013 Esmr en - 0

Environmental and Social Monitoring Report
Project Number: 43281-013

December 2022
Nepal: Tanahu Hydropower Project – Panel of

Experts’ Report
Prepared by Jonathan Hinks, Susan Novak, and Jose M. Cabral for the Government of Nepal
and the Asian Development Bank.
This environmental and social monitoring report is a document of the borrower. The views
expressed herein do not necessarily represent those of ADB's Board of Directors,
Management, or staff, and may be preliminary in nature.
In preparing any country program or strategy, financing any project, or by making any
designation of or reference to a particular territory or geographic area in this document, the
Asian Development Bank does not intend to make any judgments as to the legal or other
status of any territory or area.
TANAHU HYDROPOWER LIMITED
REPUBLIC OF NEPAL
Tanahu Hydropower Project
Third Report of Panel of Experts
December 2022
Jonathan Hinks Susan Novak Jose M. Cabral

Dam Safety Expert and Social Development Expert Environmental Expert
Panel Chair
1.
1
TABLE OF CONTENTS1
SECTION A - SUMMARY OF FINDINGS AND RECOMMENDATIONS ................... 6
1.1 Overview of Project .................................................................................................................... 6
1.2 Engineering Summary................................................................................................................ 7
1.3 Environmental Summary ............................................................................................................ 8
1.4 Summary of Social Issues ....................................................................................................... 11
SECTION B – DAM SAFETY REPORT................................................................... 15

2.1 General .................................................................................................................................... 15
2.2 Geology .................................................................................................................................... 15
2.3 Seismicity ................................................................................................................................. 15
2.4 Landslides ................................................................................................................................ 16
2.5 Hydrology ................................................................................................................................. 18
2.6 Sedimentation .......................................................................................................................... 18
2.7 Diversion Tunnels and Cofferdam ........................................................................................... 21
2.8 Dam .......................................................................................................................................... 21
2.9 Other Issues ............................................................................................................................. 26
2.10 Spillway .................................................................................................................................... 29
2.11 Power Intake ............................................................................................................................ 30
2.12 Headrace Tunnel...................................................................................................................... 30
2.13 Low Level Outlets..................................................................................................................... 30
2.14 Powerhouse ............................................................................................................................. 31
2.15 Instrumentation ........................................................................................................................ 32
2.16 Engineering Conclusions and Recommendations ................................................................... 32
2.17 Engineering References........................................................................................................... 33
SECTION C - ENVIRONMENTAL SAFEGUARDS REPORT ................................. 35

3.1 Introduction .............................................................................................................................. 35
3.2 THL Environmental, Health and Safety (EHS) Management Capacity ................................... 35
3.3 EHS Compliance Monitoring .................................................................................................... 43
3.4 Status of Preparation of Project-Wide Environmental Management Plans, and
Recommended Mitigation Approach to Project Impacts on Fish Species and Biodiversity .... 51
SECTION D – SOCIAL ISSUES .............................................................................. 53

4.1 Summary of key findings and recommendations ..................................................................... 53
4.2 Affected households that are cultivating acquired land ........................................................... 53
4.3 Non-titled households .............................................................................................................. 53
4.4 Community assets.................................................................................................................... 54
4.5 Resettlement along transmission line ...................................................................................... 54
4.6 Grievance redress mechanism ................................................................................................ 55
4.7 Project employment ................................................................................................................. 56
4.8 4.8 Livelihood Restoration Plan and Community Development Plan ..................................... 57
4.9 Emergency Preparedness Plan ............................................................................................... 57
4.10 Methodology and scope of report (see also Section 1.4 ) ....................................................... 58
4.11 Households affected by hydropower project ............................................................................ 59
4.12 Buffer zone study ..................................................................................................................... 59
4.13 POE comment .......................................................................................................................... 59
4.14 AHs continuing to cultivate acquired land ................................................................................ 60
4.15 POE comments and recommendations ................................................................................... 60
4.16 Hs without land title .................................................................................................................. 61
4.17 POE comment and recommendations ..................................................................................... 63
4.18 Community assets.................................................................................................................... 63
4.19 Transmission line - Resettlement entitlements and compensation ......................................... 64
4.20 Status of entitlements and compensation along TL ................................................................. 65
2
4.22 Grievance redress mechanism ................................................................................................ 66
4.23 Objectives and principles of an effective GRM ........................................................................ 66
4.24 Analysis of grievances ............................................................................................................. 67
4.25 POE Comments ....................................................................................................................... 69
4.26 Overview of THP Grievance Redress Mechanism .................................................................. 70
4.27 POE Comments ....................................................................................................................... 72
4.28 POE comments ........................................................................................................................ 73
4.29 POE comments ........................................................................................................................ 74
4.30 Evaluation of THP Grievance Redress Mechanism and recommendations ............................ 75
4.31 Scope and structure of the THP GRM ..................................................................................... 75
4.32 Procedures for lodging, registering, tracking, and reporting grievances ................................. 76
4.33 Grievance redress process ...................................................................................................... 77
4.34 Project employment ................................................................................................................. 78
4.36 Livelihood Restoration Plan ..................................................................................................... 80
4.37 POE comment .......................................................................................................................... 82
4.38 Community Development Plan ................................................................................................ 82
4.39 Community infrastructure ......................................................................................................... 83
4.40 Skills development training ...................................................................................................... 83
4.41 POE comments and recommendation ..................................................................................... 83
4.42 Education ................................................................................................................................. 84
4.44 Health ....................................................................................................................................... 85
4.45 Water and Sanitation................................................................................................................ 86
4.46 POE comments ........................................................................................................................ 86
4.47 Gender Equality and Social Inclusion Action Plan ( GESIAP ) ................................................ 86
4.48 Orientation and awareness raising .......................................................................................... 87
4.49 POE comments ........................................................................................................................ 87
4.50 Social media campaign ............................................................................................................ 87
4.51 POE comments ........................................................................................................................ 88
4.52 Health camp ............................................................................................................................. 88
4.53 POE comments ........................................................................................................................ 88
4.54 Training for self-help groups .................................................................................................... 88
4.55 POE comments ........................................................................................................................ 88
4.56 Social mobilizer training ........................................................................................................... 89
4.57 Social policy on GESI and Gender Based Violence – POE Comment .................................... 89
4.58 Economic empowerment.......................................................................................................... 90
4.59 POE comment .......................................................................................................................... 90
4.60 General comments about GESI ............................................................................................... 90
4.61 2013 GESIAP ........................................................................................................................... 91
4.62 “Other” CDP funding ................................................................................................................ 91
4.63 Contingency provisions ............................................................................................................ 91
4.64 Dam Safety .............................................................................................................................. 91
4.65 Rural electrification scheme ..................................................................................................... 92
4.66 References ............................................................................................................................... 93
Acknowledgements................................................................................................ 95
APPENDIX 1: LINK TO PAPER ABOUT CHAMOLI MUDFLOW ........................... 96
APPENDIX 2: EXTRACT FROM “EARTHQUAKE ENGINEERING FOR
DAMS AND RESERVOIRS” ......................................................................... 97
APPENDIX 3: 2002 PAPER ON qra FOR HYDRAULIC GATES .......................... 130
3
ABBREVIATIONS
ADB
Asian Development Bank
AH Affected Household
AIH Affected Indigenous Household
AP Affected Person
CDC Compensation Determination Committee
CDP Community Development Plan
CDS Community Development Strategy
DAO District Administration Office
DCR Demand-capacity Ratio
DDC District Development Committee

