2008

Geophysical Survey in
Archaeological Field Evaluation

Preface to the Second Edition

These guidelines are intended to help
archaeologists, particularly curators, consultants
and project managers, to better understand
and engage with the techniques of geophysical
survey, for the best results. It is hoped too that
practitioners of geophysical survey will find
them helpful and that, altogether, the guidance
can contribute to raising the consistency and
quality of geophysical survey in archaeological
field evaluation.
Geophysical survey in archaeology continues
to flourish. As of 2006, it is estimated to
be a component of at least 23.4% of all
evaluations arising from planning applications
(http://csweb.bournemouth.ac.uk/aip/aipintro.htm).
The techniques are also finding an increasing
role in the presentation and interpretation
of archaeological sites, in contributing to
archaeological and forensic research, and
in helping to satisfy the demand for media
coverage of archaeological subjects.
Geophysical survey has a wider academic and
professional forum than was the case several
years ago. Since its inauguration in 1995 at
Bradford University in the UK there has
subsequently been a succession of biennial
conferences on Archaeological Prospection,
held in Japan, Germany, Austria, Poland, Italy
and Slovakia, and attended by an ever more
cosmopolitan variety of specialists in geophysics
and remote sensing.The Environmental and
Industrial Geophysics Group (EIGG) of the
Geological Society has similarly hosted a
continuing series of biennial one-day meetings
devoted to recent research in the subject.The
journal Archaeological Prospection, initiated in
1994, has gone on to establish itself as the main
vehicle for publication of relevant research and
case studies; and an International Society for
Archaeological Prospection (ISAP) was initiated
in 2003 (http://www.archprospection.org).
Archaeological geophysics is now a component
of undergraduate teaching in at least 12
universities, although the only post-graduate
degree courses devoted to the subject are
the MSc in Archaeological Prospection at the
University of Bradford (http://www.brad.ac.uk/
archsci/msc_ap.htm) and the MSc in
Archaeological Geophysics recently offered
at Orkney College of the University of the
Highlands and Islands
(http://www.orkney.uhi.ac.uk/courses/archaeology/
geophysics-at-orkney-college-uhi).
Despite the increasing familiarity with methods
and techniques, and a growing number of
practitioners, geophysical survey can be both
a very technical subject, as well as a fertile
area for continuing innovation, commercial
2

exploitation, and integration with other

prospecting disciplines. It is clear from our
consultations that in these circumstances there
remains a need for independent guidance,
which the following document is intended
to provide not only for curators of the
archaeological resource, but also for others
who need to know about the potential and
pitfalls in more detail. Our purpose here
is above all dedicated to bettering the
consistency and quality of geophysical survey
in evaluations, especially those arising from
development proposals.
Much of what was presented in the first edition
remains valid and will be re-iterated here.
There are, however, changes reflecting shifts
in thinking and approach that have taken place
over the last few years.To take one example,
the debate on the efficacy of topsoil magnetic
susceptibility as an aid to evaluation, which was
very topical in the early 1990s, has moderated
now that it is increasingly accepted that detailed
magnetometer coverage is preferable, and more
feasible, over yet larger areas. More importantly,
there have been changes in geophysical
instrumentation, technology, methodology and
software, all of which are having an impact on
the choice and performance of geophysical
survey under varying conditions. A particular
example is the great improvement in the virtues
of ground penetrating radar (GPR), now that
software and computing power enable both
greater coverage and production of more
comprehensible display and interpretation.
Another significant development, following
the influential example of European practice,
is the increasing awareness and availability of
alkali-vapour as well as fluxgate magnetometers.
Both types of magnetometer, as well as other
types of sensor, are now being deployed as
arrays on mobile platforms, with considerable
potential to raise the versatility and speed
of ground coverage.
Other areas of rapidly advancing progress
include the further integration of geophysical
data within Geographical Information Systems
(GIS), which has in turn increased the need for
consistency of data geo-referencing and archiving.
In parallel, there are a growing number of survey
projects that seek to integrate ground-based
prospecting methods, together with remote
sensing technologies such as lidar, to maximise
interpretative and analytical potential.
That said, wetlands, alluviated and urban
environments persist as challenges to
geophysicists. While not relevant here, but
to be the subject of future guidance from
English Heritage (forthcoming 2008) it is
worth noting that the remote examination
of the shoreline and seabed is a growing

imperative now that maritime archaeological

conservation is in the ascendant.
The first edition of this guidance was published
in 1995, and this revision is offered in the hope
of maintaining a balanced and independent
view on best practice in the context of progress
since then. With the benefit of much positive
advice, comment and discussion from many
colleagues, for whose patience and advice we
are very thankful, we hope we have improved
the content, and its presentation and clarity.
As the document is available on line
(http://www.english-heritage.org.uk/upload/
pdf/GeophysicsGuidelines.pdf) we expect
to make future revisions and updates more
immediately and easily in future and would,
as ever, welcome comment and advice
towards these.

Contents
Preface to the Second Edition . . . . . . . . . 2

Part I
Standards for Geophysical Survey
Part III Guide to Choice of Methods . . 13

1 Introduction

1 Introduction . . . . . . . . . . . . . . . . . . . . . . . 13

There is currently no formalised standard

for the conduct of geophysical survey in
archaeological field evaluation. For the purpose
of this guidance, however, it is expected that
such survey will, as far as is reasonably possible,
determine the nature of the detectable
archaeological resource within a specified area
using appropriate methods and practices1.
These will satisfy the stated aims of the
survey project. Members of the Institute
of Field Archaeologists (IFA) will, and other
practitioners should, comply with the Code
of conduct, Code of approved practice for
the regulation of contractual arrangements in
field archaeology, and other relevant by-laws
of that Institution2.

Part I Standards for Geophysical Survey

2 Choice of geophysical survey . . . . . . . . 13

1 Introduction . . . . . . . . . . . . . . . . . . . . . 3

3 Costs . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
4 Urban (and brownfield) sites . . . . . . . . 14

2 Guidance . . . . . . . . . . . . . . . . . . . . . . . . 3

5 Cemeteries . . . . . . . . . . . . . . . . . . . . . . . 15

2.1 Justification for survey . . . . . . . . . . . . . . 3

6 Alluvium . . . . . . . . . . . . . . . . . . . . . . . . . . 16

2.2 Fieldwork . . . . . . . . . . . . . . . . . . . . . . . . 4

7 Wetlands . . . . . . . . . . . . . . . . . . . . . . . . . 16

2.3 Data treatment . . . . . . . . . . . . . . . . . . . . 4

8 Road and pipeline corridors . . . . . . . . . 17

2.4 Data interpretation . . . . . . . . . . . . . . . . 4

9 Wind farms . . . . . . . . . . . . . . . . . . . . . . . 17

2.5 The survey report . . . . . . . . . . . . . . . . . 4

10 Extremely large areas . . . . . . . . . . . . . . 18

2.6 Dissemination . . . . . . . . . . . . . . . . . . . . . 5
2.7 Data archiving . . . . . . . . . . . . . . . . . . . . . 5

Part IV

2.8 Competence of survey personnel . . . . 5

Practitioners Guide to Good Practice

1 Application of techniques . . . . . . . . . . 19

Part II Geophysical Survey and Planning

1.1 The survey grid . . . . . . . . . . . . . . . . . . . 19

1 Archaeology and planning . . . . . . . . . . . 6

1.2 Magnetometer survey . . . . . . . . . . . . . 20

1.3 Earth resistance

2 MoRPHE . . . . . . . . . . . . . . . . . . . . . . . . 6

(resistivity) survey . . . . . . . . . . . . . . . . . 24

2.1 Start-up and planning . . . . . . . . . . . . . . 6

1.4 Ground penetrating radar . . . . . . . . . 28

2.2 Execution . . . . . . . . . . . . . . . . . . . . . . . . 6

1.5 Electromagnetic methods . . . . . . . . . . 34

2.3 Closure . . . . . . . . . . . . . . . . . . . . . . . . . . 7

1.6 Topsoil magnetic

susceptibility survey . . . . . . . . . . . . . . .36

3 Briefs and specifications . . . . . . . . . . . . 7

1.7 Other geophysical methods . . . . . . . . 37

3.1 The brief . . . . . . . . . . . . . . . . . . . . . . . . . 7

1.8 Metal detecting . . . . . . . . . . . . . . . . . . 40

3.2 The specification . . . . . . . . . . . . . . . . . . . 8

1.9 Geochemical methods . . . . . . . . . . . . 40

1.10 Remote sensing . . . . . . . . . . . . . . . . . 41

4 The survey report . . . . . . . . . . . . . . . . 9

4.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . 9

2 Analysis of geophysical data . . . . . . . . 41

4.2 Introduction . . . . . . . . . . . . . . . . . . . . . . 9

2.1 Data processing . . . . . . . . . . . . . . . . . . 41

4.3 Methods . . . . . . . . . . . . . . . . . . . . . . . . . 9

2.2 Data display . . . . . . . . . . . . . . . . . . . . . 45

4.4 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

2.3 Data interpretation . . . . . . . . . . . . . . . 49

4.5 Conclusions . . . . . . . . . . . . . . . . . . . . . . . 9
4.6 Site location plan(s) . . . . . . . . . . . . . . . . 9

References . . . . . . . . . . . . . . . . . . . . . . . 50

4.7 Data presentation

Glossary . . . . . . . . . . . . . . . . . . . . . . . . . 54

plots and plans . . . . . . . . . . . . . . . . . . . 10

4.8 Plots of raw data . . . . . . . . . . . . . . . . . 10

Appendix I Related standards,

4.9 Plots of processed data . . . . . . . . . . . . 10

codes and guidance . . . . . . . . . . . . . . . . 56

4.10 Interpretative diagrams . . . . . . . . . . . 10

Appendix II Contacts . . . . . . . . . . . . . . . 57
Appendix III Useful websites . . . . . . . . . 58

5 Dissemination . . . . . . . . . . . . . . . . . . . 10

Appendix IV List of consultees . . . . . . . 59

5.1 Sources of information . . . . . . . . . . . . 10

5.2 Dissemination requirements . . . . . . . . 11

Contributors . . . . . . . . . . . . . . . . . . . . . 60
Acknowledgements . . . . . . . . . . . . . . . . 60

6 Archiving . . . . . . . . . . . . . . . . . . . . . . . 11
7 Legal considerations . . . . . . . . . . . . . . 11
7.1 Access . . . . . . . . . . . . . . . . . . . . . . . . . . 11
7.2 Metal detectors . . . . . . . . . . . . . . . . . . 12
7.3 Geophysical survey . . . . . . . . . . . . . . . 12

All relevant fieldwork must conform to the

Standard and Guidance set out by the Institute
of Field Archaeologists for archaeological
field evaluation.
(http://www.archaeologists.net/modules/
icontent/inPages/docs/codes/fldeval2001.pdf)
2
The IFA is the professional body for
archaeologists in the United Kingdom
(www.archaeologists.net). It exists to advance
the practice of archaeology and allied disciplines
by promoting professional standards and ethics
for conserving, managing, understanding and
promoting enjoyment of heritage. It has about
2500 members in the UK and abroad.
1

This basic requirement for geophysical survey

in archaeology is fairly straightforward, although
much will depend on the definition of what
is reasonably possible.To help address this,
we initially itemise below some more precise
requirements that must be achieved; followed
in Parts IIIV by more specific guidance on
best practice. Part II (Geophysical Survey and
Planning) and Part III (Guide to the Choice of
Methods) are aimed at those who commission
surveys; Part IV is a more in-depth description
and assessment of the main methodologies,
for those more concerned with these.

2 Guidance
2.1 Justification for survey
Prior to fieldwork, the geophysical survey
requirements must be integrated within
a written statement (the project design,
specification, written scheme of investigation, or
survey contract).This must include an explicit
justification for the choice of survey methodology,
while retaining some flexibility should this require
modification in the light of particular site
conditions at the time of fieldwork.The choice
of survey methodology will be appropriately
matched both with the archaeological and
logistical demands of the project.
3

2.2 Fieldwork

2.2.3 Earth resistance: area survey

All fieldwork should be conducted under the

principle of repeatability; in other words, that,
within reason, the data obtained should be
capable of independent duplication. Fieldworkers
must ensure that every effort is made on
site to be courteous and considerate in their
dealings with landowners, local residents and
organisations, respecting all aspects of the
environment. A high level of professionalism
is necessary at all times.

The maximum acceptable sampling interval for

area surveys is 1m along traverses separated
by a maximum of 1m.
Area surveys, using the twin probe (or twin
electrode) probe configuration, are the preferred
method of ground coverage.The square array
(often employed on cart-based systems) is also
acceptable for area surveys. Other methods
require special justification.

2.3 Data treatment

Correct observance should be made of any

legal constraints on site for instance, the
requirement of a Section 42 Licence for
survey over scheduled monuments and
other protected places, and the licence now
needed for survey on National Trust land
(Part II, 7.3).

For twin probe systems the mobile probe

spacing should usually be 0.5m; wider
separations and/or multiplexed arrays require
explanation.The equivalent spacing for a
square array would typically be 0.75m.

A copy of unprocessed raw data must be

retained and archived (see below, 6 Archiving).

2.2.1 The survey grid

This is the network of control points used to
locate the geophysical survey measurements
relative to base mapping and/or absolute position
on the Earths surface, (see Part IV, 1.1).Whether
physically marked on the ground or measured
while surveying using a global positioning
system (GPS), these must be located to
survey-grade accuracy ( 0.1m).The survey
grid must be independently re-locatable on
the ground by a third party, by measurement
to local permanent features, and/or by the use
of GPS coordinates. All locational information
must be geo-referenced. In certain cases (eg
where permanent features are absent), and with
appropriate permission, it may be acceptable to
emplace permanent survey markers.
Care must be taken to ensure that any survey
markers or other equipment are not a hazard
to people or animals.

2.2.2 Magnetometer survey

Survey must be conducted with a continuously
recording magnetometer of appropriate
sensitivity.
Area survey must be the preferred method
of ground coverage in all instances where this
is practicable.
The maximum acceptable sampling interval
for an area survey is 0.25m on traverses a
maximum of 1m apart.
Magnetometer scanning, as a method of initially
assessing the magnetic response of a site, may
be used at the discretion of surveyors who
are experienced in its application, for devising
(or advising upon) an appropriate evaluation
strategy that will use other methods.The
technique should not otherwise be included
in briefs or specifications.
4

2.2.4 Ground penetrating radar survey

Generally, this technique will be applied for
the detailed investigation of a site by individual
profiles and the visualisation of the data as
time slices. A maximum traverse spacing of
0.5m is recommended with samples taken
at intervals of 0.05m.
Specific site conditions and the aims of the
survey may require an alternative sampling
methodology to be adopted, but this must
be fully justified in any supporting
specification documents.
Determination of an appropriate sampling
interval, centre frequency of antenna(s) used
and sub-surface velocities used for depth
estimation from the resulting data must be
supported through an appropriate survey
design, including field test measurements
where appropriate.

2.2.5 Magnetic susceptibility survey

Magnetic susceptibility survey should not take
precedence over magnetometer survey where
the latter is practicable.
Areas of high topsoil magnetic susceptibility
should be complemented by detailed area
magnetometer survey. Some areas of low
or indifferent magnetic susceptibility values
should also be sampled with detailed
magnetometer coverage, to confirm that
under the prevailing site conditions, changes
in magnetic susceptibility do correlate with
archaeological potential.
The chosen method(s) of magnetic susceptibility
measurement must be appropriate to prevailing
ground surface conditions.
Measurements of topsoil magnetic susceptibility,
for area surveys and transects, must be made
at intervals not exceeding 10m.

Where possible, such topsoil measurements

must be compared and contrasted with
those obtained from subsoil and local
archaeological features.

Area surveys must be conducted, and

subsequent data treated, so as to result
in a data-set that is as uniform as possible.
Edge-effects between contiguous survey
areas should be minimised.

Raw data collected in the field must be of high

quality. Any data-collection artefacts subsequently
apparent in the survey data should be identified
and removed using appropriate data processing
(Part IV, 2.1). All such processing should be
clearly described. Any data collection artefacts
that cannot be corrected by data processing
should be described and clearly distinguished
from possible archaeological anomalies.
If data has been seriously compromised during
collection, a return to the site to re-survey
the affected area(s) should be considered.

2.4 Data interpretation

The interpretation of survey data must be
undertaken by a competent archaeological
geophysicist who is knowledgeable of the
archaeological and geomorphological conditions
prevailing on site. Consultation must also take
place with other site specialists (eg landscape
archaeologists, aerial photographers) wherever
possible.
The interpretation of magnetometer and
magnetic susceptibility data must endeavour to
distinguish anthropogenic from other causes of
magnetic enhancement on the site(s) concerned.
A clear distinction must always be made
between interpretation that is scientifically
demonstrable, and interpretation based on
informed speculation.
Any reference to negative evidence must be
fully qualified and explained. Lack of geophysical
anomalies cannot be taken to imply a lack of
archaeological features, and in such cases an
alternative evaluation procedure eg trial
trenching, or the use of a different geophysical
technique should be considered.

2.5 The survey report

All fieldwork must be followed by a report.
This will be a clear and succinct text, supported
by tables, figures, appendices and references as
necessary (see below, 2.5.1). It ought to stand
independent of supporting material and should
combine the qualities of concise technical

description linked to lucid and objective analysis

and interpretation. It must in the most part
be intelligible to specialists and non-specialists
alike. It should usually be accompanied by a
statement of the authors and contractors
professional qualifications.

there must be an explanation in the caption or

text as to the supporting analysis. If the ground
level is significantly uneven (> 0.5m) along the
survey traverse concerned, a topographically
corrected section should also be considered.
Each plot must include a key describing the
symbols and conventions used.

2.5.1 Report structure and contents

The report will normally contain the
following elements:

title page
summary or abstract
introduction
methods statement
results
conclusions
acknowledgements
statement of indemnity
references
appendices

Further detail on report content is provided

in Part II, 4.

2.5.2 Data presentation plots and plans

Depending on the geophysical methods used,
each report must include:
a survey location plan demonstrating
relationships to other mapped features
(minimum scale 1:2500);
an image of minimally processed survey
data (see Part IV, 2.2 and 2.3, preferred
minimum scale 1:1000);
where appropriate (see Part IV, 2.2) a trace
(or XY) plot of raw magnetic data (for
very large sites, a sample of data might be
supplied instead, to support the specific
interpretation of anomalies identified from
greyscale images);
specimen profiles, in the case of GPR surveys;
a greyscale plot, or dot density plot
(minimum scale 1:1000);
and one or more interpretative
plans/diagrams (minimum scale 1:1000).

2.6 Dissemination
A copy of the survey report (paper or digital,
as required) should be lodged with the Local
Authority Historic Environment Record (HER),
normally within six months of the completion
of fieldwork, but if, necessary, may be delayed
until after completion of the full project (see
Part II, 5.2).This should be a responsibility of
the commissioning body, in consultation with
the project director and the contractor.

Membership of professional institutions or

relevant associations, while not a requirement,

should also be a consideration and is

encouraged.These include:

Institute of Field Archaeologists (IFA)

European Association of Geoscientists

& Engineers (EAGE)

European GPR Association (EuroGPR)

Less experienced staff must be supervised

throughout any fieldwork, subsequent data

treatment, interpretation of the data and/or

report preparation.

Copies of any report resulting from a survey

for which a Section 42 Licence (see Part II, 7.3)
has been obtained must be sent both to the
English Heritage Regional Inspector of Ancient
Monuments and to the English Heritage
Geophysics Team, Fort Cumberland, Eastney,
Portsmouth PO4 9LD.
Details of the survey must be entered on
OASIS (see Part II, 5.2).

The location plan must be directly relatable

to the OS National Grid. Reproduction of any
part of an OS map requires copyright permission
see http://www.ordnancesurvey.co.uk/
oswebsite/business/copyright/index.html.
Each plan and/or plot must have a bar scale
(or annotated metric grid) and an accurately
oriented north arrow.
Greyscale, dot density and trace (XY) plots
must also have annotated scales indicating the
range of the variables depicted.
GPR profiles require a horizontal scale, and
a scale of two-way travel time on the Y-axis.
If an estimated depth scale is also included,

2.7 Data archiving

A minimum requirement is that a viable
digital copy of the raw survey data must be
retained for future interrogation, together
with adequate information on the location
of the survey and the survey methodology.
In addition to storage on a secure medium,
appropriate documentation of survey practice
and data files is also required.The archiving of
data associated with geophysical survey should
follow the advice provided in Geophysical
Data in Archaeology: A Guide to Good Practice
(Schmidt 2002), together with the advice in
Archaeological Archives: A Guide to Best Practice
in Creation, Compilation, Transfer and Curation
(Brown 2007).

2.8 Competence of survey personnel

All staff, including sub-contractors, must be
suitably qualified and competent for their
project roles, employed in line with relevant
legislation and IFA by-laws (where relevant).
The project manager must have:
competence in basic metric survey procedure;
experience in a supervised capacity of at
least 30 different site surveys, or a minimum
of three full years supervised experience
of archaeological geophysics;
and a degree in archaeology and/or an
appropriate science (eg MSc in
Archaeological Prospection).

Part II
Geophysical Survey and Planning

1 Archaeology and planning

Government guidance (DoE 1990) states that

where nationally important archaeological
remains, whether scheduled or not, are affected
by proposed development there should be
a presumption in favour of their physical
preservation. From this stems the necessity
for field evaluation as a preliminary stage in
the planning process.The potential contribution
of geophysical survey should be considered in
each instance where development is proposed.
As geophysical survey will often be a crucial
element in site evaluation it is most important
that it should be correctly integrated within
briefs and specifications and within subsequent
project management.

2 MoRPHE
Field evaluation, and any geophysical survey
that it includes, should be part of an integrated
programme of research. Management of
Research Projects in the Historic Environment
(MoRPHE) is a system developed to promote
this process. A typical project will often
proceed through a number of stages (Lee
2006) and the role of geophysical survey
is described broadly in relation to these.
Detailed discussion of individual aspects of
survey procedure follows in the subsequent
sections.

2.1 Start-up and planning

Consideration of geophysical survey can be
most crucial during the early stages of project
planning. Indeed, in many programmes of
archaeological evaluation the geophysical
survey will be completed and acted upon,
as a self-contained project, entirely within this
phase. In the right circumstances such survey
can provide information of great clarity on the
extent and nature of archaeological deposits
and features. Even in less perfect conditions,
survey results can be highly informative, and
therefore it is important that geophysical
methods should always be considered at the
outset of each programme of evaluation.
Most evaluations will be initiated with a desktop
study, often starting with an interrogation of
the relevant Historic Environment Record
(HER), followed by an assessment of all other
extant documentary records, including aerial
photographic (AP) coverage (ACAO 1993).
Such a study should also determine the following
information relevant to geophysical survey:

solid geology
drift geology
soil type
current land use and surface conditions

history of previous ground disturbance

history of previous geophysical survey (if any)
legal status of the site

Once this information is available, the potential

for geophysical survey should be assessed.
If geophysical survey is then agreed to be
relevant, a project design or specification can
be drawn up, calling upon expert advice in
order to avoid wasteful or misdirected outlay
of resources, or missed opportunities.

2.2 Execution
Project Execution, as defined here, includes
fieldwork, assessment of potential, archive
deposition, and dissemination (Lee 2006).

2.2.1 Fieldwork
The following stages of geophysical survey
fieldwork should be considered and planned
for, where appropriate:
(a) Pilot (test or trial) survey: it may occasionally
be necessary for a preliminary assessment to
be made of a sites response to geophysical
survey, particularly where large areas (>20ha)
are concerned.This procedure should indicate
whether local conditions are suitable for useful
results to be obtained and what techniques
and sampling methodology may be most
appropriate. Such preliminary information,
based on expert assessment, can forestall
the wasteful deployment of resources on
inappropriate techniques and on sites where
the use of geophysics is unlikely to be helpful.
A brief site visit may be all that is required.
Any pilot survey should not usually take
more than a day to achieve, and the results
should be made available immediately for
incorporation into the project design. Project
managers should ensure that they are made
aware of the geophysical potential, or lack of
it, of their site(s) at the outset; the justification
for survey must be clear.
(b)Full survey: once this justification is assured
an agreed survey strategy can proceed.This
may be full or partial coverage of the site at
high or low levels of detail, using one or more
techniques, depending on the strategy adopted.
(c) Extended coverage: in some circumstances it
may be necessary to accommodate additional
survey if earlier results (or subsequent
excavation) indicate that this would be
profitable. Where appropriate, allowance
for such contingencies should be made in
briefs and specifications.
It is particularly important at this time to establish
a secure and agreed timetable in which the above
stages of survey are correctly integrated with
the other evaluation strategies. In many instances
it will be for survey to take place after field
walking, utilising a shared grid system, but before

trial trenching or excavation.The timetable

should be sufficiently flexible to accommodate
additional contingency survey, and costing
should allow for this. Above all, the timetable
should permit adequate time for the results
of geophysical survey to be fully reported in
order to inform subsequent project planning.
Once the report has been made available,
allowance should be made for the project
team to communicate with the surveyors
to discuss any outstanding matters, especially
as these may relate to the archaeological
interpretation of the geophysical data.
Good timetabling must be linked with full and
informed cooperation between all parties.
Particularly relevant to geophysical survey is
that landowners and/or their agents and/or
tenants have been informed and given their
permissions for the survey to take place.
Obtaining such permissions, as well as details
of access and the resolving of any other local
complications, should usually be the responsibility
of the project manager rather than that of
the surveyors.
The above recommendations should be
followed wherever possible. It is acknowledged,
however, that very often practical necessity
particularly shortage of time may dictate a
different course of action. For instance, there may
be insufficient time to prepare a full report in
advance of excavation or of the development
itself, in which case survey plots produced in
the field must be acted upon directly.
Once the survey strategy has been agreed,
costed, timetabled and the relevant permissions
obtained, the fieldwork can go ahead accordingly.
Actual fieldwork procedures are discussed
more fully below in Part IV.
In the context of the full research programme,
geophysical survey will usually be incorporated
in the Initiation Stage, allowing its results to
direct the subsequent Execution Stage of
the larger programme.

2.2.2 Assessment of potential

There are two sets of instances where
assessment of the potential of the geophysical
survey data may be required as part of the
Execution Stage of the larger programme:
(a) where such data indicates that further survey
would be of significant advantage to the
realisation of specified archaeological research
objectives.There are many instances where
extended geophysical survey could significantly
enhance the value of a project by placing a
partially recorded site or sites within a wider
spatial context, in which crucial relationships

with other features, sites or the wider

landscape can be better understood.This
synthetic role of geophysical survey should
never be underestimated.
Any such additional survey should be justified
and planned for in an updated project design.
It should, if possible, employ the original team;
if other surveyors must be used then the
project manager should ensure that full
continuity and integration of survey procedure
and interpretation is achieved. If possible,
the original raw field data should be made
accessible to the incoming surveyors (see
below, 2.2.3).
(b)where the geophysical survey data, in its own
right, has significant potential for advancing
research into geophysical prospecting
techniques, or the interpretation of
geophysical data.This potential should
always be assessed at the outset of a
project, and kept under review.
In both senses (a) and (b) above, geophysical
survey data has a research potential and should
be considered alongside other more customary
post excavation data. If deemed significant by
the project team, any scope for realisation of
this potential should be included in an updated
project design.The latter will include provision
for the publication of results either within the
main project report, or as a separate paper
in a more specialised publication.

2.2.3 Archive deposition

While the full details of the geophysical survey
will be archived at the conclusion of the survey
project (see below, 6 Archiving), the project
manager and survey staff should be aware of
the necessity of recording and safeguarding
raw data, the data processing steps undertaken,
and locational information, at all appropriate
stages during the course of the project.

2.2.4 Dissemination
The results of the main research programme
will be drawn up, in draft report form, for review
and subsequent publication. However, the report
on the geophysical survey will usually have been
completed and presented to the project team
and/or commissioning body earlier. Close liaison
with the project team must continue, however,
to ensure that the geophysical data and its
interpretation is presented in appropriate
proportion to its contribution to the stated
objectives of the wider programme.
The following options can be considered for the
final presentation of the geophysical survey results:
(a) that a summary should be included in the
main report text, while the survey report

and related data is retained in archive;

(b)that a summary should be included in the
main report text, while the survey report
is included as an appendix;
(c) that the survey report should be modified
for reproduction in the main report text.
It is not acceptable for the contribution of
geophysical survey to be ignored, even if results
have been indifferent or negative. A minimum
requirement is that a summary statement is
recorded in the overall programme report.
It should be noted that under the Copyright,
Designs and Patents Act 1988 the organisation
or person undertaking field and reporting
work retains the copyright to the material,
unless stated otherwise in the contract for
the work.This position should be made clear
to all relevant parties at the outset of work
(IFA 2001, Appendix 5).
Every effort should be made to ensure that
the survey report becomes publicly accessible.
All field data and reports will be deposited
with the site archive, and the HER updated.
Where results for some reason cannot be
disclosed, a minimal record should be made
and fully updated within a reasonable time
(normally six months). A fuller discussion of
dissemination and archiving follows in sections
5 and 6 below.

2.3 Closure
Once the survey project has been concluded,
time should be planned for documentation of
any follow-on actions, unresolved issues and
lessons learned.

sub contracted; in either case particular care

is required, and advice on this can be acquired
from various sources (eg CBA 1982; Darvill
1993; Darvill and Atkins 1991; IFA 2001).
While the difficulties of working within
a developer-funded scenario are not
underestimated, it is not acceptable for
geophysical survey to be commissioned
on the hoof, after a hasty phone call.
The following sections on briefs and
specifications are a guide only, pointing to the
type and level of information usually required.
These are not meant to be inflexible, and the
documentation will need to be adapted to
the circumstances of each survey or project.

3.1 The brief

A requirement for geophysical survey may
become apparent during either the appraisal
or the assessment stage in the response to
an application for development.The earlier this
is realised and incorporated into a brief the
better. Clients and curators are encouraged
to seek specialist advice to ensure that the
content of the brief is fully appropriate to
the circumstances in each case. If necessary,
independent advice on geophysical survey can
be sought from outside the commercial sector,
for instance from the English Heritage Regional
Science Advisors or from the English Heritage
Geophysics Team (see contact details in
Appendix II).
The following information usually needs to
be provided in a brief:

Summary: a concise statement (200 words

3 Briefs and specifications
Definitions of these terms are provided in
the glossary and references can be found
in the bibliography. In particular, readers are
referred to the Standard and Guidance for
Archaeological Field Evaluation published by
the Institute of Field Archaeologists (IFA 2001:
http://www.archaeologists.net/modules/icontent/
index.php?page=15).

maximum) of the purpose of the survey,

what type of survey is required, by whom,
why, where and by when a report must
be delivered.

Background: a brief account of the relevant

context to the survey requirement. It must
include the following:
OS NGR location(s)
designations (eg Scheduled Monument
number(s))
archaeological context (eg evidence from
APs, surface remains, documents)
relevant recent history of the site
(eg landscaping)
reason for the survey
any wider project context

In a commercial tendering situation, briefs are

provided by the client, and tenders invited
(Project Start Up: see above); tenderers will
respond with a specification or project design
(Project Initiation: see above). If a tenderer feels
that a differing approach to that identified in the
brief might better suit the circumstances, then
this can be proposed as an alternative and
separately costed specification.The final
specification or project design will then be
agreed with the planning archaeologist or
curator, and will form part of a contract that
must be drawn up in writing. Being of such a
specialist nature, geophysical survey is often

Site conditions: a site description, to include

the following:

underlying solid and drift geology, and

soil type(s)
7

ground/vegetation conditions at the time

of the survey
ownership

Survey type: a statement of the geophysical

The report: a statement to the effect that

technique to be used examples might be:

all fieldwork, data processing and reporting

must follow the recommendations set out in
these guidelines. State how many copies of the
report are required, and what arrangements
are in place to deposit one of these with
the HER.

Survey location: a map of a suitable scale

to show the context, location and size of

the proposed survey area(s).

The geophysical survey requirement: this will

state the objectives of the geophysical survey
and the methodology by which these are
intended to be achieved.The detail of the
required methodology will be provided in a
separate specification (which may follow as
part of a combined Brief and Specification).
In the meantime it is sufficient to identify that
geophysical survey is required, although a
more specific methodology can be indicated,
for example:

earth resistance area survey

detailed magnetometer area survey
detailed GPR area survey
GPR profiles

fluxgate gradiometer area survey

alkali-vapour magnetometer area survey
earth resistance area survey
EM soil conductivity area survey
GPR profiling
GPR area survey

Survey instrumentation: it is not usually

necessary to specify the make or model of
equipment (however, these should be stated
in the resulting survey report).

Access: a statement of access arrangements,

GPR equipment must be suitable to meet
the required specification, specifically any
requirements for the centre frequency of the
antenna(s) to be used and the necessity for
antenna shielding. Note that, should topographic
correction be a requirement of this or any
other type of survey, care must be taken that
this is accounted for and costed as appropriate.

