of
satellite procurements, each satellite system needs its own central ground station and control
centre. Relocating a satellite can thus require more or less complicated arrangements depending
on whether it is within or beyond the visibility of its central ground station. Such a contingency
support was successfully implemented on several occasions in the 1980s and 1990s between
GOES, METEOSAT and GMS spacecraft.
4.2 THE BASELINE SPACE SEGMENT
The World Weather Watch’s Global Observing System space component is comprised of two
types of satellites: operational meteorological satellites and environmental R&D satellites.
Operational meteorological satellites are designed to be operated in either of the two following
orbit classes: geostationary equatorial orbit or sun-synchronous polar orbits. Most of the
environmental R&D satellites are also in sun-synchronous polar orbits, although not all.
Six evenly spaced geostationary satellites are required to provide permanent coverage of the
globe up to at least 55° latitude. A fully global coverage including the polar regions can only be
obtained with polar orbiting satellites; four of them on equally spaced sun-synchronous orbital
planes can provide frequent enough observations to reflect the diurnal cycle.
Other types of orbits may be adopted for specific missions depending on the coverage
requirement. For instance, a 35° inclination orbit was adopted for the tropical rainfall monitoring
mission and a 66° inclination orbit is preferred for an optimized ocean surface topography
mission. Highly elliptical orbits are also under consideration but are not planned to be used
within the Global Observing System, in the near future.
4.2.1 Sun-synchronous polar-orbiting satellites
4.2.1.1 Principle
A sun-synchronous satellite’s orbital plane keeps a constant angle with the sun throughout the
year, in order to ensure that it always passes over a given latitude at the same local solar time. This
is a clear advantage for many applications because it minimizes scene differences due to the time
of day and amount of solar illumination and hence simplifies operational utilization. Customarily
such satellites are used for precise radiance measurements as needed for temperature and water
vapour soundings, monitoring land or sea-surface temperature and radiation fluxes.
Sun-synchronism can be achieved with a low Earth orbit (LEO) inclined at about 99° to the
plane of the Equator, that is, an angle around 81° in the retrograde direction. Since the orbit
passes over both polar regions it is called a polar orbit. Meteorological sun-synchronous
satellites are usually on quasi-circular orbits with an altitude in the range of 800 km to 1 000 km,
which implies an orbital period around 101 minutes. The satellite thus circles the planet every
101 minutes, that is, about 14 times each day. Since the Earth spins on its axis while the plane
of the orbit remains almost constant, the tracks of successive orbits are displaced further to the
west of about 25.5° longitude. If the imaging swath width is at least 2 900 km, there is no gap at
equatorial latitudes between the areas covered at each revolution, and there is significant overlap
between consecutive passes at higher latitudes. Each satellite can then view the entire planet
twice in any 24-hour period, once during daylight and once at night. A sun-synchronous satellite
is classified as a morning or a.m. satellite if the daylight pass across the equator occurs in the
morning. It is said to be an afternoon or p.m. satellite if the daylight pass occurs in the afternoon.
In general the afternoon pass is an ascending orbit from South to North, and the morning pass is
a descending orbit from North to South, although this is not a rule.
Figure IV.1 gives a North Pole view of the orbital planes of sun-synchronous satellites as planned
for 2008. While the Earth is moving around the sun and rotating on itself, the orbital planes keep
a constant angle with the sun’s direction. The figures (0000, 0600, 1200, 1800) indicate particular
values of the mean local solar time (MLST). The MLST is determined by the location with respect
to the sun’s direction. The MLST is 1200 at the point of the Earth that is facing the sun, and on the
meridian of this point. Sun-synchronous orbits that have their daylight part between 0600 and
1200 MLST are called “morning orbits” while orbits that have their daylight part between 1200
and 1800 MLST are called “afternoon orbits”.
4.2.1.2 Implementation
The United States and the Russian Federation have operated polar meteorological satellites since
the 1960s, with current satellites of the NOAA-K, L, M and METEOR-3M design respectively.
China launched polar meteorological satellites FY 1-C and FY I-D in 1999 and 2002. New
generation spacecraft series are being implemented, with METOP (EUMETSAT) in 2006;
METEOR-M1 (Russian Federation) and FY-3 (China) are planned for launch in 2007 and NPOESS
(United States), at the beginning of the next decade. If these plans can be carried out without
significant delay, continuity of observation from the polar orbit will be maintained and will
benefit from greatly enhanced performance. The currently planned configuration for 2008
represented in Figure IV.1 is expected to include three morning satellites, METOP, FY-3 and
METEOR-M1; and one afternoon satellite, NOAA-18.
4.2.1.3 Observing missions
The relatively low altitude of sun-synchronous polar satellites enables instruments to be

flown that observe the atmosphere and the surface with high resolution. The core payload of
meteorological polar-orbiting satellites includes in particular, imaging and sounding instruments
Darkness Daylight
1800
Afternoon orbits
METOP
FY-3A
METEOR-M1
0000
1200
Sun
NOAA 18
Morning orbits
0600
Figure IV.I. Orbital planes of sun-synchronous satellites as seen from the North Pole, 2008
configuration
(see Table IV.1). Imaging radiometers have a high horizontal resolution and monitor the surfaces
of land, sea, ice or clouds, in spectral channels where the atmosphere is weakly absorbing
(window channels). Sounding instruments have a high spectral resolution and compare the
radiation emitted by the atmosphere in series of narrow channels along atmospheric absorption
bands (CO2 and H2O in infrared, O2 and H2O in microwaves). A variety of other instruments,
passive or active, may be part of the payload, depending on the particular mission objectives and
spacecraft design constraints.
Table IV.1. Description of the respective payloads of NOAA-N, -N’ and METOP polar satellites
NOAA-N, N’ METOP
Function
(USA) (EUMETSAT)
Atmospheric temperature sounding in the microwave spectral region;
AMSU-A AMSU-A
all-weather
Infra-red atmospheric temperature sounding (useful in clear sky

HIRS/3 HIRS/4
conditions)
New generation infra-red atmospheric sounding with improved

IASI spectral resolution; measures temperature and humidity profile with
improved vertical resolution, and tropospheric chemical constituents
Atmospheric temperature sounding from the lower troposphere to

GRAS
the stratosphere, using radio occultation of a GPS-type signal
Atmospheric humidity sounding in the microwave spectral region;

MHS MHS
all-weather
Images and radiance/temperature of clouds and surface; vegetation

AVHRR/3 AVHRR/3
monitoring; supports sounding by identifying cloud-free regions
Profiles of atmospheric ozone and other constituents in the upper

SBUV/2 GOME
atmosphere
ASCAT Wind vectors at the ocean surface; active instrument
4.2.1.4 Data dissemination missions
The data acquired by polar satellites are accessible either by direct broadcast or through
retransmission by advanced dissemination systems (ADM). Satellite products are also distributed
over the Global Telecommunication System and an increasing number of images or products can
be found on the Internet.
Direct broadcast from the spacecraft provides the user with real-time data when the satellite is
within the visibility area of its receiving station, which extends to about 2 500 km around the
station if we assume a minimum antenna elevation angle of 5 degrees. Typical visibility areas
of local receiving stations in different locations are indicated in Figure IV.2. The data accessed
from direct broadcast are relevant to the portion of the globe that is scanned by the spacecraft
at the moment of reception; they are thus referred to as local data. While the historical analogue
dissemination service, automatic picture transmission, is still available on current NOAA
spacecrafts, new generation satellite systems will include digital dissemination services only.
The standards agreed by the Coordination Group for Meteorological Satellites for digital
dissemination in L-band from polar satellites are high-resolution picture transmission (HRPT)
and low-resolution picture transmission (LRPT) for high-resolution and low-resolution data
sets, respectively. However, the latest planned satellite generations such as FY-3 and NPOESS
will also make use of X-bands in order to cope with higher data rates. The current baseline for
direct broadcast services in the 2006–2015 time frame is summarized in Table IV.2. Detailed
information can be obtained directly from the respective satellite operators.
Direct broadcast services deliver raw data (see product level definition in Table IV.4). Several pre-
processing software packages are available: the International TOVS Processing Package software
90
60
30
0
-180 -150 -120 -90 -60 -30 0 30 60 90 120 150 180
Latitude
-30
-60
-90
Longitude
Figure IV.2. Typical visibility areas of local receiving stations in various locations
for NOAA satellites is provided by NOAA/NESDIS, the AVHRR and ATOVS pre-processing package
(AAPP) that is suitable for NOAA and METOP satellite series is available from the EUMETSAT
Satellite Applications Facility on numerical weather prediction (SAF NWP) led by the United
Kingdom’s Met Office.
Table IV.2. Main direct broadcast services from operational polar-orbiting satellites in the
2006–2015 time frame
Satellite series Service Frequency (MHz) Data rate (Mb/s)

METOP LRPT 137.1 / 137.9 .072
METOP AHRPT 1701 3.5
NPOESS LRD 1706 3.88
NPOESS HRD 7812 / 7830 20
NOAA-N,N’ APT (DSB) 137.1 / 137.9 (137.3 / 137.7) .017 (0.008)
NOAA-N,N’ HRPT 1698 / 1707 .665
FY-1 CHRPT 1704.5 4.2
FY-3 AHRPT 1704.5 4.2
FY-3 MPT 7775 18.2
Meteor-M LRPT 137.1 / 137.9 0.080
Meteor-M HRPT 1700 0.665
A global data set cannot be received from a single local receiving station. Data recorded aboard
the spacecraft are downloaded by the satellite operator at one or several ground stations and,
after relevant pre-processing, made available through archive retrieval services or distributed in
near-real-time through various means. As the amount of data acquired exceeds the recording
capability of NOAA spacecraft, the global data set of AVHRR imagery is only available at reduced
spatial resolution within the global area coverage service. Advanced TIROS operational vertical
sounder (ATOVS) data are processed and distributed over the Global Telecommunication System.
Taking into account the storage time aboard the spacecraft during up to one orbital period, and
the time needed for data management, transmission and processing, the full global sounding
data set cannot be available earlier than three hours after acquisition.
Regional retransmission services are being implemented to complement core ground segment
functionalities and combine the advantage of direct broadcast (real-time availability) and of on-
board recorded data services (global coverage). Since the coverage and timeliness requirements
of regional and global numerical weather prediction became more demanding in 2001, the
EUMETSAT ATOVS Retransmission Service (EARS) was initiated by EUMETSAT. The principle of
EARS is to implement a network of local HRPT stations, to concentrate ATOVS data sets received
in real time by these stations and to re-distribute these data in a consistent format to the wider
user community. Through the addition of acquisition areas of each HRPT station, the EARS
network coverage extends to a large part of the northern hemisphere from Eastern Europe to
North America, and from the polar cap to North Africa. Data are available within 30 minutes
to end users. Implementation of similar regional ATOVS retransmission services (RARS) in the
Asia-Pacific region and South America is being planned with a view to providing sounding data
at full global coverage in a time frame adequate for assimilation in regional and global numerical
weather prediction. Extension of these services to other instrument data beyond sounding is also
being considered.
This near-real-time and global coverage functionality is part of the design of the ground segment
of the National Polar-orbiting Operational Environmental Satellite System programme, since
it is planned to reduce the latency time by implementing a network of 15 downlink stations all
around the globe. The network would ensure that the spacecrafts are almost always in the range
of a downlink station and can transmit their data with virtually no need for on-board storage.
Dissemination of data acquired through a RARS is performed either via the Global
Telecommunication System or advanced dissemination methods. Increasing emphasis is put on
the use of those methods for cost-efficient access to satellite data and products. Since this is not
specific to polar-orbiting satellites, see 4.3 for further details.
4.2.1.5 Other communications missions
Polar-orbiting satellites are well suited to support data collection systems (DCS). The payload
of NOAA and METOP includes the ARGOS DCS, which uses Doppler frequency-shift techniques
aboard the satellite to determine the location of any ARGOS transmitter or beacon anywhere
in the world with an uncertainty of some 150 metres. ARGOS can also collect data from sensors
on fixed or mobile data collection platforms (DCPs) and thousands of such DCPs are operating
around the world. Although continuous satellite coverage is not available for the relay of DCP
data except in the polar regions, the baseline subsystem provides a minimum of eight satellite
over-flights each day for every point on Earth.
A search and rescue satellite-aided tracking (SARSAT) system uses polar-orbiting and other low-
orbiting satellites to pick up distress signals from downed aircraft or ships in distress and relays
the signals to rescue forces through ground stations in cooperating countries. Locating the signal
geographically further supports the rescue operations. The polar-and low-orbiting satellites are
equipped with transceivers operating at frequencies of 121.5, 243 and 406 MHz.
4.2.1.6 Space monitoring missions
In the baseline system, NOAA satellites include a space environment monitor instrument that
measures solar proton flux, electron density and energy spectrum, and total particulate energy
distribution at spacecraft altitude. The two detectors included within this instrument are the
total energy detector and the medium-energy proton and electron detector. Meteor series
also include detectors for solar wind particles. These data are used to monitor and predict solar
events, such as sunspots and flares, and their effects on the magnetic field. Measurements of
arriving energetic particles are used to map the boundaries of the polar aurora that affects
ionospheric radio communications and electric power distribution systems.
4.2.2 Geostationary satellites
Geostationary satellites revolve about the Earth in the same direction and with the same period
as the Earth’s rotation, that is, they stay in almost fixed positions close to 36 000 km above one
point on the Equator. Since the altitude of the geostationary Earth orbit (GEO) is 40 times higher
than on a polar LEO, it is technically more difficult to perform measurements of the Earth’s
atmosphere and surface at high spatial resolution, but the advantage of the GEO is to allow a
continuous weather watch over a fixed extended part of the Earth’s globe, or the Earth’s disc.
Frequent images, typically every 15 or 30 minutes, can be generated of the full Earth disc. Smaller
areas can be scanned at even faster rate, namely rapid scanning. Full-disc and rapid scanning
images are widely used to support nowcasting and severe weather warnings, to monitor
mesoscale cloud growth, for verification of synoptic forecasts and to support TV weather reports.
Analysing the displacement of clouds, water vapour patterns or other atmospheric features
between consecutive scanning cycles makes it possible to derive wind vector fields. Atmospheric
radiation fields observed by GEO satellites complement polar-orbiting satellite data with the
effect to improve the temporal sampling of variables like sea-surface temperature or precipitation
estimates either directly or indirectly through assimilation in numerical weather predication
models. Image quality is reduced with increasing distance from the sub-satellite point due to the
Earth curvature. Data are considered useful for quantitative processing up to a zenith angle of
around 70°, which corresponds to a great-circle arc of about 60° from the subsatellite point.
Some geostationary spacecraft such as METEOSAT or FY-2 are spin-stabilized platforms, using
their own rotation to scan the Earth’s disc line by line and to maintain a stable platform attitude.
Other spacecraft designs like GOES, GOMS and MTSAT are 3-axis stabilized, which makes
accurate attitude control more difficult but allows for greater flexibility in instrument operation.
As a baseline, two geostationary meteorological satellites are maintained by the United States,
at 75°W and 135°W, while one is operated by China (105°E), EUMETSAT (0°), Japan (140°E)
and the Russian Federation (76°E). In addition, India operates satellites at 74°E and 93°E
mainly for national use. The current baseline coverage is illustrated in Figure IV.3, although the
actual coverage often differs from the baseline, for example over the Indian Ocean. The overall
configuration is subject to regular review by the Coordination Group for Meteorological Satellites
and WMO with the aim to optimize and consolidate the coverage, taking into account the
participation of new satellite operators.
4.2.2.1 Observing missions
The core mission of operational geostationary satellites is to provide continuous imagery with a
refreshment cycle of 30 minutes or less, in order to perform the following tasks:
(a) Monitor mesoscale cloud features in support of nowcasting;
(b) Allow the derivation of wind vector fields by tracking cloud or water vapour and other
features in support of numerical weather prediction.
Latitude 90
60
30
0
-180 -150 -120 -90 -60 -30 0 30 60 90 120 150 180
-30
-60
-90 Longitude
GOES-W METEOSAT GOMS MTSAT GOES-E FY-2
Figure IV.3. Nominal coverage by the current baseline geostationary satellites, with a
maximum zenith angle of 70 degrees
Several satellites provide more frequent images, either over the full Earth’s disc or a selected area.
All operational geostationary imagers include at least the following three core channels: visible,
water vapour and infrared around 0.7, 6.7 and 11 µm respectively, with a typical horizontal
resolution at the sub-satellite point of 1 or 2 km in the visible and 5 km in the infrared window
bands. In addition, more recent satellites have a 3.9 µm channel and a “split-window” 12.0 µm
and/or a 13 µm channel. The SEVIRI imager aboard METEOSAT Second Generation series
includes 12 channels. A recommendation from the Implementation Plan for Evolution of Space-
and Surface-Based Subsystems of the GOS is to improve the spatial and temporal resolution
of geostationary earth orbit imagers in those spectral bands relevant for monitoring rapidly
developing small-scale events and for wind retrieval. All geostationary satellite instruments are
evolving towards more spectral coverage and faster imaging.
Some satellites have an extended payload to measure temperature and humidity profiles by
infrared radiometry, or the Earth radiation budget. The more recent GOES satellites carry a
dedicated atmospheric sounding instrument with 8 carbon dioxide channels, 4 water vapour
channels, 4 infrared channels, and ozone, nitrogen and visible channels. Soundings are produced
hourly primarily over the United States and adjacent waters. The horizontal resolution of the
sounding radiances is 10 km.
4.2.2.2 Data dissemination missions
Geostationary spacecraft also provide direct broadcast digital data dissemination services as
described in Table IV.3. High-rate information transmission (HRIT) and low-rate information
transmission (LRIT) are the agreed CGMS standards for direct broadcast from geostationary
satellites in L-band, respectively for high and lower data rate, while analogue image
dissemination services such as WEFAX, which stands for weather facsimile, are gradually being
phased out. In addition, increasing emphasis is put on the use of advanced dissemination
methods which complement or sometimes replace direct broadcast. Since the use of these
methods is not specific to the geostationary satellites, it is described in more detail in 4.3. Derived
products such as atmospheric motion vectors are distributed over the Global Telecommunication
System for numerical weather prediction use.
Table IV.3. Main digital direct broadcast dissemination services from geostationary satellites
in 2006–2010
Satellite Service Frequency Data rate