DDR Due Diligence Report
DLRO District Land Revenue Office
DMS Detailed Measurement Survey
EA Executing Agency
EIB European Investment Bank
EL Elevation above sea level
EMP Environmental Management Plan
ESMU Environmental and Social Management Unit
ESMSP Environmental and Social Management Service Provider
FHH Female-headed household
FSL Full Supply Level
GBV Gender Based Violence
GESI Gender Equality and Social Inclusion
GESIAP Gender Equality and Social Inclusion Action Plan
GLOF Glacial Lake Outflow Flood
GON Government of Nepal
GRC Grievance Redress Committee
GRM Grievance Redress Mechanism
HH Household
IMA Independent Monitoring Agency
IP Indigenous People
JICA Japanese International Cooperation Agency
LB Left bank
LCF Local Consultative Forum
M&E Mechanical and Electrical
MCE Maximum Credible Earthquake
4
Main Central Thrust
MCT
Nepal Electricity Authority
NEA
NFTP Non-timber forest product(s)

NGO Non-governmental organization
OBE Operating Basis Earthquake
PAM Project Administration Memorandum
PIB Public Information Booklet
PIC Public Information Center
PMF Probable Maximum Flood
POE Panel of Experts
PPTA Project Preparation Technical Assistance
PSC Project Supervision PSCs
RB Right bank
RC Replacement cost
RCC Roller Compacted Concrete
RES Rural Empowerment Society
RIPP Resettlement and Indigenous Peoples Plan
ROW Right-of-way
RTS Reservoir Triggered Seismicity
SDE Social Development Expert
SEE Safety Evaluation Earthquake
SIA Social impact assessment
SPS Safeguard Policy Statement
THL Tanahu Hydropower Ltd.
THP Tanahu Hydropower Project
UpRIPP Updated Resettlement and Indigenous Peoples Plan
USD United States Dollar
5
SECTION A - SUMMARY OF FINDINGS AND RECOMMENDATIONS
1. SECTION A - SUMMARY OF FINDINGS AND RECOMMENDATIONS
1.1 Overview of Project
Tanahu Hydropower Limited (THL) has received $ 505 million funding for the project
from ADB, JICA, EIB and GoN. The construction camp and access roads are
complete and it is hoped to complete the diversion tunnels by the end of 2022.
The Panel of Experts (POE) presently comprises the following:
 Jonathan Hinks, Dam Safety Expert
 Susan Novak, Social Development Expert
 José Cabral, Environmental Expert

In addition, other experts can be added to the panel on either a temporary or long-
term basis. Jonathan Hinks presently provides the only engineering representation
and it might be worth strengthening the panel temporarily by the addition of a
geologist/landslide expert and, in view of the large number of gates, an M&E specialist.
It is understood that suitable experts with the necessary skills are available locally.
Tanahu Hydropower Ltd. (THL) was established in 2012 as the Executing Agency (EA)
for the development of the Tanahu Hydropower Project (Project or THP), located on
the Seti River about 40 km south-east of Pokhara and downstream of the town of
Damauli, the capital of Tanahun District. It is a storage type hydropower project with
an installed capacity of 140 MW. The main components are a 140m high concrete
gravity dam that will create a storage reservoir some 25 km long with a surface area
of 7.26 km2 and a gross capacity of 295.1 Mm3. An underground powerhouse is
located 6 km downstream on the right flank of the Seti valley exploiting a sharp loop
of the river.
Permanent and temporary roads provide access to the Project area. Other facilities
include a permanent THL camp near Damauli and temporary facilities for contractor's
camps, yard, office areas, etc. The THP also includes the construction of a 37-km
transmission line and a rural electrification program that will provide electricity to
communities within the Project area1.
The Project is co-funded by Asian Development Bank (ADB), Japan International
Cooperation Agency (JICA) and European Investment Bank (EIB). Loan agreements
totaling USD 419 million were signed in 2013 for the main Project components, i.e., i)
headworks, ii) waterway, powerhouse and related equipment, iii) the transmission line
and iv) the rural electrification program. Funds from the Government of Nepal (GoN)
and Nepal Electricity Authority (NEA) are used for the pre-construction infrastructure.
Construction of the 2 No. 11.4 m dia diversion tunnels is expected to be completed by
the end of 2022. The capacity of the diversion tunnels has recently been increased
from 2,305 m3 /s to 2,870 m3 /s. The right tunnel is concrete lined with an invert about
10 m lower than that of the left tunnel which is to be left unlined.
1 The transmission line and rural electrification program are outside the mandate of the Panel of Experts (POE).
6
1.2 Engineering Summary
Much progress with the engineering design has been made since the first meeting of
the Panel of Experts (POE) was held in Kathmandu from 17 to 21 December, 2018
with a site visit during that period.
A meeting involving THL, Tractebel, and POE was held on site on 14 – 15 January,
2020 in an attempt to resolve outstanding issues relating to seismic inputs to be
assumed for the extreme earthquake.
Since then the contractor has carried out extensive calculations concerning the
seismic behaviour of the dam and, as a result, POE are satisfied that the dam should
be able to withstand a seismic event with a Peak Ground Acceleration ( PGA ) of up
to 0.7 g. This event is estimated to have a return period of 8,000 years which POE
consider acceptable and generally compliant with the recommendations in ICOLD
Bulletin 148. The risk of dam failure due to seismic cracking is, therefore, estimated
as 1.25 % in a period of 100 years.
The factors of safety against sliding and overturning have also been shown to be more
than 1.3 in all blocks for a PGA of 0.75 g ( with a return period of 10,000 years ).
Generous designs and procedures are proposed for annual reservoir lowering and
sediment flushing and this is expected to give the reservoir a life of about 135 years,
after which the scheme can be operated as run-of-river, albeit with reduced power
production and turbine life.
The scope for stabilising the valley sideslopes upstream of the Tanahu dam appears
limited but the problem is potentially serious as a result of possible heavy rainfall,
earthquakes and rapid drawdown. The perceived increased risk of Glacial Lake
Outflow Floods ( GLOFs ) due to climate change also needs to be considered. As far
as the project is concerned the most serious risk seems to be that a large landslide
might block and dam the River Seti, or one of its tributaries, with debris which could
then get washed away releasing a catastrophic flood downstream. Satellite or other
monitoring of potential landslip areas should, therefore, be considered.
Spillway gates have low reliability at many dams and many have failed to operate
when required. The selected gate suppliers are all said to be reputable companies
but attention needs to be paid to standby generators ( which should be regularly
operated under load ), operator training and regular maintenance.
In the last few months the contractor has carried out extensive calculations concerning
the seismic behaviour of the dam and, as a result, POE are satisfied that the dam
should be able to withstand a seismic event with a Peak Ground Acceleration ( PGA )
of up to 0.7 g. This event is estimated to have a return period of 8,000 years which
POE consider acceptable and generally compliant with the recommendations in
ICOLD Bulletin 148 ( 2010 ). The risk of dam failure due to seismic cracking is,
therefore, estimated as 1.25 % in a period of 100 years.
The factors of safety against sliding and overturning have also been shown to be more
than 1.3 in all blocks for a PGA of 0.75 g ( with a return period of 10,000 years ).
7
Outflow Floods ( GLOFs ) due to climate change also needs to be considered. As far
as the project is concerned the most serious risk seems to be that a large landslide
might block and dam a river with debris which could then get washed away releasing
a catastrophic flood downstream. Satellite or other monitoring of potential landslip
areas should, therefore, be considered.
Spillway gates have low reliability at many dams and many have failed to operate
when required. The selected gate suppliers are all said to be reputable companies
but attention needs to be paid to standby generators ( which should be regularly
operated under load ), operator training and regular maintenance.
1.3 Environmental Summary
During the POE mission, the work in relation to environmental safeguards issues
focused on the examination of the Environmental, Health and Safety (EHS)
management capacity of THL, the level of EHS compliance by the Project, and the
status of preparation of four Project-wide Environmental Management Plans, with
particular emphasis on the recommended mitigation approach to the impacts of the
Project on fish species and biodiversity to be developed as part of the Fish
Conservation Management Plan. This subsection summarizes the main findings and
recommendations in the three areas indicated.
1.3.1 THL Environmental, Health and Safety (EHS) Management Capacity
 The vacancy for several months of the position of Director of the ESMU is
concerning, since this means that the ESMU lacks a professional at a
sufficiently high hierarchical level within the organization to interact with other
THL management units, the PSC and Contractors, to liaise with environmental
and other authorities, and to provide strategic direction and guidance to the
work of the Unit, all of which may undermine the overall effectiveness of the
ESMU, as well as the relevance and priority that EHS and social safeguards
issues must have as an integral component of the Project.
 The hiring of a Civil Engineer represents an important technical reinforcement