Survey methodology: a statement of

Timetable: a statement or tabulation of the
project timetable, emphasising the scheduling
of fieldwork and report presentation.

Further information: anything further of broad

relevance to enabling the survey work.

Digital archiving: a statement of what

arrangements are in place to ensure that
both survey documentation and digital data
are archived in line with current guidance
(see 6 Archiving).

methodology. For example: all methodologies

will follow those recommended in the
appropriate operators manuals for:

traverse/line separation
probe configuration (earth resistance surveys)
mobile probe spacing (earth resistance surveys)

providing clarity on how access to the site

is to be achieved, and any conditions on this,
together with a statement of whose responsibility
it is to obtain permission from the landowner
and/or manager.

Legal and other provisions: a statement of

any legal or other limitations relevant to the
survey (eg over Scheduled Monuments or on
National Trust property), and a clear statement
of whose responsibility it is to acquire the
relevant consents and licences in such cases,
and when this is to be done.
Timetable: a statement of time constraints
(eg for access to site), and the date by when
the report must be delivered.

3.2 The Specification

More specific and detailed survey requirements
are described in The Specification.This will
usually be separate from the preceding brief,
but if circumstances permit, the two may be
combined as part of the same document.

Sampling interval/density: the sampling regime

Feedback: a statement that the results of any

must be stated. The examples listed in Table 1

are suggested as the widest acceptable intervals
and traverse separations for evaluations
(although circumstances may dictate a denser
sampling for more detailed characterisations).

trial trenching or other excavation will be made

known to the geophysical survey contractor,
and that any subsequent commentary by the
contractor, will be included in the final project
report, if appropriate.

The specification should include the following:

Summary: a resum of the information

provided by the brief.

Table 1 Recommended sampling densities for various geophysical survey techniques.

Survey location: an annotated map or plan

Technique

Evaluation
(reading x traverse)

Characterisation
(reading x traverse)

For further
information see

magnetometer

0.25m x 1.0m

0.25m x 0.5m

Part IV, 1.2

earth resistance

1m x 1m

0.5m x 1m or
0.5m x 0.5m

Part IV, 1.3

GPR*

0.05m x 1m

0.05m x 0.5m

Part IV, 1.4

electromagnetic (EM)

1m x 1m

0.5m x 1m or
0.5m x 0.5m

Part IV, 1.5

EM for geomorphology

5m x 5m

Part IV, 1.5

topsoil magnetic
susceptibility

10m x 10m

Part IV, 1.6

indicating which areas are to be surveyed.

If different areas require differing survey
methodologies, then these should be indicated if
possible.The map can also be used to provide
other important information (eg access routes)
where necessary.

The survey grid/co-ordinate system: the

following needs to be identified:
a temporary/permanent survey grid is to
be established
responsibility for doing so (usually the
survey team)
accurate location of grid intersections ( 0.1m)
georeferencing (measurements to permanent
features to allow the grid to be exactly
re-located if necessary by a third party)

* These are general recommendations but for GPR survey appropriate reading intervals are
highly dependent on the centre frequency of the antenna used.

Further information: anything further of

Note that any pilot survey should be the

subject of separate and equivalently detailed
documentation, although this may be undertaken
in advance to inform the completion of a
final specification.

Plans/plots: survey grid location (1:2500 min)

plot(s) of raw data (1:1000 min)
minimally enhanced
XY traces of magnetic data,
where appropriate
plot(s) of enhanced data (1:1000
min)
grey tone (or dot density)
interpretation diagram (1:1000 min)

4 The survey report

4.1 Summary

The end product of any geophysical survey

is the survey report.This should be a clear
and succinct text supported by tables, figures,
appendices and references as necessary. It should
stand independently of supporting material and
should combine the qualities of concise technical
description linked to lucid and objective analysis
and interpretation. It should be intelligible to
specialists and non specialists alike. It should usually
be accompanied by a statement of the authors
and contractors professional qualifications.

This should be a prcis of the principal objectives

of the survey and the extent to which they
were achieved.

specific relevance to realising the objectives

of the geophysical survey.

The minimum requirements of such a report

are summarised in the listing below, parts of
which are then described in further detail.
Title page:

title of report
author(s)
contractor
client
report reference number
date
Summary of results: an abstract
Introduction: site location (including NGR)
site description/history
survey objectives
Methods:
survey methods used
reasons for this choice
date(s) of fieldwork
grid location
geophysical instruments used
sampling intervals
equipment configurations
method(s) of data capture
method(s) of data processing
variables used for the above
method(s) of data presentation
Results:
description
interpretation
Conclusions: assessment of achievement
(or not) of survey objectives
results summarised
implications
geophysical research value
recommendations (if appropriate)
Statement of indemnity
Acknowledgements
References: list of works referred to
Appendices: technical details of methodology
and instrumentation, data
(eg mag susc tables; grid
location measurements)

4.2 Introduction
This should provide the reasons for the survey,
set against a brief description of the site(s) or
area(s) concerned. It should include reference
to solid and drift geology, soil type and local
geomorphology.The archaeological background
(if known) should be summarised and reference
made to previous fieldwork and/or publications,
as well as to other relevant information (eg
from the aerial photographic record and/or
any related field investigations).
Other introductory items include: date(s)
of fieldwork, National Grid References, any
research objectives, legal status of site(s),
ground conditions, weather, site peculiarities,
documentary history, and any other relevant
information.

4.3 Methods
The methods statement should be a concise
account of the survey methods used, referring
to an appendix or to other appropriate source
for a more detailed description of standard
methodologies. Above all, it is important that
the instrument type is specified, how the data
was gathered and at what sampling interval.
This information should be followed by noting
the methods of data processing and software
used. Reference should be made to the plots
presented with the report, explaining reasons
for their choice, if necessary.

4.4 Results
This section is usually the most variable in
content between one survey and another, and
between different practitioners descriptions
and analyses of their respective results.
Where more than one survey technique has
been used it is usually best to describe each
set of results and their interpretation under
a separate subsection. Similarly, where non
contiguous subdivisions of the survey area
are involved, these should each be dealt
with in turn.
Much will depend on the clarity and simplicity or
by contrast the complexity, of the results

as to how the report should proceed. Some

authors may prefer to write a factual account
of the survey results, followed by a section on
their interpretation and discussion. An alternative
is to set out a blend of objective description
and explanatory interpretation that draws upon
supporting information from other sources
(eg augering, aerial photography, trial trenching,
etc). However, exhaustive narrative detail,
anomaly by anomaly, is tedious and should be
avoided; instead, the maximum use should be
made of accompanying plots and interpretation
diagram(s).Where plots and diagrams are mostly
self explanatory, the associated text should
be brief. Most importantly, explanations must
be clearly expressed and the division between
objective reasoning and more subjective
circumstantial inference made distinct.The
interpretation of archaeological geophysical
data must inevitably include surmise and
this should be encouraged but the reader
must be left in no doubt precisely where
the areas of uncertainty lie.

4.5 Conclusions
The conclusions should address the survey
results with reference to the original objectives.
The overall archaeological significance of
the survey findings can be summarised and
conclusions drawn, where necessary, about
the need for future survey or research. In
developer-funded evaluations, unless it is
specifically requested in the specification, it is
not appropriate for the contractor to launch
into discursive assessments of archaeological
importance or to make curatorial
recommendations.
The names and affiliations of the author(s) of
the report should be stated at its conclusion,
as well as the date of its final draft (or this
information could be supplied at the beginning
of the report).

4.6 Site location plan(s)

In most cases these should be based on a
large-scale OS map, displaying National Grid
eastings and northings, and for which copyright
permission must be obtained. Other base plans
may be acceptable, so long as they allow the
entire survey grid to be shown, and they include
features that can be clearly and accurately
re-located on the ground, or identified on
the appropriate OS map.
The survey grid should be superimposed on
such a base map, and the opportunity may be
taken to number the grid squares for ease of
reference from the text; or the survey areas
may be shown by outline only. In either case it
is necessary to ensure that the surveyed area
is unambiguously indicated on the location
plan. Areas of the grid covered by different
9

techniques can be indicated by differential

shading or colours. Grid location measurements
can be included on the plan, so long as clarity
is preserved, or can be tabulated in an appendix
(although it is acceptable for this information
to be retained only on archive plans or in
site notes).

4.7 Data presentation plots and plans:

Much as one may hope that readers will have
assimilated all the written detail of the report
it is probably true that the greatest attention
is paid to the summary and conclusions, and
especially to the accompanying plots and
interpretation diagram(s).These latter, then,
should be of a very high standard and should
include the following components (sections
4.8, 4.9 and 4.10 below).

additionally, where helpful.The usefulness of

colours for data images lacks the subtlety of
greyscale and so need only be used sparingly, if
at all (but colour is of course otherwise highly
advantageous in other plans and diagrams).
The above recommendations are for plots
of detailed area survey by magnetometer
or earth resistance meter. Rather different
presentations may be required for other
classes of data. Closely spaced magnetic
susceptibility, phosphate or other point data
may be presented similarly, although symbols
of proportional size, or of graded shading, are
more effective for more widely spaced survey
data. A Key should always be provided. Profile
data (pseudo-sections, GPR, etc) can be
presented in tonal plots or in colour scales.

Each survey report should include at least

one plot of minimally processed, raw data. Raw
magnetometer data is usually best displayed
in greyscale or XY trace format (but not as
wire-frame diagrams) although this may not
be practical for very large surveys. Raw earth
resistance data is better plotted in greyscale
or dot density format. Raw data may undergo
minimal processing (eg edge-matching, zigzag
correction), but should not be filtered.There
should be a statement of any processing that
has been applied.

4.9 Plots of processed data

Although many experimental attempts may
be made to enhance images of the geophysical
data from a site, only the most representative
of these need be included in the report. It
may be necessary to state in the text that this
is so, and that the interpretation provided is
a synthesis.
Each plot should be annotated with the details
of the type of enhancement used. All plots,
whether of raw or processed data, should
include scale bars, scales indicating the range
and magnitude of the data on display, north
arrows and grid coordinates (where necessary).
As far as possible, separate plots should be at
the same scale and orientation to enable direct
comparison. A scale of 1:500 is often suitable,
although scales as small as 1:1 000 are acceptable
for large surveys. Plots may need to be at a
scale sufficiently large to allow measurements
to be made from them for the subsequent
location of excavation trenches. Greyscale
plots are to be favoured for the display
of magnetometer data but should be
accompanied by trace plots where these
provide complementary information that has
influenced the interpretation cited (see Part
IV, 2.2.1). Dot density plots, contour plots, 3D
wire-frame plots and the like can be used
10

Information about geophysical surveys

undertaken in England can be obtained
from the following sources.

English Heritage Geophysical Survey Database

(http://sdb2.eng-h.gov.uk) This is an on-line
index of geophysical surveys undertaken by
English Heritage since 1972, with hypertext
links to many reports completed since 1993.
The database also includes information about
all surveys undertaken on scheduled sites as
a consequence of Section 42 consents (see
below, sections 7.2 and 7.3). A limited number
of commercial surveys are also included.

Archaeological Investigations Project (AIP)

4.10 Interpretative diagrams

4.8 Plots of raw data

5 Dissemination
5.1 Sources of information

In some cases the survey plots by themselves

are of such stark clarity that further interpretative
aid, beyond annotation, and description in
the report text, is unnecessary. However, it
is usually essential to include a diagram, or
diagrams, as a supplement to the interpretation
provided in the text. It is recommended that
such graphics are at the same scale as the
survey plot(s), for ease of direct comparison,
or can be provided at a smaller scale as an
overview of the wider picture. In some
instances, the plots themselves may be
annotated, but this can be visually confusing
and should therefore always be accompanied
by an unannotated plot for comparison.
The creation of interpretative diagrams is
not an exact science, and often involves the
translation of a synthesis of various lines of
evidence into a single graphic image. While
such a diagram will convey much that is
objectively true of the original data, it will
also, to some extent, convey more subjective
impressions. As stipulated above concerning
data interpretation (4.4 Results), it is crucial
that the distinction between fact and surmise
is clear.To achieve this it is acceptable to
provide two diagrams: one that shows an
objective simplification of all the geophysical
data, and another one that shows a more
subjective archaeological interpretation of the
first. For the second diagram, particularly if it
is the only interpretative diagram to be used,
it is important that the graphical conventions
convey the nuances of the interpretation, but
are not misleading where there is ambiguity or
uncertainty. For instance, bold lines and sharp
edges should be avoided when attempting to
delineate the oft quoted tentative anomalies/
features.The use of too many conventions
and/or colours can be extremely confusing
and should be avoided. A full, explanatory key
of any conventions, symbols, and colours and
shadings used is essential.

This includes data from archaeological

investigations in England from 1990, with
the resulting gazetteers available online.The
entries, which include a separate category
devoted to geophysical survey, comprise short
abstracts summarising the work carried out,
information about the location of the site and
investigating authority/body and bibliographic
references. (http://csweb.bournemouth.ac.uk/
aip/aipintro.htm)
Gazetteers that include some information on
geophysical surveys are also being developed
in Ireland (excavations.ie), Wales (CBA Wales)
and Scotland (Council for Scottish Archaeology).

National Monuments Record Excavation

Index, based in the NMR offices in Swindon,
is in partnership with the AIP, and its online
catalogue has a limited number of summaries of
geophysical surveys (3418 records, April 2008).

Archaeological Data Service (ADS) The

ArchSearch facility is the online catalogue of
the ADS http://ads.ahds.ac.uk/catalogue/ and
allows the searching of records provided by
the AIP, NMR Excavation Index, and the OASIS
project (see below, section 5.2), as well as the
databases of many other participating projects
and organisations. It is therefore possible to
use the facility to search for geophysical survey
information, where each survey is described
in a summarised form. A small number (13)
of survey reports from West Yorkshire can
be accessed in their entirety, together with
the geophysical data (http://ads.ahds.ac.uk/
catalogue/projArch/wyas/).

Historic Environment Records (HERs) and

Sites and Monuments Records (SMRs) These
records, increasingly known as HERs, are
maintained by each local authority (LA) and
constitute each areas fundamental stock-take
of historic environment information
(http://www.algao.org.uk/la_arch/fs_HERs.htm).

Most HERs include information about

geophysical surveys, which is currently
abstracted by the AIP and hence made
available in the AIP gazetteers, and through
the NMR, and ArchSearch. Most HERs also
hold copies of geophysical survey reports
for the LA area concerned, but with varying
consistency, coverage and formats.

Summary information on geophysical survey

is now gathered at source as part of the
OASIS project (Online AccesS to the Index
of archaeological investigationS:
http://ads.ahds.ac.uk/project/oasis/). It is
therefore a third responsibility for surveyors
and/or their clients to complete the on-line
OASIS sub-form.

Other published sources The sources described

It should be further incumbent on the

geophysical surveying community not only to
make available information on specific surveys,
but more widely to continue to raise the profile
of its research and results through education
and outreach, using all available media.

above are aimed particularly at accessing

information from the mass of so-called grey
literature. Additional information about specific
surveys or projects can of course be found
throughout much of the published domain (see,
for example, references and the list of websites
below).The leading journal for the publication
of research and case studies is Archaeological
Prospection (http://www3.interscience.wiley.com).
A combined catalogue of many libraries
holdings is accessible at http://ads.ahds.ac.uk/
catalogue/ARCHway.html.

International Society for Archaeological

Prospection (ISAP) This society was established
in 2003 and is the main forum for communication
within the discipline, including an email discussion
group and a regular electronic newsletter.
All practitioners are advised to join
(http://www.bradford.ac.uk/acad/archsci/
archprospection/).

5.2 Dissemination requirements

Geophysical surveyors, and their clients, face
a responsibility to ensure:
that a copy of the full survey report is
deposited with the relevant HER (preferably
within 6 months of completion);
and that reports on surveys over
Scheduled Monuments are submitted to
English Heritage (within 3 months of the
completion of the work: see below,
section 7.3).

These obligations will ensure that fundamental

information on surveys is made available for
consultation, and allow for the continued public
accessibility of summary information through
the sources and mechanisms listed above.
It is recognised that public dissemination may
at times not be appropriate (eg in the case of
sites vulnerable to looting, or where sensitive
planning issues are at stake), but the principle
remains that, excepting such circumstances,
survey information should be made as widely
accessible as possible. Client confidentiality
can be respected for reports associated with
a planning application, but these should be
submitted to the HER within a reasonable
time (preferably within six months of the
notification of results to the LA).

6 Archiving
This subject is dealt with comprehensively
in the Archaeology Data Service document
Geophysical Data in Archaeology: a Guide
to Good Practice (Schmidt 2002). All those
involved in the acquisition and deposition of
geophysical information should be familiar
with this guidance and implement it where
practicable as current good practice.

The only national and international facility

for digital data deposition is provided by
the Archaeology Data Service (ADS:
http://ads.ahds.ac.uk/project/collpol.html) and
all those concerned should make themselves
aware of its current policy and requirements,
and seek advice as necessary.
In conclusion, until further guidance becomes
available, the minimum requirements related to
the archiving of digital geophysical data are that:
each project has a responsible digital
archiving strategy, agreed between
contractor, client and repository;
this allows for the adequate storage, security
and long-term accessibility of both raw
and improved geophysical data (sensu
Schmidt 2002);
the survey report includes all relevant
survey and data documentation, preferably
tabulated for ease of future reference; and
ADS advice and good practice is sought
and followed.

7 Legal considerations
At present there is a minimum requirement
that a report (see above, section 5.2) on each
geophysical survey should be deposited with
the local HER.
The ADS Guide (Schmidt 2002) proposes that,
in addition and as a foundation for adequate
digital archiving, there should be a systematic
and consistent tabulation of information
about the survey. At present this is not widely
practised. However, current proposals are
seeking, through the development of the
OASIS project, to provide a single tabulation
that subsumes the various current database
requirements into a single accessible source
of information about geophysical surveys.
Until further guidance on this becomes
available the survey report represents the
minimum requirement.
All geophysical data are now digital and the
preservation of these as a viable future resource
is a major consideration for all concerned. It is
crucial that the generators of such data should
have a strategy in place, from the outset of a
project, that allows for their adequate storage,
security and long-term accessibility (Schmidt
2002, section 4). At present, requirements for
digital archiving may be imposed through the
commissioning or specification process where
conformity with a particular digital archiving
policy or agency is a requirement. Surveyors
should always make sure that a consultation
has taken place at the start of a project
to ensure that appropriate procedures for
depositing archives are incorporated in the
specification or project design.

Note: it is intended that new heritage

protection legislation, currently expressed in
the Draft Heritage Protection Bill, published
as we go to press (April 2008), will come into
effect from about 2010. Once enacted, this
new Bill will supersede previous relevant
legislation such as the Ancient Monuments
& Archaeological Areas Act 1979.The advice
that follows reflects the current situation, but
will be updated once the new legislation is
confirmed. For the moment, we anticipate that
the licensing requirements referred to below
(7.2, 7.3) will be retained for England, for
registered heritage structures (including
former Scheduled Sites), and may be extended
to include registered heritage open spaces
(registered parks, gardens and battlefields).
The draft Bill may be accessed at:
http://www.culture.gov.uk/Reference_library/
Publications/archive_2008/DraftHeritage
ProtectionBill.htm.

7.1 Access
Although geophysical survey is subject to the
usual legal constraints concerning trespass
there will be instances when a landowners
refusal to allow access can be overridden
on the legal authority of a central or local
government department. The contracted
agents of the latter may thus be granted legal
powers of entry, as stated for instance under
Section 43 of the Ancient Monuments and
Archaeological Areas Act 1979.
It should be noted that, where powers of entry
can be invoked for the purposes of conducting
an archaeological survey, these powers do not
11

allow for the breaking of the surface of the

ground. If construed literally, this ruling forbids
the use of probes, augers and grid pegs. Soil
samples may be obtained in some cases for
engineering purposes, and these may be useful
to the geophysical evaluation, but it remains
illegal otherwise to break the ground surface
without the landowners permission. In all
circumstances it is a responsibility of the
contracting body to secure the goodwill
of the landowner and thence the required
permissions.

7.2 Metal detectors

Section 42 of the Ancient Monuments and
Archaeological Areas Act 1979 states that
the use of any device designed or adapted
for detecting or locating any metal or mineral
in the ground in a protected place requires
the written consent of the Secretary of State.
Such consent, known as a Section 42 Licence,
is obtainable direct from English Heritage
and is required before the use of such
instruments in the following categories
of protected place:
the site of a Scheduled Monument or
of any monument under the ownership or
guardianship of the Secretary of State or the
Commission or a local authority by virtue
of the Act;
anywhere within an area of archaeological
importance.
It is an offence to use a metal detector
in such areas, to remove any metal objects so
detected, or to fail to comply with any of the
conditions of consent issued under a Section
42 Licence.
Further information and advice on use of
metal detectors can be found in English Heritage
(2006). Information on protected areas, including
the location of Scheduled Monuments, can be
found at: www.magic.gov.uk. Data on Scheduled
Monuments in England can be obtained on
request from [email protected].

7.3 Geophysical survey

The restraints stated above also apply to
the use of non invasive geophysical survey
equipment. When such survey in a protected
place is contemplated a written application
for a Section 42 Licence must be sent to
the English Heritage Inspector of Ancient
Monuments (IAM) for the region.
The letter of application should provide full
details of the proposed survey, including: the
name of the monument affected, a plan of the
area to be surveyed, objectives of the survey,
a statement on the technique(s) to be used,
make of instruments, names of individuals
12

who will do the work and when the work will

take place.The application will be considered
by the IAM and may also be referred to the
Geophysics Team (English Heritage, Fort
Cumberland) for approval. Survey proposals
should not usually encounter any difficulties
in receiving consent, but applicants need to
appreciate that the issue of a licence cannot
be instantaneous and should therefore apply
as far in advance as possible.
The Section 42 Licence will restrict the
consent for survey to a clearly defined area
and will be limited to named individuals
or the nominees of a named individual or
organisation. A condition of consent is usually
that a copy of the survey report is sent to
both the relevant English Heritage Regional
Office and to the Geophysics Team (Fort
Cumberland) within a fixed period (usually
three to six months) after completion of the
fieldwork. A date will be given after which
the licence is no longer valid.
With the exception of Class Consents (eg
certain agricultural or forestry activities), any
disturbance to the ground on a scheduled site,
such as augering, requires Scheduled Monument
Consent from the Secretary of State. In practice,
small scale sampling of topsoil (<100g) obtained
by augering or otherwise is usually acceptable
under terms agreed in a normal Section 42
Licence. Any proposal for larger scale
disturbance, whether to topsoil or subsoil,
should be discussed with the IAM for
further advice.
The above requirements are specific to England,
although similar conditions apply in Wales and
Scotland. For Wales, geophysical survey of
scheduled sites requires prior written consent
from Cadw, the Welsh Assembly Governments
historic environment service. In Scotland, requests
for permission should be made in writing to
Historic Scotland which acts on behalf of the
Scottish Ministers.There are no restrictions
on geophysical survey in Northern Ireland,
although ground disturbance at any
archaeological site requires a licence from
the Environment and Heritage Service.
Note that an Archaeological Licence Agreement
is required for any surveys by a third party on
National Trust (NT) property. Enquiries should
initially be directed to the NT Archaeologist
for the Region (see www.nationaltrust.org.uk/).
Proposals will then need to be accompanied
by a project design stating the aims of the
project, a description of the methodologies to
be used, its location, relevant previous research,
proposed personnel, funding and a description
of the relevance of the project together with
an indication of its future application and

publication. A risk assessment and proof of

public liability insurance will also have to be
provided. If approved, the NT Archaeologist
will then complete and send two copies of
the Archaeological Licence Agreement to the
applicant for signing. Conditions will include
the requirement that the licensee, upon
completion of the survey, will provide a
completed SMR form, a final report, copies
of any resulting publication and copies of
related records.
The operation of GPR equipment anywhere
within the UK requires an appropriate licence
issued by Ofcom and adherence to an agreed
code of practice (see Part IV,1.4.4 for full details).

Part III
Guide to Choice of Methods
1 Introduction
Geophysical survey should be thought of as
one of the main techniques of site evaluation
and its potential contribution must always be
considered in each instance where development
is proposed.
The purpose of the following section is
to provide advice that will be helpful to
archaeologists in determining whether or
not a geophysical survey is required in a
particular instance, and, if so, what techniques
and methodologies may be the most useful
to consider.

allow. Choosing an appropriate survey strategy

is never straightforward: it will depend upon
the interplay of many factors, and will therefore
vary from one site to another. It is rare that
any one strategy can be singled out to the
exclusion others, and different surveyors may
well arrive at different procedures, each of
which will have merit for different reasons.
It will be assumed that those who commission
surveys will probably take specialist advice in

each situation.The following tables are offered

as a rough preliminary guide to the options that
should be considered further.
The first guide to choice of survey offered
here (Table 2) is in the form of a key. Start at
the top of the table with the first question and
follow the directions in the right-hand column
to sections further down the table, and so
on, leading ultimately to a suitable survey
option (or options) for the problem in hand.

Table 2 Choice of geophysical survey: a key.

go to:

The choice of survey method(s) will vary with

the site conditions, logistics and time constraints
particular to each separate evaluation project.
Adequate time should be allowed for the
geophysical survey to be undertaken and
reported on once this has been identified as
a preferred evaluation technique. Clients must
be assured that the appropriate methodology
is being applied in each case.

It is obvious too, that within this broad concept

of integration, geophysical survey itself offers a
variety of approaches that can and should be
used together to their mutual advantage. All
projects need to give consideration to the full
breadth of techniques that might be applicable
to an evaluation, and to develop a specification
that maximises their joint potential. For example,
magnetometer survey may provide a distribution
of pits, ditches and industrial features, but it will
usually be necessary to combine this with more
targeted earth resistance survey and/or GPR
to identify building foundations. For the purposes
of evaluation alone, however, it will often be
sufficient for the choice of techniques simply
to give an indication of the archaeological
potential; the use of more elaborate integrated
survey strategies will be a feature of research-led
projects aimed at more detailed archaeological
interpretation and towards advancing
methodological development.

14, 24, 28

Are the archaeological features

deep (>1.0m) or
shallow (<_ 1.0m) or
very magnetic (eg kiln)
weakly magnetic?

2, 11
4
10
11

16, 20

Is the geology
metamorphic/sedimentary/drift or
magnetic (eg basalt) or
drift with magnetic pebble components?

5
6
6

15
15
15

Are the expected features mostly

masonry/stonework or
cavities or
large earth-cut features (eg channels, moats) or
industrial features (including hearths, etc) or
ordinary earth-cut features (ditches, pits, etc) or
other burnt feature (eg building)
diffuse/small (ie not major earth-cut features)
unknown?

6
2, 9
6, 10
10
10
10
10, 11
10

Try GPR (earth resistance and/or magnetometer survey may

occasionally be appropriate, if conditions permit).

Try Twin Probe earth resistance

GPR or
EM.

Is it the possible site of a

building or
a major linear feature (eg road, wall)
features at depth (>1.0m)?

3
2
7

2 Choice of geophysical survey

Geophysical survey is of course one of many
possible approaches to the evaluation of
archaeological potential, and its contribution
must be appropriately balanced with others so
as to optimise the project outcome. A typical
combination might include data derived from
aerial photography, map regression, geophysics,
field walking and test-pitting. Ideally, data-sets
such as these will be analysed and interpreted
within a GIS environment.

Is the site/area
rural, semi-urban or
urban (built-up)?

page:

24
28
34

6
6, 8
9

Try earth resistance traverses,

EM traverses or
EM area survey.

34
34

Try electrical resistance tomography (ERT)/pseudo-section profiles.

24

10 Try magnetometer area survey.

11 Consider magnetometer area survey using alkali-vapour instrument.

These guidelines are purposefully kept as brief

and concise as the complexities of this subject
13

Page numbers in the right hand column refer

to pages elsewhere in the document where
more detailed discussion is available: readers
are cautioned not to accept a survey option
without consulting the relevant sections of
Part IV. The same advice applies to Table 3,
which lists some of the most commonly
occurring types of archaeological feature, and,
alongside each, attempts to categorise the
suitability of the main survey techniques for
its detection in each case.Table 4 lists very
generalised comments on the suitability of the
major solid and drift geologies to magnetometer
survey only (the responses of other geophysical
techniques to differing geologies are less easy
to categorise simply; where possible reference
to these responses is made independently in
Part IV).
In submitting these tables we must acknowledge
that they are a considerable over-simplification
and therefore reiterate that they are intended
to serve only as a rough guide to choice of
survey technique. Professional opinion varies
on some of the attributions offered. For the
moment, and into the foreseeable future, each
situation will warrant specialist advice and this
should be sought at an early stage in any project,
once the general necessity for geophysical
survey has been established.

The depth and complexity of most urban

stratigraphy, closely constrained by modern
intrusions, metallic contamination, services and
adjacent structures, provides a near insuperable

Table 3 Matching survey method to feature type: survey options (see key below): the choice
of geophysical survey method(s) appropriate to a range of archaeological features, based on
experience from the UK. Only the most commonly used survey methods are listed.This is a
rough guide only, to which there will be exceptions, depending upon individual site circumstances
and future technical developments.
Feature type

Mag area
survey

areas of occupation

below artefact scatters

large pits (>2m diam)

smaller pits (<2m diam)

ring gullies (prehistoric)

post-holes (>0.5m diam)

hearths

kilns/furnaces

sunken-featured buildings

Earth res
survey

GPR

EM
(cond)

Mag susc

house platforms

ditches (<2m width)

large ditches (>2m width)

palaeochannels

roads/tracks

robber/bedding trenches

timber structures

masonry foundations

brick foundations

paving/floors

buried megaliths (mag)

3 Costs

buried megaliths (non-mag)

Routine archaeological surveys are usually

costed per hectare of area covered at standard
sampling intervals. Such prices are usually
inclusive of all aspects of the work and the
supply of a report (and a specified number
of copies of this). However, in some cases
particularly geotechnical surveys quotations
may not be all-inclusive and fieldwork may
be costed per day on site with separate
charges for data analysis and reporting.There
may be a reduction if multiple techniques are
carried out on a shared grid and concessions
may be available if there is a research and/or
publicity interest for the company concerned.
Prices can vary significantly between different
companies and will of course vary according
to constraints peculiar to each site. Clients
are advised to obtain a range of quotations
for scrutiny. Care should be taken to
establish whether or not VAT is included.

stone-lined drains

other cavities

The tables are followed by a more specific

discussion of the survey options for a selection
of commonly occurring evaluation scenarios.
For those who wish to follow up aspects of
technique and methodology in more detail we
recommend consulting Part IV. Furthermore,
valuable complementary information is available
in the following publications: Clark (1996);
Gaffney and Gater (1993; 2003); Gaffney,
Gater and Ovenden (2002); Linford (2006).

14

4 Urban (and brownfield) sites

On completion of the tendering process it is

good procurement practice for the client to
name the successful contractor, to declare the
range of prices received and to provide a list
of tender applicants.

graves

cremations

ridge and furrow

lynchets

waterlogged contexts
key:

Y The technique responds well in many conditions and is usually to be recommended.

y The technique can respond effectively in many conditions but is best used in conjunction

with other techniques.

? The technique may work well in some conditions, and its use may therefore be
questionable; another technique might be preferable.
n The technique may work in some conditions but is not usually recommended; another
technique is usually preferable.
N The technique is probably not effective, or its effectiveness is uncertain.

deterrent to successful geophysical survey.

An exception to this is when the survey is
intended to detect the remains of industrial
archaeology, which can often cause distinctive
and strong anomalies.
Tightly constrained sites in heavily built-up
areas do not usually offer suitable conditions
for geophysical techniques, with the possible
exception of GPR.This method is capable of
detecting some types of archaeological feature
(see Part IV, 1.4), and can also locate services
and structural detail within building fabric. It
is best applied when there is a measure of

foreknowledge of what is sought, and preferably

in conjunction with trial trenching or with coring.
Magnetometer survey over tarmac is possible
only in exceptional circumstances. It may be
possible over other types of paving but only
in relatively unusual circumstances when no
elements of the paved surface are strongly
magnetic. Earth resistance survey is not possible
over tarmac but electrical sections can be
collected over other types of paved surfaces
using plate electrodes and conductive gel or
bentonite clay (Athanasiou et al 2007). Such
surfaces are well suited to the use of GPR,

Table 4 Geology and the response to magnetometer survey.

Geology

Response to magnetometer survey

Igneous

Thermoremanent effects can preclude survey over some igneous

rock types (eg basalts); however, others (eg Cornish granites) seem
to be relatively unaffected.

Metamorphic

Experience so far suggests that thermoremanence is not usually

a significant problem and magnetometer survey can be effective
(eg over gneiss and slates); but beware of adjacent intrusions.