GVAR 1685.7 MHz 2.1 Mbps
GOES
LRIT 1691.0 MHz 128 kbps
METEOSAT (first
HRI 1694.5 MHz 166 kbps
generation)
LRIT (primary
METEOSAT(second
dissemination is by 1691.0 MHz 128 kbps
generation)
ADM)
HRIT 1687.1 MHz 3.5 Mbps

MTSAT
HRIT 1691.0 MHz 0.665–1 Mbps

Elektro-L
LRIT 1691.0 MHz 64–128 kbps
S-VISSR 1687.5 MHz 660 kbps

FY-2
HRIT
COMS 1691.0 MHz
LRIT
4.2.2.3 Data collection and search and rescue missions
The continuity inherent to geostationary satellite operations provides the capability to collect
data from fixed or moving data collection platforms according to a fixed schedule or in alert
mode.
Each operator of geostationary meteorological satellites supports a regional data collection

system, for gathering data from data collection platforms at fixed locations within the field of
view of their respective satellites. The service has proved valuable in relaying alert information
pertaining to for example, tsunami, flash floods or radiation, over a large sector of the Earth.
The International Data Collection System was established by the Coordination Group for
Meteorological Satellites to allow the collection of environmental data from mobile data
collection platforms such as those mounted on ships, planes or free drifting buoys and balloons.
Such international platforms are transmitting on a fixed frequency that is compatible with any
of the geo-stationary meteorological satellites within communications range. The Automated
Shipboard Aerological Programme (ASAP) uses the International Data Collection System to relay
atmospheric sounding data obtained from moving vessels.
A COSPAS-SARSAT geostationary search and rescue (GEOSAR) transponder is aboard GOES,

METEOSAT, INSAT, and is planned for ELEKTRO-L. Distress signals emitted by emergency
beacons at 406 MHz are thus retransmitted in real time to dedicated ground stations. Unlike
the search and rescue system on polar satellites, the geostationary satellite cannot provide the
location of the beacon, but serves as an immediate indication of an emergency situation (see
4.2.1.5 for more information relating to search and rescue systems on polar-orbiting satellites).
4.2.2.4 Space environment monitoring missions
The GOES spacecraft carry a space environment monitor consisting of three major components:
a magnetometer which measures the magnetic field at spacecraft altitude a solar X-ray sensor
which provides data on solar X-ray activity to monitor and predict solar flares and an energetic
particle sensor and high-energy proton and alpha detector designed to measure energetic
particle flux at orbital altitude. X-ray data, monitored in real time from space environment
monitor sensors, can disclose the onset of a solar flare that may seriously affect telephone and
radio communications. High-energy particles may damage solar cells, cause sensor malfunctions
and trigger spurious commands aboard spacecraft. For similar objectives, a heliogeophysical
measurement system is planned aboard Elektro-L.
4.2.3 Research and development (R&D) satellites
4.2.3.1 Primary purpose of R&D satellite missions
Weather and climate monitoring and prediction, knowledge of atmospheric processes and
environmental resources monitoring require the observation of many geophysical variables
beyond the mission objectives of core meteorological satellites described in 4.2.1 and 4.2.2. A
number of environmental satellites have been launched or are being planned for that purpose
in the framework of experimental programmes of space agencies. They are referred to as
“R&D satellites”. This category of satellite includes a wide range of missions with different
status: technology demonstrator for new instrument concepts such as space-borne lidar, proof
of concept for retrieving new variables from remote sensing, for example, soil moisture—or
missions of already proven feasibility providing data that are required in support of process
studies, for example, atmospheric chemistry missions. From a WMO standpoint, R&D satellite
missions are especially valuable through the advances that they make possible in instrument
technology, retrieving methods and process modelling that will ultimately benefit operational
programmes and providing utilization opportunities for future operational polar-orbiting and
geostationary satellites.
4.2.3.2 Relevance of R&D satellite missions to the Global Observing System
Since their primary purpose is to fulfil research and development objectives, R&D satellite data
do not necessarily comply with operational requirements of long-term continuity and real-time
data availability. Furthermore, there may be no guarantee that products will be derived from
stable and validated algorithms. Nevertheless, research and development mission data are a
valuable complement to operational data to improve the operational coverage, fill possible
gaps and support calibration or validation activities. The early use of research and development
data in an operational context is also essential as a learning process to adapt assimilation tools
and anticipate as much as possible the operational availability of such data. The capability of
numerical weather prediction models to assimilate new data streams is a key factor in reducing
the gap between operational and research and development data and optimizing the benefit of
research and development missions.
Research and development missions make a valuable contribution to the following areas:
(a) Precipitation: This is an essential variable for environmental and climate monitoring,
hydrology and weather prediction. Spatial and temporal rainfall variability and occurrence
of extreme events at regional scales require high-density observations. Assimilation of
precipitation data contributes to improving numerical weather prediction;
(b) Cloud microphysics: Understanding the distribution of cloud water content, cloud
properties and characteristics is important to parameterize and validate cloud/precipitation
processes in numerical weather and global climate models and to determine the Earth’s
radiation balance;
(c) Aerosols and trace gases: Atmospheric chemistry variables affect the radiation budget for
climate models and are important to monitor and forecast air quality and atmospheric
pollution. Key products include aerosol total column and profile, particle size and optical
properties;
(d) Surface winds and atmospheric motion vectors: These are critical to weather predictions
models, and wind is a key parameter for many areas of environmental monitoring and
prediction such as coupled oceanic and atmospheric models, tropical weather analysis and
hurricane warnings, aeronautical meteorology and fire watch;
(e) Ocean surface variables such as temperature, topography, colour and sea-ice: Ocean
surface characterization is essential for global coupled ocean-atmospheric climate models;
(f) Land surface variables: Variables such as soil moisture or vegetation status are important
in many applications such as agriculture, identification of potential famine areas, irrigation
management, land use planning and environmental monitoring, for example, erosion and
desertification. Land surface variables are essential to determine lower boundary conditions
for numerical weather prediction models;
(g) Disaster monitoring: Real-time disaster monitoring and management require high-
resolution imagery across large areas, in all-weather conditions, day and night. Examples
of disaster monitoring include flood, drought, fires, earthquake, landslide, sandstorms,
duststorms tsunami, volcanoes, snow/ice cover and red tide.
A list of R&D satellite missions—actual or planned —that are of direct interest to the Global
Observing System is available on the WMO Space Programme web page. Details on the
respective spacecraft and missions can be found from individual satellite operators.
4.2.3.3 Transition to operational status
Atmospheric or other environmental variables that were initially monitored in support of process
studies have proved to be essential in the long term for climate monitoring and modelling. The
range of geophysical variables for which sustainable observation needs to be implemented is
thus considerably extended beyond the original scope of the operational Global Observing
System. Following a successful proof of concept, research and development instruments become
candidates for being followed by an operational version and there is an expectation from the
user community that continuity be achieved by an operational follow-on without any gap.
Transition to an operational status may require implementing first a transition or preparatory
mission in partnership between research and development space agencies and operational
agencies. Strong user involvement in data and products validation activities is also advisable.
With regard to scheduling, a transition strategy may also require extending research and
development missions beyond their initial objectives to fulfil operational needs, as was the
case with ESA’s European remote-sensing satellite mission and NASA-JAXA’s tropical rainfall
monitoring mission for instance, in order to bridge the gap between the original research and
development mission and its follow-on.
Priorities for transition to operational status need to be regularly reviewed in the light of evolving
requirements and of the outcome of research and development missions and impact evaluation
of their data. For example, current priorities include:
(a) Ocean surface topography (ongoing transition through NASA-NOAA-CNES-EUMETSAT

agreement for JASON-2);
(b) Cloud microphysics and atmospheric chemistry missions to monitor trace gases and
aerosols playing a role in atmospheric radiation balance, including the greenhouse effect,
and in precipitation processes;
(c) Global precipitation monitoring (beyond the extension of the tropical rainfall monitoring
mission and future global precipitation measurement).
Other missions addressing essential geophysical variables are still at the proof of concept stage,
for instance 3D wind measurements by space-borne lidar or soil moisture monitoring.
4.3 DATA CIRCULATION AND USER SERVICES
4.3.1 General ground segment features
The ground segment consists of the facilities that are needed to operate the spacecraft, perform
data acquisition, processing, archiving and distribution and to provide user support services.
Spacecraft control normally relies on the following facilities:
(a) A command and data acquisition station capable of receiving the raw data stream
transmitted by the spacecraft and house-keeping data, and of uplinking commands;
(b) A ranging system to accurately check the location of the satellite;
(c) A satellite and mission control centre in charge of monitoring the status of spacecraft and
instruments, performing attitude control manoeuvres to maintain the spacecraft within
its specified position and addressing any incident through appropriate measures to secure
spacecraft operation.
Within a processing facility, data are pre-processed from level 0 (raw data) to Level I calibrated
and geo-located, or “navigated”, radiances) and subsequently processed to derive geophysical
products (Level II and beyond). Table IV.4 summarizes the terminology used for conventional
data levels. Core product processing by satellite operators is often complemented by distributed
processing centres having specialized skills in specific application areas, for example, EUMETSAT’s
network of Satellite Applications Facilities.
Table IV.4. Usual terminology for data-processing levels
Data level Description

0 Raw data
I Data extracted by Sub-levels for polar satellite Sub-levels for GEO

instrument, at full data I.0 Instrument counts with
instrument pixel resolution, Ia: instrument counts with Earth-location and
with Earth-location and ancillary information calibration information
calibration information Ib: instrument counts with I.5 Earth-located and
quality control and calibrated instrument
with Earth-location and radiances
calibration information
appended but not
applied
Ic: brightness
temperatures (IR) or
reflectance factor (VIS)
of instrument pixels
with Earth-location and
calibration information
Id: same as Level Ic,
with cloud flag (for
sounding data only)
II Geophysical value (temperature, humidity, radiative flux) at instrument pixel resolution
III Remapped (gridded) product based on geophysical value derived at instrument pixel
resolution
IV Composite product (multisource) or result of model analysis of lower-level data
Note: Products and Level I data are archived by satellite operators. Detailed information on archive data catalogue,
formats and retrieval media can be found on the respective operators’ websites (see References, Part IV).
Links established among satellite operators or with other entities allow data exchange from
different spacecraft and regions. Access to multi-satellite data sets can thus be facilitated.
Near real-time data dissemination of data and products relies on a range of communication
means including direct broadcast, advanced dissemination methods and Global
Telecommunication System distribution, as indicated in 4.2.1.4 and 4.2.2.4. Users should refer
to the web pages of satellite operators and the WMO Space Programme to obtain the latest
information on data access possibilities made available by satellite operators in specific regions
with details on data policy and data formats (see References, Part IV).
4.3.2 Integrated Global Data Dissemination Service
Direct broadcast from meteorological satellites provides true real-time data access independently
of any telecommunications infrastructure except the receiving station; however there are
limitations to this approach. In the case of polar satellites, direct broadcast only delivers data
related to the instantaneous field of view of the satellite (local data). Furthermore, low earth orbit
or geostationary earth orbit direct broadcast users need appropriate software and computer
facilities, which are often expensive, to perform pre-processing and product processing. In other
cases such as the first generation of METEOSAT, data are not directly broadcast by the satellite
but first downloaded to the central processing site and pre-processed up to Level I.5 before
being retransmitted to the satellite and disseminated from there. This lessens the difficulty of
pre-processing by the user. However, another limitation remains: owing to the meteorological
spacecraft data dissemination rate, it may not be possible to transmit the entire data set in full
resolution.
Emphasis was put in recent years on the use of advanced dissemination methods whereby
satellite data and products are distributed by state-of-the-art commercially available
telecommunications means instead of being restricted to the on-board telecommunication

function of the meteorological spacecraft. Data distribution can be optimized from a pure
telecommunications point of view and evolve with time in order to meet new requirements and
benefit from the most cost-efficient technology. Additionally, when the meteorological platform
does not have to support dissemination functions, the position-keeping requirements can be
relaxed, which facilitates extended lifetime operations. The most widespread type of advanced
dissemination method is file transmission with Internet Protocol by digital video broadcasting
satellite as available from various telecommunications operators around the world. This allows
data rates of tens of Mbps to be received with standard low-cost equipment. Dissemination is
performed in Ku-Band or in C-band. C-band offers more robust signal strength in intertropical
regions because of its lower absorption by water vapour and liquid water. It is however more
sensitive to local interference with radars.
ADM services can offer unified access to various data sources including LEO, GEO or R&D
satellites and multi-satellite composite products, high-level products and non-satellite
information. Satellite data dissemination via ADM is not specific to the satellite world and is a
component of the WMO Information System.
The Integrated Global Data Dissemination Service (see Figure IV.4) is the approach to satellite
data and products circulation within the WMO Information System. In pursuing that vision,
WMO supports cooperation among satellite operators towards establishing ADM in a
coordinated way, organizing global data exchange, ensuring fulfilment of the needs of WMO
Programmes and compatibility with overall WMO Information System concepts. Satellite
operators will act within the Service as Data Collection and Product Centres. The objective of
the Service is to offer integrated access to the data of all satellites available over WMO Regions
in a cost-efficient way for both users and satellite operators. The Integrated Global Data
Dissemination Service relies on the expansion of ADM coverage, together with direct broadcast
services as a complement, and the follow-on of the current Global Telecommunication System.
4.3.3 User services
Users should refer to the individual web pages of the respective satellite operators to benefit
from the available user services (see References, Part IV). This normally includes updated
information on the following points: Satellite status and operations (eclipse, manoeuvres);
product catalogue and product algorithm description; archive catalogue, format description
and retrieval modalities; technical issues for real-time data access, including specifications for
receiving devices, software for data reception, decoding, decompression and/or pre-processing;
administrative issues for real-time data access, including registration and subscription formalities
and data policy; background information, publications, training opportunities and user
conferences and user support.
All kinds of WMO programmes

data and products, mostly non-satellite
Integrated Global Data Dissemination
Service components
Data collection and National
Data from operational Other regional data: product centres centres
polar and geostationary research and development,
satellites available high-level products
within region
Interoperability
(Satellite) data collection Global information
and product centres system centre
and data exchange
Data dissemination (possibly shared with non-satellite) Other data-

communication
ADM 1 ADM 2
means
Other data
communication User User
means community 1 community 2 Users
Figure IV.4. Schematic diagram of the Integrated Global Data Dissemination Service within
the WMO Information System framework
4.3.4 Satellite meteorology user training
The WMO Strategy for Education and Training in Satellite Matters is based on cooperation
between satellite operators and some Regional Meteorological Training Centres that take
a particular responsibility as Centres of Excellence in satellite meteorology. The strategy
emphasizes the “training of trainers”. The network of Centres of Excellence is expanding and
currently covers practically all WMO Regions and provides training in five of six of the WMO
official languages, as indicated in Table IV.5.
Table IV.5. Network of Centres of Excellence and sponsoring satellite operators
Centre of Excellence Sponsoring satellite Primary language WMO region

operator
Niamey (Niger) EUMETSAT French Regional Association I
Nairobi (Kenya) EUMETSAT English Regional Association I
Muscat (Oman) EUMETSAT and IMD Arabic Regional Association II
Bridgetown (Barbados) NOAA/NESDIS English Regional Association IV
San José (Costa Rica) NOAA/NESDIS Spanish Regional Associations III and
IV
Nanjing (China) CMA Chinese and English Regional Associations II and V
Melbourne (Australia) JMA English Regional Association V
Buenos Aires NOAA/NESDIS, ESA, Spanish Regional Association III