of the ESMU, since this strengthens its capacity to assess the operation and
effectiveness of physical mitigation measures implemented by Contractors, and
to suggest technical improvements to underperforming or failing physical
mitigation measures.
 Taking into consideration the shortcoming regarding the vacancy of the ESMU
Director noted above, and some limitations in field experience in environmental
oversight as well as in ESH enforcement effectiveness elaborated in
Subsection 3.3, both the ESMU and ESMSP largely have the capacity to
undertake their compliance and effects monitoring responsibilities.
 Apparently only staffs of the ESMSP and the corresponding Contractor for the
package inspected usually take part in monthly EHS inspections, and only the
8
staffs of ESMU, ESMSP and the respective Contractor of the package
inspected usually attend monthly EHS meetings. The fact that the PSC
apparently does not take part usually in the monthly EHS inspections or the
associated meetings means that there are missed opportunities for the PSC to
gain firsthand knowledge of noncompliance issues and their severity and
duration, and ways to address them. Given that the PSC acts on behalf of THL
in the management and supervision of the work of contractors, including quality
control and remediation of technical deficiencies and contractual non-
compliances, its participation in monthly EHS inspections and meetings is
crucial.
 An additional way in which the participation of the PSC in monthly EHS

inspections and meetings would contribute to enhancing enforcement of
mitigation and management measures is the sharing by the PSC with the
ESMU and the ESMSP of the approved method statements for the
implementation of pertinent works and measures that will be subject to
inspections (e.g., slope stabilization measures, location and technical
specifications for spoil storage and disposal areas, technical specifications for
the exploitation of quarry and borrow areas, etc.). This would allow the ESMU
and the ESMSP to prepare better for the issues to look for and focus on during
site inspections.
 The persistency, level of risk and immediate urgency to remediate the non-
compliances described Subsection 3.3 regarding community safety and
hazardous materials handling in the Package 1 Contract indicate that current
EHS enforcement mechanisms are not working with the necessary efficiency.
In this sense, THL management should consider requiring the PSC to employ
more stringent enforcement tools included in works contracts, such as the
withholding of payments, the imposition of fines and ultimately the issuance of
stop work orders, for unrectified continuous non-compliances that can lead to
imminent significant negative impacts on the biophysical environment and/or
the safety and health of affected communities and Project workers. Although it
is clear that the main goal of all of the parties involved in the Project is to ensure
that it is implemented on time following best engineering and EHS and social
practices, the more stringent contractual enforcement measures may be
necessary to be implemented sparingly in the noted instances of utmost non-
compliance.
 The findings during the POE mission on community safety and hazardous
materials management in the Package 1 Contract also point to the need for the
institutional strengthening of the EHS staffs of the ESMU and the ESMSP,
composed mainly of dedicated and capable young professionals for whom, with
the exception of the ESMSP’s Personal Safety cum Risk Management
Specialist, the Project constitutes their first practical experience conducting
EHS inspections and enforcing EHS requirements for a major infrastructure
undertaking. Thus, THL and the ESMSP should consider offering training
opportunities to their EHS staffs in the oversight, mitigation and enforcement of
EHS measures and requirements in large infrastructure projects, such as those
offered virtually by specialized institutions.
9
 A good practice implemented by the ESMSP is the development of the website
“Tanahu Hydropower Project - Survey System” to centralize data in a largely
interactive digital format on relevant biophysical and socio-economic
characteristics of the Project area, including maps, and information on EHS and
social compliance and enforcement.
1.3.2 EHS Compliance Monitoring
 The most serious nonconformities, presenting the highest risks to the safety of
communities and workers, and to environmental quality, were detected in the
Package 1 Contract. In particular, the Contractor does not have a system in
place for the handling, storage, transport and disposal of hazardous materials
and wastes, and a licensed hazardous waste management company has not
been contracted for the final disposal of this type of waste. Hazardous materials
and wastes are not stored separately in an isolated area of the camp in a locked
room. Further, liquid hazardous materials and wastes that may leak or spill are
not stored in general in adequate containers placed on an impervious surface
and enclosed in a bunded area with sufficient capacity to hold the volume of
liquid that may spill. Illustrative of this are: i) the storage in metal shelves within
a storage room of gasoline in plastic water bottles close to containers with
diesel, engine oil, brake fluid, engine coolant, transmission oil and even
chemicals; and ii) the placement of barrels of new motor oil in an open area
without containment walls, mixed with barrels of chemicals such as modified
inorganic aluminum salt solution and open containers of used engine oil. There
are multiple unattended oil spills in different parts of the workshops and there
are no spill cleanup kits on site.
 Regarding community safety, an average of 30 to 40 people per day walk