Sedimentary:

Magnetometer survey can be recommended over any sedimentary

geology.There are few significant distorting factors (but see below
under drift) although a wide range of magnetic susceptibility in the
parent rock results in a very variable background response to survey.

conglomerates/
grits/pebble beds

Response is average to poor (eg over Millstone Grit), but good in

places, eg Devonian grits.

sandstones

Average response is poor, eg over some Old Red Sandstone and

Mercian Mudstone; generally good over the Greensand, New Red
Sandstone and some Tertiary formations.

limestones

Response is good, especially over Cretaceous Chalk, Jurassic and

Magnesian limestones; less so over Carboniferous limestones.

mudstones/clays

Average response (London and Oxford Clays) is ?poor (eg Mercian

Mudstone); but results can be very variable.

drift

see below

Drift:

Quaternary deposits overlying the solid geology are a primary

consideration.They often show a high degree of local variation
and the magnetic response is usually dependent on the magnetic
mineralogy of the parent solid geology.

sands/gravels

Response is very variable; good on materials derived from Jurassic

limestones and in parts of East Anglia; moderate to good in southcentral England and in the west Midlands (Severn Valley).

coversands

Response is uncertain to ?poor.

boulder clay

Response is generally poor (eg in parts of East Anglia and

northern England).

clay-with-flints

Response is good.

brickearth

Response is average to ?poor; better in SW England.

alluvium/colluviums

Response is average to poor, depending for instance on depth of

burial of features below this material (see Part III, 6).

however, and this technique can be considered

for reconnaissance survey in the first instance
where surface conditions preclude the use of
other techniques.
On more open sites rough ground, verges,
gardens, allotments, playing fields, smaller parks,
cemeteries, etc the more traditional techniques
can be applied, although experience shows
that good results, while sometimes possible,
are not often obtained. Surface obstruction
or ground disturbance can prohibit sufficient
survey coverage and mar the survey response,
or both. Geophysical survey will not be
justified in many circumstances, although
magnetometer, earth resistance and GPR
methods can be invoked when encouraged
by specific expectations (eg of kilns, voids or
wall foundations). Decisions on survey method
and the interpretation of results must depend
on as thorough a knowledge as possible of
former land use.Trial trenching, coring and/or
test pitting may well be a preferable approach
in a majority of cases.

5 Cemeteries
Survey within present-day cemeteries, for
whatever purpose, while sometimes called
upon, is rarely very successful. Earth resistance
traverses, and GPR, can be used, where space
permits, to identify or confirm the course of
features (usually wall foundations) the presence
of which may already be suspected from other
sources of information. Note that permission
needs to be obtained from the church warden
prior to survey.
A more common difficulty is the detection of
former cemeteries or individual graves. None
of the techniques described above can easily
detect individual inhumation graves or cremations
owing to their relatively small scale and lack of
physical contrast between fill and subsoil. Stone
lined coffins or cists may be detectable with
earth resistance, or with GPR (Bevan 1991),
using a narrow sampling interval (0.5m x 0.5m
for earth resistance survey; 0.05m x 0.5m for
GPR), but ordinary graves in rural situations
are perhaps best sought with a magnetometer,
also with a narrow sampling interval. The
magnetometer response to ferrous items
such as chariot fittings or individual weapons
may give away the presence of graves, but it
is not possible to tell the difference between
these responses and those from irrelevant
ferrous items.
Individual cremation burials may be detectable
magnetically but the response is not normally
distinguishable from background variations
(nor, indeed, from anomalies from other
types of feature of similar dimensions and
magnetic characteristics).
15

Ferrous and non-ferrous items such as

coffin nails and grave goods are detectable
electromagnetically with metal detectors, the
supervised use of which can be valuable in the
detailed study of sites or of individual graves
(David 1994).
Graves, cremations or cemeteries can therefore
only be detected in very favourable conditions,
often only indirectly, and when there is already
good reason to suspect such features to be
present. Geophysical evaluation, particularly
over poorly known ground, will therefore easily
overlook this important category of feature.

6 Alluvium
The detection of archaeological features at
depths of >1m, whether covered by alluvium,
colluvium, blown sand, peat or other material
remains a major problem. Archaeology under
river alluvium, in particular, has attracted
much attention (Howard and Macklin 1999;
Needham and Macklin 1992) and the
problems encountered by geophysical
techniques in these circumstances have
been addressed by Clark (1992) and Weston
(2001).The use of geophysical methods as
part of a multidisciplinary approach to the
geoarchaeological evaluation of deeply
stratified sedimentary sequences has been
addressed by a number of authors (see for
example Bates and Bates 2000; Bates et al
2007; Carey et al 2006; Challis and Howard
2006; Powlesland et al 2006).
There can be no preferred recommendation
until the merits of each individual site or area
have been assessed. A pilot survey, linked with
coring or test pitting can be invaluable in the
subsequent development of a preferred full
evaluation. Depths of alluvial cover, magnetic
susceptibility values for the major sediment
units, and local geomorphology will all have
a significant bearing. Aggregates companies
may have commissioned borehole and other
surveys that can be helpful. British Geological
Survey (BGS) (http://www.bgs.ac.uk/boreholes/
home.html) and other specialist surveys may
also be available. Information on mechanical
coring as an aid to archaeological projects has
been published by Canti and Meddens (1998)
and by English Heritage (2007).
Magnetometer survey should usually be the
method of choice (see Part IV, 1.2). Depending
upon relative magnetic susceptibility values of
the fills of smaller features, alluvium and subsoil,
and the depth of burial, archaeological sites may
be detectable up to 1m down (Clark 1992).
The deeper the archaeology, however, the
less likely to be resolved are small and poorly
magnetised features. Magnetic anomalies show
a tendency to broaden as they become more
16

deeply buried by alluvium. While larger ditches,

pits, hearths and kilns, etc may well be detectable
at depths of 1m or more, the signal from smaller
features will be too weak; many types of site
especially pre Iron Age ones and those without
significant magnetic enhancement (eg most
ritual and many ephemerally occupied sites)
can be missed altogether.
Magnetometer survey should preferably target
shallower alluviated areas, and their margins,
and should, if possible, attempt to follow
detected features into areas of deeper alluvial
cover, thereby enabling an estimate of fall off
in local detectability to be made. Close
attention to available aerial photographic
and microtopographical evidence is always
essential (see Part IV, 1.10).
Survey with alkali-vapour magnetometers,
which have an increased sensitivity over fluxgate
instruments (see Part IV, 1.2), makes it possible
to detect weaker signals from more deeply
buried features. At present there are insufficient
case studies available from UK sites to
demonstrate a clear preference for one
or other type of magnetometer. It seems
inescapable, however, that the greater sensitivity
of alkali-vapour instruments will offer an
advantage over less sensitive instruments on
sites where variations in topsoil magnetisation
are minimal, as may be the case over some
alluviated sites (Linford et al 2007).The degree
of that advantage, and its archaeological
significance, remains to be quantified and will,
of course, vary from site to site. For the time
being, the use of alkali-vapour magnetometers
should at least be a consideration in evaluations
of alluviated areas where magnetic targets are
concealed at depths of >1m.
If magnetometer survey is ineffective there
may be some justification in attempting earth
resistance survey over suspected structural
remains, but problems of resolution at depth
(>1.0 m: Clark 1992), as well as the costliness
of extensive survey, can be prohibitive. Electrical
sections, using widely spaced electrodes (>1m)
can be of value in plotting the larger-scale
features of the sub-alluvial surface (Bates and
Bates 2000), although GPR, under suitable
conditions, is probably a more flexible and
rapid method (see Part IV, 1.4).
Area survey of topsoil magnetic susceptibility can
indicate general areas of artificial enhancement
derived from shallow archaeological horizons
and may be useful for directing subsequent
magnetometer survey. Magnetic susceptibility
data may also help map the alluvial edge if
this is not otherwise evident from other data.
Augering to obtain samples from sub-surface
horizons should be done to obtain control

measurements, but this is usually too time

consuming for any extensive area survey of
magnetic susceptibility, phosphate or other soil
component. Such detailed work would, in any
case, probably be inappropriate in the majority
of commercial evaluations.
EM survey (conductivity and magnetic
susceptibility) can be used to identify features
of gross geomorphology under alluvium, but
does not yet seem capable of detecting even
the moderate detail of archaeological features
buried at depth. Low-frequency GPR
(<200MHz) can also detect features such
as palaeo-channels and gross stratigraphy but
the signal attenuation of higher frequencies in
conductive soils either prevents or seriously
inhibits the detection of smaller archaeological
features (see Part IV, 1.4).
In summary, alluvial and other types
of superficial deposits present serious
difficulties for geophysical prospecting.These
are accentuated at depths in excess of a
metre. For large areas, a pilot survey can be
conducted, testing the suitability of various
techniques, although the emphasis may often
turn out to be on magnetometer survey.
Other survey techniques, such as GPR, can
be used more selectively but at present none
can be recommended as an adequate general
technique in these conditions. While some
archaeological sites may well be detectable
from the surface, it remains true that many
others, perhaps even the majority, will remain
elusive until made visible by more direct
intervention. However, the ability to detect
larger-scale geomorphological features, such
as palaeo-channels, and the value these may
have for indirectly predicting the presence
of archaeologically significant deposits, must
not be underestimated.
As things stand, whereas geophysics may be
helpful in some circumstances, archaeological
evaluation over deeper alluvium (>1m) should
rely on a combination of field techniques
centred on a scheme of test trenching, possibly
assisted by lidar (see Part IV, 1.10.3).

7 Wetlands
The problems of depth of burial, as above,
are accentuated by waterlogging.The only
technique that at present seems to offer any
potential is GPR over low mineral content
peat. At low frequencies (eg 100MHz) the
peat/mineral interface of peat basins is detectable
at depths up to about 10m (Theimer et al
1994; Utsi 2001), and reflections have also
been recorded from substantial objects such
as bog oaks (Glover 1987). Some case studies
(eg Clarke et al 1999) indicate that GPR is
also capable of detecting potentially significant

anomalies within peat, and there are reports

that wooden trackways or other structures
may be detectable (Jorgensen 1997; Utsi 2001).
Although such accounts are promising, there
is a need for further experimentation, and
reference to ground-truth, before GPR can be
recommended as a routine approach in these
circumstances. In other types of wetland, in clay
or saline situations, GPR and other techniques
are ineffective at locating organic structures.
Geophysical techniques can, as yet, have little
part to play in wetland evaluation. Structural
remains (such as pile dwellings, trackways, etc)
in organic sediments, in particular, are often
undetectable.Traditional dry land geophysical
techniques are best attempted in areas of
relative dryness and shallow overburden (islands
or wetland margins) and features so detected
may then have some indirect bearing on the
likely location of significant sites elsewhere
obscured. Aerial photography, lidar and remote
sensing (Cox 1992; Donoghue and Shennan
1988), linked with augering and test trenching
can offer the best overall evaluation, geophysics
being drafted in for the examination of specific
shallow or marginal sites.
It should be noted that magnetic susceptibility
readings on waterlogged material can be
suppressed by chemical changes (Thompson
and Oldfield 1986). Magnetic susceptibility
signals will persist in some lacustrine and
intertidal deposits, however (eg Linford 2003).

8 Road and pipeline corridors

The need to evaluate linear corridors traversing
many kilometres of countryside in advance of
the building of pipelines, new roads or the
upgrading of existing routes, continues to create
considerable demand for non destructive
evaluation (Lawson 1993). Geophysical survey
thus has a crucial role, and although the general
rules of survey as outlined elsewhere in these
guidelines apply, the special problems of survey
logistics, and the choice of an appropriate
balance of survey methodology, suggest that a
separate consideration is needed. Specifically,
while linear corridors may be comparable in
total area to the very large development areas
described below in section 10, their narrow
lateral extent makes them particularly amenable
to detailed survey over the entire development
area using modern survey methodologies.
Hence the considerations in this section override
those described below for extremely large
development areas in general.
It is stressed that the following recommendations
are general and do not attempt to set out a
rigid procedural blueprint. As for any call upon
geophysical survey, individual site conditions
will dictate a survey procedure that must be

expected to vary from one instance to another.

Inevitably, too, different survey specialists will
favour slightly different approaches.The following
attempts to set down basic considerations that
should be common to all.
Linear developments are complicated by the
large and extended area of land affected and
by the variety of geological and soil conditions
through which the route will inevitably pass.
Geophysical survey may often play a unique
role in the evaluation of archaeological remains
threatened by linear developments and should
be conducted at an early stage in the planning
process, when consideration of the results may
mitigate the route of the development to take
account of significant archaeological features.
While it is acknowledged that the destruction
caused by the linear development is the main
concern, consideration should also be given to
the impact of the development on obtaining
geophysical data in the future. In particular,
ferrous pipelines will produce a large area of
magnetic disturbance, up to 20m either side of
the pipe, which will compromise the subsequent
acquisition of magnetic and electromagnetic data.
A balance must be met between the cost of
obtaining adequate geophysical coverage, the
impact of the proposed development and
the anticipated benefits of the survey results.
The following specific points should
be addressed:
(1)The proposed geophysical methodology
should be appropriate for the location
of archaeological remains along the route
of the linear development.The results of
previous geophysical surveys conducted
under similar conditions should be
considered when recommending both
instrumentation and sample intervals;
note that a single technique may not be
suitable for the entire length of the
proposed development.
(2)Detailed area survey over a closely
sampled grid is to be preferred over any
unrecorded (eg magnetometer scanning)
or low sample density recorded methods
(eg topsoil magnetic susceptibility). In
all cases single long traverses should
be avoided.
(3)The area covered by such detailed survey
should be sufficient to encompass the
entire easement of the development and
any additional areas where damage to
underlying archaeological deposits may
occur (eg plant access routes).
(4)If possible, the survey transect should
also be of sufficient width to characterise
adequately the archaeological potential of
significant geophysical responses, particularly

linear anomalies, traversing the route.This

may save the need for any subsequent
requirement for additional survey to
further define enigmatic anomalies.
(5)The recent introduction of multi-sensor
geophysical instruments and platforms,
combined with GPS, has significantly
increased the rate of field data acquisition.
As a result, areas that in the past would
have been considered so large that they
could only be partially sampled, are often
now amenable to rapid and cost-effective
detailed magnetometer survey in
their entirety.
Providing no overriding geophysical contraindications exist (eg unfavourable geology
or soils, preponderance of modern ferrous
interference, etc), then magnetometer survey
should provide the most cost-effective method
of evaluation. A sample density of at least
0.25m x 1m should be used, which can be
collected rapidly in the field using a multisensor instrument.
Other geophysical techniques would not
usually be deployed blind over large parts of
a linear development and the considerations
for their effective use are the same as for any
exceedingly large evaluation area (see below,
section 10).
The width of the corridor to be evaluated
using geophysics will depend on the particular
linear development in question. However, in
the case of pipeline developments, given the
typical easement width and the area excluded
from subsequent survey by the presence of
the ferrous pipe or embankments, a minimum
linear transect width of 30m would commonly
be suitable. For road corridors the maximum
width is normally between 40m and 100m,
and this should always be completely covered.
Agreement should be reached with the client
as to whether or not a broader coverage to
either side of the corridor may be allowable
in order to place features within their broader
context. Broader coverage may also be
of benefit to the development, identifying
potential alternative routes to be planned
around areas where archaeological remains
are identified. Clients or their agents should
certainly be strongly encouraged to allow
for such contingencies, following appropriate
consultation.

9 Wind farms
Wind farms are a relatively new form of
development designed to generate electricity
from a sustainable resource.They require the
construction of a group of turbines usually
on a site in an elevated, exposed rural area.
Owing to their nature, it is necessary for the
17

turbines to be dispersed relatively widely

across the landscape and each needs a
firm foundation set into the ground. When
considering geophysical evaluation of wind
farm sites, it is preferable that the entire area
over which the turbines are to be distributed
is surveyed in detail using magnetometer
survey. If areas of particular archaeological
potential are identified, targeted follow-up
survey with more intensive techniques such
as earth resistance and GPR can then be
used, as for other forms of development.With
a full survey over the whole site it should be
possible to select individual turbine positions
so that the most archaeologically sensitive
areas are avoided.
However, if turbine positions are constrained
and the area of the entire site is so large that
it is considered unreasonable to survey it all
in detail, then consideration should be given
not only to the physical foundation of each
turbine but also to its magnetic footprint when
installed. Wind turbines are typically tall steel
structures that cause strong local magnetic
fields, which will influence sensitive magnetometer
measurements made in their vicinity. Once
in place it will not be possible to detect
archaeological remains using a magnetometer
within a radius of about 30m of the turbine.
It is thus recommended that, at minimum,
detailed magnetometer surveys of 100m by
100m areas be carried out centred on each
turbine position before their emplacement.

10 Extremely large areas

In some cases the total extent of a development
area may exceed the area that can be reasonably
surveyed in detail (that is, at a maximum sampling
interval of 0.25m x 1m for magnetometer survey
and 1m x 1m for earth resistance survey).
What is deemed to be reasonable will of
course represent a finely balanced equation
between several factors, not least of which will
be the available resources and the previously
established archaeological sensitivity of the
areas in question.The archaeological factor
should always be the prime consideration.
A preliminary essential is that the survey
potential of any area is assessed in the light
of existing desktop knowledge. Where the
priorities for survey are not then obvious, and
especially in the case of large areas (>20ha),
a pilot magnetometer survey can be carried
out before any further commitment to major
outlay of resources. Further preliminary field
trials to assess response to local conditions
may also be warranted, and these could include
magnetic susceptibility sampling, magnetometer
scanning or sample earth resistance survey,
as well as tests with EM or other specialised
survey techniques (eg GPR).
18

In rural, semi rural, and many other open areas,

where magnetic interference is not prohibitive,
there is usually no good reason not to undertake
detailed magnetometer survey over the entire
evaluation area, covering at least the ground
that will be destroyed or damaged by the
development. Increased archaeological sensitivity,
or other pressures, may demand that larger
areas are covered in detail.
Earth resistance survey will not normally be
undertaken blind over large areas, and will
only be applied where such survey is clearly
called for on the basis of independent
evidence (Part IV, 1.3).
In exceptional circumstances, where full detailed
survey is deemed not to be practicable, a
compromise between this and less intensive
sampling may be required, justified by the
commissioning body. In these cases, again,
magnetometer survey should usually be a
priority consideration.The following approaches
may apply:
(1)detailed geophysical survey of priority sites
already identified by desktop study; then,
once such sites have been accounted for,
(2)magnetic susceptibility survey of blank
areas (or the entire area if necessary),
at a maximum sampling interval of 10m,
followed by selected magnetometer
or earth resistance survey of areas of
magnetic enhancement;
(3)magnetic susceptibility survey with
systematic sampling by detailed
magnetometer coverage;
(4)augering should be allowed for, both to
obtain samples and to aid the direct
interpretation of specific anomalies.
Survey procedures should follow the
recommendations given in Part IV, 1.
The following cautions apply:
(1)Single long traverses of magnetic
susceptibility measurements must be
avoided and several parallel traverses
separated by a distance similar to the
measurement interval along the traverse
should always be recorded.
(2)Single earth resistance or magnetometer
traverses are not acceptable.
(3)Magnetic susceptibility measurements must
always be followed up by complementary
and more detailed survey in areas of
enhancement. Some areas lacking
enhancement must also be tested in
this way to demonstrate that, for the
area in question, variations in magnetic
susceptibility are primarily caused by the
presence or absence of archaeological
remains and not by changes in other

factors such as geology or recent land

use.To assist interpretation, magnetic
susceptibility values for different subsoil
types should be obtained for comparison
against topsoil values. If necessary this
relationship can be further examined by
comparison with fractional conversion
measurements.
(4)If there are any exceptional circumstances
making it impossible to evaluate an entire
development area using detailed area survey
over a closely sampled grid, it is still desirable
for at least 50% of the total area to be
sampled with detailed measured survey,
with the remainder sampled by one or
more of the methods referred to above.

Part IV Practitioners Guide to

Good Practice
1 Application of techniques
1.1 The survey grid
Geophysical fieldwork relies on the presence
of an accurately plotted network of control
points extending across the area to be worked
on and this is usually referred to as the survey
grid. An internally accurate and correctly
georeferenced grid is crucial to all subsequent
survey and to the whole project outcome: close
attention to this fundamental stage of fieldwork
is therefore essential. Recent developments
involving mobile sensor platforms incorporating
real time global positioning system (GPS)
sensors mean that it is no longer always
necessary to establish a conventional grid of
fixed markers over the surface of the area to
be surveyed (see below, section 1.1.2). When
employing such technologies, survey teams
should recognise that a grid of control points
capable of accurately defining the boundaries
of the area surveyed is still required even if
not actually laid out with ground markers
before or during the geophysical survey.
However the survey grid is located, during
fieldwork a record should be made relative to
it of surface conditions and sources of modern
geophysical interference that might have a bearing
on subsequent interpretation of field data.

1.1.1 Conventional survey grids

Establishing and marking out the survey grid
are usually the responsibility of the project
manager, although this should be discussed
and clarified with the geophysical survey team
involved.The grid can be laid out by any
suitably qualified personnel with the agreement
of (and, if necessary, following the instructions
of) the geophysical surveyors. Considerations
of geophysical methodology or ground response
may well dictate a preferred grid alignment,
particularly when the alignment of linear features
is already known. In this regard, Gaffney and
Gater (2003, 858) provide a concise review
of the issues common to most archaeological
geophysical surveys.
Where deadlines are tight, a previously surveyed
grid will allow the incoming geophysical survey
team to concentrate their specialist time to
greater effect. Where more time is available,
they may otherwise wish to provide the
survey grid themselves. Whoever lays out the
grid, it is important that its internal accuracy
and measurements to fixed topographic points
are rigorously and independently checked.
Geophysical survey teams are advised always
to check the accuracy of previously surveyed
grids and to take independent measurements
for grid location.There can be no excuse
whatsoever for any subsequent mismatches

between different parts of a geophysical survey,

or other positional confusion. It is preferable
and convenient, but not essential, for the
geophysical survey grid to match exactly
the Ordnance Survey National Grid (see
for example English Heritage 2003) or a site
grid devised for other purposes, such as field
walking.The need to fit a survey into existing
boundaries may dictate the use of a different
grid, however. If more than one grid must be
used, accurate location of each will be critical
for the subsequent integration of results.
A unit of either 20m or 30m for the side of
each grid square is usual (although some survey
methodologies may use a different optimal
base survey unit), with grid intersections located
on the ground using wooden pegs or other
temporary markers, which must be non-magnetic
for magnetometer surveys. Because of the many
hazards involved, not least of which concern
the safety of people and animals, the choice
of markers and their duration in the ground
needs careful forethought as well as the
agreement of the landowner and/or tenant
(see also Part I, 7.1).
The grid must be laid out using currently
accepted conventional methods (eg Bettess
1992; Bowden 1999; Clark 1996).
For long grid lines, in excess of 100m, the use
of a theodolite, EDM total station or GPS is
advisable. For smaller grids, the use of an optical
square is acceptable (eg English Heritage 2002).
English Heritage (2003, 89) provides a useful
summary of the different types of measurement
accuracy associated with survey grids, defining
relative, map and absolute accuracy. Using any
of the aforementioned techniques it should be
possible to locate the grid control points on
the ground to a relative accuracy of 0.1m.
GPS equipment is becoming increasingly available
and English Heritage (2003) addresses the

issues associated with its use for archaeological

survey, classifying the various types of GPS
system according to the positional accuracy
that can be achieved (navigation-grade, mapgrade and survey-grade). Survey-grade GPS,
capable of absolute positional accuracy of 0.1m
(either in real time or with post-processing),
is the only grade suitable for locating survey
grid control points. It should be noted that
the positional accuracy of existing base maps
may be lower, depending on how they were
originally created (see English Heritage 2003,
89). Bearing this in mind, it is advisable when
using GPS to locate the survey grid to measure
the positions of some fixed local landmarks or
boundaries recorded on the area base map
and not just record the temporary survey
grid points. Any discrepancies between GPS
positioning and local base mapping can then
be compensated for and it is also possible to
re-establish the grid independently relative to
the measured landmarks.

1.1.2 Interfacing with GPS

Recent developments in GPS technology mean
that it is now possible to interface geophysical
survey instruments directly to continuously
logging mobile (portable) GPS sensors, enabling
the position of each measurement to be
accurately located as it is taken (Fig 1). A
differential GPS system may be employed to
position measurements rapidly relative to a
field-based control station, which is subsequently
georeferenced to provide absolute accuracy
through post-survey processing. However, the
most recent real-time GPS systems can provide
immediate survey-grade absolute accuracy
by receiving broadcast signals from real-time
correction signal transmitters calculated from
a network of fixed control stations. With both
types of GPS system, it is possible to carry out
an accurately positioned geophysical survey
without first establishing a physical grid of
ground markers. It is important for the users

Fig 1 (above) The GEEP towed mobile sensor platform with built-in GPS (photograph courtesy of Ian Hill, University of Leicester).

19

of such systems to be aware that the same

considerations apply with respect to the
georeferencing of the survey area, as when
GPS is used to position a conventional survey
grid. For instance, the speed of data acquisition
might dictate that it is not possible to position
every geophysical measurement directly to
survey-grade GPS accuracy. Because of such
considerations the boundaries of the survey
area must be accurately georeferenced to the
same standard as would be expected when
a conventional survey grid is employed.
Portable GPS sensors mounted in a backpack
or on a mobile sensor platform (section 1.7.7)
afford freedom from the need to establish a grid
of fixed control points and then surveying a
series of regular parallel traverses between them.
However, with respect to geophysical survey
two considerations should be borne in mind.
Many geophysical instruments have a response
that is conditioned by their direction of travel
(eg magnetometers) and subtle archaeological
anomalies may not be distinguishable in a survey
where random measurement errors are
introduced by frequent changes of direction.
For this reason, a completely random walk
data-acquisition strategy is usually inappropriate
for geophysical surveys. An even density of
measurements should also be achieved over
the whole survey area, avoiding dense clusters
of measurements in some parts and very wide
gaps between measurements in others.
One way to avoid both problems is to emulate
the parallel, evenly spaced, traverses employed
in conventional surveys either by using a
portable navigational control linked to the GPS
system or by establishing a series of regularly
spaced aiming points at the edges of the survey
area. When employing such methods to ensure
even coverage, care should be taken to avoid
veering too far off-line when surveying each
traverse as this could result in overly wide gaps
between adjacent traverses resulting in a lack
of geophysical coverage in the intervening area
(Fig 2a).To demonstrate that an even coverage
has been achieved when not using a conventional
grid, the point cloud of measurement positions
should be plotted on the base map in the
survey report (eg Fig 2b).

1.2 Magnetometer survey

1.2.1 Choice of magnetometer survey
Magnetometer survey offers the most rapid
ground coverage of the various survey techniques
and responds to a wide variety of anomalies
caused by past human activity. It should thus
be the first technique considered for detailed
survey of an area and other, slower, techniques
should usually follow afterwards, targeting
smaller areas of interest identified by the wider
magnetometer survey.
20

a)

a)

d)

b)

b)

c)

e)

Fig 2 (above top) Some preliminary field trial data collected at Wroxeter Roman city using the system pictured in Fig 1: (a)
Greyscale plot of the caesium magnetometer results, which clearly show part of the Roman city plan; this data compares well
with hand-held magnetometer data collected over the same area using a more traditional, but much slower, survey methodology.
(b) Plot of the on-board GPS measurements showing the track of the system around the field; this was a rapid trial to test

different survey methodologies and the southern corner of the survey, where gaps are visible between the magnetometer

transects in (a), highlights the importance of ensuring even data coverage when not surveying on a regular grid (data courtesy

of Ian Hill, University of Leicester).

Fig 3 (above bottom) Handheld magnetometer systems: (a) Geoscan FM36; (b) Geoscan FM256 in dual sensor configuration

(photograph courtesy of Roger Walker, Geoscan Research Ltd); (c) Bartington GRAD601 dual channel fluxgate system; (d)

Scintrex SM4G Caesium magnetometer; (e) Foerster FEREX 4-channel fluxgate system (photograph courtesy of Norman

Bell, Allied Associates Geophysical Ltd).

Fig 4 (right) Cart mounted magnetometer systems: (a) Four Scintrex SM4 caesium sensors mounted at 0.5m intervals; (b)

two Geometrics G858 sensors mounted at a 1.0m interval (photograph courtesy of ArchaeoPhysica Ltd); (c) Foerster Ferex

4.032 4-channel fluxgate system with sensors mounted at 0.5m intervals (photograph courtesy of Institut Dr Foerster);

(d) two sets of SQUID gradiometers mounted at a 0.5m interval.

Fig 5 (far right) Greyscale plots of caesium (a) and fluxgate (b) gradiometer data acquired over the same series of Roman

enclosures at the same sample intervals (0.5m traverse spacing and 0.125 measurement intervals along traverses).

Instrumentation: Scintrex SM4G and Bartington Grad601 sensors in 1m vertical gradiometer configuration.

It can identify thermoremanently magnetised

features such as kilns and furnaces as well as
in-filled ditches and pits and areas of industrial
activity (both recent and ancient). Unless
composed of materials that contrast magnetically
with the surrounding soil (eg bricks carrying a
thermoremanent magnetisation), magnetometers
do not usually detect wall footings directly and
in this regard it is complemented by earth
resistance survey.

1.2.2 Instrumentation
The prime workhorse for routine magnetometer
survey in UK archaeological evaluation is the
fluxgate gradiometer.This instrument combines
sensitivity of the order of 0.1nT with lightweight
design and rapid measurement rates, and
several commercial systems are now available
in the UK. However, alkali-vapour magnetometers
are now becoming popular having long been
routinely used in continental Europe.These
instruments may also be named opticallypumped or caesium magnetometers (although
at least two other alkali metals potassium and
rubidium can also be used).They offer
sensitivities of the order of 0.05 to 0.01nT
and can make measurements at similar rates
to fluxgate systems.The commercial fluxgate
and alkali-vapour systems most commonly
employed in the UK are listed in Table 5 and
a number are pictured in Fig 3.
The main practical difference between the two
types of instrument is that an alkali-vapour
magnetometer measures the total absolute
magnitude of the local magnetic field, while
a fluxgate gradiometer measures the relative

a)

c)

difference between the magnitude of the

vertical component of the local field measured
by two sensors positioned one above the
other (separated typically by a distance of
0.5 or 1m). When required, it is possible to
configure an alkali-vapour magnetometer as a
gradiometer by differencing the measurements
made by two separate, appropriately mounted,
sensors (although in this case it is the vertical
gradient of the total magnetic field that is
measured). In general, alkali-vapour instruments
are more sensitive (Becker 1995) but it is usually
necessary to mount them on some form of
mobile platform or cart (Fig 4ab) thus
reducing sources of random measurement
errors to take full advantage of their enhanced
sensitivity. In practical terms, for a typical site
at UK latitudes, differences between the two
instrument types in resolving the primary
archaeological features appear to be relatively
minor (Fig 5) and the principal requirement
is that the magnetometer should have a
measurement sensitivity of 0.3nT or better.
It may be remarked that other types of
magnetometer are also available (eg proton,
Overhauser); however, their use for routine
survey would require special justification.
Whatever type of magnetometer is employed,
the operator should be fully familiar with the
manuals supplied with it (and any updates
provided by the manufacturer) and should
rigorously apply the recommendations for
equipment maintenance and survey procedure.
A number of manufacturers have adapted
their systems to allow multiple sensors to be

mounted horizontally in parallel.This enables

two or more traverses of data to be collected
simultaneously, increasing the speed at which
surveys may be carried out. While this is a
relatively recent innovation in the case of most
fluxgate systems (Fig 3b, c and e), multiple alkali
vapour systems, often deployed on custombuilt carts have been in existence for some
time (Fig 4ab). Cart-mounted arrangements
are also now being developed for some fluxgate
systems (Fig 4c). For any type of magnetometer,
these offer the benefits of reduced random
measurement noise and rapid area coverage
(a larger number of sensors may be mounted
in parallel, typically enabling four to six multiple
traverses to be measured simultaneously,
potentially with an integral GPS for positioning).
Set against this, carts can be more restricted in
the types of terrain in which they can operate as
compared to light-weight, hand-held instruments,
especially where the survey area is small and
constrained, so a range of field conditions can
mitigate in favour of the latter (Gaffney and
Gater 2003, 6872).