(Argentina) NASA
A core component of this strategy is the Virtual Laboratory for Education and Training in Satellite
Meteorology adopted by the Coordination Group for Meteorological Satellites and WMO. It
develops and maintains a range of training materials and tools that are accessible on-line via
the Virtual Laboratory Resource Library. It organizes conventional face-to-face training events
(“training of trainers”) and on-line lectures with remote instructors. Cooperation among the
Centres of Excellence and with sponsoring satellite operators is supported by regular on-line
interactive sessions dealing with live meteorological situations. Up-to-date information on
the Virtual Laboratory is found on its web page that is accessible from any of the sponsoring
institutions’ web pages, as well as those of WMO and the Coordination Group for Meteorological
Satellites.
4.4 DERIVED PRODUCTS
4.4.1 Calibration issues
Instrument counts can only be converted to radiances with calibration coefficients that
characterize the instrument’s response as a result of its geometry and detector sensitivity.
Industrial pre-launch measurements provide an initial model of the instrument’s response and
long-term behaviour, but only in-orbit commissioning and regular monitoring can provide
accurate calibration coefficients as required for any quantitative product derivation. Some
instrument designs include on-board calibration devices relying on a black body source for
infrared sensors and moon, sun or deep space viewing. Each calibration cycle then allows
updating the nominal calibration coefficients that are included in the disseminated data flow.
When no on-board calibration system is available, or in order to check such on-board systems
when available, retrospective calibration is achieved by atmosphere radiance measurements
over stable and homogeneous targets such as deserts and cloud-free oceans. Results of this
retrospective calibration process, known as vicarious calibration, are routinely published by
satellite operators.
Monitoring climate change requires the detection of trends in terms of small variations, for
example, a few tenths of a degree of temperature over a decade, which requires particularly
accurate calibration ensuring consistency of global data sets from different sensors over a
very long period. The instrument’s nominal calibration thus needs to be followed by an inter-
calibration of different sensors that are simultaneously in orbit, comparing data series of carefully
space-time co-located scenes with similar viewing angle, in order to provide normalized
calibration coefficients.
The use of particularly accurate reference sensors, including airborne calibration campaigns,
makes it possible to approach absolute calibration that is ultimately required for climate
modelling. Provisions are being made within WMO and the Coordination Group for
Meteorological Satellites towards establishing a global space-based intercalibration system that
would enable to perform intercalibration on an operational basis rather than retrospectively.
4.4.2 Product categories
Some of the products derived from the satellite observations processed at major central facilities
are exchanged internationally over the Global Telecommunication System. Some products are
broadcast through ADM services together with Level I data. These products are also available
on File Transfer Protocol servers hosted by the respective satellite operators. A consolidated list
of operational products can be via through the web pages of satellite operators (see References,
Part IV).
The CGMS Directory of Meteorological Satellite Applications describes the generation and use of
a wide range of products that can be derived from satellite observations, including in particular
the following categories:
(a) Radiances and imagery products
Imagery products result from calibration, geolocation, remapping and dynamics adjustment of
Level I data. Imagery and radiance products are a necessary step for derivation of higher-level
products. In particular, an accurate cloud mask product is a prerequisite for deriving meaningful
quantitative surface and sounding products. Cloud detection can be performed through
comparison with infrared brightness temperature thresholds or visible reflectance thresholds,
with adjustments depending on the underlying surface (sea or land), the latitude and the
season.
Imagery products are also used on a routine basis by direct interpretation either as single-
channel images or after further processing such as multispectral compositing, temporal
combination of animated sequences, or multisatellite mosaics.
(b) Cloud characteristics
Cloud characteristics products derived from satellite imagery provide essential support to
nowcasting and regional short-term forecasting. Multispectral discrimination techniques applied
to visible and infrared imagery allow identification of cloud types. Cloud top temperature or
pressure level can be derived from infrared imagery.
(c) Atmospheric temperature and humidity soundings
Vertical temperature and humidity soundings by polar satellites are mainly derived from infrared
sounder data, in clear-sky, and from microwave sounders in cloudy areas. Sounding data from
the NOAA/ATOVS instrument package are operationally available from NESDIS on the Global
Telecommunication System in SATEM code at reduced resolution. ATOVS data have been shown
to have a significant positive impact on numerical weather prediction in both hemispheres.
Advanced numerical weather prediction centres are increasingly assimilating radiance data
directly, rather than the temperature and humidity soundings retrieved from them. Information
on the error characteristics of retrieved products or radiances, including biases, is essential for
their successful assimilation. Regional sets of locally received ATOVS radiance data are available
through the regional ATOVS retransmission service as described in 4.2.1.4. Hyperspectral infrared
sounders of the AIRS–IASI generation will allow considerable progress in accuracy and vertical
resolution.
(d) Atmospheric motion winds
Such winds are generated automatically by applying a correlation algorithm to sequences of

two or three images. They are extracted in the conventional manner by tracking the movement
of cloud fields from a geostationary satellite. A similar approach can be used in the absence of
clouds, in particular with water vapour features detected in water vapour absorption channel
images or ozone field patterns in ozone absorption channel images. Such geostationary winds
are computed within 60° latitude and longitude of the subsatellite point, at least four times daily
at 0000, 0600, 1200 and 1800 UTC, and distributed in SATOB and/or BUFR code by geostationary
satellite operators. In general, operators are moving towards the production of higher-resolution
products and the use of quality indicators to assist in their use. For polar regions, advantage
can be taken of frequent overpasses by polar-orbiting satellites for the derivation of winds, as
routinely demonstrated with water vapour radiances from the MODIS instruments aboard
NASA‘s AQUA and TERRA satellites.
(e) Land and sea-surface temperature
Sea-surface temperature may be derived from infrared images of both polar and geostationary
meteorological satellites in cloud-free areas. The primary advantages of the polar data are that
they provide global coverage and have generally better spatial resolution. The advantage of the
geostationary satellites is that their frequent data coverage gives a better chance to find cloud-
free pixels. It also allows frequent temporal sampling necessary to monitor diurnal variations.
Regional advanced very high-resolution radiometer (AVHRR)-based products are available at
high resolution—typically 2 km—or global products at lower resolution, for example, 10 km.
With METOP it should be possible to derive operationally high-resolution products on a global
scale. High-accuracy sea-surface temperatures can be derived from AATSR (ENVISAT) and
MODIS (AQUA and TERRA). Microwave passive imagery can also support measurements of that
variable.
Infrared imagers also allow derivation of land surface temperature in combination with visible
imagery to take into account the effect of vegetation.
(f) Snow and ice
Snow and ice areas are identified by visible, infrared and microwave imagery (AVHRR, MODIS,
SSM/I). Active microwave sensors such as scatterometer or synthetic aperature radar imagers are
useful to characterize the superficial snow or ice layer.
(g) Vegetation
Daylight AVHRR imagery provides an essential indicator of the overall vegetation status, the
normalized difference vegetation index based on the difference between reflectance of AVHRR
channel 1 (0.6 µm) and channel 2 (0.9 µm). The index is used to assess fire risk and estimate
vegetation growth and crop yields. More sophisticated products are the leaf area index and the
fraction of absorbed photosynthetically active radiation. Vegetation products are also generated
from more recent imagers like MODIS. Characterizing the land surface is important to determine
lower boundary conditions for numerical weather prediction and is a major component of
climate monitoring from space.
(h) Ocean surface
Active microwave sensors are essential to monitor the ocean surface: altimetry data provide sea-
level information for ocean current monitoring and oceanographic modelling, but also provide
information on sea state, such as significant wave height and wind intensity. Scatterometer
data, for example, from SeaWinds on QuikSCAT and ASCAT on METOP, provide ocean surface
wind vectors that are an essential input to numerical weather prediction and tropical cyclone
monitoring.
4.5 TRENDS IN SPACE-BASED SUBSYSTEMS
Space-based observation capabilities are changing rapidly in many ways. One striking element
is the increasing number of countries contributing to the space-based observing system, either
by operational satellites or R&D satellites. When geostationary satellites from China, EUMETSAT,
India, Japan, the Republic of Korea, the Russian Federation and the United States become
operationally available at the same time, new possibilities will arise to optimize the global
baseline configuration and ensure its robustness. For polar satellites, more operational spacecraft
will allow a better time sampling from the low earth orbit. The growing capability of operational
users to take advantage of R&D satellites and arrangements for real-time availability of selected
R&D satellite data will also contribute to enhance the Global Observing System.
The performance of space-based instruments is dramatically improving, allowing higher

temporal and horizontal resolution, and considerably higher spectral resolution. At some stage,
geostationary satellites will all be equipped with hyper-spectral infrared sensors allowing
frequent temperature and humidity sounding that so far can only be performed from the
low earth orbit a few times a day. Radio-occultation measurements of navigation satellites
signals from the Global Navigation Satellite System will complement the observation of the
stratosphere and upper troposphere. Active and passive infrared measurements from the polar
orbit will improve observations of aerosols and atmospheric chemical components. Precipitation
monitoring by radar and microwave imagery will be effective on the global scale. Doppler lidar
on polar satellites will provide 3D global wind fields. Lightning detectors aboard geostationary
satellites will improve the monitoring of active convective cells in areas not sufficiently covered
by ground-based systems. Provided that the relevant frequency bands are kept free from man-
made interferences, the potential of microwave passive radiometry will be further developed
for monitoring soil moisture and ocean salinity and better monitoring of cloud microphysical
properties.
The increased resolution and number of instruments will result in an explosion of the data rate
acquired from satellites, requiring an updated and flexible dissemination infrastructure that
the Integrated Global Data Dissemination Service is expected to facilitate. The data volume,
together with the multiple numbers of data sources should lead users to focus their development
efforts and strengthen their cooperation in order to be able to rely more on common products
generated by regional specialized centres according to agreed requirements, rather than
individually processing low-level data only. The need for accurate and consistent data series will
also emphasize the importance of sensor calibration and inter-calibration in line with agreed
standards.
Further details on the long-term vision of the space-based subsystem can be found in The Role of
Satellites in WMO Programmes in the 2010s (WMO/TD-No. 1177).
References
CEOS Handbook http://www.eohandbook.com/eohb05/ceos/part3_3.html

The Observing Systems Capability Analysis and Review Tool (OSCAR) http://www.wmo-sat.info/oscar/
spacecapabilities
Guide to Meteorological Instruments and Methods of Observation (WMO-No. 8), Part IV (in preparation)
ESA/Earth Observation [Observing the Earth] http://www.esa.int/Our_ Activities/Observing_the_Earth
EUMETSAT https://www.eumetsat.int/website/home/index.html
EUMETSAT Numerical Weather Prediction Satellite Applications Facility (AAPP software) https://www.
eumetsat.int/website/home/Satellites/GroundSegment/Safs/NumericalWeatherPrediction/
index.html
Information on Meteorological and other Environmental Satellites (WMO‑No. 411)
Japan Aerospace Exploration Agency http://global.jaxa.jp/

Japan Meteorological Agency (MTSAT Series) http://www.jma.go.jp/jma/jma-eng/satellite/introduction/
MTSAT_series.html
Meteorological satellite systems of the Russian Federation http://sputnik.infospace.ru/
NASA Earth Observing Missions https://www.nasa.gov/topics/earth/index.html
National Satellite Meteorological Center (NSMC) (China) http://www.nsmc.org.cn/en/NSMC/Channels/
welcome.html
NOAA/NESDIS https://www.nesdis.noaa.gov/
Satellite Data Archiving, SAT-14 (WMO/TD-No. 909)
The Role of Satellites in WMO Programmes in the 2010s, SP-1 (WMO/TD-No. 1177)
WMO Space Programme http://www.wmo.int/pages/prog/sat/index_en.html
PART V. REDUCTION OF LEVEL I DATA
5.1 GENERAL
Various WMO manuals and guides define Level I data (primary data or instrument readings) and
Level II data (meteorological parameters, that is, nominal values) and provide the appropriate
recommendations concerning data reporting requirements.
Level I data, which comprise instrument readings and unprocessed measurements, are in general
instrument readings or sensor signals expressed in appropriate physical units and referred to
Earth coordinates. They require conversion to the meteorological variables specified in the data
requirements given in the Manual on the Global Observing System (WMO-No. 544). In general,
the conversion of Level I data into the corresponding meteorological variables is achieved by
applying the calibration functions and all systematic corrections. In some cases the process
involves more complex procedures.
Level II data, which include meteorological variables and processed data, are data obtained
directly from many types of simple instruments, or derived from Level I data.
Data exchanged internationally are supposed to be Level II data and Level III data (derived
meteorological variables). If Level I data meet the data-reporting requirement as defined in
WMO Manuals and Guides, no adjustment is needed. In such a situation both Level I and Level II
data have identical values.
In some cases, the WMO recommendations set out in the Manual on the Global Observing System
(WMO‑No. 544) and Part I of the Guide to Meteorological Instruments and Methods of Observation
(WMO‑No. 8) require adjustment of Level I data into Level II data with regard to more than one
aspect. The instrument value can be adjusted for the following reasons: representative height of
sensor above local ground, surface roughness, wind speed, temperature, evaporation or wetting
losses.
5.2 REDUCTION PROCESS
Meteorological stations installed by Members should meet the requirements for the variables
most commonly used in synoptic, aviation, and marine meteorology, and in climatology
summarized in the Guide to Meteorological Instruments and Methods of Observation (WMO‑No. 8).
Chapter 1, Part I, Annex 1.B, and 1.7.1.2, Chapter 1, Part III, contains information on automatic
weather stations; 1.3.2.6, Chapter 1, Part II, lists requirements and Chapter 3, Part III, provides
general information on data reduction. Details concerning corrections and reductions of
instrument values for meteorological variables can be found in Part I, whereas sampling
meteorological variables is fully discussed in Chapter 2, Part III. Automatic weather stations are
further discussed in 1.3.2, Chapter 1, Part II.
The introduction of new technology necessitates more than ever before a standardization of the
conversion of raw data into Level I data or Level I data into Level II data. Owing to the lack of such
approved standards and procedures, many commercial companies produce their own algorithms
and consider these as firmware, with no access by the users. Situations in which instrument
specialists lack necessary information, in particular in case of malfunction of microprocessors or
black boxes, should be avoided.
5.3 AVERAGING OF MEASURED QUANTITIES
Although it is common practice to report observational data, averaged over a specified time,
clear arguments in favour of averaging or a definition of the mathematical technique for
averaging have not yet been provided. The averaging of some variables, however, is defined
in Chapter 1, Part II, and in 3.6, Chapter 3, Part III, of the Guide to Meteorological Instruments and
Methods of Observation (WMO‑No. 8).
Two typical reasons may exist for averaging:
(a) To present a value that is more reliable in case of fluctuating, noisy measurements, whether
natural or artificial;
(b) To present a value with a higher measure of spatial representativeness.
In both cases, different mathematics may be chosen. For (a), a typical resistor-capacitor filtering
method, not an arithmetical mean based on a time window, reduces the noise. For (b), an
arithmetical mean based on a time window might be favourable, although the use of a constant
weighting factor is questionable. Moreover, the use of the median value for observations within
a period is favourable in some cases; therefore, calculating the arithmetic mean should not be
recommended in all cases. Averaging observational data to obtain Level II data is required in
many circumstances.
In the absence of a clearly defined regulation, the averaging of each observed value for
further reporting should be based on a well-defined method, to be substantiated with sound
arguments. Mathematical calculus to be used should be described and explained in detail.
References