through the diversion tunnels from the villages of Kundhi, Setang, Lokma and
Bajo to go mainly to the markets in Damauli and back to their communities,
including during hours of high construction activity. This situation occurs daily,
even though the Contractor for Package 1 provides a truck in the mornings and
afternoons to drive villagers from and to their communities. Villagers indicated
that the pickup and drop-off schedule is inconvenient and that therefore
sometimes they do not wait for transportation and just walk. A solution must be
proposed by the Contractor, which may entail increasing the frequency of trips
to two in the mornings from the villages and two in the afternoons to the villages.
The seriousness of the risk to community safety that the situation described
poses cannot be overstated and a solution has to be found as a matter of
urgency. In addition, the characteristics of the platform on the back of the truck
where villagers ride is not ideal, in particular for the elderly and children, and
thus the truck should be appropriately retrofitted to facilitate accessing and
exiting the platform; further, adequate seating and grab handles for comfort and
security should also be provided. This situation relates to the obstruction by the
Project of direct access by the affected communities to other villages and the
main settlements in the area such as Damauli, an impact that has not been
adequately addressed and for which there are no clear plans to remedy.
 The position of Environmental Officer has been vacant since the last week of
August 2022 in Package 1 and therefore the Contractor does not have the
required expertise among its staff to undertake inspections specifically focused
10
on environmental issues. In addition, the main Contractor only has about 5
members of its key and non-key staffs on site, including the Project Manager
and the Safety Manager, with most of the civil works being undertaken by local
subcontractors. This may be an underlying reason for some of the non-
compliances observed.
1.3.3 Status of Preparation of Project-Wide Environmental Management Plans and
Recommended Mitigation Approach to Project Impacts on Fish Species and
Biodiversity
 The ESMSP anticipates that the final versions of the following three Plans will
be submitted before the end of November 2022: Consolidated Environmental
Management Plan, Forest and Watershed Management Plan, and Wildlife
Conservation and Awareness Plan.
 Regarding the Fish Conservation Management Plan, a virtual meeting was held
with the participation of the ADB Senior Environmental Specialist, the POE
Environmental Expert, THL senior management, and the staffs of the ESMU
and the ESMSP, in order to discuss and reach agreement on a
recommendation on the most effective mitigation approach to the barrier effect
created by the dam to the upstream migration of long-ranging species, as well
as the impacts of degradation of fish habitats, and decrease of ichthyofauna
diversity and populations of some species. The meeting analyzed the technical
feasibility, design and implementation complexity, cost and likely effectiveness
of the following four proposed mitigation measures: a fish pass, a fish hatchery,
a catch and haul system, and the protection of a section of the Madi River.
 The discussion concluded that the best mitigation approach would be to