1.2.3 Methodology
Before beginning a survey the magnetometer
must be correctly prepared for use. Most
magnetometers require some warm-up
period before they settle into stable
operation.This is typically of the order of
five minutes for alkali-vapour instruments but
fluxgate gradiometers, being more sensitive
to differences in temperature, typically require
about twenty minutes to adapt fully to site
conditions. Most fluxgate gradiometers must
then be balanced (aligning the two fluxgate

a)

b)

d)
b)

21

sensors along the vertical axis) and zeroed

(calibration of the measurement scale for the
local conditions).This procedure should usually
be done over an area of uniform magnetic field,
preferably using the same location throughout
the survey. Particular care must be taken in the
selection of this location when calibrating dualor multi-sensor instruments as a proportionally
larger area free of local magnetic field
perturbations is required.
The operator must remove all sources of
magnetic interference from his or her clothing
(note: coins cannot now be assumed to be
non-magnetic). Particular care must be taken
to ensure that footwear is not magnetic and
that even small extraneous ferrous items (staples,
studs, tags, springs in zippers) are not present
in clothing. Note also that magnetic material
(including excessive amounts of soil) can become
attached to footwear (and sometimes even to
the instrument itself) during the survey and
can adversely influence the magnetometer
signal where the soil is strongly magnetic.
Clients should appreciate that there are some
circumstances (eg soil on footwear) that
cannot be easily avoided and may therefore
result in a slight deterioration in data quality.
Field conditions may dictate the type and
configuration of magnetometer that it is most
practical to employ. A cart-based system may be
of limited use in a confined area. Gradiometers
discriminate more strongly than total-field systems
in favour of anomalies in close proximity to the
sensors (Breiner 1999, 501).This property
can limit the maximum depth at which features
can be detected and total field systems are
perhaps more suited when remains are expected
to be deeply buried (eg alluviated environments).
However, gradiometers can survey in closer
proximity to modern ferrous objects such as wire
fences or pylons. Indeed, this configuration is
often the only way to carry out a magnetometer
survey near a busy road as it reduces the effect

of transient magnetic anomalies caused by

passing vehicles, which cannot be readily filtered
out by post-processing. Most archaeological
features will produce weak magnetic anomalies,
so magnetometers with several range settings
should be set at their most sensitive and
certainly ought to be configured to measure
differences of the order of 0.10.3nT. However,
in some instances (eg when surveying over
industrial archaeology or substantial kilns or
furnaces) reduced sensitivity may be necessary
to avoid saturating the sensors when mapping
very high magnitude anomalies.
Given the relative rapidity (and thus costeffectiveness) of modern magnetometers,
the preference should be for a detailed
magnetometer survey of the entire area
subject to evaluation.The area to be surveyed
is typically divided into a series of regular square
or rectangular blocks or sub-grids (see above,
section 1.1.1) and each is then methodically
surveyed by conducting a series of equally
spaced parallel traverses across it with the
magnetometer. Measurements are recorded
at regular, closely spaced, intervals along each
traverse. This is usually achieved by setting
the instrument to take readings at fixed time
intervals and using an audible time signal to
ensure an even pace, or by recording fiducial
markers at regular distances so that variations
in pace can be subsequently corrected for.
However, as noted in section 1.1.2 (above) some
recent magnetometer systems can integrate
directly with a GPS system to log the position
of each measurement directly and obviate the
need for a pre-established survey grid.
For detailed area survey the maximum
separation between measurements along
a traverse should be no more than 0.25m.
Clark (1996, 801) considers the sample
resolution necessary to discriminate between
near surface ferrous objects and more deeply
buried archaeological features and concludes

that a sample separation of 0.25m enables

full characterisation of anomalies with minimal
distortion to their shape. Schmidt and Marshall
(1997) examine the same problem from the
perspective of the sampling theorem.They
conclude that the sampling interval should
not exceed the burial depth of the features
being searched for. As the shallowest features
may be in the topsoil, typically some 0.20.3m
beneath the magnetometer sensor, a sample
interval of 0.25m is again recommended.
Modern magnetometers such as those listed
in Table 5 can sample rapidly (~10 times per
second), have data loggers with large internal
memory capacities, and can quickly transfer
stored data to a computer. Hence, sample
density along traverses has relatively little
impact on the time taken to survey an area.
However, the same is not true of the separation
between traverses where the time taken is
inversely proportional to the traverse separation
(closer traverse separation increases the number
of times the magnetometer must be traversed
across the area to achieve the necessary
coverage). Practicality dictates that some
compromise is necessary and, for evaluation
surveys, where the primary goal is to establish
the presence or absence of archaeological
remains, the maximum acceptable separation
between traverses is 1m. Clark (1996, 81) notes
that dense measurement along traverses is usually
effective for characterising the nature of features
so that it is then often sufficient only to establish
their extent in the cross-traverse direction.
However, where it is necessary for the survey
to identify smaller discrete features, such as
postholes, a closer traverse separation should
be used and 0.5m is recommended. Fig 6
illustrates the resolution of magnetometer
surveys at a variety of sampling densities over
two circular arrangements of postholes.The
most dramatic increase in the ability to resolve
the anomalies caused by these small features

Table 5 Selected magnetometer models and manufacturers by type.

Manufacturer

Model

WWW URL

Resolution

Multi-sensor?

Bartington Instruments

Grad601

http://www.bartington.com/grad601.htm

~ 0.3nT

single and dual sensor versions

Foerster

FEREX 4.021

http://www.foerstergroup.com/UXO/ferex.html

~ 0.3nT

4 sensor frame available

Geoscan Research

FM36 &
FM256

http://www.geoscan-research.co.uk/page71.html
http://www.geoscan-research.co.uk/page28.html

~ 0.3nT

single and dual sensor versions

fluxgate gradiometers

alkali-vapour magnetometers

22

Geometrics

G858-G

http://www.geometrics.com/858-d.html

~ 0.03 to 0.01 nT

single and dual sensor versions

Scintrex

SM4G

~ 0.03 to 0.01 nT

single and dual sensor versions

is achieved when the traverse separation is

reduced to 0.5m.The commensurate increase
in survey time required to cover areas at this
greater traverse density can be reduced by
the use of multi-sensor systems such as those
pictured in Fig 4. Multiple alkali-vapour sensors
can be mounted at separations of 0.5m, while
a similar effect can be achieved with fluxgate
sensors fixed 1m apart by the use of interleaved
traverses (eg see Gaffney and Gater 2003, 65).
Boundaries such as hedges and fences will
often constrain the orientation of the survey
grid. However, where possible, it is preferable
for traverses to be walked at right angles to
the direction of recent ploughing to minimise
any adverse effects of the latter on subsequent
plots. Where the alignment of anticipated
linear archaeological features can be predicted
in advance (perhaps from air photographic or
earthwork evidence), it is again preferable
to avoid orienting traverses in this direction.
Linear anomalies parallel to magnetometer

traverses can be inadvertently removed by

processing to counter the directional sensitivity
of the instrument. At all latitudes the greatest
peak-to-peak magnetic anomaly is obtained in
the northsouth direction (Breiner 1999, 41).
So, when employing a sampling interval along
the instrument traverses narrower than the
separation between them and if there are no
other constraints on traverse orientation, a
northsouth orientation will achieve optimal
benefit from the anisotropic sample density.
Fluxgate magnetometers can exhibit excessive
sensitivity to motion-induced errors when
oriented in a particular direction to the Earths
magnetic field, the direction being specific to
each instrument.Taking traverse direction into
account, care should be taken to avoid surveying
with the magnetometer while oriented in
this adverse direction, changing the way the
instrument is carried if necessary. A similar
consideration applies with respect to alkali
vapour sensors, which are insensitive to magnetic

fields in directions aligned too closely to a

particular direction dictated by the sensors
geometry known as the tumble angle. Again, care
should be taken to ensure sensors are aligned
appropriately for the local magnetic field direction.
Remedies for sensor orientation problems
tend to be instrument specific and the relevant
manufacturers manual should be consulted
in each case.
Instrument traverses may be recorded in either
zigzag or parallel fashion (Gaffney and Gater
2003, fig 10), with data logger settings and
subsequent data handling varying accordingly.
While zigzag traverses enable the most rapid
ground coverage, there can be a tendency for
the response of alternate traverses to be offset
with respect to one another.This can occur when
the magnetometer is not held in the correct
relative position or because of an incorrect
walking pace relative to an odometer rate.The
effect is often most pronounced when traverses
run up and down slopes and results in linear
anomalies at right angles to the traverse direction
being staggered and producing a herring-bone
pattern.The worst effects of this problem can
be eliminated by post-processing, but are often
difficult to remove entirely. Hence for portions
of a survey over particularly difficult terrain,
parallel traverses should be considered and
in all cases care should be taken to eliminate
the effect as far as possible by correct data
collection procedures.
Continuous-reading magnetometers may also
be used for scanning.The instrument is carried
along traverses spaced and oriented according
to local requirements without logging the signal
(Clark 1996, 8391; Gaffney and Gater 2003,
934). Its output is observed by the operator
and anomalies marked, then further investigated
by more intensive scanning or by detailed
recorded survey and/or augering. However,
the method depends for much of its success
on the experience of the operator, and even
the most skilled surveyors are unlikely to be
able to detect, by scanning alone, dispersed or
weakly magnetised features (and their patterns),
which may nonetheless be of considerable
significance.The technique can be a useful
preliminary exercise in assessing magnetic
response and locating well defined and strongly
magnetic features, but it should not be relied
upon as the sole geophysical method used to
evaluate an area.This applies also to variants
of the method that log the instrument signal,
using grid location and/or GPS navigation.

Fig 6 (above) Caesium magnetometer and fluxgate gradiometer data collected at varying sample intervals, illustrating the effect of
increasing traverse density for detecting discrete anomalies.

The latter is presently rarely practised and

the reliability of output from all methods
of scanning is difficult to assess. Advice
on this may change as methods become
more refined.
23

While most magnetometers now boast non

volatile storage capacities capable of storing
more than a days worth of surveying, it is
advisable to transfer data frequently to a
portable computer to avoid excessive data
loss in the event of an instrument malfunction.
Frequent checks of the data being collected
are also advised to ensure that adverse site
conditions or faulty instrumentation are not
compromising quality. Surveyors need to be
alerted to factors such as the incorrect balancing
of the instrument and the possible presence
of magnetic contamination on the operator,
as both can significantly distort data. If the
magnetometer is responding poorly to local
conditions then adjustments to the survey
procedure should be made to compensate for
these.To guard against unexpected failure of
the portable computer, data should also be
backed up to a suitable secondary storage
medium at the end of each days surveying.

1.3.1 Choice of earth resistance survey

about half that possible using a magnetometer,

so survey costs per unit area are generally
higher. It is thus particularly important that
earth resistance survey is used economically
and in circumstances suited to its particular
strengths.

The rate of coverage using earth resistance

survey is limited by the need to make direct
electrical contact with the ground by the
insertion of electrodes. A number of
developments, such as mounting electrodes on
a fixed frame as well as automated measurement
and data recording have greatly increased the
speed at which this can be done. Nevertheless,
the rate of ground coverage typically remains

Earth resistance survey can often identify

ditches and pits because they retain more (or
sometimes less) moisture than the surrounding
soil. However, in many instances the chances
of detecting these with a magnetometer are
higher and this more rapid technique should
be preferred. Exceptions might be considered
in areas of extreme magnetic interference or

guidelines do not aim to provide detailed

theoretical or methodological information but
simply set out to establish basic parameters
of good practice.

a)

1.2.4 Units of magnetic measurement

Magnetometers measure changes in the Earths
magnetic field and the SI unit of magnetic field
strength is the tesla (T) (Moskowitz 1995;
Payne 1981;Taylor 1995). However, this unit is
inconveniently large with respect to the weak
magnetic anomalies caused by archaeological
anomalies, so magnetometer measurements
are normally quoted in nanotesla (nT)
where 1nT = 10-9T. Gradiometers measure
the difference between two magnetic
measurements separated by a fixed distance.
Units of magnetic field gradient nT/m might
be deemed appropriate, but a true gradient
is only measured when the decay in magnetic
field strength is linear between the two sensors
and this is generally not the case unless the
nearest causative features are at a distance
much greater than the sensor separation.
Hence, most magnetometer manufacturers
simply quote the difference in nT between
the measurements made by each of the two
sensors and do not normalise for the sensor
separation. It has thus become accepted practice
in UK archaeological geophysics to quote
gradiometer measurements in nT. However,
where this convention is used, it is important
that the sensor separation is also noted, as
measurements made over the same anomaly
will differ depending on the sensor separation
of the gradiometer used.

b)

1.3 Earth resistance (resistivity) survey

While research continues to produce many
refinements to the electrical prospecting
technique, for most field evaluations standard
earth resistance survey is required. Details
of theory and field procedures have been
extensively aired in the literature (eg Clark
1996; Gaffney and Gater 2003) and instruction
manuals (eg Walker 1991). Hence, the following
24

Fig 7 (above) Caesium magnetometer (a) and earth resistance (b) survey of the same area of a Roman site in Hampshire.
Both detect ditches but the earth resistance survey reveals wall footings in clear plan where the magnetometer survey shows
just magnetic noise from ceramic debris.

where soil and geological conditions are not

conducive to the development of anthropogenic
magnetic anomalies. Conversely, earth resistance
survey should be favoured where building
foundations and other masonry features are
suspected, for instance over Roman villas,
ecclesiastical and other medieval buildings,
defensive works, etc. When applying earth
resistance survey there should already be a
strong presumption that such features exist
within the survey area. In this sense, earth
resistance is not a primary prospecting technique
and its application in many evaluations will
be secondary (Level II: Gaffney and Gater
2003, 8891).
Magnetometer and earth resistance survey
complement each other (Fig 7) and, for large
evaluations, it is often best to assess the area
magnetically first, followed by selected earth
resistance survey of areas identified as likely
to contain building remains. Choice of survey
method is rarely so simplistic, however, and will
depend upon a balanced expert consideration
of each separate situation.Those who
commission geophysical evaluation should
ensure that the particular works proposed are

a)

b)

adequately justified prior to the settlement of

the contract. It is especially important to be
certain whether or not earth resistance survey
is appropriate.

1.3.2 Instrumentation
While earlier resistance meters such as the
Bradphys and Martin-Clark systems are still in
use, they do not provide the pace of operation
or data handling facilities of more modern
instruments.The most commonly employed
resistance meters for contemporary area
surveys are listed in Table 6.These systems
make measurements automatically when
electrical contact is made with the ground and
can automatically record readings to on-board
electronic memory.The Geoscan RM15 system
(Fig 8) is particularly versatile, with optional
modular extensions creating a frame mounting
up to six multiplexed electrodes. Under
favourable conditions several measurements at
different electrode separations may be made
each time the frame contacts the ground; one
application of this facility is to speed data
acquisition by collecting two parallel traverses
of data simultaneously. Recent innovations have
allowed earth resistance meters to be used

c)

Fig 8 (above) Geoscan RM15 earth resistance meter in use (a) in standard twin electrode configuration; (b) with a multi-electrode
array addressed via an MPX15 multiplexer (photograph courtesy of Roger Walker, Geoscan Research Ltd); (c) mounted on an
MSP40 square array cart with a fluxgate gradiometer also attached.

with cart-based platforms on which spiked

wheels replace the traditional electrodes.
These platforms offer faster rates of ground
coverage and it is possible to mount other
instruments, such as GPS receivers or
magnetometers, for simultaneous coverage.

1.3.3 Methodology
The type and standards of grid layout are the
same as for magnetometer survey. For area
evaluation surveys the twin electrode (or twin
probe) configuration (Clark 1996, 38) will
normally be employed. Using this configuration,
the vast majority of buried features are detected
as simple single-peaked anomalies, and anomaly
shape is only weakly dependent on the
orientation of the electrode array (Aspinall
and Lynam 1970). Cart-based systems may,
alternatively, use the square array, which has
similar response characteristics but avoids the
need for fixed remote electrodes. However,
it should be noted that three different
measurement configurations may be used
with a square array (usually termed alpha, beta
and gamma) and each is preferentially sensitive
to anomalies running in a particular direction
(Aspinall and Saunders 2005). Hence, it is
recommended that both alpha and beta
measurements are made over a survey area
when using the square array.
Clark (1996, 57) considers optimum electrode
separation for the detection of features buried
at different depths. However, it is rare that the
precise burial depth of archaeological features
is known in advance and, for the twin electrode
array, a mobile electrode separation of 0.5m is
now standard and detects features up to 1m
beneath the surface.Where deeper overburdens
are expected, a separation of 1m is commonly
employed. Electrode separations much greater
than 1m tend to result in multiple-peaked
anomalies and unacceptable loss of definition.
Modern multiplexers and modular frames enable
measurements at several different electrode

Table 6 Earth resistance meters commonly used for UK archaeological surveys.

Manufacturer

Model

WWW URL

Type

ABEM

Terrameter LUND
imaging system

http://www.abem.se/products/sas4000/sas4000.php

64 channel multiplexed system for

electrical imaging surveys.

Campus

Tigre

http://www.campusinternational.co.uk/campus_tigre.html

32, 64 or 128 channel multiplexed

system for electrical imaging surveys.

Geoscan Research

RM15-D

http://www.geoscan-research.co.uk/page15.html

Lightweight meter for use with mobile

electrode frame. MPX15 6 channel
multiplexer available as addition.

TR Systems

Resistance Meter

http://www.trsystem.demon.co.uk/html/
resistance_meter.html

Lightweight meter for use with

mobile electrode frame.

25

separations to be collected simultaneously.

The combined results can provide a degree
of vertical characterisation for buried features
(Fig 9) or be used to filter out geological
trends and accentuate near-surface archaeological
features (Clark 1996, 1556).
Different geologies, soils, and differences in soil
moisture and chemical content can all affect
the magnitude of the earth resistance anomaly
caused by a buried feature; the optimum range
setting and measurement resolution will
therefore usually have to be determined for
each site at the time of the survey. Under
typical UK conditions measurements might
range between 0 and 200 ohms in which case
a resolution of 0.1 ohm would be suitable.
However, in dry conditions much higher earth
resistances can be encountered and a
measurement range of 0 to 2000 ohms might
be needed, in which case a resolution of 1
ohm would be acceptable.

that a stable measurement can be achieved.

Indeed, it may not be possible to survey for
up to several metres either side of such
underground electricity cables and pipelines.
In nearly all circumstances area earth resistance
surveys should be conducted rather than
isolated traverses.The use of isolated search
traverses, or widely spaced traverses, is only
acceptable when attempting to trace known
features, eg wall alignments, defences, ditches,
roads, etc. However, using such traverses blind,
in isolation from supporting data is not advised
because they are difficult to interpret. Now
that instrumentation enables relatively rapid
collection of data there is little excuse for the

key-hole type of survey unless site conditions

dictate it. Rather, the survey should be as
extensive as resources and good sense permit,
allowing a full appreciation of background
conditions against which to interpret anomalous
indications. Recent developments in cart-based
systems suggest that, in future, earth resistance
coverage of much larger areas may be
practicable within relatively short timescales.
Surveyors and their clients should of course
be aware that the resistivity response depends
on moisture contrasts in the soil, and that
these are in turn interdependent with climatic
regime, vegetation, soil and feature type. For
optimum results it is necessary to take these

The standard reading interval for earth

resistance surveys in field evaluations is
1m x 1m, and this sample density should
be adequate to detect the presence of
archaeology in most circumstances. Increasing
sample density to 0.5m x 1m or 0.5m x 0.5m
can produce sharper detail (Fig 10) but increases
the time required to survey the area (Clark
1996, 162) although modern multiplexed
systems can minimise the additional time
required. At the standard interval it should
be possible to cover about 0.75 to 1ha
per day.
Area survey with the twin electrode system
involves positioning two fixed remote electrodes
at a distance of some 15m to 30m (~30 times
the mobile electrode separation) from the
mobile frame and connected to it by a cable.
As the survey progresses it will become
necessary to reposition the remote electrodes
so that the survey can continue and care
should be taken to normalise measurements
between the new and old remote electrode
positions to avoid discontinuities in the
measured survey data (Gaffney and Gater
2003, 324).The need for normalisation may
be greatly reduced, or even eliminated, by
separating the remote electrodes from each
other by a large distance (Dabas et al 2000)
but at the expense of maintaining a greater
separation from the mobile frame (necessitating
a longer cable) and increased sensitivity to
electrical interference. Underground electricity
cables and pipelines with cathodic corrosion
protection can produce significant electrical
interference and, when working in their vicinity,
care should be taken to set the resistance
meters current frequency filters and
measurement averaging times to ensure
26

a)

b)

Fig 9 (above top) Earth resistance survey conducted using six different electrode separations over a Roman building at Wroxeter,
Shropshire.The closer separations detect near-surface features, such as the footings of internal partition walls, while the wider
separations preferentially detect the footings of the external, structural walls indicating that these continue to a greater depth
below the surface (data courtesy of Roger Walker, Geoscan Research Ltd).
Fig 10 (above bottom) Earth resistance surveys at Freens Court, Herefordshire, with readings at 1m x 1m sample density (a) and
0.5m x 0.5m density (b), illustrating the improved resolution of the latter, which resolves two rows of discrete post pad anomalies
in the eastern (bottom) part of the survey area.

factors into account and, preferably, to conduct

the survey at a time when moisture contrasts
are at their most accentuated, or to resurvey
the site at different times of year (Fig 11 and
Clark 1996, 4856). Regrettably, such approaches
will be unrealistic within the time constraints
of most development programmes and any
such limitations should be noted in the
subsequent report.

1.3.4 Electrical sections

Earth resistance measurements are most
sensitive to features buried at a particular
depth, which, as mentioned above, is influenced
by the electrode separation of the array used.
By repeating measurements at each point on
the surface using a number of different electrode
separations it is possible to obtain rudimentary
information about the variation of earth

resistance with depth a simple example using

six different separations is illustrated in Fig 9.
However, more detailed depth information
may be determined by laying out a large linear
array of electrodes (often 25 to 64) and
connecting them to a multiplexed earth
resistance meter with multi-core cables, so that
measurements at all possible separations and
positions are made (Milsom 2002, 114ff).
Each measurement uses four of the electrodes
in the array, selected by the multiplexer and
conforming to one of the standard electrode
configurations (eg Wenner or dipole-dipole).
By ascribing each measurement to a horizontal
location beneath the centre position of the
four electrodes used and a depth proportional
to their relative separation, a vertical slice
through the ground known as an electrical
pseudo-section can be built up. Such pseudosections contain distortions resulting from the
often complex interaction between the electric
current flow and resistive anomalies in the
subsurface (Aspinall and Crummett 1997),
but a more accurate electrical section may
be created using computer post-processing
with iterative inversion algorithms (see below,
section 2.1.3 and Loke and Barker 1996) a
technique often referred to as electrical imaging
or tomography. An example showing the use
of electrical imaging to characterise buried
wall footings is shown in Fig 12.

a)

b)

c)

Fig 11 (above top) Earth resistance surveys over the same area at Stanwick Roman Villa, Northamptonshire repeated at monthly intervals for eighteen months, illustrating the seasonality of the
response of archaeological features to this technique. High resistance (white) anomalies are clearest in winter when the soil has a high moisture content, while low resistance (black) anomalies
are clearest in the summer months, when there is a high soil moisture deficit.
Fig 12 (above bottom) Earth resistance survey at Basing House, Hampshire: (a) 0.5m twin electrode earth resistance area survey identifying a portion of the medieval curtain wall footings
(strong white linear anomalies); (b) pole-pole pseudo-section showing the earth resistance of a vertical profile along the line indicated in (a); and (c) inversion of the data shown in (b) clearly
showing the buried wall footing in cross section.
27

Electrical imaging has been employed with

some success to characterise archaeological
anomalies and three-dimensional surveys can
be constructed by measuring a sequence of
parallel sections and stacking the results
(Collier et al 2003). However, the technique
is slow compared to area survey methods, as
a large number of electrodes need to be prepositioned for each section. Electrical sections
are therefore usually employed to improve the
characterisation of anomalies rather than for
their initial discovery. For this reason they have
been little used in UK archaeological evaluation
surveys and should only be considered when
there is an agreed need to further characterise
potential archaeological anomalies after initial
discovery by area survey techniques.
Nevertheless, they are increasingly employed in
geomorphological studies to provide details of
buried landscapes associated with archaeological
activity. In this application, large geological-scale
sections are measured at strategically targeted
locations, typically using more widely
separated electrodes than for direct analysis
of archaeological-scale anomalies (Bates and
Bates 2000; Bates et al 2007).

Where electrical sections are employed, an

inter-electrode spacing suited to the scale and
depth of the expected anomalies should be
chosen.This might be as narrow as 0.51m
when imaging archaeological features, but
may be much wider (2m, 5m or more) for
geomorphological studies. Different electrode
configurations (Wenner, dipole-dipole, etc)
have different response characteristics (Loke
2004), so the configuration used and the reasons
for its selection should be noted in the survey
report. Care should also be taken to minimise
the contact resistances of each electrode in
the array (typically to <1000 ohms) before
initiating the measurement sequence. Most
data acquisition software for electrical sections
will include a facility to test the contact
resistances of each electrode and, where
contact resistance is found to be unacceptably
high, the insertion point can be moistened
with water and the electrode re-inserted to
improve the electrical contact.

1.4 Ground penetrating radar

Collectively, the term ground penetrating radar
(GPR) has been applied at an administrative
level within Europe to all methods of geophysical

survey utilising electromagnetic radiation in

a range from 30MHz to 12.4GHz to image
buried structures.This encompasses a wide
range of applications and the term is used
here to describe the more common,
commercially available GPR systems suitable
for archaeological surveys (Conyers and
Goodman 1997; Daniels 2004; Reynolds
1997;Vaughan 1986).

1.4.1 Choice of GPR survey

GPR can often be more costly than conventional
methods of area geophysical survey (eg magnetic
and earth resistance techniques), but does
present some unique capabilities to provide
estimates of the depth to target features and,
under suitable conditions, present threedimensional models of buried remains. GPR
can also be the only practical method to apply
on certain sites, or within standing buildings,
where the presence of hard surfaces and
above-ground ferrous disturbance precludes
the use of other geophysical techniques.
However, the resolution of vertical stratigraphy
is limited and highly dependent on both
site conditions and the instrumentation
deployed.

Table 7 Summary of expected GPR response over various types of site and features.

28

Type of site or feature

Expected response

Comments

building remains, foundations

and wall footings

good

Generally very well resolved; previous earth resistance survey may indicate
sufficient conductivity contrasts.

services

good

Modern services, particularly metal pipes, can be readily distinguished. Small bore
plastic services may be more difficult to image. More significant stone-lined drains
and conduits can also be resolved.

site stratigraphy

moderate

Providing adequate physical contrast between adjacent layers and features exists,
stratigraphy can be resolved within the limits of spatial resolution for the antenna
(Table 9).

voids and cavities

good

The contrast between air-filled voids and surrounding soil produces a strong
reflection. Distinctive polarity reversals of the incident wave form may also be
discerned. Partially filled voids containing rubble or water may also be resolved.

standing structures,
historic buildings

good

Specific architectural questions, such as the presence of hidden void spaces within a
wall, may be resolved. High frequency antennas are often required and are effective
for locating metallic features.

wetlands

moderate/good

Response may be highly site-dependent and influenced by the presence of high-conductivity

clays. Success has been reported for imaging targets in peat and below fresh water.

geomorphology

moderate/good

Lower-frequency antenna may be required in the presence of alluvial clays, but

palaeochannels and other large scale features can still be located.The depth of
overburden can also be mapped.

pits, ditches, post-holes

moderate

Site-dependent, but successful surveys demonstrate the suitability of GPR to these

feature types. Physical contrast and feature size can limit detection.

graves

poor

Dependent on the nature of interment and depth of the feature; stone-lined

coffins should provide a strong reflector.

A wide range of site surfaces may be considered

for GPR survey, including concrete, tarmac and
even fresh water, although the technique is
limited by the attenuation of the signal in
conductive media. In practice, this will largely
be determined by the concentration of clay
and the moisture content of the soil at the
site. Highly conductive media, such as metal
objects or salt water will prove largely opaque
to the GPR signal. Strong reflectors in the nearsurface will also reduce the energy transmitted
to immediately underlying targets and this may
include the local water table (or other nearsurface interface). Ferrous reinforcement
bars in concrete are also readily imaged by
GPR but their presence will not, necessarily,
preclude the identification of underlying
reflectors.
For normal ground-coupled antenna, good
physical contact with the site surface is necessary
to ensure adequate coupling of the radar energy
with the soil. As far as possible, vegetation
and any other surface obstructions should be
removed from the site prior to the survey.
High-frequency, air-launched horn antennas
are designed to be operated from above the
ground surface for civil engineering applications
(eg road deck investigations), but do not have
sufficient depth penetration for archaeological
surveys. Air-launched antenna may prove useful
for surveying delicate architectural features
(eg plaster mouldings, wall paintings or mosaic
pavements) when it is desirable to have no
physical contact between the instrument and
the surface under investigation.

While the use of GPR for detailed large area

surveys (>1ha) has increased it is often applied
as a complementary technique, following the
acquisition of magnetic or earth resistance
data, to target specific archaeological anomalies
identified over a more limited area of the site.
Care must be taken to ensure that GPR
survey is appropriate to a site, particularly
if it is the only technique to be applied.The
proximity to sources of radio-frequency (RF)
interference that may affect the data quality
such as mobile telephone transmitter base
stations or the radio modem of an on-site
differential GPS system should be considered.

1.4.2 Instrumentation
GPR systems utilise an electromagnetic source,
generated by a transmitter antenna on the
ground surface, and record the amplitude and
time delay of any secondary reflections from
buried structures.These secondary reflections
are produced when the GPR pulse is incident
upon any media with contrasting conductivity
( ) or (dielectric) permittivity ( ), or both, to
the medium above.The magnetic permeability
() of the sub-surface will also influence the
propagation of a radar wave, but for most
practical considerations it may be ignored.
In general, the GPR response will be largely
determined by the local variation of water
content in the sub-surface.The maximum
depth of penetration for a GPR is governed
by a combination of signal scattering and
attenuation within the subsurface, through the
dissipation of radio-frequency energy as eddy
currents within conductive media.

The majority of archaeological materials and

soils are semi-transparent to the GPR signal
and this is able to penetrate to some depth,
creating a series of secondary reflections from
buried objects distinguished by an increasing
time delay.The resulting time-amplitude data is
displayed as a two-dimensional profile with the
X-axis indicating the horizontal location of the
antenna on the ground surface and the Y-axis
representing the increasing time delay (depth)
from the initial impulse. While radar waves
propagate more slowly in the ground than in the
air, velocities are still extremely high and the
receiver electronics must be capable of recording
events separated by less than a nanosecond
(10-9s).The recorded delay represents the total
time required for an incident pulse to travel
from the transmitter to the target and then
for the reflection to return to the receiver.
This dual pathway is known as a two-way
travel time and can be converted to provide
the approximate depth of buried targets
where an accurate estimate of the sub-surface
velocity can be made.
GPR systems consist of an antenna unit housing
the transmitter and receiver, an electronic control
unit, a data console and a power supply.
Different configurations of these components
are offered by the major manufacturers and
each may have advantages in particular survey
conditions (Table 8; Fig 13).

Antenna units The GPR impulse covers a

comparatively broad band of frequencies,
usually defined by a nominal centre frequency.

Many site-specific variables must be considered

when using GPR, but in general it will respond
to a wide range of archaeological features
(Table 7), and is often successful over sites
where earth resistance survey has proved
fruitful (eg presence of masonry walls, void
spaces, etc). GPR is sensitive to the interface
between differing materials and some target
features produce highly distinctive GPR anomalies
(eg hyperbolic responses from point reflectors).
However, the identification of complex material
properties, for example distinguishing either
human or animal bone from the surrounding
substrate, is considered to be beyond the
capabilities of the technique under typical
field conditions.
Precise depth estimation from GPR surveys is
often difficult to achieve, yet is a critical process
for the successful presentation of results.
Unprocessed GPR data, expressed in terms
of the time delay of returned reflections, can
always be recalibrated in the light of additional
information (eg trial excavation results) to
present a more accurate physical depth estimate
for other unexcavated targets.

Fig 13 (above) Annotated photograph of a Sensors and Software Pulse Ekko 1000 GPR system.The sledge accommodates
either a 900 MHz, 450 MHz or 225MHz centre frequency antenna and maintains good coupling with the ground surface
through its flexible plastic skid plate.
29

Because of the increased attenuation of higher

frequencies, low-centre-frequency antennas
will provide a greater depth of penetration.
However, the longer wavelengths produced
by low-centre-frequency antennas will reduce
the vertical and lateral resolution of buried
targets and only physically large structures will
be resolved at depth (Table 9).The footprint of
the subsurface illuminated by the approximately
conical spreading of radar energy in the ground
is also frequency-dependent and increases
with depth (Annan and Cosway 1992, and
Fig 14).This may limit the effective depth
of investigation for certain targets and also
introduce reflections from objects buried to
either side of the instrument traverse.
The majority of commercial GPR systems allow
operation with a number of interchangeable
antenna units with different centre frequencies
to suit the soil conditions, depth of penetration
and resolution required. For near-surface
archaeological surveys a bistatic antenna

Fig 14 (above) The vertical and horizontal resolution of a GPR can be estimated from the centre frequency of the antenna
(fc) and the relative permittivity (r) of the ground from which the wavelength () can be derived.The footprint of the
conically spreading energy increases with depth (D) reducing the effective horizontal resolution (figure adapted from
Annan and Cosway 1992).

Table 8 Manufacturers of current GPR equipment used for archaeological surveys.