PART VI. DATA QUALITY CONTROL
6.1 GENERAL
Meteorological observations are exchanged between countries on a worldwide basis. Users need
to be confident that the observations they receive from other countries are made according to
agreed standards set by WMO. The accuracy of the data is of primary importance to many kinds
of analyses, computations and scientific investigations. Therefore, the need for quality control
of observational data is linked with the fundamental importance of obtaining consistent and
accurate data for their optimum use by all possible users, including the World Weather Watch
Programme, and international research programmes.
Data quality is a measure of how well data serve the purpose for which they were produced.
All data are produced for a purpose, and their quality is directly tied to whether they meet the
requirements of that purpose. Although data quality addresses the appropriateness of the data
for a specified use, there is no reason why the data cannot be put to a different use as long as the
data user understands the original requirements and is confident that the data can meet current
application requirements.
Data quality assessment is conducted during production against the producer’s specifications.
Data quality assessment is inherently complex and cannot be represented by a simple numeric
value. Rather, it should be indicated by the sum of bits of information about the data that are
captured during the data production process and made available to the data user as metadata.
There are two keys to the improvement of data quality: prevention and correction. Error
prevention is closely related to the collection of the data and the entry of the data into a
database. Although considerable effort can and should be given to error prevention, the fact
remains that errors in large data sets will continue to exist, and data validation and correction
cannot be ignored.
It is better to prevent errors than to have to rectify them later, and it is by far the cheaper option.
Making corrections retrospectively can also mean that the incorrect data may have already been
used in a number of analyses before being corrected. Preventing errors does not affect those
errors already in the database; however, data validation and cleaning remains an important part
of the data quality process.
All means of automatic error monitoring should be used to recognize errors in advance before
they affect the processed values.
The basic characteristics of quality control and general principles to be followed within the
framework of the Global Observing System are described briefly in the Manual on the Global
Observing System (WMO‑No. 544).
The purpose of Part VI is to provide supplementary information and describe in more detail the
practices, procedures and specifications which Members are invited to follow for quality control
of observations made by their respective National Meteorological Services.
Recommendations provided below should be used in conjunction with the relevant WMO
documentation dealing with data quality control.
Details of quality control procedures and methods to be applied to meteorological data intended
for international exchange are described in Chapter 6 of the Guide on the Global Data-processing
System (WMO‑No. 305).
Minimum Global Data-processing and Forecasting System standards for data quality control are
defined in section 2, Volume I, Part II, of the Manual on the Global Data-processing and Forecasting
System (WMO‑No. 485).
Basic steps for quality control of data relating to automatic weather stations are given in Chapter
1, Part II, of the Guide to Meteorological Instruments and Methods of Observation (WMO‑No. 8);
more general instructions are given in Part III of the aforementioned publication.
6.1.1 Levels of application of quality control procedures
Observational data must be quality controlled at different levels of data pre-processing and
processing and transfer in real time and non-real time, using various procedures.
The levels of quality control procedures are as follows:
(a) The observing site, starting with data acquisition by manual or automatic meteorological
stations;
(b) Data collection centres, prior to the transmission of observational data over the Global
Telecommunication System;
(c) GTS centres (standard telecommunication procedures, for example, control of timeliness
and data format);
(d) GDPFS centres and other available facilities.
Within the framework of the Global Observing System, quality control is restricted to items (a)
and (b) above; therefore, the instructions and guidance provided in the Guide are concerned
with observing sites and collection centres only.
Although the Global Observing System deals solely with Level I data and their reduction and
conversion into Level II data, quality control should be performed at all stages until the data are
transmitted over the Global Telecommunication System.
The reliability and accuracy of meteorological observations, the causes of observation errors
and methods of preventing such errors are within the scope of the areas of quality control under
discussion.
There are two levels involved in checking the quality of real-time observational data:
(a) Quality control of raw data (Level I data)—basic quality control performed at the observing
site. This quality control level is relevant during acquisition of Level I data and should
eliminate errors of technical devices, including sensors, systematic or random measurement
errors and errors inherent in measurement procedures and methods. Quality control at this
stage includes the following tasks: a gross error checks, basic time checks and basic internal
consistency checks. Application of these procedures is extremely important because some
errors introduced during the measuring process cannot be eliminated later.
(b) Quality control of processed data—extended quality control, partly performed at an

observing site, but mainly at a National Meteorological Centre. This level is relevant during
the reduction and conversion of Level I data into Level II data and Level II data themselves.
It deals with comprehensive checking of temporal and internal consistency, evaluation of
biases and long-term drifts of sensors and modules, malfunction of sensors and the like.
Quality control levels are outlined as follows:
(a) Basic quality control procedures to be carried out at a station:

(i) Automatic quality control of raw data:
a. Plausible value check: gross error check on measured values;
b. Check on a plausible rate of change: time consistency check on measured values;
(ii) Automatic quality control of processed data:
a. Plausible value check;
b. Time consistency check:
PART VI. DATA QUALITY CONTROL 185
i. Check on a maximum allowed variability of an instantaneous value, or step

test;
ii. Check on a minimum required variability of instantaneous values or
persistence test;
iii. Calculation of standard deviation;
c. Internal consistency check;
d. Technical monitoring of all crucial parts of a station;
(b) Extended quality control procedures to be conducted at a National Meteorological Centre:

(i) Plausible value check;
(ii) Time consistency check:
a. Check on a maximum allowed variability of an instantaneous value or step test;
b. Check on a minimum required variability of instantaneous values or persistence
test;
c. Calculation of standard deviation;
(iii) Internal consistency check.
Quality control procedures of Level II data and Level III data should be implemented at the
National Meteorological Centre to check and validate the integrity of data, namely completeness,
correctness and consistency of data. The checks performed at the observing station must
be repeated at the Centre, but in a more elaborate, sophisticated form. This should include
comprehensive checks against physical and climatological limits, time consistency checks for a
longer measurement period, checks on logical relations among a number of variables, or internal
consistency of data, and statistical methods for analysing data.
Quality control procedures and techniques, pre-processing checks, checks for surface data and
upper-air data, flagging, computer program design and combined quality control for data Levels
II and III are described in detail in the Guide on the Global Data-processing System (WMO‑No. 305).
Quality control can be carried out manually and automatically. In principle, all the necessary
quality control procedures can be applied manually, but in general the time needed to do so
is unacceptably long. No appreciable delay should be caused by quality control because the
data must be transmitted in real time for operational use. The real-time quality control at the
observing point is, however, of paramount importance since many of the errors introduced
during the observation process cannot be eliminated later.
The real-time activity of quality control within the Global Observing System includes data up
to one month old for land or sea stations. This applies particularly to the monthly CLIMAT and
CLIMAT TEMP messages and BATHY/TESAC reports.
6.1.2 Observational errors
The entire range of observational errors can be divided into the following three main groups:
(a) Errors of technical devices, including instruments;

(b) Errors inherent in observing procedures and methods;
(c) Subjective random or systematic errors on the part of observers and collecting operators.
Several types of errors involving measured data can arise:
(a) Random errors are distributed more or less symmetrically around zero and do not depend
on the measured value. They sometimes result in overestimation and underestimation of
the actual value. On average, the errors cancel each other out.
(b) Systematic errors on the other hand, are distributed asymmetrically around zero. On
average these errors tend to bias the measured value above or below the actual value.
Systematic errors can be caused by a long-term drift of sensors, or a sensor without valid
calibration.
(c) Gross or large errors are caused by malfunctioning of measurement devices or by mistakes
made during data processing; errors are easily detected by checks.
(d) Micrometeorological (representativeness) errors are the result of small-scale perturbations

or weather systems, for example, turbulence, affecting a weather observation. These
systems are not completely observable by the observing system owing to the temporal or
spatial resolution of the observing system. Nevertheless when such a phenomenon occurs
during a routine observation, the results may appear strange compared to surrounding
observations taking place at the same time.
(e) Measurement errors in observations cannot be eliminated completely; the problem is

reducing them to an acceptable level. A measurement error can be regarded as the sum of
the above-mentioned types of errors. For further reference, see 1.6.1.2 and 1.6.2, Chapter 1,
6.2 PROCEDURAL ASPECTS OF QUALITY CONTROL
6.2.1 Responsibility and standards
The primary responsibility for the quality control of observational data and the determination of
their quality rests with the National Meteorological Service from which the data originate. Data
producers should ensure that the following procedures are observed:
(a) Quality control procedures are implemented and exercised during data acquisition;
(b) Data and data quality are adequately and accurately documented;
(c) Validation checks are routinely carried out on all observational data;
(d) Validation checks are fully documented;
(e) Data are available in a timely and accurate manner with documentation that allows users to
determine fitness for use;
(f) Feedback from users on the data quality is dealt with in a timely manner;
(g) Data quality is maintained at the highest level at all times;
(h) All known errors are fully documented and made known to users.
Therefore, it is crucial that Members make adequate provision for data quality control to ensure
that they are as free from error as possible and the data quality is known at every level of the
data-obtaining process.
According to the Manual on the Global Observing System (WMO‑No. 544), Members must
implement minimum standards of real-time quality control at all levels for which they are
responsible, for example, observing stations and National, Regional and World Meteorological
Centres. In addition, the Manual on the Global Data-processing and Forecasting System
(WMO‑No. 485) recommends that they do so before data received via telecommunication links
are processed.
Recommended minimum standards of real-time quality control at the observing station and the
National Meteorological Centre levels are similar to those provided in section 2, Volume I, Part II,
of the Manual on the Global Data-processing and Forecasting System (WMO‑No. 485).
6.2.2 Scope of quality control
The quality control system implemented by the National Meteorological Service should include
data validation, data cleaning and quality control monitoring.
(a) Data validation
Validation is a process used to determine whether data are inaccurate, incomplete, inconsistent
or unreasonable. The process may include completeness checks, plausible value checks and
time and internal consistency checks. These processes usually result in flagging, documenting
and subsequent checking of suspect records. Validation checks may also involve checking for
compliance against applicable standards, rules and conventions. A key stage in data validation is
identifying the causes of the errors detected and preventing those errors from re-occurring.
The quality of data should be known at any point of the validation process and can change
with time as more information becomes available. Quality control flags should be used for that
purpose.
(b) Data cleaning, or remedial actions
Data cleaning refers to the process of “fixing” data errors that have been identified during the
validation process. The term is synonymous with “data cleansing”, although some use data
cleansing to encompass both data validation and data cleaning. It is important that data is not
inadvertently lost in the data cleaning process, and that changes to existing information be
carried out with care. It is reasonable to retain both the original data and the corrected data
side by side in the database so that if mistakes are made in the cleaning process, the original
information can be recovered.
Data cleaning includes defining and determining error types, searching and identifying error
instances, correcting errors, documenting error instances and error types and modifying data
entry procedures to reduce future errors.
(c) Quality control monitoring
See 3.1.3.14, Part III, of the present Guide on network performance monitoring, which can be
applied and implemented by National Meteorological Services for all types of observational data,
and Appendix VI.1.
The procedures mentioned above should be partly applied at observing stations, but primarily at
the relevant collection centres.
6.2.3 Implementation
Standardization of quality control procedures to be implemented at observing sites within

the framework of the World Weather Watch is far from complete. The growing volume of
international exchange of observational data calls for urgent measures to ensure that the
meteorological information originating in different countries is of comparable quality. The
practical steps to be taken for this purpose must aim at the maximum possible standardization.
See Appendix VI.2, Guidelines for quality control procedures applying to data from automatic
weather stations.
To ensure that national obligations within the framework of the World Weather Watch system
are met, regulatory material is required on tasks of meteorological stations and weather offices
with regard to checking and correcting synoptic observational data. Both manual and automatic
quality control methods may be used. Standard software modules are being developed to
support computer-based quality control at observing sites or collection centres.
6.2.3.1 Manual methods
If manual quality control methods are used, a Member should include the following procedures
in its national instructions or regulations on real-time monitoring of surface and upper-air
observations before transmitting them over the Global Telecommunication System:
(a) Surface synoptic observations
Quality control and monitoring of the observing programme is to be carried out at the National
Meteorological Centre for all stations concerned, at least those included in the regional basic
synoptic network at main and intermediate standard times.
(b) Upper-air observations
Upper-air observations from stations of the regional basic synoptic network are processed and
encoded either at the station itself or centrally. Quality control and monitoring should be carried
out starting at the station and at the other centres responsible for processing or transmitting the
data.
Data checking is applied to the following areas:
(a) Occurrences of hyperadiabatic vertical gradients of temperature in the free atmosphere;
(b) Comparison of some values in selected points against temporal change since the last
observing time and against values obtained by interpolation from data of neighbouring
stations;
(c) Checking of the vertical wind profile against inhomogeneities.
Clearly, erroneous TEMP data should be fully or partially rejected from distribution, depending
on where errors occur. Should minor errors be detected after checking, these TEMP data are
corrected manually. In any case, corrections should be distributed nationally marked as corrected
messages. Checked and controlled TEMP data are distributed internationally in accordance
with transmission schedules and are therefore only marked as retard messages if data cannot be
corrected and distributed in time.
After the data have been examined and checked at the station, quality control and correction—if
necessary and possible—of all national meteorological data prior to their transmission over the
Global Telecommunication System should be carried out manually at the relevant centres.
Monitoring the process of observations, timing, encoding and transmission is performed by