combine the implementation of the Madi River Protection Area with the catch
and haul system, with all meeting participants agreeing on this, including the
senior management of THL. It was further agreed that the ESMU and the
ESMSP would update the Fish Conservation Management Plan with a detailed
analysis of the viability and likely effectiveness of the four measures examined
and the justification of the measures supported. It was indicated that ADB’s
Safeguards Division must give its No-Objection to the mitigation approach
identified before it can be implemented as part of the Fish Conservation
Management Plan.
1.4 Summary of Social Issues
1.4.1 General
The following are key findings and recommendations that have a bearing on
compliance with social safeguards for the THP. Other findings and recommendations
are included in the Social Report ( Section 4 ) that, in the opinion of the POE social
expert, can contribute to enhancing the work to address the needs and priorities of
directly and indirectly affected people. The recommendations are numbered according
to the relevant section of the Social Report.
11
1.4.2 Methodology and Scope of POE Review
The methodology for the POE review of social safeguards issues for the Tanahu
Hydropower Project (THP) has combined document review, meetings with key
personnel from THL and its consultants, and site visits throughout the hydropower
(HP) project area including informal discussions with members of affected households
and other community members.
Prior to arriving in Nepal, the POE social expert conducted an extensive home-office
review of project-related documents that had been provided in the intervening time
since the previous POE mission in December 2018, by THL and/or ADB, as well as
other documents that were accessed from the THL website. The review produced
extensive notes on the status of social safeguard and gender equality and social
inclusion (GESI) aspects of the THP, along with the identification of issues to query
for clarification, updating, etc., during the POE mission.
The POE social expert arrived in Nepal on 4 November 2022. In Kathmandu, meetings
were held with the THL Managing Director, members of the THL Environment and
Social Management Unit (ESMU) and Environmental and Social Management Service
Provider (ESMSP) teams, and the consultant for external monitoring of social
safeguards issues. These meetings were invaluable to, first, meet people who had not
been in place at the time of the previous POE mission; and to clarify some initial
questions about the progress being made in the implementation of the Resettlement
and Indigenous Peoples Plans (RIPPs) for the HP and transmission line (TL)
components of the THP, from the perspective of people involved in the implementation
and external monitoring of the resettlement process.
From 8 November through 16 November, the POE social expert was based at the THL
camp to conduct site visits in the project area. The majority of this time was spent in
site visits to communities to talk with members of project-affected households and
other community members and/or to see some of the community infrastructure
projects that THL has implemented. Throughout this field work, the POE social expert
was supported by staff of THL, ESMU and ESMSP, in particular, ESMSP’s Deputy
Team Leader, Social Expert, and GESI Expert.
The POE social expert has been apprised of the issues that have led two separate
groups of households to appeal directly to the Asian Development Bank (ADB)
Accountability Mechanism and the European Investment Bank (EIB) Accountability
Mechanism. This includes one group of 15 directly affected AHs in Paltyang, as well
a group of 12 Dalit households in Wangtankitar and 5 Magar households in Jalbire
that may be affected by the establishment of a buffer zone around the reservoir. As
negotiations are ongoing through the ADB and EIB processes, the POE social report
does not address the issues directly. However, in the course of field work, the POE
social expert did visit Wangtankitar to discuss other issues and, also, had occasion to
talk with residents of Paltyang and Jalbire, again on separate matters.
The POE social report briefly reviews the planning and status of implementation of key
elements of the social safeguards, along with POE social expert comments,
suggestions and, as relevant, recommendations.
12
1.4.3 Affected households that are cultivating acquired land:
A number of households continue to cultivate agricultural land that has been acquired
by THL, as per THL policy. It is important to assess the continuing vulnerability of these
households as candidates for the Livelihood Restoration Plan (LRP) and to ensure
that they understand clearly and in a timely manner when they will have to stop
cultivating these lands.
1.4.4 Non-titled households
Among households cultivating acquired land, there are a number that did not have title
to the land and, in some cases, were deemed ineligible for an allowance for non-titled
affected households; as above, they should be targeted in vulnerability assessments
for the LRP. Also, other non-titled households that may be affected by the proposed
buffer zone have initiated the process of registering their land; it is recommended that
THL support them in this process but without being directly involved in order to ensure
impartiality and transparency.
1.4.5 Community assets
While public resting places affected by the reservoir have been replaced, most other
affected community assets have not. Of particular concern to local communities are
the loss of cremation sites, access to foot paths, and to activities on both sides of the
river (reservoir). The POE social expert recommends that steps are taken to move
forward to finding solutions to these issues.
1.4.6 Resettlement along transmission line
Some affected households along the transmission have received compensation and
work is ongoing to determine rates for other types of affected assets. However, at the
same time, the civil works contractor is actively working to construct tower foundations
and erect towers. Also, a portion of the transmission line passes through forests but
there has been no determination if Community Forest User Groups (CFUGs) are
affected. Therefore, the POE social expert recommends actions to expedite
resettlement activities.
1.4.7 Grievance redress mechanism
The POE social expert made an extensive analysis of the THP Grievance Redress
Mechanism (GRM) and found a number of issues related to widely accepted principles
and best practice. The following is a summary of recommendations that are further
elaborated in the main Social Report.
1.4.8 Project employment
The THP has met a basic target of employing at least 1,000 people as set out in the
Design and Monitoring Framework. However, the THP has not met targets for
employment of local people, i.e., people living in Tanahun District, or women. The
target that 30% of project employees are women may not be realistic, however, given
the high expectations people have about project-related employment, the POE social
expert recommends action on existing commitments and other initiatives to expand
employment opportunities.
13
1.4.9 Livelihood Restoration Plan and Community Development Plan
The Livelihood Restoration Plan (LRP) and Community Development Plan (CDP) are
in development or recently developed. The POE social expert has reviewed the
research that has gone into the development of the plans as well as the early
formulations of them; and has made suggestions rather than recommendations. The
following recommendations, however, reflect lessons learned from activities that have
been undertaken and/or the information received from informants in local
communities.
1.4.10 Emergency & Preparedness Plan
At present, this plan addresses issues downstream of the dam. The SOE social expert
believes that it is important that it be amended to reflect safety issues associated with
dam operation and annual flushing of the reservoir, for people and communities in
vicinity of the reservoir.
1.4.11 Recommendations
Recommendations in respect of the above issues are given in Section D of this report.
14
SECTION B – DAM SAFETY REPORT
2. SECTION B – DAM SAFETY REPORT
2.1 General
No dam is ever 100 % safe but, despite the risks associated with high seismicity in the
area, POE consider that the risks are As Low As Reasonably Practicable ( ALARP )
and acceptable for the project. Justification for this conclusion is given in the following
sections.
2.2 Geology
The geology at the dam site, power tunnel and underground power station are
generally considered favourable although karstic limestone is present and lugeon
values are high in the abutments. No evidence has come to light to suggest that there
could be an active fault along the river valley at the damsite ( ie one that has moved
in the last 11,000 years ). One was, however, discovered when the foundations were
opened up at the Clyde dam in New Zealand and the possibility of something similar
happening at Tanahu cannot be completely ruled out.
The possibility of landslides, triggered either seismically or by heavy rainfall, into the
reservoir, also needs to be considered (see also under seismicity below); and, also
the geology of the plunge pool and its ability to withstand high velocities previously
estimated as being up to 32 m/s on the surface and 23 m/s at the bed.
The adit driven from near the tailrace to the roof of the power cavern passes through
Dolomite and Slates which appear to be favourable to the proposed tunnelling.
Excavation of the cavern for the power station is now complete with many rockbolts in
evidence
2.3 Seismicity
The Main Himalayan Thrust is the boundary between the two plates along the range
front of the Himalayan Arc. This splits into three parallel faults (Figure 1): the Main
Central Thrust; The Main Boundary Thrust; and the Main Frontal Thrust. The Main
Central Thrust runs along the range front of the High Himalayas. The main Frontal
Thrust is understood to be the main active fault and is located 100 km south of the
range front.
4
Figure 1. North-South cross-section across the Himalayan arc, showing the major
thrust faults; the Main Frontal Thrust (MFT), the Main Boundary Thrust (MBT) and the
Main Central Thrust (MCT), which join at depth into the Main Himalayan Thrust (MHT).
The MFT, which is the main active fault, is `creeping' or steadily slipping aseismically
15
below a depth of about 15-20 km. Shallower than this depth, the fault is `locked' by
friction, and stress builds up in the surrounding rock over decades to hundreds of
years.
Following a site meeting on 15 November 2022 POE have assumed a 10,000 year
event of MW = 8.0 vertically beneath the dam on the MFT/ MHT at a depth of 30 km.
The average of 6 commonly used attenuation formulae yield a PGA of 0.75 g for this
event. There are other ways of estimating the 10,000 year PGA which yield higher
values but POE consider it acceptable to cap the PGA in the 10,000 year event at 0.75
g since no mechanism has been identified that would give a higher value.
POE’s own estimate is that for the inner C17-76 concrete we can assume a 12 month
dynamic tensile strength at lift joints of 2.40 MPa ( see Section 8 below )
Similarly POE estimate a 12 month dynamic tensile strength at lift joints of 2.76 MPa
for the 3.0 m thick downstream outer facing. The upstream outer facing is a slightly
richer mix with a 12 month dynamic tensile strength at lift joints of more than 3.11 MPa.
2.4 Landslides
Because of the high seismicity and steep valley sideslopes the possibility of large
landslides into the reservoir needs to be carefully considered. The classic example of
a disaster of this type is the 260 Mm3 landslide into the Vaiont reservoir in Italy in
October 1963. Damage to the 262 m high arch dam was restricted to the bridge along
the crest, which was washed away, but a huge wave overtopping the dam killed about
2,000 people in Logarone and other villages downstream. There were certainly widely
felt earth tremors but it seems likely that these were the result of the landslide rather
than its cause.
The triggering of landslides in mountainous regions by earthquakes depends also on
the weather conditions. For example, on April 10, 2016 a magnitude 6.6 earthquake
with focal depth of ca. 210 km occurred in Afghanistan which triggered several mass
movements in Kashmir at epicentral distances of approx. 350 km. One of the major
landslides involved 80 buildings. It is obvious that the level of ground shaking was very
low but as this large mass was close to failure, it needed only a small trigger to be
activated. Therefore, the combination of saturated slopes and ground shaking
triggered several slides.
More recently there was the 27 Mm3 Chamoli debris/mudflow in Uttarakhand Province
in Northern India on 7 February 2021 for which details are given in Appendix 1 to this
report. The 13.2 MW Rishiganga hydropower scheme downstream was destroyed,
200 people were killed and the 520 MW Tapovan hydropower scheme was buried in
mud.
More locally 243 people were killed as a result of the Langtang landslide/avalanche,
north of Kathmandu, triggered by the Magnitude 7.8 Ghorkha earthquake of 25 April,
2015. It has long been recognised that large landslides into storage reservoirs in the
Nepal Himalayas could pose a serious threat to the safety of the dams. However,
experience gained to date is limited.
16
Figure 2. Landslides into the Zipingpu reservoir area caused by the May 12, 2008
Wenchuan earthquake in China (photo taken from Zipingpu dam crest ).
The scope for stabilising the valley sideslopes upstream of the Tanahu dam does not
appear great but the problem is potentially serious. The perceived increased risk of
Glacial Lake Outflow Floods ( GLOFs ) due to climate change also needs to be
considered ( see below ). Attention is also drawn to the risk of rapid drawdown failures
of the reservoir sideslopes following the planned summer drawdowns of the reservoir
each year.
Figure 3. Upper catchment of River Seti
17
2.5 Hydrology
Peak unrouted floods into the reservoir are quoted as follows:
Return Period ( Years ) Peak Inflow ( m3/s )