30

Manufacturer

Models

WWW URL

Type

ERA Technology

SPRscan

http://www.era.co.uk

cart-mounted system offering interchangeable 250, 500,

1000 and 2000 MHz antennas

GSSI

SIR,TerraVision

http://www.geophysical.com/

systems offering a wide range of fully inter-changeable

antennas from 40MHz to 2.2GHz options for cart mounting,
borehole and multi-channel use. Multi-channel (14) 400MHz
array (TerraVision)

IDS

RIS-ONE,
RIS-PLUS

http://www.ids-spa.it/

single- and multi- (8) channel systems using a range of

interchangeable antennas from 80 to 2000MHz and also
integrated multi-frequency units

MALA Geoscience

Ramac

http://www.malags.com/

compact system with a range of fully inter-changeable antennas

from 25, 50, 100, 200, 250, 500, 800MHz, 1.0, 1.2, 1.6GHz
options for cart-mounting, borehole and multi-channel use

Sensors & Software

Pulse Ekko
PRO, NogginPlus

http://www.sensoft.ca/

systems offering a wide range of mainly separable antennas

from 12.5, 25, 50, 100, 200, 250, 500, 1000MHz options for
cart-mounting, borehole and multi-channel use. NogginPlus
uses 250, 500 and 1000MHz antenna

3d-Radar

GeoScope

http://www.3d-radar.com/

digital stepped-frequency continuous wave radar system

operating in a frequency range from 30MHz to 2GHz; multi
channel integrated antennas (1 to 63 channels) allow highly
detailed data collection

Utsi electronics

GroundVue
1 to 5

http://www.utsielectronics.co.uk/

wide frequency range of twin-array beam focused antennas

covering 30 100, 125 500, 200 600MHz, 1, 1.5 and
4GHz, cart-mounting, borehole, and multi-channel (4) use

Radar Systems Inc

Zond-12e

http://www.radsys.lv/

dual channel GPR system with a range of ground-coupled

antennas between 100, 300, 500, 900MHz, 1.5 and 2GHz; two
air-launched antennas are also available operating at either a
single frequency of 750MHz or a 38/75/150MHz combined unit

unit, consisting of a separate transmitter and

receiver will be used, although these may be
enclosed within a common housing. Most midto high-centre-frequency antennas will also be
shielded to minimise unwanted reflections.
More specialised antenna units designed for
specific requirements such as borehole surveys
or high-frequency air-launched systems for
road pavement analysis are also available. Of
greater interest to archaeological surveyors are
multiple parallel antenna arrays, which allow rapid
acquisition of densely sampled data-sets.

Electronic control unit These units provide

the driving signal to the antenna and sample
the received response at a sufficiently high
frequency. Modern systems digitise the receiver
data directly, enabling detailed post-acquisition
processing. Some units may apply an analogue
gain directly to the signal prior to digitisation,
to improve the discrimination of later reflections,
but it is important to avoid clipping the response
beyond the maximum amplitude value recorded
by the system. Older analogue instruments,
producing only a graphical record of the GPR
traces, are not appropriate for archaeological
surveys because it is not possible to apply
any post-acquisition processing or visualistion
to the data.
Increasingly, GPR systems offer multi-channel
operation where two or more sets of antennas
can be recorded in a near-simultaneous manner.
This might allow a site to be covered with a
range of centre frequencies, imaging both nearsurface and deeper-lying targets, or a parallel
array of antenna units can be used for the
rapid acquisition of densely sampled data.

A laptop computer running suitable control

software can often suffice for this purpose,
using an internal hard disk drive for data storage
and a high speed transport bus to cope with
the large volume of data produced by the
GPR system. Integration with a co-located
GPS receiver or robotic EDM enables the
simultaneous collection of positional and
topographic data (eg Leckebusch 2005).

Power supply GPR systems require a

considerable power supply to function
adequately throughout the working day.This
power is usually supplied from a 12v lead acid
battery but a direct supply may be possible from
a vehicle mounted system. Lead acid batteries
can pose a Health and Safety risk because
of the weight of high power units and from
potential liquid acid leakage. Gel acid batteries
considerably reduce the risk of leakage.

System mounting More recently, integrated

GPR systems have been designed for single
user operation with all of the components
mounted on a compact, collapsible wheeled
cart.These systems are readily portable and
may be deployed on more confined sites
where the absence of trailing cables between
the various subunits can greatly speed the rate
of data acquisition.Transport of the antenna
units may be improved by mounting these in a
sledge with a flexible, plastic skid to ride over
uneven terrain while maintaining good coupling
with the ground surface. A GPR system may
also include an odometer wheel to automatically
trigger the unit at set distance intervals, although
these may require calibration when operated
over sites with uneven terrain.

1.4.3 Methodology
Data console The function of the data console
is to set the instrument parameters on the
control unit, to view the receiver output in real
time and to record the digitised data securely.

This section considers only the use of impulse

GPR operating in a common offset antenna
configuration. Alternative applications of GPR
are considered in section 1.7 below.

Initial field tests are recommended to confirm

that the equipment is functioning properly, and
that instrument parameters are correctly set.
Antennas of differing centre frequencies should
be trialled to determine an appropriate balance
between resolution and depth of penetration
(Fig 15). Operators should ensure that mobile
telephones and any other RF transmitters
in the immediate vicinity of an impulse GPR
antenna are switched off.The survey may have
to be conducted with more than one centre
frequency of antenna, either because of rapidly
changing site conditions (eg an increasing depth
of overburden) or the need to resolve targets
of differing physical size and depth of burial
(eg on a deeply stratified urban site).
If the instrument trials prove unsuccessful, or
suggest marginal data quality, then the survey
should be aborted at a pre-agreed fee. This
may be unnecessary for small surveys, where
data acquisition is unlikely to exceed a single
day in the field.
The requirement for the survey grid is similar
to other geophysical techniques but operation
on standing buildings may impose special
requirements for recording the position of
the antenna over the face of a wall or ceiling.
Survey transects should, where possible, be
positioned parallel to any surface irregularities,
for example kerb stones, to maintain good
antenna coupling with the ground surface.
Strong radar reflectors (eg metal fences, walls
or vehicles) present at the surface of the site
may produce spurious reflections in the data
caused by uncoupled energy leaking from the
transmitter.This may occur over sites with
uneven terrain where the antennas do not
make good physical contact with the ground
surface. Such air wave anomalies can be
distinguished in the data as characteristicly high
velocity (~0.3m/ns) and of limited attenuation
compared with sub-surface reflectors.

Table 9 Approximate values for the variation of GPR penetration depth and resolution with centre frequency for typical soils, encompassing a
range of values for dielectric constant and soil conductivities.The horizontal resolution will decrease with depth and is given for the maximum
penetration depth assuming a dielectric constant, r = 15.These values are intended as a guide and may be improved when a more detailed
estimate of the site conditions and target parameters are available.
Centre Frequency
(MHz)

Depth penetration
for typical soils (m)

Wavelength () in
soil r = 15 (m)

Horizontal resolution width of Fresnel zone

at maximum depth (m)

Vertical resolution
/4 (m)

1000

~1.0

0.08

0.2

0.02

500

~2.0

0.16

0.4

0.04

200

~3.0

0.39

0.8

0.10

100

~5.0

0.77

1.4

0.19

50

~7.0

1.55

2.4

0.39

31

Near-surface horizontal reflectors, such as

concrete surfaces or metal manhole covers,
may also cause the incident radar pulse to
reverberate repeatedly between the antenna
and the surface, resulting in high amplitude
multiple reflections (ringing) down the profile.
There are three main modes of GPR

data acquisition:

(1) Scanning GPR instruments provide a

real-time visual display of the recorded data
and may be used to locate known or suspected
features, perhaps during invasive works in the
field. Cart-based systems may be reversed along
the survey line while scrolling the data backwards
to identify the location of an anomaly.

the survey area for a given two-way travel

time (or depth estimate).Visualising the GPR
results in this format may greatly assist the
interpretation of complex data-sets (although
some types of anomaly, for example from
dipping reflectors passing through several
time slices, may not be adequately resolved).
Additional modes of display and data analysis,
including examination of the individual profiles,
are also recommended.The use of falseperspective, three-dimensional representations
of the data, such as cut-away solid models or
iso-volumes, may enhance the visualisation of
certain data-sets or anomalies, but should not
be used as the sole method of visualisation
(eg Leckebusch 2003; Linford 2004). Fig 16
provides examples of the various means of
GPR data display.

(2) Individual recorded profiles Single profiles

may be recorded over the suspected location
of known features or to investigate anomalies
identified by other geophysical techniques; for
example, to estimate the depth to a particular
target or to determine the course of a linear
feature over an extensive area where the
route may be interpolated between widely
spaced traverses.

(3) Detailed area survey Area survey over

a regular grid of closely spaced traverses is
strongly recommended for detailed GPR
investigations. Ideally, to avoid spatial aliasing,
traverse spacing should be less than the
approximate footprint of the radar energy
at the required depth of investigation (Fig 14
and Table 9). Under typical conditions for a
500MHz centre-frequency antenna any traverse
spacing above 0.25m will be spatially aliased.
However, as such densely sampled surveys are
difficult to achieve over large areas unless a
multi-channel instrument is available, a traverse
separation of 0.5m is suggested where spatial
aliasing will not be detrimental to the
interpretation of the target features.The
non-symmetric radiation pattern from a GPR
antenna causes the orientation of targets
with respect to the direction of the profile
to influence the anomaly produced. Repeat
survey over orthogonal traverses, or very
closely spaced parallel traverses (0.1m or finer)
will improve the definition of features running
closely parallel to the original orientation of
the data profiles (eg Conyers 2004, 67).
Profiles collected over a regular grid may be
acquired in either a parallel or zigzag fashion,
providing sufficient care is taken with the
positioning of the antenna to avoid any
offset between alternate lines.
The resulting high-density data are best
presented as a series of time slices where each
successive time slice represents the horizontal
variation of reflector strength (energy) across
32

The number of traces (scans) to be recorded

along each profile, the time window through
which reflections are measured for each trace
and the number of times each trace is repeated
at a particular sample point (stacking), should
be set to appropriate values to image the
targets under investigation. Because of the low
signal-to-noise ratio of the GPR signal, oversampling is recommended where this does not
adversely slow data acquisition. For a typical
archaeological survey, with a mid-centre

frequency antenna (500MHz), traces should be

recorded at least every 0.05m along a profile.
An increased trace density may be appropriate
for more detailed survey with a higherfrequency antenna. Establishing the correct
time window through field trials is, perhaps,
more important as this will determine the
maximum depth to which the GPR will
record data.
Any time-to-depth estimate should be supported
with details of how the sub-surface velocity
was determined and applied to the data, taking
into account any significant alteration of soil
type across the site or variation in moisture
conditions that may occur during the course
of the survey.This may be achieved through
either calibration between a recorded reflection
and a known-depth target, analysis of the shapes
of diffraction hyperbolas, common mid-point
(CMP) measurements made in the field (Fig 17)
or direct determination using time-domain
reflectometry.
Most GPR acquisition assumes the profile
is collected over a planar surface. Where
significant topographic variation exists this
should be recorded and an appropriate
elevation correction applied to the GPR data.
Under conditions of gently undulating terrain

Fig 15 (above) Trial GPR transect collected over peaty soil repeated with 450MHz (a) and 225MHz (b) centre frequency
antennas. At this site the lower frequency antenna has successfully identified three deeply buried anomalies that are only
partially represented in the higher frequency data.
Fig 16 (opposite top) Examples of modes of display for three-dimensional GPR data: (a) time slices showing the variation of
reflector amplitude at selected depths; (b) cut-away false perspective view of the whole data volume; (c) topographically
corrected data volume showing underlying anomalies (greyscale); (d) iso-volume representation of stone-lined passages leading in
to a souterrain feature; (e) buried land surface across a dry valley extracted from the GPR data beneath the (semi-transparent)
DTM; and (f) a volumetric interpretation of a Roman building abstracted from time slice data overlaid with a cloud of plough
damaged material.
Fig 17 (opposite bottom) An estimate of the average subsurface velocity (v) can be obtained by conducting a common mid-point
(CMP) survey in the field. (a) The distance (X), between the GPR transmitter and receiver is gradually enlarged about a central
point increasing (b) the travel time (T) of both the air wave passing directly between the two and the ground coupled wave
travelling through the very near surface, and any reflections, if present, from more deeply buried objects.The velocity of the
waves can be determined from (c) the slope of the reflections on a CMP profile, which can be further enhanced by the use
of (d) semblance analysis. In this case the velocity of the reflected waves from buried objects is approximately 0.075m/ns,
slightly lower than the ground coupled wave (~0.125m/ns).

a)

b)

c)
d)

e)

(within 0.5m) the elevation correction

may be applied directly to the GPR profile
as a static shift to each trace. However, more
severe gradients will also require a tilt-angle
correction to be applied to the data to avoid
discrepancies in the apparent location of
subsurface reflectors (eg Goodman et al 2006;
Leckebusch and Rychener 2007).The degree
of horizontal displacement will depend on the
slope angle of the surface and the depth of
investigation. For example, anomalies identified
at a depth of 1m below a slope inclined by
20 will be shifted horizontally by approximately
0.34m from the surface location of the
GPR antenna.
Detailed GPR survey will create large volumes
of data that will initially be stored on the
internal hard disk of the data console or
laptop computer. However, data back-up
at regular intervals to suitable high-volume
secondary storage media is recommended.

f)

a)

b)

c)

d)

Results from a GPR survey, whether visualised

as an individual profile or as a horizontal time
slice, should indicate the time delay and include
an appropriate greyscale or colour key to show
the variation in the amplitude of the reflections.
The recommended sub-unit for the two-way
travel time delay is the nanosecond (ns) and
the amplitude of the reflections will initially
be recorded as a potential measured by the
receiver antenna in the millivolt range, although
results following post-acquisition processing are
generally presented in arbitrary, relative units.

1.4.4 Radio licensing and emissions legislation

Owing to the increased demand for wireless
communications and the need to avoid
interference between electronic equipment,
legislation governing the use of the radio
spectrum and electromagnetic compatibility
(EMC) issues has been introduced and further
regulations are currently under development
at both a national and international level.
GPR equipment must, obviously, adhere to
the relevant legislation, but presents some
unique considerations that do not readily
fall into common categories of other similar
electronic devices, such as cellular telephones
or computer equipment.
Impulse GPR systems use a mobile, ultra-wide
band (UWB) transmitter operating at a lowpower output that is specifically designed to
emit this energy into an absorptive earthmaterial medium, typically the ground. For
archaeological applications of GPR this radiated
energy generally falls between 30MHz and
12.4GHz, a portion of the radio spectrum that
for administrative purposes is currently subject
to legislation at a European level through
standards set by the technical authority of
33

the European Telecommunications Standards

Institute (ETSI).The use of GPR equipment
meeting the ETSI standard has traditionally
been controlled at a national level but will
be harmonised by the Europe-wide regulatory
authority, the European Conference of Postal
and Telecommunications Administrations (CEPT).

CEPT informed by a public enquiry period that

closed in September 2006.The production of
draft regulation by CEPT is currently ongoing,
although some important technical issues,
requiring the harmonisation of ETSI standards
with regulatory decisions, are still to be resolved.
Generally, the areas of most concern are:

The use of GPR equipment within the UK must

conform to all current radio licensing and EMC
requirements. A formal licensing scheme for the
UK was introduced by Ofcom from 1 September
2006 to regularise the situation existing under
the previous waiver granted to members of
the EuroGPR Association. From this date lawful
operation of GPR equipment in the UK can only
be achieved within the terms of the Ofcom
licence, which requires full compliance with the
EuroGPR code of practice, including the use of a
site log for operation (see www.eurogpr.org and
www.ofcom.org.uk for more details). Equipment
rental pools will record site log details and
licensing arrangements for occasional users
hiring GPR instruments.These arrangements will,
eventually, be superseded by wider European
regulations following the recommendations of

airfields
prisons
defence establishments, including military
training grounds
radio astronomy sites

Most recently manufactured GPR equipment

will have been designed to meet current
EMC legislation and operate at a lower power
than previous comparable instruments.These
requirements also permit the operation of wideband pulse techniques following the current
entry in the UK Frequency Allocation Table on
a non-interference basis between 150 and 4000
MHz, excluding the use of certain antennas at
both ends of this range. Advances in antenna
design and integral electronics often result

in these modern systems surpassing the

performance, in terms of depth penetration
and signal-to-noise ratio, of the earlier generation
of instruments that they have replaced
(eg Sirri et al 2005). However, there are some
applications where the original, high voltage,
transmitters would provide the only means
to obtain sufficient energy for imaging deeply
buried targets in highly absorptive media.
Additional concerns for the GPR user
community are:
operation beyond the agreed bandwidth
(150 and 4000 MHz for the UK Frequency
Allocation Table);
compliance of older legacy equipment with
new regulations;
restrictions on the development of future
equipment; and
transmission surveys / vertical faces (control
of energy absorption).

1.5 Electromagnetic methods

A range of geophysical instruments make use
of electromagnetic (EM) waves, distinguished
by the frequency and duration of the source

Table 10 Manufacturers of current EM equipment used for archaeological surveys.

Manufacturer

Models

WWW URL

Type

CF Instruments

CM-031

http://www.alliedassociates.co.uk/files/cm31.html

Similar to the Geonics EM31, the CM-301 has a 3.74m coil

separation and operates at a frequency of 9.766kHz.

Dualem

various

http://www.dualem.com

A range of complete instruments and individual sensors is

available, operating at a frequency of 9kHz that allows the
simultaneous measurement of conductivity and magnetic
susceptibility from two coil orientations; also coil separations
of 1, 2 or 4m are available with some multiple spacing
instruments for depth sounding.

Geonics

EM38, EM31

http://www.geonics.com/

The EM38 has a 1m coil separation for near-surface surveys

and operates at a fixed frequency of 14.6kHz.Two variants of
the basic instrument are also available offering simultaneous
measurement of either both field components (EM38B) or
coil orientations (EM38DD).
The EM31 has a coil separation of 3.66m and operates at a
frequency of 9.8kHz, providing a depth of investigation to 6m.
A 2m coil separation is also available (EM31-SH) for
intermediate depths up to 4m.

34

Geophex

GEM-2

http://www.geophex.com/

The GEM-2 is a multi-frequency instrument operating over a

range of 300 Hz to 96 kHz with a coplanar coil configuration
separated by 1.66m. Geophex also manufacture a gradiometer
EM system (GEM-5) designed for increased immunity to
ambient environmental EM noise.

GSSI

EMP-400

http://www.geophysical.com/

The EMP-400 records 3 user selected frequencies simultaneously

from a range between 1 to 16kHz and has a coil separation
of 1.219m.

that they utilise. While such a broad definition

should include GPR, magnetic susceptibility
meters and metal detectors, these special
cases are considered individually elsewhere.
This section therefore considers only inductive
EM instruments, also known as slingram or
conductivity meters (Table 10).These emit a
continuous low-frequency (<300 kHz) EM
signal from a transmitter coil, that will in turn
generate a secondary field within any electrical
conductors present in the near-surface (eg
Wait 1955). A separate tuned receiver coil
records the modulated signal, where it is found
that the in-phase component is proportional
to the magnetic properties of the subsurface
and the out of phase, or quadrature, response
to the electrical conductivity.Theoretically, as
conductivity is the reciprocal of resistivity, this
modulated signal enables an EM instrument to
simultaneously collect data-sets comparable to
both the earth resistance and the (induced)
magnetic response (eg fluxgate gradiometer
survey) from a site.
While initial research demonstrated the ability
of EM instruments to identify archaeological
features (eg Scollar 1962;Tabbagh 1986;Tite
and Mullins 1969), the technique is not, at
present, widely used in the UK for archaeological
evaluation. In principle, as the coils of an EM
instrument do not necessarily have to make
contact with the ground surface they offer
the advantage of rapid field data acquisition,
combined with the simultaneous collection of
magnetic and conductivity data-sets. However,
considerable inter-site variability of the EM
response may be encountered, depending
on underlying geology and soils, requiring
calibration against more conventional methods
of geophysical survey. EM instruments are also
sensitive to conductive objects in the nearsurface that may preclude their use, for example
metal fences, rubbish, buried pipes, etc, and to
electrical interference from both cultural (eg
power lines) and atmospheric sources.
For most archaeological applications an
EM instrument with an inter-coil spacing of
approximately 1m will suffice, collecting data
at a reading interval of 1m x 1m (Fig 18). Field
operation and calibration will vary between
instruments, but it should be possible to convert
the recorded signal (often expressed as parts
per thousand or ppt) to units of apparent
conductivity in millisiemens per metre
(mS/m) and volume magnetic susceptibility
(dimensionless).The effective depth of
penetration is largely dependent on the
separation between the transmitter and
receiver, analogous to expanding the electrodes
of an earth resistance array, although the physical
orientation of the coils allows even a fixed spacing
instrument to provide a shallow (horizontal)

a)

b)

c)

a)

b)

d)

e)

c)

Fig 18 (above top) Compact EM instruments with an inter-coil spacing of c 1m are well suited to archaeological surveys:
(a) hand operated Geonics EM38B (14.6kHz) with integrated GPS recording both conductivity and magnetic properties
of the subsurface; (b) single channel analogue EM38 mounted in a sledge; (c) Deeper penetrating Geonics EM31 (9.8kHz)
with a 3m coil separation mounted onboard the GEEP multi-instrument sledge system together with two towed EM38s
(photograph courtesy Ian Hill, University of Leicester).
Fig 19 (above bottom) Comparison over a buried Roman wall between twin electrode earth resistance data collected with a
0.5m mobile electrode spacing (a) and conductivity data collected with a Geonics EM38 in shallow (horizontal) (b) and
deeper penetrating (vertical) (c) coil orientations. Fluxgate magnetometer (d) and in-phase, vertical coil orientation, EM
magnetic susceptibility data (e) over the same area are also shown.

35

and a deeper penetrating (vertical coil

orientation) mode of operation (eg Keller and
Frischknecht 1966).The frequency of operation
will also influence the penetration depth and
response of the instrument, depending on site
conditions. Comparative studies with instruments
such as the Geonics EM38 demonstrate a good
correlation with twin probe earth resistance
and magnetic surveys (Fig 19; Cole et al 1995;
Huang and Won 2000; Kvamme 2003). However,
certain combinations of site conditions, coil
orientation, operating frequency and phase
may produce a complex signal that is not
directly proportional to a single physical
property of the sub-surface (eg Linford 1998;
Tabbagh 1986;Tite and Mullins 1973).
More widely spaced traverses may be of
use when a deeper penetrating (wider coil
separation) instrument is used to identify
geomorphological features, such as
palaeochannels, or map changes of soil
magnetic susceptibility across an expansive
landscape. Rates of coverage will vary depending
on the precise instrument and sample interval
in use, but should be similar to earth resistance
covering approximately 1ha per day for a
1m x 1m sample interval survey.Vehicle
mounted instruments with integrated GPS
measurements are more rapid and enable
several hectares to be covered in a day at a
coarse sample interval (eg 10m x 10m).

1.6 Topsoil magnetic susceptibility survey

Archaeological settlement activity often results

in a localised concentration of soils and

sediments with an enhanced magnetic
susceptibility, because of the alteration of
naturally occurring iron minerals (Clark 1983;
Clark 1996; Cole et al 1995; Dalan and Banerjee
1998; Evans and Heller 2003; Fassbinder and
Stanjek 1993; Linford 2005;Thompson and
Oldfield 1986). Measurements are generally
made in the field (although soil samples may
be recovered for laboratory determination)
at a coarse sample interval of 10m, utilising
suitable instrumentation (Figs 20 and 21;Table
11). Care must be taken to account for the
presence of masking deposits, the influence
of recent land use and field conditions at
the time of the survey that may reduce the
contact between a field coil and the ground
surface. Laboratory determination may make
possible more detailed sample preparation
and additional measurements (eg frequency
dependence of susceptibility or fractional
conversion). Units of volume specific magnetic
susceptibility () used for measurements made
with a field loop are dimensionless within the
SI system and laboratory determination from
recovered soil samples should be corrected to
values of mass specific magnetic susceptibility
() in dimensions of m3kg -1.
Usually, a wider survey extending beyond the
evaluation study area should be considered, to
allow any regional correlation between magnetic
susceptibility with geology and soil type to be
distinguished from possible anthropogenic
enhancement (eg Dearing et al 1996, fig 1).

Even under ideal field conditions topsoil

magnetic susceptibility survey remains an
indicative technique that is unable to establish
the definitive presence, or absence, of
archaeological remains without the support
of additional methods of evaluation.Topsoil
magnetic susceptibility survey alone is,
therefore, not recommended and evidence
of an indifferent response to this technique
should not be used to discount the potential
presence of archaeological features.The
comparatively greater influence of ground
surface conditions and masking deposits such
as alluvium create anomalous areas of both
increased and depleted topsoil magnetic
susceptibility and should therefore always
be investigated through subsequent detailed
magnetometer survey.
Careful consideration should always be given
to the benefits of total coverage by detailed
magnetometer survey, as the enhanced level
of interpretation drawn from the results may
often outweigh the increased costs involved.
However, topsoil magnetic susceptibility
results are considered to be of value when
either interpreting magnetometer data
(eg Fig 20), or when assessing the suitability
of varying soil types and geology in advance
of conducting a detailed survey.Topsoil
susceptibility measurements over stripped
excavation surfaces and sectioned features
have also proved to be useful at an intra-site
level (eg Bayley et al 2001, fig 5; Linford 2003;
Linford and Welch 2004), and borehole

Table 11 Manufacturers of current magnetic susceptibility equipment used for archaeological surveys.
Manufacturer

Models

WWW URL

Type

Bartington
Instruments

MS-2

http://www.bartington.co.uk/

a highly versatile, portable magnetic susceptibility meter,

offering dual frequency operation (0.465 and 4.65kHz), and
a range of field and laboratory sensor coils.

Geofyzika

MFK1 Kappabridge http://www.agico.com/

various models of high sensitivity laboratory instrumentation

operating over a range of frequencies (0.976, 3.904 and
15.616 kHz)

SatisGeo

KT-6 Kappameter

http://www.satisgeo.com/

integrated hand-held sensor and meter operating at a single

frequency of 10kHz with a penetration depth of ~20mm

ZH Instruments

SM-30, SM-400

http://www.giscogeo.com/pages
/maggysm2.html

The SM-30 is a compact hand held sensor and meter operating

at 8kHz with a penetration depth of ~20mm. An automated
borehole measurement system SM-400 is also available.

http://www.heritagegeophysics.
com/Magnetic_Susceptibility/
SM-30_SM-100.htm
http://www.gfinstruments.cz/
http://www.zhinstruments.cz/
Geo Instruments

36

GMS-2

http://www.fugroinstruments.
com/html/inst/prod_magsus.htm

compact hand-held sensor and meter operating at 0.76kHz

measurements have been used to successfully

determine significant anomalies beneath
surface deposits across wider landscapes
(eg Dalan and Banerjee 1996).

producing an alternating current in the

subsurface. Similar dipolar pairs of insulated
electrodes are then used to measure the
potential created by the ground current.

1.7 Other geophysical methods

Multiple potential electrodes can be towed

at different spacings behind the current
electrodes to simultaneously measure the
apparent resistivity at varying depths and can
be inverted to provide a pseudo-section of
the ground surface (see above, section 1.3.4).
However, at higher source frequencies,
attenuation of the signal may prove to be a
limiting factor and the depth of investigation
restricted by the electrical skin depth. Currently,
these instruments seem to be ideally suited to
rapid, large-scale, reconnaissance surveys for
the detection of more deep-lying archaeological
or geomorphological features, but may yet
challenge the quality of traditional earth
resistance data for very near surface targets.
Rough or uneven ground conditions can be
problematic, causing poor coupling between
the insulating electrodes and the subsurface.

Despite offering limited use for traditional

applications of archaeological evaluation a wide
range of additional geophysical techniques is
available that may, under certain conditions, be
applicable. Some of the techniques discussed
in this section are highly specific for example
the use of micro-gravity for the detection of
buried voids while other techniques propose
new means for obtaining data-sets comparable
with more traditional methods. Most of these
latter techniques are currently at a stage
of development between research and full
commercial deployment, but may well be
adopted as the technology matures in the
near future.

1.7.1 Capacitative arrays

These systems are designed for the rapid
acquisition of apparent resistivity data and use
a series of electrodes mounted on individual
insulating mats that may be towed rapidly
across a site without the need to obtain a
direct contact with the ground surface.The
electric potential produced by the charges
on the electrodes causes the movement of
charged particles in the ground resulting in a
brief capacitative coupling, continuing only until
an equal and opposite reverse potential has
been established in the subsurface. Use of a
sufficiently high frequency source will reverse
the flow of charged particles in the ground,

a)

1.7.2 Seismic methods

Seismic methods use low energy acoustic
waves transmitted by vibration through the
host medium and can be used effectively in
both marine and terrestrial environments.
Velocities of seismic waves vary from c 200m/s
in soil up to 7000m/s in solid geological units
and, at the frequencies deployed, can result
in relatively long wavelengths, generally >1m.
This restricts the archaeological application
of seismic methods to relatively large-scale
features. For shallow, terrestrial, investigations

Fig 20 (below) Area magnetic susceptibility survey (a)

showing increased response over an area of dense
magnetometer anomalies (b). Low responses to the NE
correlate with recent soil dumping, but some finer detail is
not represented, such as a cemetery (c).
Fig 21 (above) Bartington MS2 magnetic susceptibility meter
in use: (a) in the laboratory with collected 10g soil samples
and; (b) on site with a field coil.

b)
c)

37

the energy source can be as simple as a

sledgehammer striking a flat metal plate on the
ground, with the resulting vibrations measured
by a line of regularly spaced geophone sensors.
Each geophone is secured to the ground by
a metal spike and consists of a suspended coil
wound around a high magnetic permeability
core in the field of a strong permanent magnet.
Vibrations are then transferred through the
spike to the coil to produce a proportional
electric current. Multi-core cables are then
used to connect the entire array to a suitable
multi-channel seismograph that amplifies
the signal and records the time when the
vibrations reach each geophone.
Energy from a seismic source travels as both a
direct wave spreading out through the surface
layer and also into successively deeper layers in
the subsurface. On meeting an interface between
two layers, part of the energy is reflected back
to the surface and the remainder continues at
a refracted angle. Assuming the lower layer has
a higher velocity, an angle of critical refraction
exists where the incident wave will travel
parallel to the interface at this higher velocity,
with some of its energy returning to the ground
surface as an expanding head wave before the
slower direct wave.
This difference in travel path forms the basis
of the seismic refraction method, where the
travel time of the refracted wave is measured
from the first received energy for each
geophone along the spread from the source,
and subsequently used to estimate the depth
to the subsurface interface.The seismic refraction
method requires the velocity to increase with
each subsequently deeper layer, a condition that
may not always be met for typical archaeological
surveys (Ovenden 1994), but has been
successfully applied during the investigation
of certain archaeological features, such as the
vallum south of Hadrians wall in Northumberland
(Goulty et al 1990).
An alternative approach is to use the seismic
reflection technique that, analogous to GPR,
records the amplitude variation of the received
signal at each geophone over a suitable time
window. For each source location seismic
reflection data is often recorded at several
locations that share a common mid-point
(CMP) between the source and receiver.
Multiple observations of buried reflectors are
then recorded at successively later travel times
and the data reduced to a single trace with a
much improved signal-to-noise ratio. Field
acquisition with this method is relatively slow
compared to other near-surface techniques,
but has been successfully applied for a number
of archaeological surveys (eg Vafidis et al
2003).The potential advantages of multi-fold
38

CMP data acquisition have also been investigated

for GPR survey (eg Pipan et al 1999) together
with the application of powerful seismic
processing software, developed for oil
exploration, which may equally be applied
to GPR surveys over archaeological sites
(eg Lehmann and Green 1999).
For shallow terrestrial imaging, seismic methods
are disadvantaged by the need to produce
high-frequency acoustic sources, to create
short wavelengths in the soil, while coupling
the source energy effectively to the ground
surface.The spacing between the receiver
geophones also needs to be reduced to obtain
an appropriate sample interval, but this may be
restricted where the amplitude of the source
can potentially cause damage to the geophone.
Attempts have been made to improve the
applicability of acoustic techniques (eg Frazier
et al 2000; Hildebrand et al 2002) particularly
using swept-frequency sources that may
well prove fruitful for imaging archaeological
features buried under conditions unsuitable for
other techniques, such as highly conductive
alluvial soils (eg Metwaly et al 2005).

1.7.3 Borehole methods

Many geophysical techniques are compromised
by either the depth to the target archaeological
features or, particularly on urban sites, the
presence of considerably disturbed near-surface
deposits (eg building rubble). One approach is
to introduce the geophysical equipment into
the ground through a borehole cored from
the surface.This may, for example, take the
form of a specially designed GPR transmitter
that can be lowered down the borehole and
measurements made to a receiver mounted
on either the surface or in a second borehole.
Equally, seismic sources and geophones, earth
resistance electrodes or even magnetometers
may be used for borehole investigations. Active
source-receiver instruments (eg GPR) allow
transmission tomography methods to be
applied from deviations of the travel path
as the transmitter and receiver are lowered
down two separate boreholes.
The major disadvantage with these techniques is
the necessity to introduce an invasive borehole
into the site that may damage the buried
archaeological remains. In some cases the
information gained from a borehole geophysical
survey may outweigh these concerns,
particularly when boreholes have to be sunk
for other invasive geotechnical investigations.