the National Meteorological Centre in real time and non-real-time, the latter aiming at ensuring
improved quality for non-real time data. Checks and appropriate remedial actions are conducted
in the event of missing or delayed messages, observational errors and format errors.
6.2.3.2 Automated methods
Automated quality control of large quantities of meteorological data has become essential and is
now possible via computer systems and quality control programmes.
The main advantages of automatic quality control procedures within the scope of their natural
limits are as follows:
(a) Objectivity and repeatability;
(b) Uniformity;
(c) Possibility of using complex control parameters and practically unlimited specifications;
(d) Elimination of tedious checking of huge amounts of correct data;
(e) Close supervision of the quality control results on display units by experts so that any
possible errors can be rapidly analysed.
The principles of organizing automated quality control of meteorological data depend on the
development stage of algorithm methods for quality control and automatic data acquisition.
The algorithms used by Members for quality control at an observing station are similar. In most
cases they are based on physical and/or climatological interdependence and some statistical
relations. There is, however, a requirement for further increasing the efficiency of the algorithms
and programmes in use, and it is recommended that Members exchange information on their
experience with the methods used in order to benefit others.
6.3 OTHER QUALITY CONTROL PROCEDURES
A number of minor improvements can be made in an observing station to help ensure that the
observers carry out their duties correctly. The following is not intended to be an exhaustive list,
but can serve as a guide to the many ways in which quality control can be expected, especially at
single-observer stations.
6.3.1 Availability of statistics on variables
In practice, checking programmes are designed to reveal all gross and substantive errors which
recur regularly. The detection of rare and irrelevant measurement errors is not worthwhile, as the
reliability of tests varies in inverse proportion to the size of errors they are designed to detect. In
other words, the degree of accuracy of the check tests varies—there is always the risk of missing
an error or mistaking a correct value for a doubtful one. Most tests have been compiled on the
basis of experience; they are the result of practical intuition or statistical analysis.
The statistics on variables must be made available at the observation site in order to enable the
observer to compare the current observation with the statistics on past events at the station
concerned. Such statistics may be essential in detecting equipment malfunctions. There will
always be a need for a complete check of the overall results before the message is dispatched,
even for a fully automatic observing system.
6.3.2 Use of accepted abbreviations
In order to reduce subjective errors during the recording of visual observations and instrument
readings, the human observer should make use of accepted abbreviations. They should be
standardized within a National Meteorological Service and laid down in national observing
instructions.
6.3.3 Pictorial representations and diagrams
A plot of hourly values of the main variable can be used by the observer to detect gross errors.
Variables subject to this type of error are pressure, temperature and dew point; however, the
plotting of other variables may also be useful in this respect. Care should be taken to keep a
certain number of plots of the different variables on the log sheet at any time and to use the same
sheet the next day. Simultaneous plots of not less than six entries are required to maintain an
adequate history.
A chart showing the average diurnal change in temperature representative of a recent period
of several years and relevant to the observational data can also be useful to the observer as
a safeguard against gross errors. In areas where large variations from such averages do not
occur, the succession of hourly readings as indicated in the previous paragraph can provide
an additional check. If the observed value appears incompatible with the average value, the
observer should take a further independent reading. The existence of these averages can also
stimulate additional interest in the observer, who will then be able to recognize those occasions
which are not typical or are likely to be significant.
In tropical areas, a chart showing the average diurnal pressure change will provide a useful
guide, familiarizing the observer with the magnitude of the changes to be expected.
In making cloud reports, the observer is confronted with a wide variety of cloud types and
complicated regulations. The inexperienced observer should use flow diagrams to enable the
correction of a report. These diagrams are comparatively simple to construct and are available in
Volume I of the International Cloud Atlas (WMO‑No. 407).
Observers should have access at all times to publications dealing with observing and coding
procedures. Furthermore, the procedures should be displayed whenever possible as visual
reminders of their responsibilities. Pictorial representations and diagrams are generally more
effective, but when lists or tables are necessary they should be placed in a prominent position.
6.3.4 Simplified mathematical checks
Experience has shown that the greatest number of errors are detected at the pre-processing
stage, which precedes the analysis. Mathematical checks are made of upper-air and surface
synoptic data received on an operational basis over communications channels by data-
processing centres.
Simplified mathematical checks are used during the comparison of observed values with their
approximated values. The comparison may be carried out both vertically over the observation
site and in a horizontal plane between data for the same level.
The following are used as approximated values:
(a) Values of the meteorological variables computed using the hydrostatic equation assuming
polytropy of the atmosphere or linear variability within layers, and so forth (hydrostatic
control);
(b) The value and sign for temperature in adjacent layers, assuming a maximum gradient;
(c) Interpolated values which may vary depending on how they were observed;
(d) Forecast values for the actual period;
(e) Approximated values, mainly polynomials of the second and third degree;
(f) Average value for groups of adjacent stations;
(g) Local statistical parameters of meteorological variables calculated on the basis of empirical
relationships (regression equation).
Many control procedures make it possible not only to detect errors but to correct them or to
reconstruct omitted observations.
References
Guide to the Applications of Marine Climatology (WMO‑No. 781)

Guidelines on Climate Metadata and Homogenization (WCDMP-No. 53; WMO/TD-No. 1186)
Guide on the Global Data-processing System (WMO‑No. 305)

Guide to Practices for Meteorological Offices Serving Aviation (WMO‑No. 732)
International Cloud Atlas (WMO‑No. 407)
Manual on the Global Data-processing and Forecasting System (WMO‑No. 485)
Manual on Marine Meteorological Services (WMO‑No. 558)
APPENDIX VI.1. DATA QUALITY CONTROL
1. QUALITY CONTROL OF DATA FROM THE SURFACE-BASED SUBSYSTEM
1.1 General
Various quality control methods are used for surface synoptic data, that is, data related to
standard observational times. These include horizontal, vertical, three-dimensional, time and
hydrostatic checks and a combination of these methods.
1.2 Tests using statistical structure parameters within an interpolation

scheme
1.2.1 Horizontal checks
Horizontal checks can be performed with the aid of methods used for objective analysis by
means of optimum interpolation for every station according to data from several—usually four
to eight—surrounding stations and comparison of the interpolated value against the observed
value. The interpolation is made according to the following formula:
n
f0' = ∑ ρ i f i ' (1)
i =1
where f0' is the interpolated value of deviation ƒ' of the element ƒ from its normal value at the
station in question; fi are the observed deviations of this element from its normal value at the
surrounding stations; ρi are the weight multipliers found by solving the equation system.
n
∑m ρ
j =1
ij j + δ 2 ρ i = mio (2)
where mij are covariances describing statistic relation of the element ƒ values at different points,
namely:
mij = f i ' f j' (3)

(the bar above indicates statistical averaging), δ is the mean square of the observational errors. It
is rational to perform control by proceeding not from the absolute value of residue f0i − f 0'
between the interpolated and observed values, but from the ratio:
f0' − f 0'
R= (4)
∆0
where Δ0 is the mean square difference between f0' and f 0' which can be computed according to
the formula below:
n
∆ 0 = moo − ∑ ρ i moi + δ 2 (5)
i =1
after weights ρi definition. If Δ0 does not exceed some critical value K, the data are recognized
to be correct; otherwise an error is assumed to exist. According to available data, K = 4 may be
used; this ensures that the correct values are never called into question. In light of the above
assumption, it is perhaps advisable to use smaller values.
The inequality Δ0 > K indicates with a rather high probability that the presence of errors can arise
not only from an erroneous value ƒ0 at the checked station, but also from an erroneous value
at the surrounding stations, especially in the case where the weight ρi for this value is large.
Therefore, it is first necessary to perform a similar check for the surrounding stations to ensure
that the erroneous value has been detected. In most cases this procedure ensures the indication
of the erroneous value and its replacement by the interpolated value. When the values are
erroneous at several surrounding stations, this method is not applicable.
1.2.2 Vertical checks
Vertical checks are also based on a comparison of the observed value with the interpolated one.
However, this interpolation is not carried out with data from the neighbouring stations related
to the same level, but rather with data from the same station related to other levels. As the
covariances in the vertical do not possess characteristics of homogeneity and isotropy, the values
of mij (3) entering formulae (2) and (5) do not depend on the distance between levels, but rather
on the height (pressure) of both levels.
Vertical checking and horizontal checking with respect to upper-air data is used mainly for the
geopotential. However, with information on covariance matrices it is easy to tell how this method
might be used for other meteorological elements as well.
1.2.3 Three-dimensional checks
Three-dimensional checks are performed by comparing the observed value with the value
interpolated from data on several levels, both at the station in question and at neighbouring
stations. Since the data on the three-dimensional statistical structure of a number of basic
meteorological fields are available, the use of this procedure should pose no difficulty.
1.2.4 Time checks
Time checks involve data from the current and previous observational times. However, this
control method requires time extrapolation instead of interpolation. It is therefore expedient to
use e-data computed from previous observations as a reference for checking the results of the
numerical prognosis for the time under consideration.
1.2.5 Hydrostatic checks
Hydrostatic checks are based on the use of the static equation or the geopotential barometric
formula to show that the geopotential and temperature at different isobaric surfaces are
internally consistent. The essence of this method is as follows:
Integrating the hydrostatic equation:

∂H R T
=− (6)
∂r 9.8 r
for the layer located between two adjacent isobaric surfaces ρn and ρn + 1 and turning from the
absolute temperature T to the temperature in degrees Celsius, result in:
R ρn +1
9.8 ∫ρn
H n +1 − H n = An + td ln ρ (7)
where R is the gas constant for air, An is the layer thickness at 0°C.
Assuming the mean temperature t in the layer is equal to the arithmetic mean of its boundary
values, (7) can be reduced to the following form:
H n +1 − H n = An + Bn ( tn + tn +1 ) (8)
The numerical values of coefficients An and Bn are listed in the table, Estimate of static check
potentialities.
As the upper-air reports contain information on both the geopotential and temperature of
isobaric surfaces, ratio (8) can be used to check this information. For this purpose the difference
between the values of the left- and right-hand sides of relation (8) need to be computed for each
layer located between adjacent mandatory isobaric surfaces.
δ n = H n +1 − H n − An − Bn ( tn + tn +1 ) (9)
In addition, the values of δn need to be compared against its tolerable values Δn The latter can be
estimated empirically by handling a large number of observations and theoretically.
These estimates, as well as the estimates of the given method in general, depend to a large
extent on the method used for defining the temperature and geopotential of the isobaric
surfaces at each station.
A theoretical estimation of the tolerable discrepancies in the hydrostatic check is more difficult
than for the methods described above, for these discrepancies have a variety of causes, namely:
random errors of temperature measurement, random and systematic deviations of the vertical
temperature profile t (ln p) from the linear one and rounding-off errors in the geopotential
computation. The joint action of random observational errors and random deviations t (ln p) from
the linear variation can be estimated with the aid of the formula:
moo δ2 1
∆1 = 1+ − 2 4 − ( 2 − ln q ) q (10)
2

2 moo ln q
where Δ1 is the mean square of the corresponding discrepancy δ, q is the correlation coefficient
between temperature values of the particular adjacent surfaces.
Estimate of static check potentialities
p hPa 1000 850 700 500 400 300 200

An, m 1300 1553 2692 1786 2302 3244 5546
Bn, m/ 2.38 2.84 4.03 3.27 4.21 5.94 10.15

degree
KΔ1, m 20 24 35 9 19 70 64
KΔ, m 1 1 6 10 10 10 10
KΔ1 + Δ2, m 20 25 41 19 29 80 74
Δ, m 30 30 40 30 40 100 120
The values of KΔ1 cited in the above table are computed for the winter season; they are somewhat
smaller for the summer season. The coefficient K is taken rather large (K = 3.5), because in
addition to large-scale disturbances of the temperature vertical profile, taken into account in
formula (10), mesoscale disturbances can take place.
The values Δ2 given in the Table above represent maximum discrepancies resulting from
rounding-off geopotential values, while Δ are the tolerable discrepancies of equation (8) defined
empirically. In comparison with KΔ1 + Δ2, natural deviations of t (ln p) from its linear form due to
great curvature of the profile of t near the Earth’s surface take place in the 1 000–850 hPa layer
and in the 300–200 and 200–100 hPa layers due to the presence of the tropopause. This clearly
explains the difference between Δ and KΔ1 + Δ2 for the indicated layers. As for the 500–400 and
400–300 hPa layers, the values of Δ can perhaps be smaller.
In addition to the detection of errors exceeding tolerable discrepancies, the static checking
procedure makes it possible to determine the source of the error and consequently to correct it.
This is true because errors from different causes result in diverse combinations of discrepancies.
For example, an error in a geopotential value due to transmission garbling will result in
discrepancies in equation (8) for the two adjacent layers, which are equal to this error and have
opposite signs. Garbling of a temperature value will cause two discrepancies of the same sign
proportional to the Bn coefficients. In addition, an error in computing the thickness of a layer
results in a discrepancy for that layer only.
1.2.6 Combined checks
Combined quality control checks are carried out by means of various control methods not merely
consecutively, but in close interrelation. This is necessary, first because no single control method
is sufficient to detect and correct all the erroneous information and second, because different
methods react to the errors in different ways, depending on the source.
The potential success of quality control increases significantly if several methods are used in
combination, that is, if conclusions are drawn as to the character and the value of a certain error
using results obtained from all the methods. This enables the detection of the source of the
error; its localization, i.e. determining which of the suspected values is erroneous; definition
of its numerical value and its correction. For example, the error localization obtained by the
combined use of the horizontal and vertical checks is attained by the following means: the value
of the tolerable discrepancy for a station changes during a horizontal check, based on the results
of a previous, vertical check, for example. If the vertical check indicates an error, the tolerable
discrepancy for the horizontal check must be diminished.
2. QUALITY CONTROL OF DATA FROM THE SPACE-BASED SUBSYSTEM
2.1 General
Indirect sounding data from satellites refer to a rapid succession of non-standard observational
times, that is, the data are asynoptic. In order to check, take into account and assimilate asynoptic
information for the objective analysis of meteorological fields, data should be used which refer to
various points in space and time. In other words, a conversion from a space (three-dimensional)
analysis of meteorological fields to a four-dimensional space-time analysis is required.
At the same time, indirect sounding data possess at least three more features that distinguish
them from upper-air sounding data. First, the former give space–mean values, i.e. the scale of
averaging is considerably larger than for values obtained by means of conventional upper-air
sounding. Second, satellite-based instruments operate under more complicated conditions in
comparison with upper-air sounding instruments, and conversion from spectral intensity to
temperature and geopotential is approximate. Therefore, the errors of indirect-sounding data are
greater than those of radiosondes. Third, all satellite measurements are performed by only one
set of instruments during the satellite’s lifetime; the errors of indirect sounding at various points
should therefore be intercorrelated.
The above-mentioned properties of indirect sounding data allow the validity of the data to be
estimated during—not before—the four-dimensional analysis and control.
2.2 Estimation of data reliability
The methods for estimating the reliability of the four-dimensional analysis for data-control
purposes, taking into account both the asynoptic nature of the indirect sounding data and the
value of sounding errors and degree of their correlation, are examined below. It is assumed that
four-dimensional assimilation of the information is made with the aim of optimum time–space
interpolation. If the observational errors are intercorrelated and do not correlate with the true
values of the observed meteorological elements ƒ, the equation of the optimum interpolation
method for determining the weight multipliers ρi takes the following form:
n
∑(µ
j =1
ij i jν )ρ j = µoi ( i = 1, 2, n ) (11)
+ ηη
where μij is the correlation coefficient between the true values of ƒ of two stations with indexes i
and j; μoi is the correlation coefficient between the true value of ƒ of the station with index i and
the unknown value of this element at the point o; n is the quantity of data used for interpolation;
ηi2 is the mean square of the observational error of this meteorological element divided by its
dispersion σ 2; vij is the correlation coefficient between observational errors of two stations with
indexes i and j.
Having computed the weights ρi by solving equation (11), it is easy to perform the interpolation
according to the following formula:
n
f'0 = ∑ ρ i fi ' (12)
i =1
where the values with the prime sign imply deviations of ƒ from its mean of climatological value
(normal) ƒ, the sign ~ refers to the observed values of the element, in contrast to the true values,
and ^ refers to the result of the interpolation, also in contrast to the true values.
Having solved system (11), the mean square interpolation error can be estimated by using the
following formula:
n
ε 2 = 1 − ∑ µoi ρ i (13)
i =1
where ε is the degree of interpolation error, i.e. the mean square of the interpolation error
2
divided by σ 2; the value σ is supposed to be constant.
The spatial correlation functions of basic meteorological elements can be considered

homogeneous and isotropic in the horizontal plane or along isobaric surfaces, i.e.
µij = µ( rij ) (14)

may be assumed, where rij is the distance between two stations with indexes i and j; μ(r) is the
function of the given form.
Formula (14) is correct if both observations refer to the same isobaric surface and are performed
at the same time. The more general hypothesis on time–space (horizontal-time) homogeneity
and isotropy can therefore be implemented with sufficiently accurate results. If the distance
between two stations is equal to rij and the time interval between observations is Tij, the
following formula may be used:
µij = µ ( )
rij2 + c 2τ ij2 (15)
where c is the constant denoting velocity. For the surface pressure c ≅ 35 km h-1. This value of c
will be used further on.
It can now be assumed that part of the basic data, for example, data from stations with indexes
i = 1, 2, ... k, is obtained by means of conventional radiosondes, while the rest (i = k+1, k+2, ... n) is
obtained indirectly. The errors of radiosondes are considered to be white noise, that is, they do
not correlate between each other or with indirect sounding errors:
1 at j = 1;
ν ij = 0 at i = 1, 2, k ; j ≠ 1, 2, n; j ≠ i; (16)
0 at j = 1, 2, k ; i ≠ 1, 2,n; i ≠ j;
The mean square, and consequently the degree of radiosonde error, is taken to be equal for all
the radiosonde points:
η1 = η2 = ηR = η (17)
The degree of indirect sounding error is also identical to (17), but different from direct-sounding
errors:
ηR +1 = ηR + 2 = ηn = η ' (18)
System (11) then becomes:
n
∑µ ρ
j =1
ij j + η 2 ρ i = µoi ( i = 1, 2 R )
k n
∑ µij ρ j + ∑ ( µij + η 2ν ij ) ρ j = µij (19)
j =1 j = k +1
( i = k + 1, K + 2, n )
For example, if k = 2 and n = 5, the matrix of coefficients of system (19) (taking into account that
μji = μij and νji = νij ) is of the form:
1 + η 2 µ12 µ13 µ14 µ15 