1,000 4,287
2,000 4,636
5,000 5,097
10,000 5,446
PMF 7,377
The catchment area at the dam is understood to be 1,502 km2 which implies a
Francou-Rodier value of 5.31 for the inflow flood of 5,446 m3/s with a return period of
10,000 years and 5.58 for the Probable Maximum Flood (PMF) of 7,377 m3/s. It would
be useful to compare these values with those assumed for other dams, such as
Kulekani and Dudh Koshi, constructed or planned in Nepal.
Glacial Lake Outburst Floods (GLOFs) have been known to greatly exceed the PMF
in some instances but it is understood that there are only six glacial lakes in the
catchment and that none of these are considered critical by PSC. However climate
change is a concern as “South Asian mountain ice [is] dissolving into massive new
lakes, raising the spectre of catastrophic flooding” ( National Geographic Magazine,
2022 ).
2.6 Sedimentation
Average sediment inflow to the reservoir has been estimated as 7.97 Mm3/yr which
compares with a reservoir capacity of 295.1 Mm3. A very simplistic calculation
suggests a design life of 37 years for the reservoir but this ignores flushing of sediment
through the 2 No 5.0 m x 5.0 m flushing outlets. It also assumes the same sediment
inflow each year whereas very large amounts of sediment can enter a reservoir in a
single large flood.
Practice with regard to dealing with sedimentation in reservoirs has changed
significantly in recent years and now tends to follow the philosophy set out in the World
Bank publication “Extending the Life of Reservoirs” (Annandale, Morris and Karki,
2016).
Following this philosophy, with the sediment flushing regime described below, it is
estimated that the design life of the reservoir can be extended to 135 years. After this
the scheme would, presumably, be operated as run-of-river with significantly lower
annual generation and abrasion damage to the turbines.
It should, however, be noted that there will be significant environmental impacts of
large releases of sediment into the river downstream and possible future
consequences for the planned Lower Seti scheme 15 km downstream.
The PSC’s Technical Memo No.16 of 4 April, 2019 stated as follows:
1. Introduction
18
The Panel of Experts (PoE) have conducted the first mission to Nepal from 16
to 22 December 2018 and subsequently to this visit, the “First Report of Panel
of Experts” was submitted as a draft version. During this mission the PoE visited
THL’s premises as well as the project site. The comprehensive report presents
and discusses various issues with sediments being one of which. This present
document is to respond to the PoE’s observations and recommendations as to
the management of sediments.
2. PSC’s Response as to Sediments
Antecedent
A sediment flushing study has been conducted in the preparational phase of

the project in 2012. The results were presented in a 31-page report. The model
covered a reach between the dam and a point 25 km U/S of the dam. Input data
for the study was a total of 68 cross-sections.
 With six different simulation cases investigated in the study, the pattern
finally recommended is the following one:
 The live storage of the reservoir is replenished during the annual flood
season and drawn down during the dry period in the rest of the year. The
reservoir water level will follow the green and red curves as presented in
Figure 3:
 With the start of the hydrological cycle after the Monsoon season (end
August / beginning September) the reservoir’s live storage is fully
replenished and the FSL of 415 m asl is reached.
 Depending on the annual reservoir inflow, the full supply level will be
maintained until around end of November.
 Thereafter, the water level is drawn down to the minimum operating level
(MOL) of 378 m asl, which is typically reached around the middle of June.
 Thereafter, power production is discontinued, and the water lowering outlets

are being opened to further draw down the water level down to about 345
m asl before eventually the flushing operation commences by opening of
the flushing ducts.
 After a flushing period of around 40 days (end July) the live storage is being
replenished to the FSL of 415 m asl (early September). While raising to the
FSL, power production is resumed at a level of 378 m asl (early August).
19
Figure 4: Reservoir operation curve (green and red line) with tentative dates
depending on availability of water (ignore blue line)
Current Process
At the present stage, reservoir operation is envisaged as intended in earlier

studies as described in PSC’s Project Inception Report. This operating method
was developed in the previous project phases and approved by the Client.
During the preparation of the tender document phase, PSC recommended that
THL resume the sediment sampling to establish a basis for a future sediment
study, which was also recommended by PSC. The recommendation is still valid.
PSC would like to emphasize that PoE’s recommendation to conduct a holistic
sediment management plan is fully supported. PSC does have the knowledge,
experience and capacity to conduct such study, however, this is beyond the
scope of the present assignment.
With regard to the choice of outlets for reservoir drawdown PSC present two options:
Use only sediment flushing facilities to discharge reservoir water, even although
a) water level exceeds EL. 345 m
b) Use the water level lowering equipment and close sediment flushing gates,
when water level exceeds EL. 345 m
The above all seems satisfactory although, as noted above, POE are concerned about
the quality of river water downstream and the stability of the reservoir sideslopes in
the rapid drawdowns.
20
2.7 Diversion Tunnels and Cofferdam
The two 11.4 m diversion tunnels are nearing completion. The right tunnel is about 10
m lower than that on the left and will be lined. It is not presently intended to line the
left tunnel.
The construction period flood will be capable of passing the 20 year flood which has
recently been increased from 2,305 m3/s to 2,870 m3/s. This increase was made in
response to new data received recently.
POE understand that river diversion is programmed for end 2022. Completion of the
diversion tunnels will be on the critical path for project completion.
2.8 Dam
2.8.1 Stability
With an upstream slope of 0.1(H):1(V) and a downstream slope of 0.8(H):1(V) the
proposed cross-section appears quite conservative.
As far as sliding and overturning are concerned the Factors of Safety have been shown
to be acceptable for a PGA of 0.8 g. As the PGA in the Safety Evaluation Earthquake
has now been reduced to 0.75 g ( see Section 3 above ) the situation is considered to
be acceptable.
2.8.2 Seismic Cracking
During the mission of POE in November 2022 new information has come to light and
it is considered appropriate to revisit the question of seismic cracking as follows:
POE have not been advised of the cumulative stress exceedances in the SEE so it is
not possible to determine the Demand-capacity Ratio ( DCR ) using the methodology
in US Army Corps of Engineers Engineering Manual EM 1110-2- 6051 ( 2003 ).
However POE have been advised that the fundamental frequency of vibration of the
dam is 2.07 Hertz. The second mode of vibration has a frequency of 2.83 Hertz. For
such low frequencies Herzog ( 1980 ) recommends a DCR of 1.2 while ( Chopra 2020
) recommends a figure of 1.25 “unless evidence is available to justify a larger value”.
POE prefer the later figure both because of the later date of publication and because
Professor Chopra is probably the world’s leading expert in the field.
Seismic cracking is likely to propagate inwards from the faces of the dam towards the
interior. The question, therefore, arises as to whether it might be reasonable to take
account of the higher tensile strength in the 21 MPa downstream facings which are
scaled from the drawings to be 3.0 m thick. POE are not aware of anything in the
literature that would allow a definitive answer to this question but have used the higher
tensile strength in the following approximate calculation.
Assuming a minimum compressive strength of 21 MPa at 91 days ( for outer facing
concrete on the downstream slope) this implies a static tensile strength of 0.32 x ( fc)2/3
= 2.43 MPa. We can scale this up to 1.07 x 2.43 = 2.60 MPa for strength at 365 days
( see Neville, 1963 ). This is static tensile strength so we can scale it up by a further
factor of 1.25 to get a dynamic tensile strength of 3.25 MPa.
21
For “well prepared” lift joints the above dynamic tensile strength would have to be
scaled down by a factor of 0.85 to 2.76 MPa. ( USBR, 2022 ). This figure is only 4 %
less than that assumed by the contractor for his calculations summarised below;
The above calculation can be replicated for the inner concrete ( 17 MPa ) and for the
outer facing concrete on the upstream face ( 25 MPa ) to give the results quoted in
Table 1 below. These compare with a dynamic tensile strength at lift joints of 2.88
MPa used by the Contractor ( with the prior agreement of POE ) to perform the
complex calculations the results of which are summarised in Tables 2 and 3 below.
Table 1 - Dynamic tensile strength at lift joints for concrete at 365 days age ( MPa )
Contractor POE POE POE