1.7.4 Micro gravity

Variations in the local acceleration of the Earths
gravitational field, caused by the contrast in
density of the underlying structures, have been
successfully used at an appropriate scale to

investigate civil engineering or archaeological

features (Arzi 1975; Di Filippo et al 2000).
By far the greatest success has been achieved
using appropriate high-sensitivity gravimeters
to locate air-filled void features, which, by
definition, must demonstrate a considerable
density variation from the host structure
(eg Blkovsky 1979; Butler 1984; Fajklewicz 1976;
Linford 1998; Linnington 1966).
In essence, a gravimeter consists of a springsuspended weight and a means to record
accurately any varying deflection in the presence
of the local gravitational field. Practical instruments
must be highly sensitive and compensate both
for changes in the ambient temperature and for
vibrations at the sampling point.The resulting data
must then be further reduced to account for a
range of variables, including the diurnal variation
of the Earths gravitational field and even the
micro-topography of the site under investigation.

1.7.5 Radiometric methods

Radiation detectors can be used to determine
the location and concentration of certain
commercially viable radioactive ore bodies,
such as uranium. While the attenuation of
radioactive particles is relatively high in soil or
rock, particularly for alpha and beta particles
that will only travel short distances, gamma
photons offer more promise (Ruffell and Wilson
1998).The most common sources of gamma
radiation are the elements potassium, uranium
and thorium that may be found in the constituent
minerals forming many archaeological sites. Any
contrast or greater concentration of these
radioactive elements should, theoretically, be
detectable with a scintillation counter of high
enough sensitivity. For example, measurements
of gamma-ray emissions within the walled
Roman town at Silchester, Hants, revealed a
much lower count rate from the flint and chalk
building remains than the substantially higher
background value caused by the presence of
40
K in the soil. Mapping the response with
a towed scintillation counter demonstrated
significant variations, possibly indicating both
the location of building remains and the
differing depth of soil cover across the site.

1.7.6 SQUID magnetometers

While relatively common within laboratorybased instruments that measure extremely
weak magnetic fields, superconducting quantum
interference devices (SQUID) are challenging
to deploy in the field because they require
very low operating temperatures. Zakosarenko
et al (2001) demonstrated that SQUID-based
systems can also be used for measuring
magnetic field gradients and have developed a
field instrument specifically for archaeological
prospection (Chwala et al 2001; Chwala
et al 2003).

This system is based on a cart-mounted liquid

helium cryostat that is able to maintain a
N iobium SQ UID at a working temperature
of 4.2K configured as a special planar intrinsic
gradiometer where the two effective sensors
are extremely close together (Fig 4d).The
magnetic field resolution of the SQ UID is
approximately 0.00002 nT, about 200 times
greater than currently available alkali-vapour
magnetometers, and possibly exceeds the
sensitivity required to map even the weakest
archaeological anomalies encountered in the
field. However, this sensitivity is essential for
operating as a gradiometer with such closely
separated sensors, where the measured gradient
will be extremely small, but will also be less
affected by local distortions in the earth magnetic
field.This, for example, allows the SQUID sensors
to be transported in relatively close proximity
to a towing vehicle with any residual field
removed through post-acquisition processing.
W hile SQ UID sensors make sampling much
faster than conventional magnetometers
( ~ 1000Hz), making them ideal for rapid data
acquisition over large areas when operated as
a vehicle-towed array, the short gradiometer
baseline appears to limit the detection of
more deeply buried features.

1.7.7 Multi-channel instruments and

sensor platforms
The use of vehicle-towed sensor platforms,
utilising differential GPS and fluxgate compasses
for navigational and positional information, has
recently been explored and enables deployment
of a combination of multi-channel instruments
for the rapid survey of large areas.The University
of Leicester has developed a prototype system
(Fig 1), configured with an array of caesium
magnetometer and electromagnetic sensors.
Results from this trial system compare favourably
with data collected with a hand-operated
caesium magnetometer cart and were completed
in a fraction of the time required for the more
conventional survey.
Towed multi-channel GPR antennas are also
now available, offering the ability to capture
very dense data-sets, equivalent to a traverse
separation of approximately 0.1m, from a 2m
wide instrument swathe.W hile the initial cost
of these systems is beyond most archaeological
researchers at present, the benefits of such
instrumentation are clear when considering
the very large-scale application of GPR survey
(eg N eubauer et al 2002).

1.7.8 Continuous-wave radar

The majority of commercial ground penetrating
radar instruments utilise an impulse source to
introduce energy into the ground. More
recently, systems using a continuous source
have been introduced, where the transmitted

frequency is either swept (frequency

modulated continuous wave) or held at a series
of steps (synthesised or stepped-frequency)
over a range of transmitter frequencies. An
inverse Fourier transform is then applied to
the recorded frequency domain data to
produce a response similar to an impulse GPR.

ground-contacting thermocouples have also

been investigated, but the heat generated
by friction when inserting the probe into
the ground was found to slow data acquisition
with this method of survey (Bellerby et al
1990).

1.7.11 Self-potential
Somers et al (2005) demonstrate an alternative
approach to continuous-wave radio-frequency
imaging by introducing a source transmitter
beneath the intended target through a smalldiameter borehole.The energy from the buried
source then passes back up to the ground
surface having been modified, in terms of
both amplitude and phase, by the illuminated
archaeological features.These variations are
recorded by a mobile receiver over the site
surface and may be processed with appropriate
image reconstruction algorithms.The system is
analogous to an optical microscope with the
RF source acting as a below-stage lamp and the
site surface as the lens plane.The reconstruction
algorithm can then be adjusted to focus the
resulting image on a particular depth of the
target beneath the surface.

1.7.9 Random-signal radar

If the duration of a transmitted radar wavelet
is reduced to an extremely short pulse its
energy is distributed over a very wide bandwidth
compared to either traditional impulse or
continuous wave techniques. A series of ultra
short pulses can be transmitted continuously
to form a (pseudo-) random waveform to
illuminate buried target objects.The range to
the target can then be obtained by correlation
of the received signal with the transmitted
waveform (Horton 1959). O ne major advantage
of random-signal radars is their very good
electromagnetic compatibility, which at low
power levels is indistinguishable from background
noise. Given the level of regulation applied to
the frequency spectrum and electromagnetic
compatibility, this technology may eventually
replace traditional RF devices. Ground
penetrating random-signal radars have been
demonstrated and will, no doubt, find suitable
archaeological application (eg X u et al 2001).

1.7.10 Thermal sensing

Variations in ground surface temperature
can be influenced by the presence of buried
archaeological features and are usually recorded
by airborne infrared scanners that are able
to cover large areas in a single swathe. Some
attempts at ground-based thermal mapping
have also been made (eg Clark 1996, fig 11),
but these have been most successful for
investigating historic building fabrics rather than
for buried archaeological remains (eg Brooke
1987; Kooiman and de Jongh 1994). Direct
measurements of soil temperature with

Electrolyte flow in ground water, and across

any chemical potential gradient, can cause subtle
variations in naturally occurring background
potentials, for example across a gradient formed
in a concentration of ferric and ferrous ions
produced by localised burning of iron oxides
in the soil.The application to archaeological
prospection was initially investigated by W ynn
and Sherwood (1984) and is attractive for
its relative simplicity and the low cost of the
equipment required.
Field measurements are made between two
non-polarising electrodes connected to a
suitable high-impedance volt meter. However,
care must be taken to account for the influence
of topographic changes, buried metal (eg
pipelines), stray currents from power sources,
ground water movements and changes in
temperature, as any of these factors will affect
the local self-potential. Even the bioelectrical
activity of large plants and trees is sufficient
to create a detectable anomaly (Telford et al
1976, 293).
Drahor (2004) provides a summary of the
possible sources of self-potential anomalies
with regard to archaeological features and
demonstrates the success of the technique
for detecting burnt structures. However, the
advantages of the low equipment costs for
this method must be considered against the
slow rate of acquisition and the difficulty in
obtaining useable field data, and subsequently
the often complex interpretation required.
Burnt features are also readily detected by the
more rapid magnetic techniques that should
usually be considered in the first instance.

1.7.12 Induced polarisation

The effect of polarisation during the ionic
conduction of an electrical current through
the soil is a recognised constraint when using
direct current for an earth resistance survey
(see Part IV, 1.3). Electrode polarisation will also
be influenced by subtle membrane polarisation
effects associated with buried features and may
be measured using a modified earth resistance
array.Time-domain measurements are made
by applying a square wave signal to the current
electrodes, and then recording the decay
of any induced polarisation voltage over a
period of time shortly after the applied field
has been removed. Higher-frequency alternating
waveforms, generally between 0.0625Hz and
39

1000Hz, may also be used for measurements

of phase shift in the frequency domain.
Aspinall and Lynam (1968) recognised the
possible application of induced polarisation
methods for archaeological survey, and
subsequent field experiments demonstrated
the potential for identifying a buried humusfilled ditch and bank that compared favourably
with results from a simultaneous earth
resistance survey (Aspinall and Lynam 1970,
fig 57). A more recent application of this
technique used frequency-domain measurements
(also known as spectral induced polarisation)
to locate a Bronze Age trackway, constructed
from wooden planks, found in the Federsee
bog near Lake Konstanz, Germany (Schleifer
et al 2002).The well preserved cell structure
of the waterlogged wood exhibited a strong
polarisation effect producing a peak phase
shift at a frequency of approximately 5Hz
that located the feature.
While these techniques (sections 1.7.1 to
1.7.12) would not be recommended generally,
on specific sites they may find a particular
application where other methods fail.

1.8 Metal detecting

Metal detectors are EM instruments (see Part
II, 7.2 and Part IV, 1.5), but mention of them is
separated out here because their applications
are significantly different to other specialised
EM techniques, and because their use solely to
find and recover metal objects is contentious.
Depending on the instrumentation used, metal
detectors emit a pulsed or continuous EM
signal that generates detectable and characteristic
eddy currents in target conducting metals.
Depending on their sophistication, metal
detectors can be sensitive to signals from
small objects such as individual coins at
depths up to about 0.3m to larger items at
greater depths; also, detectors can be tuned
to screen out unwanted responses and to
discriminate in favour of certain metals.

English Heritage policy and good practice for

portable antiquities and surface-collected
material in the context of field archaeology
and survey programmes, including the use
of metal detectors, is stated in Our Portable
Past (English Heritage 2006) and can be
accessed at http://www.english-heritage.org.uk/
upload/pdf/Our-Portable-Past.pdf
Metal detectors are not usually used during
the initial evaluation of development sites
for their archaeological content. It is possible,
however, that desk-based assessment (eg from
the HERs and/or PAS database) will reveal that
previous finds of metal objects indicate the
potential presence of an archaeological site.
If this is the case, then metal detector survey
might be included in the subsequent field
evaluation (integrated with other relevant
prospecting methods, as appropriate). In
some circumstances it may be that a significant
metal detector find is itself the incentive for
the evaluation.
As metal detecting usually involves the recovery
and removal of artefacts, it is imperative that
this form of site evaluation is fully justified, is
integrated within an agreed project design,
and includes the use of appropriate field
methodologies, subsequent conservation,
reporting and deposition to an acceptable
standard (English Heritage 2006).
Unless used as part of an excavation (see
below), metal detecting should normally only
take place on land under arable conditions,
and as part of a properly structured field
survey project.
Metal detectors should only be used to recover
material from the contemporary plough-zone,
and not from undisturbed contexts; however,

metal detecting may also be an acceptable

technique on sites or find spots under pasture,
where there is unequivocal evidence that the
area was subject to arable cultivation in recent
years, provided that the recovery of material is
restricted to the former modern plough zone.
To be effective, a metal detector survey should
use skilled operators with suitable instruments,
working consistently and systematically over
a pre-surveyed grid composed of at largest
10m units (Fig 22). Recovered material should
normally be recorded to within these units, or
located individually using GPS or electronic
measurements.
Metal detecting may in some circumstances
be justified over areas that are destined for
development and/or excavation, and that
have been stripped of topsoil; in these cases
controlled metal detecting can be an asset
both during the excavation and in the
recovery of artefacts from spoil.
Note that metal detecting is not permitted on
Scheduled Monuments without a Section 42
Licence (see Part II, 7.2), and that restrictions
may also apply on other designated sites
(http://www.naturalengland.org.uk/conservation/
designated-areas/default.htm), and on land
under Environmental Stewardship Schemes
(http://www.defra.gov.uk/erdp/schemes/es/
default.htm). Detecting on land owned by the
National Trust must be subject to a Licence
Agreement (www.nationaltrust.org.uk/main/
md_policy-2.doc).

1.9 Geochemical methods

Apart from magnetic susceptibility survey,
(see above, section 1.6), geochemical methods
(phosphate analysis, multi-element analysis and
lipid analysis) are not generally used for
evaluations. Instead, they are either themselves

Despite initial military incentives, such

developments have been driven in part by
demand from hobbyists.
Present estimates suggest that there are in
excess of 8,000 metal detectorists in England.
However, these guidelines refer to the use
of metal detectors for archaeological field
evaluation, rather than as a hobby. Nonetheless,
all metal detector users are strongly advised
to abide by the voluntary Code of Practice for
Responsible Metal Detecting in England and
Wales.This and other valuable information
relating to the use of metal detectors can be
found through the website of the Portable
Antiquities Scheme (PAS) at www.finds.org.uk.
40

Fig 22 (above) Systematic metal detector survey of an area that has been divided into 10m grids.

the subject of methodological research, or they

are used to assist interpretation at an intra-site
scale of investigation. A review of geochemical
methods is provided by Heron (2001); see
also English Heritage (2007).

1.10 Remote sensing

Remote sensing is defined as the imaging of
phenomena from a distance (Shennan and
Donoghue 1992) and is here considered to
be distinct from the ground-based remote
sensing methods so far discussed.

1.10.1 Aerial photography

Aerial photography (AP) is the most familiar
remote sensing technique (Bewley 1993;
Bewley and Raczklowski 2002; Palmer and Cox
1993;Wilson 2000) and the aerial photographic
record should always be consulted as part
of site evaluation.This record is often highly
complementary to that obtained by geophysical
methods. In many circumstances the AP record
will dictate where ground-based methods may
be deployed and the latter will often provide
exact ground location as well as additional
definition and detail. Geophysical methods may
be able to respond positively where the AP
record is negative, or where surface conditions
are unsuitable for photography.

et al (1997), Shennan and Donoghue (1992),

and Winterbottom and Dawson (2005).

1.10.3 Remote surface mapping

Modern remote imaging systems are now able
to capture increasingly detailed and accurate
topographic information at a variety of scales
relevant to archaeological prospection. While
digital aerial photogrammetry has seen some
application (Stone et al 2004), attention is
currently focused on the considerable potential
offered by lidar (light direction and ranging),
and interferometric synthetic aperture
radar (IFSAR).
Of these, lidar currently offers the higher level
of vertical and horizontal resolution (Crutchley
2006; Holden et al 2002), and the resulting
digital surface models (DSMs) can be
manipulated to enhance the recognition of
very slight but significant surface topography
(Bewley et al 2005). A significant advantage
over aerial photography is the potential ability
to digitally remove tree cover to create digital
terrain models (DTMs) of underlying earthwork
features (Devereux et al 2005).The value to
archaeological evaluation of IFSAR is presently
less clear, although elevation data for all of
Britain is available at http://www.intermap.com/
corporate/greatBrit.cfm.

1.10.2 Multi-spectral scanning

Despite the increasing availability of higherresolution data achieved by airborne multi
spectral scanning (MSS) from sensors mounted
on satellites and on aircraft this has still not
made much impact in the day-to-day evaluation
in the UK of sites for their archaeological
content. A review of the subject has been
provided by Donoghue (2001), and examples of
case studies include Fowler (2002), Powlesland

2 Analysis of geophysical data

2.1 Data processing
Once geophysical data has been collected it
is necessary to process it for interpretation
and presentation.The advent of powerful and
affordable personal computing equipment has
revolutionised this aspect of archaeological
geophysics over the last fifteen years and several
specialised software packages are now available

(Tables 12 and 13). Detailed discussion of

the reasons for and application of numerical
processing algorithms can be found in a number
of textbooks and software manuals (Gaffney
and Gater 2003; Scollar et al 1990; Walker
2005).Two guiding principles that underlie
such discussions bear restatement. Numerical
processing can never be a substitute for poor
raw data and the surveyors aim should always
be to collect the highest quality measurements
in the field. Furthermore, every numerical
modification of the original field data should
be carried out for a clear purpose and no
processing algorithm should be used blind
without a full understanding of its implications.
The majority of numerical processing algorithms
encountered in archaeological geophysical
surveys fall into one of three categories:
(1) those designed to mitigate for artefacts
introduced into the data by the
prospecting instrumentation and/or
strategy;
(2) those that employ generic digital image
processing methods to enhance features
of interest within the data-set; and
(3) those that use mathematical descriptions
of the geophysical measurement process to
model or infer information about causative
features from the measured anomalies.

2.1.1 Mitigating data collection artefacts

Magnetometer data Scollar et al (1990, 4405)
identify a number of sources of error in
magnetometer data resulting from field procedure
and environmental factors. Computational
procedures have been developed to detect
and eliminate the effects of many of these and
maximise the clarity of archaeological anomalies

Table 12 Some of the more commonly used processing software packages available for archaeological geophysics.
Manufacturer

Software

WWW URL

Comments

Geoscan Research

Geoplot 3.00

http://www.geoscan-research.
co.uk/page9.html

wide range of processing options specific to

archaeological geophysics

DW Consulting

Archeo
Surveyor 2

http://www.dwconsulting.nl/
archeosurveyor.htm

very up-to-date interface; good processing and

display options

David Staveley

Snuffler

http://www.homeusers.prestel.co.uk
/aspen/sussex/snuffler.html

free software; aimed primarily at earth resistance

survey processing

Geosoft Inc.

OASIS Montaj

http://www.geosoft.com/

offers a wide range of advanced processing options

aimed at all forms of geophysics

GeoQuest
Associates

Insite v3

No Web site, contact:

[email protected]

offers an intuitive and easy-to-use interface

Geotomo Software

Res2Dinv 3.55
and Res3Dinv 2.15

http://www.geoelectrical.com/

specialist software for inversion of data for

electrical sections

41

present in the data-set.The most common

corrections are discussed below and illustrated
in Fig 23 (alternate terms for a procedure are
listed in parentheses after each heading):

Edge matching (equalising sub-grid shifts,

micro-levelling): A large survey will typically be
composed of a mosaic of rectangular survey
blocks or sub-grids surveyed at different times.
One of the first procedures carried out after
data collection is to combine these individual
sub-grids into a single composite data-set.
However, differences in temperature and other
environmental conditions as well as recalibration
of the magnetometer during the survey
can result in sub-grids exhibiting different
background measurement levels leading to
visible discontinuities between the edges of

adjacent sub-grids. Adjusting the mean or

median of each sub-grid to a common value
(often zero) by addition of a constant to each
measurement value within the sub-grid is usually
sufficient to eliminate edge discontinuities in
magnetometer data (Eder-Hinterleitner et al
1996). Only in extreme cases, such as the
proximity of large modern ferrous structures,
should more sophisticated methods based upon
analysis of the local statistics of measurements
close to each sub-grid edge be required (eg
Haigh 1992).

similar artefacts. Such distracting measurements

may be distinguished by their large difference
from neighbouring values within the survey
sub-grid.Typically a thresholded median or
mean filter is used to detect and replace such
extreme values (Scollar et al 1990, 492).
Methods that treat spikes as statistical outliers
from the overall data distribution have also
been developed and offer the advantage that
they can be applied to randomly collected
data before interpolation onto a regular grid
(eg Ciminale and Loddo 2001).

Spike removal (despiking): Magnetometer

sensor instability can occasionally cause
isolated extreme readings, or spikes, in the
survey data, and small pieces of highly magnetised
iron lying on the ground surface can cause

Where spike removal has been used to suppress

anomalies caused by surface iron objects, care
should be taken with subsequent interpretation
of the data. It is possible for the despiking
operation to remove the high-magnitude

Table 13 Some of the more commonly used processing software packages available for GPR data processing.

42

Manufacturer

Software

WWW URL

Comments

Geophysical
Archaeometry Lab.

GPR-SLICE 5.0

http://www.gpr-survey.com/
gprslice.html

comprehensive processing software supporting

all major data formats, GPS integration and data
visualisation in 2D/3D

GSSI

Radan 6.5

http://www.geophysical.com/
software.htm

advanced processing software supporting GPS

integration with additional modules available to
extend data interpretation and visualisation in 3D.
Supports GSSI data format

IDS

GRED-3D,
GRED-AGS

http://www.ids-spa.it/

advanced 2/3D processing software supporting IDS GPR

systems; a specialised archaeological software package
with pattern recognition algorithms is also available

MAL

RadExplorer,
REFLEXW,
Easy 3D,
ObjectMapper

http://www.malags.com/software/

packages to support data processing and visualisation

and interpretation in 2D and 3D; supports data input
from other manufacturers native formats together
with the RAMAC format

Grandjean and
Durand (1999)

Radar UNIX

http://www.iamg.org/CGEditor/
index.htm

freely available software running under UNIX for

non-commercial use for processing GPR profile data.
Supports GSSI and SEGY formats

Radar Systems Inc.

Prism 2.01,
Prism Layers 3D,
Prism Easy 3D

http://www.radsys.lv/

software for the acquisition and post-processing

of data collected with the Zond GPR system, also
compatible with SEG-Y and GSSI data formats

Sandmeier

Reflexw

http://www.sandmeier-geo.de/

extensive GPR and seismic data processing and

visualisation tools; supports GSSI, Mala, PulseEkko,
SEG-Y and other seismic data formats

Sensors & Software

EKKO_View,
EKKO_Mapper 3,
EKKO_3D

http://www.sensoft.ca/products/
pulseekko/p_software.html

packages to support the processing and display of

profile, time/depth slices and 3D visualisation of GPR
data captured in the PulseEkko data format

Tzanis (2006)

MATGPR

http://users.uoa.gr/~atzanis/
matgpr/matgpr.html

a suite of free GPR processing routines and a GUI

interface for the Matlab numerical computing
environment; compatible with GSSI, Mala, PulseEkko,
SEG-Y and Seismic Unix data formats

positive peak of such small dipolar anomalies

but leave the adjacent values associated with
the negative pole, which are often of smaller
absolute magnitude. Without the positive
peak to provide context, these latter can
be mistaken for negative archaeological
anomalies.

Destriping (unbunching): Magnetometer

surveys collected in zigzag mode can exhibit
striping where successive traverses appear as
alternating light and dark bands when the data
is plotted.This is because magnetometers can
exhibit directional sensitivity (sometimes called
heading error): a change in the value measured
by the magnetometer depending on the
direction it faces relative to magnetic north.
In fluxgate gradiometers it is usually caused
by slight differences in alignment between the
two differential sensors, and optically pumped
magnetometers may also exhibit an inherent
directional sensitivity.
The standard method of correction is to assume
that the bias caused by this effect is constant
over an entire traverse and to subtract a
constant value from all readings on the traverse,
such that their mean is set to zero or to a value
common to all traverses (eg Ciminale and
Loddo 2001). Such techniques also simultaneously
remove the long-term zero drift exhibited by
most types of magnetometer, providing the
time taken to complete each traverse is
short relative to the rate of instrument drift.
However, where traverses are long (~100m),
more sophisticated linear regression techniques
may be required instead (Tabbagh 2003).

a)

d)

b)

When destriping, care should be taken

that linear anomalies parallel to the traverse
direction are not erroneously removed by the
process, particularly when their length is close
to or greater than the traverse length and their
magnitude is similar to the biases caused by
the directional sensitivity. Eder-Hinterleitner et
al (1996) describe a destriping method that
can protect such parallel anomalies against
erroneous removal but only if they are wider
than the survey traverse separation. Hence,
every effort should be made to reduce
instrument directional sensitivity in the field
rather than relying on post-acquisition
processing to remove severe striping.

variations in pace occur during a single traverse,

re-interpolation of the sample interval may
also be necessary (Eder-Hinterleitner et al
1996). Such methods can only estimate the
displacement that has occurred by making
assumptions about how anomalies appearing
on adjacent traverses should match up.
Particular care should be taken to ensure that
linear anomalies running diagonally to the
traverse direction are not altered so that they
appear perpendicular to the traverses after
this operation.Thus, diligent field procedure
should always be employed to minimise the
need for post-acquisition correction of line
displacement errors.

Correcting line displacement errors

(destaggering): Magnetometers are often

Earth resistance data: Scollar et al (1990,

345ff) outline the problems that can occur
with earth resistance measurements.The
majority are best avoided by careful attention
during data collection. However, two types
of error are often impossible to eliminate
completely and are susceptible to mitigation
by numerical procedures.

set to take readings at regular time intervals

and the position along the traverse at which
each reading was taken is calculated on the
assumption that travel speed was constant.
However, variations in traversal rate can occur
(because the operator encounters a steep
incline and has to slow down, for example)
and this can result in the sensor not being at
the correct position when a reading is taken.
When traverses are walked in zigzag fashion,
deleterious effects can be pronounced with
linear anomalies crossing the traverses having
their peak positions displaced in opposite
directions on alternate traverses, leading to
a staggered appearance in plots of the data.
Often, shifting each traverse to maximise
cross-correlation with the two neighbouring
traverses will correct for the effect (eg Ciminale
and Loddo 2001); however, where significant

c)

e)

Fig 23 (above) (a) Composite plot of four sub-grids combined with no corrections; (b) the same four sub-grids combined,
following edge matching, whereby discontinuities between sub-grids are reduced; (c) then with additional spike removal where
distracting dipolar responses are lessened; (d) then after destriping, which had been most evident on the right half of the area;
(e) after correcting line displacement errors with the most obvious effect on the circulinear anomaly, although other anomalies
have also been clarified.This manipulation of the data is evidenced by the positional adjustment of the incomplete lines in
the top left corner.

Edge matching (equalising sub-grid shifts,

micro-levelling): Weather conditions may
change during the course of a large earth
resistance survey, causing changes in the soil
moisture content. Such changes will influence
the average resistivity of the sub-surface and
it is possible that adjacent survey sub-grids
measured on different days will exhibit a
discontinuity along their common edge.Where
changes in soil moisture conditions have been
relatively minor, corrective procedures similar
to those used for magnetometer surveys
usually suffice. However, more severe variations
in conditions may require more complex
pre-treatment to individual sub-grids such as
re-scaling the data range or the removal of a
first order trend. In extreme cases it may not be
possible to entirely remove edge discontinuities
caused by changes in field conditions.

Spike removal (despiking): Surface conditions

such as concentrations of stones or uneven
topography may result in poor electrical contact
between the ground and one or more of the
earth resistance electrodes.This can result in
anomalously high or low resistance values
being measured. As such measurements will
exhibit large differences from neighbouring
values it is possible to detect and remove
them using the same types of procedures used
to remove spikes in magnetometer surveys.
However, wherever possible, such instances
of high contact resistance should be detected
and re-measured in the field as the survey
progresses because post-acquisition removal
of large numbers of such spikes reduces the
number of truly independent measurements
in the resulting data-set.
43

Ground penetrating radar data: The level of

post-acquisition processing required for GPR
data will depend, in part, on the specific aims
of the survey (eg for the production of individual
profiles or multiple traverse data-sets, or for
display as amplitude time or depth slices)
and, perhaps to a lesser extent, the type of
radar equipment in use. Useful summaries of
appropriate GPR data processing techniques
can be found in Annan (2004) and Daniels
(2004), and more specific archaeological
applications are considered in Conyers and
Goodman (1997), Conyers (2004) and
Leckebusch (2003).
As with other geophysical methods good field
technique will minimise many data acquisition
artefacts and particular care should be taken
to maintain good antenna coupling with ground
surface.The GPR data processing procedures
discussed below represent general considerations
arising under typical field conditions and should
be read in conjunction with sections 1.4 (above)
and 2.1.3 (below).

Individual trace repositioning and interpolation

(rubber-banding): The majority of GPR data

Time varying signal gain: An appropriate gain

should be applied to amplify lower amplitude,
later reflections caused both by the attenuation
of the signal in the propagation medium and by
the spreading loss of the expanding radar wave
front with depth (eg Jol and Bristow 2003, 20).

Frequency filtering: Both low-frequency energy,

associated with antennaground interactions,
and high-frequency noise can be suppressed
by the application of suitable frequency filters,
generally matched to the centre frequency of
the specific antenna in use.
2.1.2 Image enhancement

Zero offset removal (DC shift or dewow):

Image enhancement is usually unnecessary for

magnetic gradiometer data, but it should be
considered for earth resistance data where
archaeological anomalies are often superimposed
upon larger-scale trends caused by geological
and hydrological changes.Where such techniques
have been applied it is essential that they are
identified and explained. Reference to standard
texts on the subject is acceptable, although the
choice of any variable parameters should be
detailed. All such algorithms accentuate some

Time zero alignment: Some temporal downtrace variation of the first recorded signal on
each trace may occur from electronic drift
across a data-set.This drift can be corrected

aspects of the data at the expense of suppressing

others, and many have the potential to produce
spurious processing artefacts, which may then
be misinterpreted by either the contractor or
the client.To guard against this eventuality the
survey report should explain why a particular
series of processes was necessary, summarising
the benefits to interpretation. It is misleading to
conceal the poor quality of the original data by
applying merely cosmetic enhancements.

2.1.3 Modelling and inversion

An appropriate down-trace time window
should be chosen, which may include the air
wave response to improve resolution of very
near-surface non-planar reflections; but care
should be taken to avoid the suppression of
significant horizontal reflectors, if present
(eg Conyers 2004, fig 6.3).

will be collected at a high density along-line

sample interval using either a system triggered
by a distance measuring odometer wheel or
continuous time-based trace acquisition with
additional positional information.This positional
information may be provided through the
manual insertion of fiducial markers as the
antenna passes distance markers along the
survey guide rope or, for more recent
instruments, simultaneous GPS measurements.
Regardless of the system in use it is often
necessary to reposition and interpolate the
raw GPR traces to account for slight variations
in the collected sample density because of
changes in the speed of acquisition, odometer
wheel slippage or calibration error, or the
lower density of GPS or fiducial data compared
to the rate of GPR capture. Despite the inherent
error associated with all (semi-) automated
methods of positional control, adequately
processed data-sets contain few, if any, positional
artefacts and offer considerable advantages
in speed of data acquisition compared to
manually triggering each trace.

This process corrects the mean value of each

trace to a near zero value to account for any
DC offset that may have been introduced by
the sampling electronics during the period of
data acquisition.

44

by aligning the common direct-wave response

present in every trace, often through picking
and adjusting to a single minimum amplitude
threshold (eg Conyers 2004, 901).

In some situations image enhancement

methods can be employed to accentuate
anomalies of interest within the survey
data while suppressing the effects of those
considered less archaeologically relevant. A
wide variety of such algorithms exists, many of
which are not specific to geophysical data-sets
but are generic to all types of digital image.
Scollar et al (1990, 488ff ) review a number
of those most relevant to archaeological
geophysics. Perhaps the most commonly
applied are convolution operators that
calculate a weighted local average around
each data value then either deduct it from
or substitute it for the original value (often
termed high- and low-pass filtering, respectively).
Low-pass filtering can be used to suppress the
effects of uncorrelated measurement noise
between adjacent readings while high-pass
filtering can remove the effects of large-scale
geological trends within the data allowing
archaeological anomalies to be discerned
more clearly (Fig 24).