 
 µ12 1 + η µ23 µ 24 µ24
2

 µ13 µ23 1 + η 12 µ34+η12ν µ35+η12ν 
 34 35

 µ14 µ24 µ34+η12ν 1+η 12
µ45+η12ν 
 34 45

 µ15 µ25 µ35+η12ν µ45+η12ν 1 +η 12

 35 45 
As regards the correlation coefficients νij between indirect sounding errors, the latter are
accepted as being dependent only on the distance between points. Here the limiting case is of
interest when these errors are of the “black noise” character, i.e. when for all i and j exceeding k:
ν ij = 1 ( i, j = k + 1, k + 2, n ) (20)
Therefore, the criteria of the quality control performed during the four-dimensional analysis or
assimilation of information do not, for all practical purposes, differ from those stated in 1.2.1 of
this Appendix, based on the application of the optimum interpolation. The essence of the quality
control itself consists, as previously, in the comparison of the residue between the interpolated
and reported values against the tolerable discrepancy and—depending on their relation—the
determination of the validity or erroneousness of the checked data.
APPENDIX VI.2. GUIDELINES FOR QUALITY CONTROL PROCEDURES
APPLYING TO DATA FROM AUTOMATIC WEATHER STATIONS
INTRODUCTION
Quality control of data is the best known component of quality management systems. It consists
of the examination of data at stations and at data centres with the aim to detect errors. Data
quality control must be applied as real-time quality control performed at automatic weather
stations (AWSs) and data processing centres (DPCs). In addition, it must be performed as near
real-time and non real-time quality control at DPCs.
There are two levels of real-time quality control of AWS data:
(a) Quality control of raw data (signal measurements), which is basic quality control performed
at an automatic weather station site. This level is relevant during acquisition of Level I data
and should eliminate errors of technical devices, including sensors; systematic or random
measurement errors and errors inherent in measurement procedures and methods. Quality
control at this stage includes gross error checks, basic time checks and basic internal
consistency checks. Application of these procedures is extremely important because some
errors introduced during the measuring process cannot be eliminated later.
(b) Quality control of processed data, which is extended quality control, partly performed at an
automatic weather station site, but mainly at a data processing centre. This level is relevant
during the reduction and conversion of Level I data into Level II data, and Level II data. It
deals with comprehensive checking of temporal and internal consistency, evaluation of
biases and long-term drifts of sensors and modules, and malfunction of sensors, among
others.
Quality control levels are outlined below:
(a) Basic quality control procedures (automatic weather stations)

(i) Automatic quality control of raw data
a. Plausible value check, or gross error check on measured values
b. Check on a plausible rate of change, or time consistency check on measured
values
(ii) Automatic quality control of processed data
a. Plausible value check
b. Time consistency check
i. Check on maximum allowed variability of an instantaneous value, or step
test
ii. Check on minimum required variability of instantaneous values, or
persistence test
iii. Calculation of standard deviation
c. Internal consistency check
d. Technical monitoring of all crucial AWS components
(b) Extended quality control procedures (data processing centres)
(i) Plausible value check
(ii) Time consistency check:
a. Check on a maximum allowed variability of instantaneous value, or a step test
b. Check on minimum required variability of instantaneous values, or persistence
test
c. Calculation of standard deviation
(iii) Internal consistency check.
When quality control procedures are applied to AWS data, the data are validated and flagged,
and if necessary, estimated or corrected. If original value is changed as a result of quality control
practices, it is strongly advisable that it be preserved with the new value. A quality control system
should include procedures for returning to the source of data, or original data, to verify them and
prevent recurrence of errors. All possibilities for automatic monitoring of error sources should be
used to recognize errors in advance before they affect the measured values.
The quality of data should be known at any point of the validation process, and the quality
control flag can be changed throughout the process as more information becomes available.
Comprehensive documentation on quality control procedures applied, including the

specification of basic data processing procedures for a calculation of instantaneous—one
minute—data and sums, should be a part of standard documentation at automatic weather
stations.
The guidelines deal only with quality control of data from a single automatic weather station;
therefore spatial quality control is beyond the scope of this publication. The same also applies
to checks against analysed or predicted fields. Furthermore, quality control of formatting,
transmission and decoding errors is beyond the scope of this publication owing to a specific
character of these procedures, as they are dependent on the type of message and transmission
method used.
Notes:
1. Recommendations provided in the guidelines should be used in conjunction with the relevant WMO
documentation dealing with data quality control.
2. Basic characteristics of quality control and general principles to be followed within the framework of the Global
Observing System are briefly described in the Manual on the Global Observing System (WMO-No. 544). Quality
control levels, aspects, stages and methods are described in the Guide.
3. Basic steps of quality control of AWS data are given in the Guide to Meteorological Instruments and Methods of
Observation (WMO-No. 8), especially in Chapter 1, Part II.
4. Details of quality control procedures and methods that have to be applied to meteorological data intended
for international exchange are described in Chapter 6 of the Guide on the Global Data-processing System
(WMO‑No. 305).
5. The Global Data-processing System’s minimum standards for quality control of data are defined in Volume I of the
Manual on the Global Data-processing and Forecasting System (WMO-No. 485).
1. DEFINITIONS
Quality assurance, quality control
Quality assurance and quality control are two terms that have many interpretations because of
the multiple definitions of the words “assurance” and “control”.
Quality assurance: All planned and systematic activities implemented within the quality system,
and demonstrated as needed, to provide adequate confidence that an establishment will meet
quality requirements.
The primary objective of the quality assurance system is to ensure that data are consistent, meet
data quality objectives and are supported by comprehensive description of methodology.
Quality control: The operational techniques and activities that are used to fulfil requirements for
quality.
The primary purpose of quality control of observational data is missing data detection, error
detection and possible error corrections.
Quality control of observational data consists of examination of data at stations and data centres
to detect missing data and errors; data are validated and flagged and if necessary, estimated
or corrected, in order to eliminate the main sources of errors and ensure the highest possible
standard of quality for the optimum use of these data by all users.
To ensure the high quality of AWS data, a well-designed quality control system is vital. Efforts
shall be made to correct all erroneous data and validate suspicious data detected through quality
control procedures. The quality of AWS data shall be known.
Types of errors
Several types of errors, described below, can occur with regard to measured data. They shall be
detected by implementing quality control procedures.
Random errors are distributed more or less symmetrically around zero and do not depend on the
measured value. Random errors sometimes result in overestimation or underestimation of the
actual value. On average, the errors cancel each other out.
Systematic errors are distributed asymmetrically around zero. On average these errors tend to
bias the measured value either above or below the actual value. Systematic errors can be caused
by a long-term drift of sensors or sensor without valid calibration.
Gross errors are caused by malfunctioning of measurement devices or by mistakes made during
data processing; errors are easily detected by checks.
Micrometeorological (representativeness) errors are the result of small-scale perturbations or

weather systems affecting a weather observation. These systems are not fully observable by
the observing system owing to its temporal or spatial resolution. Nevertheless when such a
phenomenon occurs during a routine observation, the results may appear strange compared
with surrounding observations taking place at the same time.
2. BASIC QUALITY CONTROL PROCEDURES
Automatic data validity checking (basic quality control procedures) shall be applied at automatic
weather stations to monitor the quality of sensors’ data prior to their use in computation of
weather parameter values. This basic quality control is designed to remove erroneous sensor
information while retaining valid sensor data. In modern automatic data acquisition systems, the
high sampling rate of measurements and the possible generation of noise necessitate checking of
data at the levels of samples and of instantaneous data, generally one-minute data. Basic quality
control procedures shall be applied at each stage of the conversion of raw sensor outputs into
meteorological parameters. The range of basic quality control procedures strongly depends on
the capacity of AWS processing units. The basic quality control outputs should be included in
every AWS message.
Types of basic quality control procedures are as follows:
(a) Automatic quality control of raw data (sensor samples) intended primarily to indicate
sensor malfunction, instability and interference in order to reduce potential corruption of
processed data; the values that do not pass this quality control level are not used in further
data processing;
(b) Automatic quality control of processed data intended to identify erroneous or anomalous
data; the range of this control depends on the sensors used.
All AWS data should be flagged using appropriate quality control flags. These are used as
qualitative indicators representing the level of confidence of the data. At the basic quality control
level, a simple flagging scheme of five data quality control categories suffices. The quality control
flags are as follows:
(a) Good – accurate; data with errors less than or equal to a specified value;
(b) Inconsistent – one or more parameters are inconsistent; the relationship between different
elements does not meet defined criteria;
(c) Doubtful – suspect;
(d) Erroneous – wrong; data with errors exceeding a specified value;
(e) Missing data.
It is essential that data quality be known and demonstrable; data must pass all basic quality
control checks. In case of inconsistent, doubtful and erroneous data, additional information
should be provided; in case of missing data, the reason of missing should be transmitted. In case
of BUFR messages for AWS data, BUFR descriptors 0 33 005 (quality information AWS data) and
0 33 020 (quality control indication of following value) can be used.
(a) Automatic quality control of raw data

(i) Plausible value check – gross error check on measured values
The aim of the check is to verify whether the values are within the acceptable range
limits. Each sample shall be examined if its value lies within the measurement range of
a pertinent sensor. If the value does not pass the check, it is rejected and is not used in
further computations of relevant parameters.
(ii) Check on a plausible rate of change – time consistency check on measured values
The aim of the check is to verify the rate of change (unrealistic jumps in values). The
check is best applicable to data of high temporal resolution (a high sampling rate), as
the correlation between the adjacent samples increases with the sampling rate.
After each signal measurement, the current sample shall be compared to the
preceding one. If the difference between these two samples is more than the specified
limit, the current sample is identified as suspect and is not used for the computation of
an average. However, it is still used to check the temporal consistency of samples, and
the new sample is still checked with the suspect one. The result of this procedure is
that in case of large noise, one or two successive samples are not used to compute the
average. In case of sampling frequency of five to ten samples per minute (sampling
intervals, 6–12 seconds), the limits of time variance of successive samples (the absolute
value of the difference) implemented at automatic weather stations can be as follows:
a. Air temperature: 2°C;

b. Dew-point temperature: 2°C;
c. Ground (surface) and soil temperature: 2°C;
d. Relative humidity: 5 per cent;
e. Atmospheric pressure: 0.3 hPa;
f. Wind speed: 20 m s–1;
g. Solar radiation (irradiance): 800 Wm–2.
At least 66 per cent or two thirds of the samples should be available to compute an instantaneous
(one-minute) value; for wind direction and speed, at least 75 per cent of the samples are
necessary to compute a 2- or 10-minute average. If less than 66 per cent of the samples are
available in one minute, the current value does not meet the quality control criterion and is not
used in further computations of relevant parameters; the value should be flagged as missing.
(b) Automatic quality control of processed data

(i) Plausible value check
The aim of the check is to verify whether the values of instantaneous data (one-
minute average or sum; where wind is concerned, 2- and 10-minute averages) are
within acceptable range limits. Limits of different meteorological parameters depend
on the climatic conditions of an AWS site and the season. At this stage, they can be
independent of them and they can be set as broad and general. The following fixed-
limit values implemented at automatic weather stations are possible:
a. Air temperature: –90°C – +70°C;
b. Dew-point temperature: –80°C – +50°C;
c. Ground (surface) temperature: –80°C – +80°C;
d. Soil temperature: –50°C – +50°C;
e. Relative humidity: 0–100 per cent;
f. Atmospheric pressure at the station level: 500–1100 hPa;
g. Wind direction: 0–360 degrees;
h. Wind speed: 0–75 m s–1 (2-minute, 10-minute average);
i. Wind gust: 0–150 m s–1
j. Solar radiation (irradiance): 0–1600 Wm–2;
k. Precipitation amount (1-minute interval): 0–40 mm.
Note: The fixed-limit values listed above can be adjusted to reflect climatic conditions of the region more precisely, if
necessary. If the value is outside the acceptable range limit, it should be flagged as erroneous.
(ii) Time consistency check

The aim of the check is to verify the rate of change of instantaneous data (detection of
unrealistic spikes or jumps in values or dead band caused by blocked sensors).
a. Check on a maximum allowed variability of an instantaneous value, or a step
test: if the current instantaneous value differs from the prior one by more than a
specific limit or step, the current instantaneous value does not pass the check and
it should be flagged as doubtful or suspect. The possible limits of a maximum
variability, the absolute value of the difference between the successive values, are
provided below:
Parameter Limit for suspect Limit for erroneous

Air temperature 3°C 10°C
Dew-point temperature 2–3°C; 4–5°Ca 4°C
Ground (surface) temperature 5°C 10°C
Soil temperature 5 cm 0.5°C 1°C
Soil temperature 50 cm 0.3°C 0.5°C
Soil temperature 100 cm 0.1°C 0.2°C
Relative humidity 10% 15%
Atmospheric pressure 0.5 hPa 2 hPa
Wind speed (2-minute average) 10 m s–1 20 m s–1
Solar radiation (irradiance) 800 Wm–2 1000 Wm–2

a
If dew-point temperature is directly measured by a sensor, the lower limit is to be used. If dew point is calculated
from measurements of air temperature and relative humidity, a larger limit is recommended, taking into account
the influence of the screen protecting the thermometer and hygrometer. A screen usually has different system
response times for air temperature and water vapour, and the combination of these two parameters may generate
fast variations of dew point temperature, which are representative of the influence of the screen during fast
variations of air temperature and relative humidity, not of a sensor default.
In case of extreme meteorological conditions, an unusual variability of the

parameter(s) may occur. In such circumstances, data may be flagged as suspect,
although correct. They are not rejected and are further validated during
extended quality control implemented at data processing centres regardless of
whether they are correct or incorrect.
b. Check on a minimum required variability of instantaneous values during a
certain period (persistence test), once the measurement of the parameter has
been taken for at least 60 minutes. If the one-minute values do not vary over the
past 60 minutes at least by more than the specified limit or threshold value, the
current one-minute value does not pass the check. Possible limits of minimum
required variability are as follows:
i. Air temperature: 0.1°C over the past 60 minutes;
ii. Dew-point temperature: 0.1°C over the past 60 minutes;
iii. Ground (surface) temperature: 0.1°C over the past 60 minutes;1
iv. Soil temperature may be very stable, so there is no minimum required
variability;
v. Relative humidity: 1 per cent over the past 60 minutes;2
vi. Atmospheric pressure: 0.1 hPa over the past 60 minutes;
vii. Wind direction: 10 degrees over the past 60 minutes;3
viii. Wind speed: 0.5 m s–1 over the past 60 minutes.3
If the value fails the time consistency checks it should be flagged as doubtful
(suspect).
A calculation of a standard deviation of basic variables such as temperature, pressure,

humidity and wind at least for the last one-hour period is highly recommended. If the
standard deviation of the parameter is below an acceptable minimum, all data from
the period should be flagged as suspect. In combination with the persistence test, the
standard deviation is a very good tool for detecting a blocked sensor or a long-term
sensor drift.
(iii) Internal consistency check