fc = 17 MPa 17 MPa 21 MPa 25 MPa
ft = 2.12 MPa 2.12 2.43 2.74
x 1.07 = 2.26 MPa 2.26 2.60 2.93
X 1.5 = 3.39 MPa 2.83 3.25 3.66
X 0.85 = 2.88 MPa 2.40 2.76 3.11
The contractor’s calculations assume a Demand-capacity Ratio of 1.5 as previously

agreed with POE. However it has recently become clear that the fundamental
vibration frequency of the dam will be only 2.07 Hertz and that, as noted above, the
DCR should not, therefore, be assumed to exceed 1.25. This is the figure used in the
POE calculations reported in Table 1 above.
With the upstream facing concrete being Class C 25-38 and the downstream facing
concrete being class C21-76 the POE suggests that a dynamic strength at lift joints
of 2.76 MPa should be assumed throughout the cross-section. This is only 4 % less
than the figure used by the contractor to generate Tables 2 and 3 and it is felt that this
small difference is not significant.
Table 2 Whittier Earthquake
Percentage areas of stress exceedance
Block 0.8 g 0.75 g 0.7 g

1 20 % 17 % 15 %
2 29% 25 % 21 %
3 20% 17 % 14 %
4 13% 11 % 9 %
5 14 % 12 % 10 %
6 24 % 21 % 18 %
7 18 % 15 % 13 %
8 18 % 15 % 13 %
9 18 % 15 % 12 %
10 15 % 13 % 12 %
11 7 % 5% 4 %
Table 3. Manjil Earthquake

Percentage areas of stress exceedance
22
Block 0.8 g 0.75 g 0.7 g
1 14 % 10 % 6%
2 9% 7% 5%
3 14% 12 % 10 %
4 11% 9% 7 %
5 17 % 15 % 13 %
6 24 % 21 % 18 %
7 17 % 14 % 12 %
8 12 % 10 % 8%
9 8% 7% 6%
10 8% 6% 4%
11 6 % 4% 3 %
Note: Stress exceedance areas exceeding 15 % at 0.75 g are highlighted

2.8.3 Discussion
In the above tables stress exceedances in excess of 15 % of the area of the block
cross sections are highlighted only for the central columns which relate to PGA’s of
0.75 g. The percentages in the central columns are the means of the percentages in
the outer columns.
POE are not clear as to whether it is more appropriate to scale the acceleration time-
histories of the Whittier earthquake of 1 October 1987 or of the Manjil earthquake of
21 June 1990 so results for both are given in Tables 2 and 3 above.
POE have assumed a concrete strength appropriate to the outer 3 m on the
downstream slope rather than that of the inner concrete which will be 15 % lower. The
3-D stress plots prepared by the contractor are impressive but somewhat difficult to
interpret. Nevertheless the plot below of peak tensile stresses anticipated in the MCE
at Platanovryssi dam in Greece suggests that the highest tensile stresses are likely to
be concentrated near the faces of the dam.
23
Figure 5. Peak tensile stresses for the MCE at Platanovryssi dam in Greece
Something similar was reported by Chopra who carried out detailed seismic analyses
for Koyna dam in India following the earthquake of 10 December, 1967.
Figure 6. Koyna dam - estimated peak tensile stresses in lb/in2 ( after Chopra )
At the Koyna dam peak acceleration in the Magnitude 6.5 earthquake of 10 December
1967 was 0.38 g with peak tensile stresses of 4.1 MPa ( 600 lb/in2 ) on the upstream
face ( and slightly higher at the heel ) and 6.9 MPa ( 1,000 lb/in2 ) at the change of
slope on the downstream face. These peak stresses would have been 2 or 3 times
the tensile strength of the concrete which was estimated at 2.4 MPa ( Chopra, 2020 ).
Serious cracking took place.
24
Following the earthquake, buttresses were retrofitted on the downstream face, resin
grouting was carried out at the major cracks and the taller non-overflow monoliths were
prestressed down to a level 21 m below the level of the major cracks.
Figure 7. Koyna dam following remedial works after the 1967 earthquake.
2.8.4 Return Period for dam failure due to seismic cracking

Based on the above let us assume dam failure by cracking in PGA of 0.70 g. At this
acceleration 27 % of Block 2 will be overstressed in the Whittier earthquake and 21 %
of Block 6 in the Manjil earthquake. These assumptions seem reasonable and tie in
with the damage sustained by the Sefid Rud dam in Iran where the dam nearly failed
in the June 1990 event with a PGA estimated at 0.714 g
A good deal of engineering judgement has gone into the above calculation but it is
based on very extensive calculations by the contractor. If we assume that it is correct,
that the 10,000 year event has a PGA of 0.75 g and that the dam fails at 0.70 g the
return period for dam failure by shaking can be estimated using the following formula
from Eurocode 8. This yields a return period ( TR ) for dam failure by shaking of 8,000
years:
( TR / 10,000 ) 0.33 x 0.75 g = 0.70 g
The above implies that the probability of dam failure as a result of seismic shaking will
be about 1.25 % for a dam life of 100 years. The local experts consulted by POE
consider that the probability will be considerably less than this but POE prefer to base
their advice on suitably conservative ( and possibly overconservative ) assumptions.
Comfort may also be taken from the fact that no large concrete dam is ever thought to
have failed as a result of seismic shaking ( although the following four are thought to
have come close to doing so.)
 Hsingfengkiang buttress dam in China, 1962
 Koyna gravity dam in India, 1967
25
 Sefid Rud buttress dam in Iran, 1990
 Pacoima arch dam in USA, 1994
Figure 8. Severe cracking at Sefid Rud dam in Iran in 1990 earthquake. The PGA at
the damsite is thought to have been 0.714 g.
2.9 Other Issues