Data modelling considers idealised forms of

the types of buried archaeological feature that
might be detected in a geophysical survey and,
by describing mathematically the physical
processes by which such features influence
surface geophysical measurements, predicts the
form of geophysical anomaly that should result.
By comparing a set of synthetic anomalies with
those detected in real survey data it is possible
to estimate parameters such as the shape and
burial depth of archaeological features. Data
inversion attempts to predict causative
archaeological features directly from survey
data by applying the mathematical inverse of
the operators used for synthetic modelling
to the field measurements.
Such techniques are usually not necessary for
standard archaeological area surveys where
the layout of archaeological features can be
determined from a plan view of the
geophysical anomalies. However, the anomalies
generated by vertical electrical sections are
often complex and the shapes and burial depths
of causative features cannot always be directly
inferred from the geophysical measurements.
For this type of data, numerical inversion
techniques may be applied to clarify the
vertical definition of any buried archaeology.
The process often proceeds iteratively,
first inverting the data, then modelling the
measurements that would be expected,
given the inferred features, and then using a
comparison between the modelled and real
data to improve the inversion.This process is
repeated until the modelled measurements
match the real measurements to an acceptable
degree (eg Fig 25; Loke and Barker 1996).
While not generally required for standard
archaeological surveys where the objective
is to identify the presence of archaeological
features, modelling techniques can also be
applied to magnetic data to estimate
characteristics of the causative archaeological
features. As magnetic anomalies are not unique
to one particular causative feature (Blakely
1996, 216), it is usually not possible to apply
inversion methods such as those used for
electrical sections. However, by making a number
of reasonable simplifying assumptions it is

a)

b)

c)

possible to model the geometry of the buried

features likely to have caused a particular
detected anomaly (eg Eppelbaum et al 2001;
Neubauer and Eder-Hinterleitner 1998).
Forward modelling of GPR data is both
complicated and computationally intensive
compared to the inversion of earth resistance
or magnetic data (eg Conyers and Goodman
1997, plate 2a; Daniels 2004, 3767, C3).
However, attempts are often made to reduce
the complex transmitted signal, or wavelet,
produced by a GPR to an ideal impulse
response function through wavelet optimisation
or deconvolution techniques.This process is
often further complicated by the time variant
attenuation of the incident wavelet as it passes
through the subsurface, but deconvolution can
often prove effective for the suppression of
certain repetitive down-trace signal artefacts
such as antenna ringing over near-surface
conductive objects (eg Conyers 2004, 1268).
In addition, the use of wave-front migration
techniques to collapse the hyperbolic response
from point-reflectors caused by the
progressively spreading pattern of radar energy
through the ground may also be considered
as a form of data modelling (eg Conyers 2004,
1289; Linford 2006, 22378). However, migrated
GPR data-sets are rarely, if ever, confirmed by
the application of a suitable forward model
and subsequent comparison against the original
data. Migration can often aid the resolution of
detailed structure within complex anomalies
caused by the overlapping response of many
individual point-source targets, but may not
be beneficial to every data-set.

2.2 Data display

Graphical presentation of geophysical survey data
is an essential step in visualising, understanding
and interpreting the results. Appropriate data
plots should be provided in the survey report
to support the interpretations made by the
practitioner and to help both specialist and
non-specialist readers to follow the reasoning
set out in the report text. A number of different
display formats have been developed for

Fig 24 (left top) Earth resistance data over a U-shaped

long barrow ditch defined by low resistance over a variable
background response, showing the effect of high pass and
low pass filtering: (a) raw data showing variable background
resistance across the surveyed area; (b) removal of variable
background using a 3m radius Gaussian high-pass filter; (c)
main archaeological responses in the data further emphasised
by smoothing with a 1m radius Gaussian low-pass filter.
Fig 25 (left bottom) Inversion of an electrical section over
a ditch section, which shows as a low resistance (dark blue)
anomaly.The top picture shows the pseudo-section created
from the raw electrical measurements, while the bottom
picture shows the best-fitting subsurface model calculated
by inversion of these measurements.The middle picture
shows the estimated pseudo-section that should have been
measured for the modelled subsurface.
45

geophysical data and the benefits and limitations

of each are summarised below. For most
survey reports, greyscale plots are the primary
presentation format, supported by some of
the plot types discussed below where these
aid exposition and interpretation.

2.2.1 Trace plots (XY traces, stacked traces)

Before the development of portable digital
computers, trace plots were a common
method for displaying magnetometer surveys,
as the analogue output from the magnetometer
could be directly connected to an XY chart
recorder, which displayed the data as it was
collected (Clark and Haddon-Reece 19723).
Each instrument traverse is depicted as an
approximately horizontal line but the line
trace deviates above or below a base (zero)
level in proportion to the magnitude of the
magnetometer measurement at that position
(Fig 26a). Subsequent traverses are plotted
parallel to the first, offset at increasing
distances up or down the page.
In its simplest implementation the trace plot
has only one variable plotting parameter
the vertical scale which specifies how far
the trace should deviate above or below the
base level in response to a unit change in
measurement.Thus the trace plot has a
relatively low degree of operator subjectivity
and anomalies of widely varying magnitudes
can all be discerned on the same plot.
Additionally, unlike other common techniques
that display the data in plan, the trace plot
depicts vertical profiles across anomalies, which
makes the distinctive signatures of some types
of anomalies readily apparent (such as the
distinctive kiln anomaly in Fig 26). Hence,
they provide a useful initial impression of
the relative overall variation in magnitude

a)

of anomalies in an unprocessed data-set and,

particularly when used to plot small areas
extracted from the overall survey, can greatly
aid interpretation of specific anomalies. However,
for the trace plot to be useful, it is essential
that a graphical indication be provided showing
the vertical scale used to represent variations
in the measured values.
A drawback of the profile view is that an
excessive number of extreme measurements
(especially spikes) in the data-set can render
the plot visually unintelligible. In this case it is
necessary to truncate (or clip) such values
before display.The very large magnetometer
surveys that are now practical with modern
multi-sensor instruments can also cause
problems as the sheer number of traverses
needing to be displayed means that there is
not enough space in the plotting area to
distinguish one from the next.Thus, it is now
not always practical for a survey report to
provide a trace plot of the unprocessed survey
data in its entirety, although plots of sub-areas
containing distinctive anomalies can still be
advantageously employed to support
interpretations.
Elaborations to the basic trace plot have
been introduced to create a more solid threedimensional appearance.Traverses plotted near
the bottom of the plot are considered to be
closer to the viewer than those farther up,
and a straightforward method to give a visual
impression of depth is to hide line segments
in the background that would be obscured by
anomalies rising up in the foreground (hidden
line removal) (Fig 26b).The impression can be
strengthened by laterally offsetting traverses in
proportion to their distance from the viewer
to provide a pseudo-isometric view (Fig 26c).

b)

2.2.2 Contour plots

Contour plots display the survey data in plan
using a series of contour lines (or isopleths)
to show the positions where the magnitudes
of the geophysical quantity being measured
cross one of a predetermined set of threshold
values (Fig 27) (Davis and Sampson 1986,
chapter 5). If the survey data contains mainly
localised variations from a base level that is
relatively constant over the whole area, it is
possible to produce an effective contour plot

a)

b)

c)

c)

Fig 26 (above top) (a) Basic trace plot of a magnetometer survey over a kiln feature; (b) the same trace plot with hidden lines removed to give an impression of solidity; and (c) replotted with
successive traverses increasingly offset to give a pseudo three dimensional effect.
Fig 27 (above bottom) Colour contour plots: (a) magnetometer data where the 1nT and 4nT contours outline the linear footings of timber buildings and adjacent enclosure ditches; (b) earth
resistance data set on a varying regional background where the choice of contouring has been less successful at isolating the anomalies; and (c) smoothly varying magnetic susceptibility data-set
with elevated readings coinciding with the location of a Roman villa and lower values associated with an adjacent river floodplain.

46

that outlines the important archaeological

anomalies (Fig 27a). However, the choice of the
particular data thresholds to contour is critical,
so contour plotting involves a high degree of
subjectivity. Where the background data level
varies across the plotting area, many different
contour values are needed to emphasise
localised details against all the different base
levels. Furthermore, whatever algorithm is used
to create continuous contours from the data,
the process intrinsically involves a degree of

a)

low-pass filtering, which will tend to smooth

out the smaller-scale anomalies that are typically
of most interest in archaeological surveys.
The net result of trying to select enough
contours to counteract these problems can
be a very busy, visually unintelligible, plot
(see for example Fig 27b; and Scollar et al
1990, fig 8.35). Hence, contour plots tend
not to be suitable for depicting detailed area
surveys containing complex archaeological

b)

anomalies. However, for low-resolution datasets where the measured geophysical property
varies smoothly across the survey area (Fig 27c),
or to emphasis the large scale regional trends
in a more densely sampled survey, contour plots
can still be an effective means of presentation.
They can also be deployed advantageously to
highlight very high magnitude thermoremanent
anomalies in magnetometer surveys. Wherever
contour plots are used, it is essential that the
contour values are labelled, as otherwise it is
impossible to determine which are the peaks
(highest values) and which the troughs (lowest
values) in the plot.

2.2.3 Dot density plots

c)

d)

Dot density plots (Fig 28a) also plot the

survey area in plan and were a popular means
of displaying data-sets prior to the advent of
affordable high-resolution computer graphics
when computer monitors were monochrome
and printers did not have high resolution half
tone or colour printing capabilities.The plotting
area is divided into small sub-rectangles each
corresponding to the footprint of one
geophysical measurement. Black dots are
placed randomly within each sub-rectangle
with the total number assigned being
determined according to the magnitude of
the geophysical measurement at that point.
The effect approximates to that of a printed
greyscale plot, albeit one in which the half tone
is readily visible. Dot density plots share many
of the advantages of greyscale plots outlined
below. However, the random assignment of
dots means that the same plot, using the same
plotting parameters, can appear different each
time it is generated, possibly affecting which
anomalies are highlighted or suppressed. Also,
the need to sub-divide the plotting area into
relatively large sub-rectangles, coupled with the
fact that randomly placed dots do not create
the same visual effect as a continuous periodic
half-tone pattern, can emphasise discontinuities
between adjacent measurements and lead
to a blocky appearance.

e)

2.2.4 Greyscale plots (greytone plots)

Fig 28 (above) Different display options for magnetometer data: (a) dot density plot; (b) linear greyscale or half-tone plot (no
interpolation); (c) linear greyscale plot of interpolated data; (d) equal area greyscale plot and; (e) plot produced using a colour palette.

Greyscale plots (Fig 28bd) are now the most

commonly used and versatile method of
displaying geophysical data in plan. As with
dot density plots the survey area is divided
into sub-rectangles each corresponding to the
footprint of one field measurement, but in this
case the rectangles are filled with a shade of
grey related to the magnitude of the geophysical
reading at that point. With modern computer
graphics capabilities a large palette of grey
shades can be used (typically between 100
and 256), providing a continuous variation in
tone between white and black.This continuous
gradation suppresses the perception of
discontinuities between adjacent measurements,
47

allowing the eye to concentrate on trends

across the survey area; and the effect can be
strengthened by interpolating the data to a
higher resolution, so that each shaded subrectangle corresponds to one pixel on the
display device being used.
The greyscale can be assigned with white
representing the lowest measured values,
progressively darker shades of grey
corresponding to higher values and black
representing the highest values of all; or the
assignment can be reversed, with black used
for the lowest values and white for the highest
(practitioners familiar with dot density plots
often favour the former while those from an
image processing background tend to use the
latter). Furthermore, the thresholds between
the measured values shaded with different
levels of grey can be assigned in different
ways, the most common choice being a linear
mapping from the survey datas range of
values, although log-linear and equal-area
(or histogram-equalised) assignments are also
useful, depending on the statistical characteristics
of the data being plotted. From the foregoing
it should be clear that it is essential for every
greyscale plot be accompanied by an
assignment key to show how the measured
values map to the shades of grey in the plot.
Greyscale plots of archaeological geophysical
data often look similar to vertical black and
white air photographs, a form of presentation
readily familiar even to those with no experience
of geophysical data interpretation. A variant
of the basic plot, the shadow plot, strengthens
this effect by pre-processing the survey data to
accentuate edges and sharp gradients running
in a pre-selected direction.The effect is similar
to an air photograph of earthworks taken in
strong oblique sunlight and can be effective in
emphasising linear anomalies sharing a common
alignment. A second variation is to replace the
greyscale with a palette of different colours to
produce a false-colour plot (Fig 28e), similar to
the way that differing land surface elevations
are colour coded in an atlas. However, it should
be noted that the eye will tend to be drawn
to the interfaces between contrasting colours,
so that the overall visual effect will be that of a
coloured contour plot. As with contour plots,
careful choice of the colour thresholds can
produce results that dramatically emphasise
particular anomalies while other details are
suppressed in the process. It is thus strongly
recommended that where colour plots are used,
a greyscale plot of the same data is also shown.

three-dimensional impression. Introduction of a

second set of parallel lines orthogonal to the
instrument traverses then creates a wire-frame
surface plot (Fig 29a) and the quadrilaterals so
formed can be coloured and shaded (Fig 29b)
to render the data as a solid three-dimensional
surface (see for example Foley et al 1991,
chapter 15). An extension to this type of
surface plot is the drape, where the shape of
the plotted surface is determined by the actual
topography of the area surveyed, whereas its
colouration is determined by the geophysical

measurements effectively a greyscale or

false-colour plot is draped over the surface
topography of the site (Fig 29c). Where the
plotted surface represents site topography,
exaggeration of the scale of the vertical axis
is often an effective way to highlight subtle
changes in elevation. In this case it is important
that the plot key makes clear the factor by
which the vertical axis has been scaled relative
to the two horizontal axes, in addition to the
usual requirement for a grey/colour scale
assignment key.

a)

b)

c)

2.2.5 Three-dimensional views

The isometric trace plots mentioned above
can incorporate diminution towards a horizon
point to provide perspective and enhance their
48

Fig 29 (above) Three-dimensional representations of geophysical data: (a) a wire frame plot (with vertical scale exaggerated);
(b) a shaded surface plot (with vertical values truncated to 20nT); and (c) a plot of the data draped over a digital terrain
model (with vertical scale exaggerated).

A different type of three-dimensional view can

be used where a 3D volume of data has been
imaged (as is often measured with GPR or
ERT equipment).The resulting data can be
displayed as either a false-perspective cut-away
model or as an iso-surface where a threshold
value is chosen and all parts of the volume
where the geophysical value is below this
threshold are considered transparent, while
those parts above the threshold are rendered
opaque (see Fig 16d). Iso-surface plots can
assist in elucidating spatial relationships between
anomalies associated with individual causative
features, although the selection of an appropriate
threshold level requires careful judgement.
All types of three-dimensional rendering can
provide visually striking representations of the
survey data but it should be borne in mind
that they will emphasise anomalies in the
foreground of the view while obscuring those
farther back.Thus the choice of viewpoint
when creating the plot will determine which
details are visible, and it should be recognised
that plots from more than one different
viewpoint may be necessary to adequately
display all parts of the survey area.
Where a computer display screen is being
used rather than a hard copy device, it is
possible to interactively change the viewpoint or
animate a sequence of views as a fly-through
to overcome this difficulty, although it is not
possible to reproduce this type of interactive
presentation in the printed report which will
be the authoritative reference for the survey
project. Hence, while three-dimensional views
can be used to good effect to highlight specific
details within a geophysical data-set, they
should not be the only type of graphical plot
presented, but should be supported by more
traditional plan representations, such as
greyscale plots.

2.3 Data interpretation

Raw geophysical data can be obtained,
processed and presented, one way or another,
by following instruction manuals and course
notes. However, the interpretation that follows
generally requires a wider experience
encompassing an understanding of the site
conditions and their history, the principles of
archaeological geophysics, as well as the foibles
of instruments and survey methodologies.
A good knowledge of archaeology is of
course important, as well as of geology and
geomorphology. Ideally an interpreter will
already have such experience, and will preferably
have conducted and/or directed the fieldwork
concerned personally (although it need not
follow that the fieldworker is thereby
automatically qualified in the subsequent
interpretation of the data).

The factors that require consideration in

arriving at an interpretation will vary from site
to site, but should normally include at least
the following:
natural
solid geology
drift geology
soil type
soil magnetic
susceptibility
geomorphology
surface conditions
topography
seasonality

artificial
landscape history
known/inferred
archaeology
agricultural practices
modern interference
survey methodology
data treatment
any other available
data

Any interpretation must normally take into

account each of these factors, the emphasis
varying according to circumstance, and should
include consultation with colleagues and other
relevant specialists where necessary. For
instance, experience shows that where there
is even the most meagre earthwork survival,
the combination of field survey and geophysical
survey is highly beneficial to their joint
interpretation.The degree of usefulness of the
former will increase according to the condition
of the earthworks and the intensity of the field
survey. Likewise, where earthworks have been
completely ploughed out, comparison with
aerial photographic analysis and evidence
from historic maps will also yield useful
interpretative data.

geophysical anomalies an acceptable additional

option only on the clear understanding that
such ratings are partly subjective and potentially
fallible assessments, applicable only to the
specific survey data concerned.
Refinement of the interpretation of geophysical
surveys is, to a significant degree, dependent
upon the feedback of ground-truth following
the survey fieldwork. Wherever possible every
effort should be made to encourage such
feedback and its subsequent dissemination into
the general pool of accumulated experience
(see Part II, 5).To aid this process, curators can
stipulate that trial trenching and excavation
reports are copied to the geophysical contractor,
that mitigation and publication briefs make
allowance for the results of geophysical surveys,
and that reporting includes the post-excavation
comments of the geophysical contractor
(if appropriate).

Arriving at an interpretation that takes into

account so many factors can be a finely balanced
process and the outcome will be coloured by,
and depend significantly upon, the experience
of the interpreter. Above all it is crucial that
any interpretation draws a clear line for the
reader between demonstrable fact that is
securely supported by the data, and less secure
inference. Here, we would only warn against
a tendency to see and attribute significance to
every detail in other words, to over-interpret.
Minutely annotated plots with laborious textual
referencing of every apparently significant
anomaly stretch the credibility and wear down
the patience of readers. Generally speaking, it
is preferable to exercise as much objectivity
and restraint as possible, and to err towards
under-interpretation, resisting the embellishment
of plots with wishful patterns and details.
While much importance is given to the
graphical presentation of results (see Part II,
4.9), and it is often this, not the text, that holds
the clients attention, it is important that the
graphics are supported and complemented by
precise written discussion as well. Occasionally,
contractors have risked applying percentage
confidence ratings to the interpretation of
49

References
ACAO 1993 Model Briefs and Specifications for
Archaeological Assessments and Field Evaluations.
Association of County Archaeological Officers
Annan, A P 2004 Ground Penetrating Radar
Principles, Procedures and Applications. Ontario:
Sensors and Software
Annan, A P and Cosway, S W 1992 Simplified
GPR beam model for survey design, in Society
of Exploration Geophysicists, 62nd Annual
Meeting 1992, New Orleans: Society of
Exploration Geophysicists, 3569
Arzi, A A 1975 Microgravimetry for
engineering applications. Geophys Prospecting
23, 40825
Aspinall, A and Crummett, J G 1997 The
electrical pseudosection. Archaeol Prospection
4, 3747
Aspinall, A and Lynam, J T 1968 Induced
polarization as a technique for archaeological
surveying. Prospezioni Archeol 3, 913
Aspinall, A and Lynam, J T 1970 An induced
polarisation instrument for the detection of
near-surface features. Prospezioni Archeol 5,
6775
Aspinall, A and Saunders, M K 2005
Experiments with the square array. Archaeol
Prospection 12, 11529
Athanasiou, E N,Tsourlos, P I,Vargemezis,
G N, Papazachos, C B and Tsokas, G N 2007
'Non-destructive DC resistivity surveying using
flat-base electrodes'. Near Surface Geophysics 5,
26374
Bates, M R and Bates, C R 2000 Multidisciplinary
approaches to the geoarchaeological evaluation
of deeply stratified sedimentary sequences:
examples from Pleistocene and Holocene
deposits in southern England, United Kingdom.
J Archaeol Sci 27, 84558
Bates, M R, Bates, C R and Whittaker, J E 2007
Mixed method approaches to the investigation
and mapping of buried Quaternary deposits:
examples from southern England. Archaeol
Prospection 14, 10429

50

Bellerby,T J, Noel, M J and Brannigan, K 1990 A

thermal method for archaeological prospection:
preliminary investigations. Archaeometry 32,
191203

Challis, K and Howard, A J 2006 'A Review of

trends within archaeological remote sensing in
alluvial environments'. Archaeol Prospection 13,
23140

Bettess, F 1992 Surveying for Archaeologists

(rev edn). Durham: University of Durham

CBA 1982 Guidelines for the Preparation

of Contracts for Archaeological Excavations.
London: Council for British Archaeology

Bevan, B 1991 'The search for graves'.

Geophysics 56, 131019
Bewley, R H 1993 Aerial photography for
archaeology, in J Hunter and I Ralston (eds),
Archaeological Resource Management in the UK:
an Introduction. Stroud: Alan Sutton, 197204
Bewley, R H, Crutchley, S P and Shell, C A
2005 New light on an ancient landscape: lidar
survey in the Stonehenge World Heritage Site.
Antiquity 79, 63647
Bewley, R H and Raczklowski, W (eds) 2002
Aerial Archaeology: Developing Future Practice.
Nato Science Series I: Life and Behavioural
Sciences, 337. Amsterdam: IOS Press
Blakely, R J 1996 Potential Theory in Gravity and
Magnetic Applications. Cambridge: Cambridge
University Press
Blzkovsky, M 1979 Processing and applications
in microgravity surveys. Geophys Prospecting 27,
84851
Bowden, M (ed) 1999 Unravelling the Landscape:
An Inquisitive Approach to Archaeology. Stroud:
Tempus
Breiner, S 1999 Applications Manual for Portable
Magnetometers. San Jose: Geometrics
Brooke, C J 1987 Ground-based remote sensing
for archaeaological information recovery in
historic buildings. Internat J Remote Sensing 8,
103948
Brown, D H 2007 Archaeological Archives: A
Guide to Best Practice in Creation, Compilation,
Transfer and Curation. Birmingham: IFA on
behalf of the Archaeological Archives Forum
Butler, D K 1984 Microgravimetric and gravity
gradient techniques for detection of subsurface
cavities. Geophysics 49, 108496

Bayley, J, Dungworth, D and Paynter, S 2001

Archaeometallurgy. Centre for Archaeology
Guidelines 2001/01. London: English Heritage

Canti, M G and Meddens, F M 1998

'Mechanical coring as an aid to archaeological
projects'. J Field Archaeol 25, 97105

Becker, H 1995 From nanotesla to picotesla, a

new window for magnetic prospecting in
archaeology. Archaeol Prospection 2, 21728

Carey, C J, Brown,T G, Challis, K C, Howard, A

J and Cooper, L 2006 'Predictive modelling of
multiperiod geoarcheological resources at a
river confluence: a case study from the TrentSoar, UK'. Archaeol Prospection 13, 24150

Chwala, A, Ijsselsteijn, R, May,T, Oukhanski, N,

Schler,T, Schultze,V, Stolz, R and Meyer, H-G
2003 Archaeometric prospection with HighTC SQUID gradiometers. IEEE Trans Applied
Superconductivity 13, 76770
Chwala, A, Stolz, R, Ijsselsteijn, R, Schultze,V,
Ukhansky, N, Meyer, H-G and Schler,T 2001
SQUID gradiometers for archaeometry.
Superconductor Sci Technol 14, 111114
Ciminale, M and Loddo, M 2001 Aspects of
magnetic data processing. Archaeol Prospection 8,
23946
Clark, A J 1983 The testimony of the topsoil,
in G S Maxwell (ed), The Impact of Aerial
Reconnaissance on Archaeology. London: CBA
Res Rep 9, 12835
Clark, A J and Haddon-Reece, D 19723 An
automatic recording system using a Plessey
fluxgate gradiometer. Prospezioni Archeol 78,
10713
Clark, A J 1992 'Archaeogeophysical
prospecting on alluvium', in S Needham and M
G Macklin (eds), Alluvial Archaeology in Britain.
Oxbow Monogr 27. Oxford: Oxbow
Clark, A J C 1996 Seeing Beneath the Soil 2nd
edn. London: Batsford
Clarke, C M, Utsi, E and Utsi,V 1999 'Ground
penetrating radar investigations at North
Ballachulish Moss, Highland Scotland'. Archaeol
Prospection 6, 10721.
Cole, M A, Linford, N T, Payne, A P and
Linford, P K 1995 Soil magnetic susceptibility
measurements and their application to
archaeological site investigation, in J Beavis
and K Barker (eds), Science and Site: Evaluation
and Conservation, Proceedings of the Conference
Held 810 September 1993 Bournemouth:
Bournemouth University Occas Pap 1, 11462
Collier, L, Hobbs, B, Neighbour,T and Strachan,
R 2003 Resistivity imaging survey of Capo Long
Barrow, Aberdeenshire. Scottish Archaeological
Internet Report 6 http://www.sair.org.uk/sair6/
index.html [visited 26/02/2008]

Conyers, L B 2004 Ground Penetrating Radar for

Archaeology. Walnut Creek, CA: AltaMira Press
Conyers, L B and Goodman, D 1997
Ground Penetrating Radar: An Introduction for
Archaeologists.Walnut Creek, CA: AltaMira Press
Cox, C 1992 'Satellite imagery, aerial
photography and wetland archaeology an
interim report on an application of remote
sensing to wetland archaeology: the pilot study
in Cumbria, England'. World Archaeol 24, 24967
Crutchley, S 2006 Light detection and ranging
(lidar) in the Witham Valley, Lincolnshire: an
assessment of new remote sensing techniques.
Archaeol Prospection 13, 2517
Dabas, M, Hesse, A and Tabbagh, J 2000
Experimental resistivity survey at Wroxeter
archaeological site with a fast and light recording
device. Archaeol Prospection 7, 10718
Dalan, R A and Banerjee, S K 1996 Soil
magnetism, an approach for examining
archeological landscapes. Geophys Res Letters
23, 1858
Dalan, R A and Banerjee, S K 1998 Solving
archaeological problems using techniques of
soil magnetism. Geoarchaeol 13, 336

Di Filippo, M, Ruspandini,T and Toro, B 2000

The role of gravity surveys in archaeology,
in M Pasquinucci and F Trment (eds), NonDestructive Techniques Applied to Landscape
Archaeology. Archaeology of Mediterranean
Landscapes 4. Oxford: Oxbow, 14854
DoE 1990 Planning Policy Guidance Note 16:
Archaeology and Planning (PPG 16). Department
of the Environment. London: HMSO
Donoghue, D N M 2001 Remote sensing, in
D R Brothwell and A M Pollard (eds), Handbook
of Archaeological Sciences. Chichester: Wiley,
55563
Donoghue, D N M and Shennan, I 1988 'The
application of multi-spectral remote sensing
techniques to wetland archaeology', in P Murphy
and C French (eds), The Exploitation of Wetlands.
BAR, Brit Ser 186, 4759
Drahor, M G 2004 Application of the selfpotential method to archaeological prospection:
some case studies. Archaeol Prospection 11,
77105
Eder-Hinterleitner, A, Neubauer, W and
Melichar, P 1996 Restoring magnetic
anomalies. Archaeol Prospection 3, 18598

Daniels, D (ed) 2004 Ground Penetrating Radar.

London: IEE

English Heritage 2002 With Alidade and Tape:

graphical and plane table survey of archaeological
earthworks. Swindon: English Heritage

Darvill,T 1993 'Working practices', in J Hunter

and I Ralston (eds), Archaeological Resource
Management in the UK: an Introduction. Stroud:
Alan Sutton, 16983

English Heritage 2003 Where on Earth are

We? The Global Positioning System (GPS) in
archaeological field survey. Swindon: English
Heritage

Darvill,T and Atkins, M 1991 Regulating

Archaeological Work by Contract. IFA Technical
Pap 8. Birmingham: Institute of Field
Archaeologists

English Heritage 2006 Our Portable Past: a

statement of English Heritage policy and good
practice for portable antiquities/surface collected
material in the context of field archaeology and
survey programmes (including the use of metal
detectors). Swindon: English Heritage

David, A 1994 'The role of geophysical survey

in early medieval archaeology'. Anglo-Saxon
Studies in Archaeology and History 7, 126
Davis, J C and Sampson, R J 1986 Statistics
and Data Analysis in Geology 2 edn.
Chichester: Wiley
Dearing, J A, Hay, K L, Baban, S M J, Huddleston,
A S, Wellington, E M H and Loveland, P J 1996
Magnetic susceptibility of soil: an evaluation of
conflicting theories using a national data-set.
Geophys J Internat 127, 72834
Devereux, B J, Amable, G S, Crow, P and Cliff,
A D 2005 The potential of airborne lidar for
detection of archaeological features under
woodland canopies. Antiquity 79, 64860

English Heritage 2007 Geoarchaeology: using

earth sciences to understand the archaeological
record. Swindon: English Heritage
English Heritage, forthcoming 2008 Marine
Geophysics Data Acquisition and Processing
Guidance Notes for Submerged Archaeological
Sites. Swindon: English Heritage.
Eppelbaum, L V, Khesin, B E and Itkis, S E 2001
Prompt magnetic investigations of archaeological
remains in areas of infrastructure development:
Israeli experience. Archaeol Prospection 8,
16386

Evans, M E and Heller, F 2003 Environmental

Magnetism Principles and Applications of
Enviromagnetics. San Diego: Academic Press
Fajklewicz, Z J 1976 Gravity vertical
measurements for the detection of small
geologic and anthropogenic forms. Geophysics
41, 101630
Fassbinder, J W E and Stanjek, H 1993
Occurrence of bacterial magnetite in soils
from archaeological sites. Archaeol Polona 31,
11728
Foley, J D, van Dam, A, Feiner, S K and Hughes,
J F 1991 Computer Graphics: Principles and
Practice 2 edn. Wokingham: Addison Wesley
Fowler, M J 2002 Satellite remote sensing and
archaeology: a comparative study of satellite
imagery of the environs of Figsbury Ring,
Wiltshire. Archaeol Prospection 9, 5570
Frazier, C H, Cadalli, N, Munson, D C and
OBrien, W D 2000 Acoustic imaging of
objects in soil. J Acoustic Soc America 108,
14756
Gaffney, C and Gater, J A 1993 'Practice and
method in the application of geophysical
techniques in archaeology', in J Hunter
and I Ralston (eds), Archaeological Resource
Management in the UK: An Introduction.
Stroud: Alan Sutton, 20514
Gaffney, C and Gater, J A 2003 Revealing the
Buried Past: Geophysics for Archaeologists.
Stroud:Tempus
Gaffney, C, Gater, J A and Ovenden, S M 2002
The Use of Geophysical Techniques in
Archaeological Evaluations. IFA Technical Pap 6.
Reading: Insititute of Field Archaeologists
Glover, J M 1987 'The use of sub-surface radar
for shallow site investigation'. London: Kings
College, University of London PhD thesis
Goodman, D, Nishimura,Y, Hongo, H and
Higashi, N 2006 Correcting for topography
and the tilt of ground-penetrating radar
antennae. Archaeol Prospection 13, 15761
Goulty, N R, Gibson, J P C, Moore, J G and
Welfare, H 1990 Delineation of the vallum at
Vindobala, Hadrians Wall, by shear-wave seismic
refraction survey. Archaeometry 32, 7182
Grandjean, G and Durand, H 1999 Radar
Unix: a complete package for GPR data
processing. Computers and Geosciences 25,
1419

51

Haigh, J G B 1992 Automatic grid balancing

in geophysical survey, in G Lock and J Moffett
(eds), Computer Applications and Quantative
Methods in Archaeology 1991. BAR S577, 1916
Heron, C 2001 Geochemical prospecting, in D
R Brothwell and A M Pollard (eds), Handbook of
Archaeological Sciences. Chichester:Wiley, 56573
Hildebrand, J A, Wiggins, S M, Henkart, P C
and Conyers, L B 2002 Comparison of seismic
reflection and ground-penetrating radar imaging
at the Controlled Archaeological Test Site,
Champaign, Illinois. Archaeol Prospection 9, 921
Holden, N, Horne, P and Bewley, R H 2002
High resolution digital airborne mapping and
archaeology, in R H Bewley and W Raczklowski
(eds), Aerial Archaeology: Developing Future
Practice. Nato Science Series I: Life and
Behavioural Sciences 337. Amsterdam: IOS
Press, 17380
Horton, B M 1959 Noise-modulated distance
measuring system. Proceed Inst Radio Engineers
147, 8218
Howard, A J and Macklin, M G 1999 'A generic
geomorphological approach to archaeological
interpretation and prospection in British river
valleys: a guide for archaeologists investigating
Holocene landscapes'. Antiquity 73, 52741
Huang, H and Won, I J 2000 Conductivity
and susceptibility mapping using broadband
electromagnetic sensors. J Environ Engineering
Geophys 5, 3141
IFA 2001 Standard and Guidance for Archaeological
Field Evaluation 2 edn. Birmingham: IFA
Jol, H and Bristow, C 2003 GPR in sediments:
advice on data collection, basic processing
and interpretation, a good practice guide, in
C Bristow and H Jol (eds), Ground Penetrating
Radar in Sediments. London: Geological Society
Special Pub 211
Jorgensen, M S 1997 'Looking into the
landscape'. Aarhus Geoscience 7, 15766
Keller, G V and Frischknecht, F C 1966 Electrical
Methods in Geophysical Prospecting. New York:
Pergamon
Kooiman, G and de Jongh, I G 1994 Thermal
revelations, in J Taylor (ed), The Conservation
and Repair of Ecclesiastical Buildings. London:
Cathedral Communications, 1011
Kvamme, K L 2003 Multidimensional
prospecting in North American Great Plains
village sites. Archaeol Prospection 10, 13142
52

Lawson, A J 1993 'The assessment of trunk

road schemes'. The Field Archaeologist 18,
3515

Linnington, R E 1966 The test use of a

gravimeter on Etruscan chambered tombs at

Cerveteri. Prospezioni Archeol 1, 3741

Leckebusch, J 2003 Ground penetrating radar:

a modern three-dimensional prospection
method. Archaeol Prospection 10, 21341

Loke, M H 2004 Tutorial: 2-D and 3-D

electrical imaging surveys [web page].

http://www.geoelectrical.com/coursenotes.pdf

[Accessed 13/06/2007]

Leckebusch, J 2005 Precision real-time

positioning for fast geophysical prospection'.
Archaeol Prospection 12, 199202
Leckebusch, J and Rychener, J 2007 Verification
and topographic correction of GPR data in
three dimensions. Near Surface Geophysics 5,
395403
Lee, E 2006 Management of Research Projects
in the Historic Environment The MoRPHE Project
Managers Guide. Swindon: English Heritage
Lehmann, F and Green, A G 1999
Semiautomated georadar data acquisition in
three dimensions. Geophysics 64 , 71931
Linford, N [T] 1998 Geophysical survey at
Boden Vean, Cornwall, including an assessment
of the microgravity technique for the location
of suspected archaeological void features.
Archaeometry, 40, 187216
Linford, N 2003 Magnetic Susceptibility, in
M Brennand and M Taylor (eds), The Survey
and Excavation of a Bronze Age Timber Circle
at Holme-next-the-Sea, Norfolk, 19989.
Proceed Prehist Soc 69, 184
Linford, N 2004 From hypocaust to hyperbola:
ground penetrating radar surveys over mainly
Roman remains in the UK. Archaeol Prospection
11, 23746
Linford, N 2005 Archaeological applications of
naturally occurring nanomagnets. International
Conference on Fine Particle Magnetism, 2022
September 2004, London. J Physics: Conference
Ser 17, 12744
Linford, N 2006 The application of geophysical
methods to archaeological prospection.
Reports on Progress in Physics 69, 220557
Linford, P and Welch, C M 2004
Archaeomagnetic analysis of glassmaking sites
at Bagots Park in Staffordshire, England. Physics
of the Earth and Planetary Interiors 147, 20921
Linford, N, Linford, P, Martin, L and Payne, A
2007 'Recent results from the English Heritage
caesium magnetometer system in comparison
to recent fluxgate gradiometers'. Archaeol
Prospection 14, 15166

Loke, M H and Barker, R D 1996 Practical

techniques for 3D resistivity surveys and data

inversion. Geophys Prospection 44, 499523

Metwaly, M, Green, A G, Horstmeyer, H, Maurer,

H, Abbas, A M and Hassaneen, A G 2005

Combined seismic tomographic and ultrashallow

seismic reflection study of an early dynastic

Mastaba, Saqqara, Egypt. Archaeol Prospection

12, 24556

Milsom, J 2002 Field Geophysics. Chichester:

Wiley

Moskowitz, B M 1995 Fundamental units and

conversion factors, in T J Ahrens (ed), Global

Earth Geophysics: A Handbook of Physical

Constants. AGU reference shelf 1. Washington:

American Geophysical Union

Needham, S and Macklin, M G 1992 Alluvial

Archaeology in Britain. Oxford: Oxbow

Neubauer, W and Eder-Hinterleitner,

A 1998 3D-interpretation of postprocessed

archaeological magnetic prospection data.