The basic algorithms used for checking internal data consistency are based on the
relation between two parameters. The following conditions shall apply:
a. Dew-point temperature ≤ air temperature;
b. Wind speed = 00 and wind direction = 00;
c. Wind speed ≠ 00 and wind direction ≠ 00;
d. Wind gust (speed) ≥ wind speed;
e. Both elements are suspect4 if total cloud cover = 0 and amount of
precipitation > 0;5
f. Both elements are suspect4 if total cloud cover = 0 and precipitation
duration > 0;6
g. Both elements are suspect4 if total cloud cover = 100 per cent and sunshine
duration > 0;
h. Both elements are suspect4 if sunshine duration > 0 and solar radiation = 0;
i. Both elements are suspect4 if solar radiation > 500 Wm–2 and sunshine
duration = 0;
j. Both elements are suspect4 if amount of precipitation > 0 and precipitation
duration = 0;
k. Both elements are suspect4 if precipitation duration > 0 and weather
phenomenon is different from type of precipitation.
1
For ground temperature outside the interval (–0.2°C +2°C). Melting snow can generate isothermy, during which
the limit should be 0°C (to take into account the measurement uncertainty).
2
For relative humidity < 95 per cent (to take into account the measurement uncertainty).
3
For 10-minute average wind speed during the period > 0.1 m s–1.
4
Possibly used only for data from a period not longer than 10–15 minutes.
5
Or greater than the minimum resolution of the rain-gauge, to take into account the deposition of water by dew or
other factors.
6
With the exception of snow pellets, which can occur with cloud cover = 0.
If the value does not pass the internal consistency checks, it should be flagged as
inconsistent.
Technical monitoring of all crucial AWS components, including sensors, is an intrinsic

part of the quality assurance system. It provides information on data quality through
the technical status of the instrument and information on the internal measurement
status. Corresponding information should be exchanged together with measured
data; in case of BUFR messages for AWS data, technical monitoring can be carried out
by using BUFR descriptor 0 33 006 – internal measurement status (AWS).
3. EXTENDED QUALITY CONTROL PROCEDURES
Extended quality control procedures should be applied at the national data processing
centre to check and validate the integrity of data: completeness of data, correctness of data
and consistency of data. Checks that are performed at AWS sites should be repeated at data
processing centres, but in a more elaborate form. This should include, inter alia, comprehensive
checks against physical and climatological limits, time consistency checks for a longer
measurement period, checks on logical relations among a number of variables (internal
consistency of data) and statistical methods to analyse data.
Suggested limit values (gross-error limit checks) for surface wind speed, air temperature, dew-
point temperature and station pressure are outlined in, Chapter 6, Quality control procedures,
of the Guide on the Global Data-processing System (WMO‑No. 305). The limits can be adjusted on
the basis of improved climatological statistics and experience. In addition, the aforementioned
publication also describes internal consistency checks for surface data, where different
parameters in a SYNOP report are checked against each other. Where other types of report
containing AWS data, such as BUFR, are involved, the relevant checking algorithms must be
redefined; in case of BUFR, corresponding BUFR descriptors and code/flag tables should also be
redefined.
Internal data consistency checks
An internal consistency check on data can result in corresponding values being flagged as
inconsistent, doubtful or erroneous when only one of them is suspect or incorrect. Therefore
further checking by other means should be performed so that only the suspect or incorrect value
is correspondingly flagged and the other value is flagged as correct.
In comparison with basic quality control performed at automatic weather stations, extended
quality control procedures will require more quality control categories, for example: data verified
(at basic quality control: data flagged as suspect, incorrect or inconsistent; at extended quality
control: validated as correct using other checking procedures) and data corrected (at basic
quality control: data flagged as incorrect or suspect data; at extended quality control: corrected
using appropriate procedures).
The different parameters in the AWS N-minute data report (N ≤ 10–15 minutes) are checked
against each other. In the description below, the suggested checking algorithms have been
divided into areas where the physical parameters are closely connected. The symbolic names
of parameters with the corresponding BUFR descriptors used in the algorithms are explained
below.
(a) Wind direction and wind speed

Wind information is considered to be erroneous in the following cases:
(i) Wind direction without any change and wind speed ≠ 0;
(ii) Wind direction is changing and wind speed = 0;
(iii) Wind gust (speed) ≤ wind speed.
(b) Air temperature and dew-point temperature

Temperature information is considered to be erroneous in the following cases:

(i) Dew-point temperature > air temperature;
(ii) Air temperature > dew point temperature > 5°C and obscuration is from {1, 2, 3} (BUFR
descriptor 0 20 025).
(c) Air temperature and present weather

Both elements are considered suspect when:
(i) Air temperature > +5°C and type of precipitation is from {6, …, 12};
(ii) Air temperature < –2°C and type of precipitation is from {2};
(iii) Air temperature > +3°C and type of precipitation is from {3};
(iv) Air temperature < –10°C and type of precipitation is from {3};
(v) Air temperature > +3°C and obscuration is from {2} or (obscuration is from {1} and
character of obscuration is from {4}) (BUFR descriptors 0 20 021, 0 20 025, 0 20 026).
(d) Visibility and present weather

The values for visibility and weather are considered suspect when:
(i) Obscuration is from {1, 2, 3} and visibility > 1 000 m;
(ii) Obscuration is from {7, 8, 9, 11, 12, 13} and visibility > 10 000 m;
(iii) Visibility < 1 000 m and obscuration is not from {1, 2, 3, 8, 9, 10, 11, 12, 13} and type of
precipitation is not from {1, … , 14};
(iv) Obscuration = 7 and visibility < 1 000 m;
(v) Visibility > 10 000 m and type of precipitation is missing and obscuration is missing
and weather phenomenon is missing (BUFR descriptors 0 20 021, 0 20 023, 0 20 025).
(e) Present weather and cloud information

Clouds and weather are considered suspect when total cloud cover = 0 and type of
precipitation is from {1, …, 11, 13, 14} or weather phenomenon is from {2, 5, … , 10} (BUFR
descriptors 0 20 021, 0 20 023).
(f) Present weather and duration of precipitation

Present weather and duration of precipitation are considered suspect when type
of precipitation is from {1, … , 10, 13, 14} and precipitation duration = 0 and type
of precipitation is not from {1, … , 10, 13, 14} and precipitation duration > 0 (BUFR
descriptor 0 20 021).
(g) Cloud information and precipitation information

Clouds and precipitation are considered suspect when total cloud cover = 0 and amount of
precipitation > 0.7
(h) Cloud information and duration of precipitation

Clouds and duration of precipitation are considered suspect when total cloud cover = 0 and
precipitation duration > 0.
(i) Duration of precipitation and other precipitation data
Precipitation data are considered suspect when amount of precipitation > 0 and
precipitation duration = 0.
(j) Cloud information and sunshine duration

Clouds and sunshine duration are considered suspect when total cloud cover = 100 per cent
and sunshine duration > 0.
For each check, if the checked values do not pass the internal consistency check, they should be
flagged as erroneous or suspect depending on the type of the check and inconsistent. Further
checking by other means should be performed so that only the suspect or incorrect value is
correspondingly flagged and the other value is flagged as correct.
The symbolic name and the corresponding BUFR descriptor (as reference) used in quality control
algorithms (a) – (j) are as follows:
Symbolic name BUFR descriptor

Wind direction 0 11 001
Wind speed 0 11 002
Wind gust (speed) 0 11 041
Air temperature 0 12 101
Dew-point temperature 0 12 103
Total cloud cover 0 20 010
Visibility 0 20 001
Type of precipitation 0 20 021
Precipitation character 0 20 022
Precipitation duration 0 26 020
Weather phenomenon 0 20 023
Obscuration 0 20 025
Character of obscuration 0 20 026
For further treatment of data, it is necessary to keep the results of the extended quality control
data quality control and the information on how suspect or incorrect data were treated, using
sophisticated system of flags. The output of the quality control system should include quality
control flags that indicate whether the measurement passed or failed, as well as a set of summary
statements regarding the sensors.
Every effort should be made to fill data gaps, correct all erroneous values and validate doubtful
data detected by quality control procedures at the data processing centre.
4. QUALITY CONTROL MONITORING
Real-time quality control procedures have their limitations and errors, such as sensor drift or bias,
and errors in data transmission can go undetected. Therefore, performance monitoring at the
network level is required at meteorological data processing centres and by network managers.
Effective real-time quality control monitoring as an integral part of a quality control system
should include checks of the following items: completeness of observations at the meteorological
station, data quality, and completeness and timeliness of the collection of observational data at
the centre concerned.
Quality control monitoring is intended to identify deficiencies and errors, monitor them and
activate appropriate remedial procedures. Some assessment can be and should be performed
in real time, whereas other evaluations can only be accomplished after sufficient data has been
gathered over a longer period of time.
Quality control monitoring requires the preparation of summaries and statistics. Therefore, it is
necessary to build a quality control monitoring system to collect statistics on observational errors
of individual meteorological variables through a series of flags indicating the results of each
check, and generate hourly, daily, weekly, monthly and yearly summaries of the following data:
(a) The total number of observations scheduled and available for each variable (completeness
of data);
(b) The total number of observations which failed the quality control checks for each variable
(quality of data) in case of:
(i) Plausible value check;
(ii) Time consistency check;
(iii) Check on a maximum allowed variability of an instantaneous value;
(iv) Check on a minimum required variability of instantaneous values;
(v) Internal consistency check;
(vi) Percentage of failed observations (quality of data);
(vii) Error and threshold values for each failed observation (reason for failure);
(viii) Root mean square error/mean error/percentage failure for failed observations for each
station (daily/weekly/monthly/yearly) (quality statistics).
Large percentages of failed observations at stations are most likely due to hardware or software
failures or inappropriate maintenance. These should be referred back to the network manager.
The quality control monitoring system must keep station monitoring statistics on the frequency
and magnitude of observation errors encountered at each station. The statistics provide
information with a view to monitoring quality of station performance, locating persistent
biases or failures in observations, evaluating improvement of quality of observation data and
performance and maintenance of station/network.
References
Automated Surface Observing System (ASOS) User’s Guide http://www.nws.noaa.gov/asos/aum-toc.pdf

Implementing an Enhanced and Integrated Quality Assurance and Quality Control System within the MSC’s New
Data Management Framework, L. Dale Boudreau, A. Zucconi. https://ams.confex.com/ams/
Annual2006/techprogram/paper_100879.htm
Quality Management and Quality Assurance, Vocabulary, ISO 8402, International Organization for
Standardization, Second Edition
Guide to Meteorological Instruments and Methods of Observation (WMO-No. 8)
Guide on the Global Data-processing System (WMO-No. 305)
Manual on Codes (WMO-No. 306), Volume I.2
Manual on the Global Data-processing and Forecasting System (WMO-No. 485)
Manual on the Global Observing System (WMO‑No. 544), Volume I
Quality Control of Meteorological Observations, Automatic Methods Used in the Nordic Countries, Report 8/2002,
Flemming Vejen (ed)
PART VII. MONITORING THE OPERATION OF THE GLOBAL OBSERVING
SYSTEM
7.1 GENERAL
The World Weather Watch Programme comprises monitoring of the operational performance
of its components in order to evaluate their efficiency, identify deficiencies and take corrective
action. The main objective of WWW Monitoring is to maintain the overall efficiency and
effectiveness of the Programme on a global, regional and national level.
Since the operations of the three core components of the World Weather Watch—the
Global Observing System, Global Data-processing and Forecasting System and Global
Telecommunication System—are closely interrelated, they cannot be monitored independently.
Therefore, in order to monitor the World Weather Watch as an integrated system, identify
deficiencies and take quick corrective action, close coordination between all centres concerned
and the WMO Secretariat is essential.
The Plan for Monitoring the Operation of the World Weather Watch is reproduced in the
Manual on the Global Data-processing and Forecasting System (WMO-No. 485) and in the Manual
on the Global Telecommunication System (WMO-No. 386). According to this Plan, monitoring is
performed on a real-time and non-real-time basis. Explanations of these terms and the procedure
for follow-up actions are given in the Plan.
7.2 IMPLEMENTATION OF THE MONITORING PROCEDURES
7.2.1 Quantity monitoring of the operation of the World Weather Watch
Three types of quantity monitoring are coordinated by the WMO Secretariat within the
framework of the World Weather Watch Programme: Annual Global Monitoring, special main
telecommunication network monitoring and data quality monitoring.
7.2.1.1 Annual Global Monitoring
Annual Global Monitoring is carried out in October each year. World Weather Watch centres are
invited to monitor SYNOP, TEMP, PILOT, CLIMAT and CLIMAT TEMP reports from the regional
basic synoptic network stations in accordance with the responsibility taken for the exchange of
data on the Global Telecommunication System:
(a) National Meteorological Centres should monitor data from their own territory;
(b) Regional Telecommunication Hubs should monitor data from their associated National
Meteorological Centres at least, and possibly from their own Region;
(c) World Meteorological Centres and Regional Telecommunication Hubs located on the Main
Telecommunication Network should monitor the complete global data set.
Each year about 100 World Weather Watch centres send their monitoring results to the WMO
Secretariat through the Internet, on diskette or on paper.
The results of the Annual Global Monitoring make it possible to compare the availability of the
reports received from regional basic synoptic network stations at the National Meteorological
Centre responsible for inserting the data in the Regional Meteorological Telecommunication
Network, at the associated Regional Telecommunication Hubs and at Main Telecommunication
PART VII. MONITORING THE OPERATION OF THE GLOBAL OBSERVING SYSTEM 209
Network centres. The differences in the availability of data between centres are due to the
following main reasons: differing requirements in data reception, shortcomings in the relay of
the data on the Global Telecommunication System, data not monitored and differences in the
implementation of the monitoring procedures at centres.
Annual Global Monitoring has the following limitations:
(a) It provides monitoring information over a limited period each year;
(b) It delivers information at the report level but no information at the bulletin level for regional
basic synoptic network Stations;
(c) Differences in the implementation of the monitoring procedures at centres lead to

differences in the availability of reports between centres.
7.2.1.2 Special main telecommunication network monitoring
Special main telecommunication network monitoring was implemented with a view to

complementing Annual Global Monitoring. Taking into account the limited resources available
at World Weather Watch centres to carry out the monitoring activities, it was agreed to share
the workload of special main telecommunication network monitoring between the main
telecommunication network centres.
One of the main features of special main telecommunication network monitoring is that the sets
of messages, also called raw data, provided by the various main telecommunication network
monitoring centres are processed by a pre-analysis centre, unique for each type of data. This
feature aims to eliminate the discrepancies in the availability of data reported by monitoring
centres caused by differences in the implementation of monitoring procedures, similarly to
Annual Global Monitoring, primarily due to different methods of counting the reports. The
purpose of pre-analysis is to prepare files that have a data-base structure and contain the
information extracted from all the sets of messages provided by the monitoring centres. The
pre-analysis files represent a unique reference for each type of data for further analysis. One
advantage of special main telecommunication network monitoring is that it is always possible
to access the raw data and read the complete text of the bulletins as received by the monitoring
centres. Special main telecommunication network monitoring provides complete monitoring
information at the report and bulletin levels for further analysis.
Special main telecommunication network monitoring is carried out four times each year:
1–15 January, April, July and October. The responsibilities taken by the Main Telecommunication
Network centres are given in Tables VII.1 and VII.2.
Table VII.1. Responsibilities of special main telecommunication network monitoring centres
Centres preparing pre-

Data set Centres providing raw data
analysis of raw data
Surface data from fixed stations: SYNOP Algiers, Melbourne, Tokyo
reports Offenbach, Toulouse, Tokyo
Upper-air data from fixed stations: Parts A of Melbourne, Nairobi, Toulouse, Tokyo
TEMP, PILOT reports; proposed extension: Tokyo
BUFR wind profiler
Climate data: CLIMAT and CLIMAT TEMP Cairo, Melbourne, New Delhi, Cairo
reports Toulouse
Data from marine stations: SHIP, TEMP SHIP, Cairo, Melbourne, Offenbach, Offenbach
PILOT SHIP, BUOY, BATHY/TESAC/TRACKOB Toulouse
reports
Centres preparing pre-