Archaeol Prospection 4, 191205

Neubauer, W, Eder-Hinterleitner, A, Seren, S

and Melichar, P 2002 Georadar in the Roman

civil town Carnumtum, Austria: an approach

for archaeological interpretation of GPR data.

Archaeol Prospection 9, 13556

Ovenden, S M 1994 Applications of seismic

refraction to archaeological prospecting.

Archaeol Prospection 1, 5364

Palmer, R and Cox, C 1993 Uses of Aerial

Photography in Archaeological Evaluations.

IFA Techn Pap 12. Birmingham: IFA

Payne, M A 1981 SI and Gaussian CGS

units, conversions and equations for use in

geomagnetism. Physics of the Earth and

Planetary Interiors 26, 1016

Pipan, M, Baradello, L, Forte, E, Prizzon, A and

Finetti, I 1999 2-D and 3-D processing and

interpretation of multi-fold ground penetrating

radar data: a case history from an archaeological

site. J Applied Geophysics 41, 27192

Powlesland, D J, Lyall, J and Donoghue, D N M

1997 Enhancing the record through remote
sensing: the application and integration of multisensor, non-invasive remote sensing techniques
for the enhancement of the Sites and
Monuments Record. Heslerton Parish Project,
N Yorkshire, England. Internet Archaeol 2
Powlesland, D J, Lyall, J, Hopkinson, D, Donoghue,
D N M, Beck, M, Harte, A and Stott, D 2006
'Beneath the sand remote sensing, archaeology,
aggregates and sustainability: a case study from
Heslerton, the Vale of Pickering, North Yorkshire,
UK'. Archaeol Prospection 13, 2919
Reynolds, J M 1997 An Introduction to Applied
and Environmental Geophysics, Chichester:Wiley
Ruffell, A and Wilson, J 1998 Near-surface
investigation of ground chemistry using
radiometric measurements and spectral
gamma-ray data. Archaeol Prospection 5, 20315
Schleifer, N, Weller, A, Schneider, S and Junge, A
2002 Investigation of a Bronze Age plankway
by spectral induced polarization. Archaeol
Prospection 9, 24353
Schmidt, A 2002 Geophysical Data in Archaeology:
A Guide to Good Practice. Oxford: Oxbow
Schmidt, A and Marshall, A 1997 Impact of
resolution on the interpretation of archaeological
prospection data, in A G M Sinclair, E A Slater
and J A J Gowlett (eds), Archaeological Sciences
1995. Oxford: Oxbow, 3438
Scollar, I 1962 Electromagnetic prospecting
methods in archaeology. Archaeometry 5, 14653
Scollar, I,Tabbagh, A, Hesse, A and Herzog, I
(eds) 1990 Archaeological Prospecting and
Remote Sensing.Topics in Remote Sensing 2.
Cambridge: Cambridge University Press
Shennan, I and Donoghue, D N M 1992
Remote sensing in archaeological research,
in A M Pollard (ed), New Developments in
Archaeological Science. Proc Brit Acad 77, 22332
Sirri, S, Eder-Hinterleitner, A, Melichar, P and
Neubauer, W Comparison of different GPR
systems and antenna configurations at the
Roman site of Carnuntum, in S Piro (ed),
6th International Conference on Archaeological
Prospection 2005. Rome: Inst Technologies
Applied to Cultural Heritage, 17680
Somers, L, Linford, N, Penn, W, David, A, Urry, L
and Walker, R 2005 Fixed frequency radio wave
imaging of subsurface archaeological features:
a minimally invasive technique for studying
archaeological sites. Archaeometry 47, 15973

Stone, J, Clowes, M, Dennison, E, Lee, G and

Wilson, P 2004 Mapping Roman earthworks,
photogrammetry: Cawthorn Camps. Cons Bull
45, 1011
Tabbagh, A 1986 Applications and advantages
of the Slingram electromagnetic method for
archaeological prospecting. Geophysics 51,
57684
Tabbagh, J 2003 Total field magnetic prospection:
are vertical gradiometer measurements
preferable to single sensor survey?. Archaeol
Prospection 10, 7581
Taylor, B N 1995 Guide for the use of the
International System of Units (SI), in NIST
Guide to SI Units. National Institute of Standards
and Technology Special Pub 811. Gaithersburg:
National Institute of Standards and Technology
Telford, W M, Geldart, L P, Sheriff, R E and
Keys, D A 1976 Applied Geophysics. Cambridge:
Cambridge University Press
Theimer, B D, Nobes, D C and Warmer, B G
1994 'A study of the geo-electrical properties
of peatlands and their influence on groundpenetrating radar surveying. Geophys Prospecting
42, 179209
Thompson, R and Oldfield, F 1986
Environmental Magnetism. London: Allen
and Unwin
Tite, M S and Mullins, C E 1969
Electromagnetic prospecting, a preliminary
investigation. Prospezioni Archeol 4, 95102

Wait, J R 1955 Mutual electromagnetic

coupling of loops over a homogeneous
ground. Geophysics 20, 6308
Walker, A R 1991 Resistance Meter RM15
Manual version 1.2. Geoscan Research
Walker, A R 2005 Geoplot Version 3.00 for
Windows: Instruction Manual.Version 1.97.
Geoscan Research
Weston, D 2001 'Alluvium and geophysical
prospection'. Archaeol Prospection 8, 26572
Wilson, D R 2000 Air Photo Interpretation for
Archaeologists 2 edn. Stroud:Tempus
Winterbottom, S J and Dawson,T 2005
Airborne multi-spectral prospection for
buried archaeology in mobile sand dominated
systems. Archaeol Prospection 12, 20520
Wynn, J C and Sherwood, S I 1984 The
self-potential (SP) method: an inexpensive
reconnaisance and archaeological mapping
tool. J Field Archaeol 11, 195204
Xu,Y, Narayanan, R M, Xu, X and Curtis, J O
2001 Polarimetric processing of coherent
random noise radar data for buried object
detection. IEEE Trans Geoscience and Remote
Sensing 39, 46778
Zakosarenko,V, Chwala, A, Ramos, J, Stolz, R,
Schultze,V, Ltjen, H, Blume, J, Schler,T and
Meyer, H-G 2001 HTS dc SQUID systems for
geophysical prospection. IEEE Trans Applied
Superconductivity 11, 8969

Tite, M S and Mullins, C E 1973 Magnetic

viscosity, quadrature susceptibility and multi
frequency dependence of susceptibility in
single domain assemblies of magnetite and
maghemite. J Geophys Res 78, 8049
Utsi, E 2001. The investigation of a peat moss
using ground probing radar, in S Vertrella,
O Bucci, C Elachi, C Lin, M Rouz and M Sato
(eds), Remote Sensing by Low Frequency Radars
Workshop 2001. Naples: European Association
of Remote Sensing Laboratories
Vafidis, A, Manakou, M, Kritikakis, G,Voganatsis,
D, Sarris, A and Kalpaxis,T 2003 Mapping the
ancient port at the archaeological site of Itanos
(Greece) using shallow seismic methods.
Archaeol Prospection 10, 16373
Vaughan, C J 1986 Ground-penetrating radar
surveys used in archaeological investigations.
Geophysics 51, 595604

53

Glossary
area survey the gathering of geophysical
data over an area, usually across a pre-defined
survey grid, resulting in a two-dimensional plan
image of the results the term thus excludes
isolated survey transects; detailed area survey
refers to surveys where data is gathered at
intervals of 1m x 1m, or less
alkali-vapour magnetometer a type of
magnetometer capable of making very
sensitive measurements of a magnetic field
by observing changes in the quantum energy
states of electrons exposed to it.The method
employed is most readily applied to alkali metals
in the gaseous state, as these chemical elements
have a single unpaired electron in their outer
shell. Also known as optically pumped
magnetometers (see Part IV, 1.2).
appraisal a rapid reconnaissance of site
and records to identify (within the planning
framework) whether a development proposal
has a potential archaeological dimension
requiring further clarification (IFA 2001)
brief an outline framework of the archaeological
circumstances that have to be addressed,
together with an indication of the scope of
works that will be required
brownfield any land that has been previously
developed
caesium magnetometer currently the most
common type of alkali-vapour magnetometer
centre frequency a nominal value for a GPR
antenna describing the dominant operating
frequency that will influence the depth of
penetration and resolution (see Part IV, 1.4)
conductivity () the ability of a material to
carry an electric current measured in units of
millisiemens; also defined as the reciprocal of
volume resistivity
contact resistance in an earth resistance
survey, the contribution to the total electrical
resistance caused by the interface between the
electrodes and the soil. It is difficult to make
good electrical contact between a temporarily
inserted electrode and dry soil, so this is
typically the largest contribution to the
overall resistance (Part IV, 1.3)
curator a person or organisation responsible
for the conservation and management of
archaeological evidence by virtue of official
or statutory duties (IFA 2001)

54

digital elevation model (DEM) a topographic

model of the bare earth that can be manipulated
by computer programmes and stored in a
grid format
digital surface model (DSM) a topographic
model of the Earths surface (including terrain
cover such as buildings and vegetation) that
can be manipulated by computer programmes
digital terrain model (DTM) a topographic
model of the bare earth that can be
manipulated by computer programmes
eddy currents electrical current induced in
a conductive feature by a changing magnetic
field, which in turn produces a secondary
electromagnetic field that can be detected
by a geophysical instrument
electrical skin depth depth to which the
alternating electric current induced by an
electromagnetic field will extend into a
conductive object or soil.This material
property is dependent on the frequency of
the incident electromagnetic field and restricts
the range of soil conductivity meters when
operated at high frequencies over
conductive sites.
fiducial (fiduciary) marker a marker introduced
into a sequence of time-triggered measurements
that can be related to a fixed position on the
ground.The position of each measurement
made by a moving instrument can then be
deduced by comparing its time-stamp to that
of the closest (in time) fiducial markers.
fluxgate magnetometer a solid-state
magnetometer that measures the strength
of an ambient magnetic field by observing
the effect it has on two oppositely wound
solenoids.The solenoids are both magnetised
by the same alternating electric current and
are placed so close together that, in the absence
of any external magnetic field the alternating
magnetic fields they generate would cancel
each other out (see Part IV, 1.2).
fractional conversion a ratio of magnetic
susceptibility before and after laboratory
heating of a soil sample to a maximum
possible value. High values may be suggestive
of occupation processes (burning) that may
otherwise be masked through changes in
background geology.
frequency dependence of susceptibility variation
of magnetic susceptibility measured from soil
samples in an alternating field at two or more
frequencies. High values may indicate the
presence of very fine magnetic particles often
associated with archaeological settlement activity.

georeferencing the process of fixing the location

of a field survey grid on the surface of the
Earth, thus making it possible to re-established it
at a later date.This can be achieved by making
measurements to landmarks with known
positions or by direct co-registration (often
using a GPS system) to a standard map
coordinate system such as the Ordnance
Survey National Grid.
geotechnical survey any subsurface
investigation, geophysical or (semi-) invasive,
conducted to assist with the technical rather
than archaeological aspects of a proposed
development or extraction scheme. Such data
(eg from an auger survey) may also prove
highly useful to archaeological geophysicists.
grid see survey grid and sub-grid
gradiometer any instrument that records
differences in a measured property between
two sensors set at a fixed distance apart, rather
than the total value of the property measured
using a single sensor.This configuration is usually
encountered in magnetometers (see Part IV, 1.2).
grey literature literature that is produced by
all levels of government, academics, business
and industry, in print and electronic formats,
but which is not controlled by commercial
publishers. Most geophysical survey reports
fall into this category.
ground-truth the real physical circumstances as
directly measured or observed at the ground
surface, or from direct interventions such as
coring, test-pitting, trenching or area excavation.
Ground-truth data is used to help validate,
calibrate and interpret indirect geophysical
and remote sensing responses.
HER Historic Environment Record
interpolation a method for calculating values
for new data points in between a discrete set
of measured data points. Often used to reduce
the blocky appearance of greyscale plots of
surveys where the field sample density was
relatively sparse. Interpolation does not increase
the amount of information in a data-set and
is not a substitute for employing a higher
sampling density in the field.
large area in these guidelines, any area in
excess of 20ha

magnetometer scanning the informal use of

a portable magnetometer to assess magnetic
response over a site or area, and/or to
locate specific strongly magnetic features.
The experienced operator walks across
a site, usually at widely spaced intervals
(5m +), observing the instruments output
and marking potential anomalies for more
detailed investigation.Recorded magnetometer
scanning utilises a grid and/or GPS location
together with plotting of the instrument signal.
map regression the process of using historic
mapped information (for example old OS,
tithe and estate maps), working backwards in
time from the present day, to investigate and
reconstruct the past appearance of sites,
buildings and landscapes
pseudo-section a sequence of earth resistance
measurements made along the same surface
base-line with different electrode separations
and arranged to depict an approximate vertical
profile of the variation of electrical resistance
with depth (see Part IV, 1.3.4)
reflector any object with suitable physical
properties to reflect an incident GPR signal,
often described as point, planar, dipping, linear,
complex (diffuse), etc to indicate the likely
nature of the causative feature. Hyperbolic
responses can be recorded over reflectors of
limited cross-section and show characteristic
tails, dependent on the velocity of the radar
wave, dipping to either side of an apex
immediately above the object.
Section 42 licence a licence issued in
accordance with Section 42 of the Ancient
Monuments and Archaeological Areas Act
1979 to those wishing to undertake metal
detecting or geophysical surveys over legally
protected sites. It currently takes the form of a
letter from English Heritage formally authorising
the conditional undertaking of such surveys
over specified scheduled monuments or other
protected places.
signal-to-noise ratio used in a general sense
to describe the limit of detection for an
individual instrument type or technique where
the magnitude of response from an underlying
feature is no longer discernible above the
background noise level
specification a written schedule of works
required for a particular project (by a curator,
planning archaeologist or client) set out in
sufficient detail to be quantifiable, implemented
and monitored; normally prepared by or
agreed with the relevant curator (IFA 2001)

square array one of the arrangements of

electrodes used for making earth resistance
measurements.The four electrodes are
positioned at the corners of a square, a
configuration particularly suited to fourwheeled cart systems (see Part IV, 1.3).

written scheme of investigation (WSI)

a detailed scheme for the archaeological
evaluation and/or recording of a development
site, approved by the Local Authority. In the
context of these guidelines. A WSI is equivalent
to a specification or project design.

sub-grid a square or rectangular block of

survey data.Typically an area to be surveyed
will be divided up into a mosaic of contiguous
squares, each of which will be methodically
covered in turn.When transferred to a computer
the data-set from each square is initially stored
separately and is termed a sub-grid.
survey grid the network of control points
used to locate the geophysical survey
measurements relative to base mapping
and/or absolute position on the Earths
surface (see Part IV, 1.1)
time- (depth-) slices visual representations
extracted from a volume GPR data-set
showing successive plan views of the variation
of reflector energy from the surface to the
deepest recorded response (see Part IV, 1.4.3)
thermoremanent magnetisation a persistent,
permanent, magnetisation acquired by certain
magnetic minerals after they have been heated
above a threshold temperature and then
cooled in an ambient magnetic field (such
as the Earths)
tomography In the context of geophysics, this
term usually describes the process of imaging
the subsurface as a sequence of two-dimensional
slices. Multiple parallel slices can be measured
and combined using a computer to image a
volume of the subsurface in three dimensions.
An alternative name for electrical sections
(Part IV, 1.3.4) is earth resistance tomography
(ERT) but GPR can also be considered a
tomographic technique.
travel time the time required for an incident
GPR pulse to pass from the surface to a buried
reflector, usually measured in nanoseconds (ns).
If the velocity of the radar wave in the soil is
known, then the distance to the reflector can
be calculated (see Part IV, 1.4.3).
twin electrode (twin probe) an arrangement of
electrodes for making earth resistance
measurements that is particularly suited to
archaeological geophysics.The two current
electrodes are each paired with one of the
two potential electrodes, one pair is set into
the ground at a fixed reference position while
the second is carried on a mobile frame and
inserted into the ground wherever a
measurement is to be made (see Part IV, 1.3).

55

Appendix I Related standards,

codes and guidance
The only code of practice devoted specifically
to geophysical survey in archaeology is one
that deals with geophysical data: Armin Schmidt
2002 Geophysical Data in Archaeology: A Guide
to Good Practice.York: Archaeology Data Service
Readers should also familiarise themselves
with: Chris Gaffney, John Gater and Susan
Ovenden 2002 The Use of Geophysical Techniques
in Archaeological Evaluations. Reading: IFA Techn
Pap 6
Codes of practice that otherwise have a
bearing on geophysical survey, albeit marginally
on its archaeological applications, include:
British Standards Institution Code of Practice for
Site Investigations, BS 5930:
http://www.standardsdirect.org/standards/stand
ards4/StandardsCatalogue24_view_4488.html
(1999). A guide to this Code has also been
published (2002): http://www.standardsdirect.org/
standards/standards4/StandardsCatalogue24_
view_26173.html
Darracott, B W and McCann, D M 1986
Planning Engineering Geophysical Surveys. London:
Geological Society, Engineering Geology Special
Publication Number 2.
Engineering Geophysics: Report by the Geological
Society Engineering Group Working Party 1988.
The Quarterly Journal of Engineering Geology 21
(3). London:The Geological Society
Building Research Establishment (BRE) 2002
Optimising Ground Investigation. Driscoll: BRE.
This digest informs building and construction
professionals who commission ground
investigations, especially clients and their
advisors who do not themselves have
geotechnical qualifications and experience.
It aims to raise awareness of the importance
of ground investigation for routine projects
and provides a summary of best practice.
http://www.brebookshop.com/details.jsp?id=
140242
The American Society for Testing and Materials
(ASTM: http://www.astm.org/) has produced:
ASTM D6429-99 Standard Guide for Selecting
Surface Geophysical Methods (which covers
forensic and archaeological applications).
ASTM D6429-99 Standard Guide for using the
Surface Ground Penetrating radar method for
Subsurface Investigation

56

Users of GPR (Part IV, 1.4) should be aware

of and abide by the Code of Ethics and the
Code of Conduct developed and used by the
membership of EuroGPR.The latter is a trade
association, open to all GPR practitioners, the
goals of which are to promote good practice
in the use of GPR for both commercial and
academic use throughout Europe, to act as a
forum for discussion on topical issues, and to
act as a voice for the industry in lobbying
European legislative authorities.
If not already members of the IFA, geophysical
surveyors should at least abide by its code
of conduct (IFA 1986, 1988, revised 2002:
http://www.archaeologists.net/).
Contractual arrangements should follow the
ICE Conditions of Contract for Archaeological
Investigation (2004,Thomas Telford Ltd
[www.thomastelford.com]).These are the
product of a joint working group of the
Institution of Civil Engineers (ICE), the
Association of Consulting Engineers (ACE),
the Civil Engineering Contractors Association
(CECA) and the Institute of Field Archaeologists
(IFA), and regulate the business relationship
between the Employer and the specialist
Archaeological Contractor.
Familiarity with the following codes and
manuals will also be advantageous:
Archaeological Investigations Code of Practice
for Mineral Operators 1991. Confederation
of British Industry
The British Archaeologists and Developers Liaison
Group Code of Practice 1991
ACAO 1993 Model Briefs and Specifications for
Archaeological Assessments and Field Evaluations
Dept of Transport 1993 Design Manual for
Roads and Bridges, Volume 11 Section 3 Part 2:
Cultural Heritage
IFA 1993 Standard and Guidance for Field
Archaeological Evaluations. Standards in British
Archaeology Working Party

Appendix II Contacts
Advice on geophysical survey can be obtained
from the following sources:

1 English Heritage Regional Science

Advisors:
North West (Cheshire, Manchester, former

Merseyside, Lancashire and Cumbria)

Sue Stallibrass

Department of Archaeology, Hartley Building,

University of Liverpool, Liverpool L69 3GS

telephone: 0151 794 5046

e-mail: [email protected]

North East (Northumberland, Durham,

Tyne and Wear, Hadrians Wall)

Jacqui Huntley

Department of Archaeology, University

of Durham, South Road, Durham DH1 3LE

telephone/fax: 0191 334 1137

e-mail: [email protected]

Yorkshire and Humber (Yorkshire and

former Humberside)

Andy Hammon

EH York Office, 37 Tanner Row,

York YO1 6WP

telephone: 01904 601983

e-mail: [email protected]

West Midlands (Herefordshire,

Worcestershire, Shropshire, Staffordshire,

former west Midlands and Warwickshire)

Lisa Moffett

EH Birmingham Office, 112 Colmore Row,

Birmingham B3 3AG

telephone: 0121 625 6875

e-mail: [email protected]

South West (Cornwall, Isles of Scilly, Devon,

Dorset, Somerset,Wiltshire and Gloucestershire)
Vanessa Straker
EH Bristol Office, 29 Queen Street,
Bristol BS1 4ND
telephone: 0117 975 0689
e-mail: [email protected]
South East (Kent, Surrey, Sussex, Berkshire,
Buckinghamshire, Oxfordshire, Hampshire
and Isle of Wight)
Dominique de Moulins
Institute of Archaeology, University College
London, 3134 Gordon Square, London
WC1H 0PY
telephone: 0207 679 1539
e-mail: [email protected]
London
Currently vacant (April 2008)
Up to date information is available from the
following websites:
1. HELM/Managing and Protecting/

Delivering advice/Regional science advisers

(http://www.helm.org.uk/server/show/category.

11227)

2. EH/Research and Conservation/Archaeology

and Buildings/Scientific techniques/RSA home

(http://www.english-heritage.org.uk/server/

show/nav.1273)

2 English Heritage Geophysics Team:

Paul Linford
Fort Cumberland, Eastney, Portsmouth PO4 9LD
telephone: 02392 856749
e-mail: [email protected]
www: http://www.english-heritage.org.uk/
server/show/nav.18391

East Midlands (Derbyshire, Leicestershire,

Rutland, Lincolnshire, Nottinghamshire, and

Northamptonshire)

Jim Williams

EH Northampton Office, 44 Derngate,

Northampton NN1 1UH

telephone: 01604 735451

e-mail: [email protected]

East of England (Bedfordshire, Cambridgeshire,

Essex, Hertfordshire, Norfolk and Suffolk)

Jen Heathcote

EH Cambridge Office, Brooklands House,

24 Brooklands Avenue, Cambridge CB2 2BU

telephone: 01223 582759

e-mail: [email protected]

57

Appendix III Useful websites

www.terranova.ltd.uk

Miscellaneous

A comprehensive source for relevant links

is at: http://www.brad.ac.uk/acad/archsci/
subject/archpros.htm

www.testconsult.co.uk

English Heritage Geophysical Survey Database:

http://sdb.eng-h.gov.uk

The links provided below are intended as an

informal listing of current (2008) websites that
may of interest to readers. English Heritage
does not accept any responsibility for the
accuracy of these websites or their contents,
and inclusion on the list does not mean that
English Heritage has given any approval or
accreditation to the company or individual
concerned. As time goes on, this listing will be
updated on the web version of this document
(see www.english-heritage.org.uk)

N on-Destructive Testing

www.wessexarch.co.uk

www.gbg.co.uk
www.aperio.co.uk

Manufacturers
Magnetometers
www.geometrics.com
www.bartington.com
www.geoscan-research.co.uk

C ontractors/consultants

www.scintrexltd.com

www.archaeologicalgeophysics.co.uk

www.gemsys.ca/

http://apss.soton.ac.uk

ADS Guidance. Geophysical Data in Archaeology:

A Guide to Good Practice: http://ads.ahds.ac.uk/
project/goodguides/geophys/
Journal: Archaeological Prospection:
http://www3.interscience.wiley.com/cgi
bin/jhome/15126
International Society for Archaeological
Prospection: http://archprospection.org/
Environmental and Industrial Geophysics
Group (EIGG): http://www.geolsoc.org.uk/
template.cfm?name=geogroup12
EIGG Test Site: http://www.le.ac.uk/gl/iah/
research/EIGG/eigghp.html

www.arch.wyjs.org.uk

EM

www.archaeophysica.co.uk

www.geonics.com

www.arrowgeophysics.co.uk

www.iris-instruments.com

N ERC Geophysical Equipment Facility:

http://gef.nerc.ac.uk/

GPR

GPR-SLICE software and GPR case studies:

www.gpr-survey.com

www.dur.ac.uk/archaeological.services/
geophysical_survey/
www.shef.ac.uk/arcus/
www.arch-ant.bham.ac.uk/bufau/
www.cambrian-archaeology.co.uk
www.earthsound.ie

www.sensoft.ca/
www.utslelectronics.co.uk
www.malags.com
www.era.co.uk

www.geophysics.co.uk
www.le.ac.uk/ulas/services/geophysical.html
www.metsurveys.com
www.northantsarchaeology.co.uk
http://dspace.dial.pipex.com/town/terrace/ld36/

Earth resistance
www.cix.co.uk/~archaeology/cia/resistivity/
resist.htm

www.geophysical.biz

www.trsystem.demon.co.uk/html/archaeology
_and_other_products.html

www.targetgeophysics.ie

www.geoscan-research.co.uk

www.archaeological-surveys.co.uk
www.apac.ltd.uk

Rentals/supplies

www.arcauk.com/geophys.html

www.georentals.co.uk

www.archaeologists.tv

www.allied-associates.co.uk

www.archaeological-services.co.uk
www.contextone.co.uk/geophysics.htm

Software

www.dvasltd.com

Geoplot (Geoscan Research) www.geoscan

research.co.uk/page9.html

www.nparchaeology.co.uk
www.gsbprospection.com

ArcheoSurveyor (DW Consulting)

www.dwconsulting.nl/archeosurveyor.htm

www.geofizz,net

Geosoft www.geosoft.com

www.geologyuk.com
www.sitescan-uk.com

Snuffler (basic but free) www.homeusers.

prestel.co.uk/aspen/sussex/snuffler.html

www.souterrain.biz

Surfer (surface plotting) www.goldensoftware.com

www.stratascan.co.uk
www.terradat.co.uk
58

N orth American Database of Archaeological

Geophysics (with extensive links):
http://www.cast.uark.edu/nadag/

Butser (test site available?):

http://www.butser.org.uk/

GPR case study (Petra):

http://e-tiquity.saa.org/~etiquity/title1.html
Archaeological Investigations Project:
http://csweb.bournemouth.ac.uk/aip/aipintro.htm
European GPR Association:
http://www.eurogpr.org/

Appendix IV List of consultees

Organisations
Association of Local Government
Archaeological Officers (ALGAO)
Council for British Archaeology (CBA)
English Heritage (Regional Science Advisors;
EH Standards Group)
Institute of Field Archaeologists (IFA)
Oxford Archaeology
Wessex Archaeology

Individuals
Mr T Archer (Arrow Geophysics)

Mr P Barker (Stratascan)

Mr G Bartington (Bartington Instruments)

Mr A Bartlett (Bartlett-Clark Consultancy)

Mr N Bell (Allied Associates Geophysical Ltd)

Mr A Boucher (Hereford City and County

Archaeological Trust Ltd)

Mr A Butler (Northamptonshire Archaeology)

Dr M Canti (English Heritage)

Mr P Cheetham (University of Bournemouth)

Dr K Hamilton (Norfolk Landscape Archaeology)

Mr D Hale (University of Durham)

Dr C Gaffney (University of Bradford)

Mr J Gale (University of Bournemouth)

Dr J Gater (Geophysical Surveys of Bradford)

Mr D Gurney (Norfolk Landscape Archaeology)

Mr A Johnson (Oxford Archaeotechnics)

Dr R Jones (University of Glasgow)

Dr J Last (English Heritage)

Dr J Leckebusch (Terra International)

Mr C Leech (Geomatrix Earth Science)

Mr J Lyall (Landscape Research Centre)

Mr A Oswald (English Heritage)

Mr M Papworth (National Trust)

Dr J Reynolds (Reynolds Geosciences)

Mrs A Roseveare (ArchaeoPhysica)

Mr M Roseveare (ArchaeoPhysica)

Bronwen Russell (University of Bournemouth)

Mr D Sabin (Archaeological Surveys)

Dr A Schmidt (University of Bradford)

Ms E Utsi (Utsi Electronics)

Dr R Walker (Geoscan Research)

Mr A Webb (Archaeological Services, WYAS)

59

Contributors
This revision has been prepared by
Andrew David, Neil Linford and Paul Linford,
with assistance from Louise Martin and
Andy Payne, and was brought to publication
by David M Jones.The material in Part IV, 1.7
is largely reproduced from Linford (2006) with
permission of Institute of Physics Publishing.
Acknowledgements
We would like to thank the many colleagues
who have shared their experience with us
over the years. We are particularly indebted
to those with whom we have specifically
consulted on this revision and many of whom
have troubled to supply constructive advice
and commentaries (a listing of all those consulted
in 2007 is included in Appendix IV). We are
grateful to everyone for his/her patience.
English Heritage is the Governments statutory
advisor on the historic environment. English
Heritage provides expert advice to the
Government about matters relating to the
historic environment and its conservation.
For further information and copies of this
publication, quoting the Product Code,
please contact:
English Heritage
Customer Services Department
PO Box 569
Swindon SN2 2YP
telephone: 0970 333 1181
e-mail: [email protected]

Published April 2008

Edited and brought to press by
David M Jones, English Heritage Publishing
Designed by Rowena Bayliss for Creative Services
Printed by Wyndehams
Minimum of 75% recovered fibre, the
remainder being from sustainable sources.
Product Code 51430

Back cover caption: (above) Greyscale plots of an earth resistance survey over the Roman amphitheatre at Richborough in
Kent. The upper plot shows the unprocessed data whilst the lower depicts the same data after high pass filtering to emphasise
internal details of the amphitheatres structure.
Front cover caption: Greyscale plot of a portion of a caesium magnetometer survey over an Iron Age settlement at
Flint Farm in Hampshire (left) compared with a photograph of the same area during excavation (right).