Data set Centres providing raw data
analysis of raw data
Data from aircraft: AIREP and AMDAR Melbourne, Nairobi, Toulouse, Toulouse
reports, Proposed extension: BUFR aircraft Tokyo
reports
Table VII.2. Data monitored by the main telecommunication network centre
Type of data T1T2 GGgg

SYNOP SM 0000, 0600, 1200, 1800
TEMP, PILOT US 0000, 0600, 1200, 1800
CLIMAT CS Report of the previous month
CLIMAT TEMP CU Report of the previous month
SHIP SM 0000, 0600, 1200, 1800
TEMP SHIP, PILOT SHIP US 0000, 0600, 1200, 1800
BUOY SS All bulletins
BATHY/TESAC/TRACKOB SO All bulletins
AIREP UA All bulletins
AMDAR UD All bulletins
After receiving the background information, Toulouse Regional Telecommunication Hub and the
Secretariat make an analysis of the monitoring results.
The analysis of the latest Annual Global Monitoring and special main telecommunication
network monitoring exercises can be accessed from http://www.wmo.int/pages/prog/www/ois/
monitor/index_en.html, where more information about quantity monitoring of the operation of
the World Weather Watch can also be found.
A project on integrated World Weather Watch monitoring is under development and will be
available for use in the near future.
With regard to monitoring, the Plan for Monitoring the Operation of the World Weather Watch
states that the Global Observing System is responsible for ensuring that the observations are
made according to prescribed standards, encoded correctly and presented for transmission at
the stipulated times. Monitoring of the Global Observing System is thus essentially a question of
quality control of the observations. The basic rules of quality control within the framework of the
Global Observing System are contained in Volume I, Part V, of the Manual on the Global Observing
System (WMO-No. 544). Detailed instructions about the quality control procedures which
Members are invited to follow are given in Part VI (Appendix VI.2) of the present publication.
Additional information can also be found in Chapter 1, Part III, of the Guide to Meteorological
Instruments and Methods of Observation (WMO-No. 8).
7.2.2 Data quality monitoring
7.2.2.1 Monitoring centres
To assess data quality, a number of data processing centres compare the information received
from each of the different types of observations with the first-guess numerical short-term
forecast. The participating centres produce monthly reports of the various observational data
that are of consistently low quality (Table VII.3). These lists of suspect data are exchanged
between participating centres, and communicated to the originating country for remedial
PART VII. MONITORING THE OPERATION OF THE GLOBAL OBSERVING SYSTEM 211
action. National focal points have been designated to assist in this action. This feedback leads to
improvements in the quality of observational data and ultimately to improved initial analysis and
model forecasts.
Lead centres (Table VII.4) have been established by the Commission for Basic Systems to
coordinate the monitoring results of specific types of observation. They produce six-monthly
consolidated reports of the observations with data of consistently low quality. These reports are
also known as suspect lists.
Table VII.3. Data quality monitoring centres
Centre Report
European Centre for Medium-Range Monthly report containing monthly suspect lists of
Weather Forecasts marine observations, radiosonde, aircraft and satellite
observations
Regional Specialized Meteorological Centre, Monthly report containing monthly suspect lists of land,
Bracknell marine, radiosonde, aircraft and satellite observations
Regional Specialized Meteorological Centre,
Montreal
Regional Specialized Meteorological Centre,
Tokyo
World Meteorological Centre Melbourne Monthly report containing monthly suspect lists of land,
marine and radiosonde observations
Regional Specialized Meteorological Centre, Monthly report containing monthly suspect lists of land
Offenbach observations
Regional Specialized Meteorological Centre, Monthly report containing monthly suspect lists of land,
Toulouse marine, radiosonde, aircraft observations and satellite
observations
Table VII.4. Lead centres in charge of coordinating monitoring results
Centre Data type Area of responsibility

World Meteorological Centre, Aircraft and satellite data Global
Washington
Regional Specialized Meteorological Upper-air data Global

Centre, European Centre for Medium-
Range Weather Forecasts
Regional Specialized Meteorological Surface marine data Global

Centre, Bracknell
Regional Specialized Meteorological Land surface observations Regional Association I

Centre, Nairobi
Regional Specialized Meteorological Land surface observations Regional Association II

Centre, Tokyo
Regional Specialized Meteorological Land surface observations Regional Association III

Centre, Buenos Aires
Regional Specialized Meteorological Land surface observations Regional Association IV

Centre, Montreal
World Meteorological Centre, Land surface observations Regional Association V

Melbourne
Regional Specialized Meteorological

Land surface observations Regional Association VI
Centre, Offenbach
7.2.2.2 Procedures and formats for exchange of monitoring results
Quality monitoring procedures and formats for the exchange of the monitoring results for
surface and upper-air data including marine, aircraft and satellite data have been developed,
periodically updated, and published in Attachment II.9 of the Manual on the Global Data-
processing and Forecasting System (WMO-No. 485). The six-monthly consolidated suspect reports
are distributed to Members so that they can take remedial action as required. These Members/
agencies then report to lead centres and the WMO Secretariat on their remedial efforts.
More information on data quality monitoring, monitoring procedures and report types can be
found at http://www.wmo.int/pages/prog/www/DPS/Monitoring-home/mon-index.htm.
References
Manual on the Global Data-processing and Forecasting System (WMO-No. 485)

Manual on the Global Observing System (WMO-No. 544)
Manual on the Global Telecommunication System (WMO-No. 386)
PART VIII. QUALITY MANAGEMENT
8.1 GENERAL
The key aim in modern quality management is not only to check the final product, but the entire
process as well. A fundamental approach to quality is the quality improvement circle, which
features four steps: preparation and planning, product implementation, checking the results
with, in addition, user satisfaction in mind, and reacting to that information in order to improve
further action.
The quality of observing systems can be assessed by comparing user requirements with the
capability of the systems to fulfil them. More details are given in Part II of the present Guide.
8.2 QUALITY MANAGEMENT FRAMEWORK
Fourteenth Congress decided, by adopting Resolution 27 (Cg-XIV) – Quality management,

that WMO should work toward a Quality Management Framework for National Meteorological
Services that would eventually include and develop the following distinct, albeit related
elements: WMO technical standards; quality management system(s), including quality control;
and certification procedure(s).
The structure of the existing Quality Management Framework corresponds to the following aims
of WMO:
(a) To ensure adequate uniformity and standardization in the practices and procedures used by
National Meteorological Services;
(b) To ensure quality of observational data, since the effectiveness of any National
Meteorological Service depends on the quality of data and products exchanged through
the WMO Systems;
(c) To ensure overall availability of observational data for all purposes, especially for the
numerical weather prediction.
As regards the Global Observing System, these aims are achieved by the following means:
(a) Extensive system of documented standards and recommended practices and procedures
which shall, or should, be followed by Members. These are described in the Manual on the
Global Observing System (WMO-No. 544), in the present Guide and other publications;
(b) Different levels of quality control procedures of meteorological observations (the observing
site, collecting centres, prior to transmission over Global Telecommunication System centres
and Global Data-processing and Forecasting System centres);
(c) Variety of systems for monitoring of the availability of observational data (statistics
concerning the availability of observational reports on a non real-time basis, real-time
monitoring on a global basis in leading centres);
(d) Activities for training of staff operating different components of the Global Observing
System (training courses, regional training centres).
The WMO Quality Management Framework should enable the provision of early and continuing
relevant advice to Members on developing their quality management system. The WMO Quality
Management Framework, in accordance with the statement issued by the Executive Council at its
fifty-sixth session, should focus on technical aspects of the operation of National Meteorological
Services.
8.3 WMO TECHNICAL STANDARDS AS REFERENCE DOCUMENTATION
The procedures and practices described in the Technical Regulations (WMO-No. 49) provide the
basic material for use as the reference material in national quality management systems. The
publication also contains some quality requirements and quality control and quality assurance
practices and procedures.
8.4 QUALITY MANAGEMENT SYSTEM
According to ISO 9001 terms and definitions, a quality management system is designed
to manage and control an organization with regard to quality. The idea of the ISO quality
management system is based on the precept that the quality of an organization’s final product
depends on how well each link of the chain of processes functions.
A quality management system defines the specific procedures, processes and resources required
to meet a specified standard. The ISO 9001 Standard defines the requirements for such a system.
The ultimate goal of a quality management system is to encourage and support the continual
improvement of the quality of services and products delivered.
A quality management system consists of a set of rules, procedures and practices that an
organization decides to follow in order to achieve objectives related to the quality of its products.
To do so, it is essential that clear and unambiguous procedures be used for each specific task.
In the Global Observing System, it is necessary to specify clearly and precisely the separate
processes of the quality management system for basic observational networks, and criteria for
the control of their quality, including the monitoring procedure and, where applicable, the
quality of processes or different functions of observing systems. More attention should be paid to
guidance on how to manage observational networks of stations and observing subsystems.
Good quality observational data cannot be provided without a quality management system. The
appropriate quality management system shall operate continuously at all points of the whole
observing system, from planning and installation, operations, maintenance and inspection,
test and calibration, quality and performance monitoring, training and education, to data
pre-processing, dissemination, processing and archiving; feedback and follow-up actions are
inseparable parts of this chain.
General ISO 9001 requirements applicable to the Global Observing System are as follows:
(a) Identification of the processes needed for the quality management system;
(b) Determination of the sequence and interactions of these processes;
(c) Determination of the criteria and methods to ensure the operation and control of the
processes;
(d) Provision of resources and information necessary to support the management and
operation of the processes;
(e) Monitoring, measurement and analysis of the processes;

PART VIII. QUALITY MANAGEMENT 215
(f) Implementation of actions necessary to achieve planned results and continual improvement
of the processes.
Further discussion on quality management can be found in Chapter 1, Part III, of the Guide to
Meteorological Instruments and Methods of Observation (WMO-No. 8).
Definitions
Quality assurance: Part of quality management focused on providing confidence that

quality requirements will be fulfilled; includes all the planned and systematic activities
implemented in a quality system so that quality requirements for a product or service will
be fulfilled.
Quality control: Part of quality management focused on fulfilling quality requirements; includes
all the operational techniques and activities used to fulfil quality requirements.
Quality: Degree to which a set of inherent characteristics fulfils requirements.
Quality management: Coordinated activities to direct and control an organization with regard to
quality.
Quality management system: Management tool designed to direct and control an organization
with regard to quality.
Quality policy: Overall intentions and direction of an organization related to quality, as formally
expressed by management.
References
Explanatory circular on the WMO Quality Management Framework: http://www.bom.gov.au/

wmo/quality_management/docs/QMF-circ_en.pdf
Guide to Meteorological Instruments and Methods of Observation (WMO-No. 8)
Quality Management Framework: http://www.bom.gov.au/wmo/quality_management.shtml
For more information, please contact:
World Meteorological Organization

7 bis, avenue de la Paix – P.O. Box 2300 – CH 1211 Geneva 2 – Switzerland
Communication and Public Affairs Office

Tel.: +41 (0) 22 730 83 14/15 – Fax: +41 (0) 22 730 80 27
E-mail: [email protected]
www.wmo.int
JN 17945

WMO No. 488

Uploaded by

Copyright:

Available Formats

WMO No. 488

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

WMO No. 488

Uploaded by

Copyright:

Available Formats

Guide to the Global Observing System

© World Meteorological Organization, 2010

Chairperson, Publications Board

July Part III, Modification to 3.2.1.2.6 – Station CBS-Ext. (06) Resolution 10

PART I. PURPOSE, SCOPE, REQUIREMENTS AND ORGANIZATION OF THE

PART II. OBSERVATIONAL DATA REQUIREMENTS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  7

APPENDIX II.1. EXTRACT FROM USER REQUIREMENTS AND OBSERVING SYSTEM

APPENDIX II.2. OUTCOME OF THE ROLLING REQUIREMENTS REVIEW PROCESS:

PART III. THE SURFACE-BASED SUBSYSTEM. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  18

3.1.3.13 Planning and budgeting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26

3.7.3.4 Special-purpose stations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89

APPENDIX III.1. FUNCTIONAL SPECIFICATIONS FOR AUTOMATIC WEATHER STATIONS.  116

APPENDIX III.2. BASIC SET OF VARIABLES TO BE REPORTED BY STANDARD

APPENDIX III.3. AUTOMATIC WEATHER STATION METADATA . . . . . . . . . . . . . . . . . . . . . . . .  124

APPENDIX III.4. THE WMO VOLUNTARY OBSERVING SHIP SCHEME. . . . . . . . . . . . . . . . . . .  131

PART IV. THE SPACE-BASED SYSTEM. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  162

4.3.2 Integrated Global Data Dissemination Service . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174

PART V. REDUCTION OF LEVEL I DATA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  181

PART VI. DATA QUALITY CONTROL. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  183

APPENDIX VI.1. DATA QUALITY CONTROL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  192

APPENDIX VI.2. GUIDELINES FOR QUALITY CONTROL PROCEDURES APPLYING

PART VIII. QUALITY MANAGEMENT. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  213

8.4 QUALITY MANAGEMENT SYSTEM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 214

World Weather Watch

World Weather Watch, a core WMO Programme, combines observing systems,

(b) The Global Telecommunication System (GTS), composed of integrated networks of

Purpose of this Guide

Manual on Codes (WMO‑No. 306)

Procedures for amending the Guide

1.1 PURPOSE AND SCOPE OF THE GLOBAL OBSERVING SYSTEM

The Global Observing System is a coordinated system of different observing subsystems,

Main long-term objectives

1.2 GLOBAL OBSERVING SYSTEM REQUIREMENTS

(c) Hydrological services, including flood warnings.

The formulation of data requirements is an evolving process based on experiences with

1.3 ORGANIZATION AND IMPLEMENTATION OF THE GLOBAL OBSERVING

(a) Quantitative information, derived directly or indirectly from instrument measurements;

(b) Qualitative, descriptive information.

(c) Assessment of cost-effectiveness, long-term sustainability and innovative collaborative

(b) Operational and technical feasibility;

(c) Members’ cost-efficiency requirements.

(a) The different scales of meteorological phenomena to be analysed;

There are generally three types of requirements:

2.2 ASSESSMENT AND FORMULATION OF OBSERVATIONAL DATA

2.2.1 Data sensitivity tests or observing system experiments

2.2.2 Observing system simulation experiments

2.2.3 Theoretical studies and simulations

2.2.4 Laboratory assessments

2.2.5 System design and analysis activities

2.2.6 Field site assessments

2.2.7 End-user application areas

(a) Agriculture and food production;

(j) Public weather services, health and safety;

2.3 EVALUATION OF REQUIREMENTS AGAINST SYSTEM CAPABILITIES

(a) Global numerical weather prediction;

2.3.1 The Rolling Requirements Review process

The process consists of four stages (see Figure II.2):

1. A review of user requirements for observations, within an area of application covered by

Clearly, the Rolling Requirements Review process needs to be repeated periodically as

2.3.2 User requirements and observing system capabilities database

PART II. OBSERVATIONAL DATA REQUIREMENTS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

PART III. THE SURFACE-BASED SUBSYSTEM. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18

3.1.3.13 Planning and budgeting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26

3.7.3.4 Special-purpose stations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89

APPENDIX III.1. FUNCTIONAL SPECIFICATIONS FOR AUTOMATIC WEATHER STATIONS. 116

APPENDIX III.3. AUTOMATIC WEATHER STATION METADATA . . . . . . . . . . . . . . . . . . . . . . . . 124

APPENDIX III.4. THE WMO VOLUNTARY OBSERVING SHIP SCHEME. . . . . . . . . . . . . . . . . . . 131

PART IV. THE SPACE-BASED SYSTEM. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162

4.3.2 Integrated Global Data Dissemination Service . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174

PART V. REDUCTION OF LEVEL I DATA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181

PART VI. DATA QUALITY CONTROL. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183

APPENDIX VI.1. DATA QUALITY CONTROL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192

PART VIII. QUALITY MANAGEMENT. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213

8.4 QUALITY MANAGEMENT SYSTEM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 214