2.9.1 Landslides and Glacial Lake Outflow Floods
Much effort has rightly been put into consideration of seismic risks to the dam.
However there are other major risks that are thought to have received much less
attention and which may possibly pose greater risks to the project and indeed to towns
like Pokhara upstream.
POE recommend a study into the risk of Glacial Lake Outburst Floods and landslides
bearing in mind that peak seismic accelerations to the north may be higher than at the
damsite itself. In Appendix 2 to this report there are reproduced chapters 17 and 18
of “Earthquake Engineering for Dams and Reservoirs” written by Professor John
Reynolds and due to be published in summer 2023.
The upper catchment of the River Seti was inspected by Ultralite on 21 November,
2022. The lower slopes were generally well wooded with no signs of landslips. The
Ultralite was limited to a height of 3,650 m but there are understood to be glaciers
further upstream.
Numerous landslide areas were noted on the flight from Pokhara to Bharatpur on 22
November, 2022 and on the flight from Bharatpur to Kathmandu on 24 November
2022.
26
Figure 9. Upper catchment of River Seti
2.9.2 Spillway Gates

There will be 4 No. ( 12.7 m x 15.0 m high ) spillway gates at the dam and 2 No. low
level outlets measuring 3.8 x 3.8 m ) for reservoir lowering in an emergency. There
will also be 2 No. gates measuring 5.0 m x 5.0 m for flushing sediment.
27
Figure 10. Peak seismic deformations in metres for PGA of 0.8 g at Tanahu dam.
In designing the gates close attention will need to be paid to the risk of jamming in an
earthquake. Careful thought will also need to be given to the provision of suitable
standby generation equipment as there are likely to be power outages in extreme
events
In the longer term attention will need to be paid to maintenance. Very many examples
could be cited of gate failures due to poor maintenance. For example in 2022 ( 22
years after construction ) three of the five gates at the 128 MW Kapichira dam in
Malawi failed to open, probably as a result of corrosion of the steel lifting cables. This
led to the triggering of the fuseplug which, in turn, caused the loss of 30 % of the
country’s generating capacity. Repairs there are costing an estimated USD 60 million.
28
Figure 11. Corroded cables at Kapichira dam in Malawi.
2.10 Spillway
There will be four spillway radial gates each of which will have to release 1,790 m3/s
in the PMF (routed). Three gates (n-1) will be able to release 5,370 m3/s which
approximates to the peak routed flood with a return period of 10,000 years (5,258
m3/s).
The drawings show 2 No. aeration steps on the spillway. These seem to be a very
wise precaution against cavitation and will help to aerate flows into the plunge pool.
A ski jump plunge pool is envisaged downstream of the dam with tailwater levels
determined by an auxiliary concrete dam about 30 m high located some 174 m
downstream of the main dam.
The energy density in the plunge pool will be high with flow velocities of 32 m/s and
23 m/s on the surface and bed respectively. The spillway and plunge pool do not align
well with the river downstream. Further hydraulic modelling is to take place at the
laboratories of Hydrolab in Kathmandu. As noted flows exiting the plunge pool will
have a significant change of direction with the likelihood of erosion – particularly on
the outside of the bend. It is assumed that topographic contours in the plunge pool will
be determined at the end of each test.
Strong back-currents at either side of plunge pools are common and these will need
to be measured on the hydraulic model.
29
Figure 12. Layout of proposed plunge pool and auxiliary downstream concrete dam.
2.11 Power Intake

POE are not clear as to the chosen option for the power intake. Advice on this subject
is provided in EM 1110-2-6051 ( US Army Corps of Engineers, December, 2003 ).
2.12 Headrace Tunnel
The concrete lined headrace tunnel will have a diameter of 7.4 m and a length of 1,203
m. There will be an underground surge tank of 24.5 m diameter 50 m upstream of
the vertical, steel lined penstock.
2.13 Low Level Outlets
There will be 2 No low level outlets with invert levels of 338.0 masl for lowering the
reservoir level in an emergency. These will be 3.8 m wide x 3.8 m high.
In addition to the above there will be 2 No. flushing outlets measuring 5.0 m x 5.0 m
with inverts at 320 m asl. These are quite generously proportioned.
30
Figure 13. Downstream Elevation showing Flushing outlets in Block 5 and Drawdown
Outlets in Blocks 4 and 6.
2.14 Powerhouse
The underground powerhouse will contain two Francis turbines each of approximately
70 MW capacity. These will be installed in the cavern measuring 89 m x 23 m x 44 m.
An adit was driven from the valley side to intersect the roof of the cavern where rock
quality was found to be favourable.
The Project Inception report comments as follows:
The powerhouse section of adit AP-2 is excavated in black, moderately strong
to strong Slates. Foliation is pronounced, generally dipping 45-50/235-240 (dip
amount/dip direction ) Seepage water (mainly drips) is present in the
powerhouse area.
Other than AP-1, it has to be noted that many fissures (joints, foliation, faults)
are healed by either calcite or quartz. A clear system regarding vein filling could
not be established, yet. Generally, the entire rock mass of AP-2 appeared to be
a little more disturbed as compared to AP-1.
In the absence of a detailed engineering geological survey, this impression

might be apparent, only. Faults or other deformations of the rock mass can be
traced easily due to abundant white veins
31
2.15 Instrumentation
The PSCs’ proposals for instrumentation are given on Page 2-32 of Volume II of the
Design Report and include the following:
 Thermocouples
 Construction joint monitoring
 Pendulums and Inverted Pendulums ( with dataloggers )
 Leakage Monitoring
 Piezometers and pressure gauges
 Monitoring of crest movements
 Seismographs on rock and on the crest of the dam

2.16 Engineering Conclusions and Recommendations
POE is concerned as to how the dam will behave in a large earthquake with its
epicentre close to the dam. The Project Supervision Consultants ( PSC ) are not,
however, convinced that such an event is possible:
Previous meetings and discussions (e.g., attached 2020 paper) made it clear,
however, that there is no active fault with 15 km (or even 25 km) from the
Tanahu dam. That means, the PGA=0.80g is a fictional value well exceeding
the maximum credible event at the Tanahu site. According to ICOLD
recommendations, the SEE (MCE) at Tanahu dam has a PGA=0.547g now
which shall be used for safety evaluation ( email from PSC dated 21 October,
2022 ).
However POE notes that the Main Central Thrust ( MCT ) is only 37 km from the
damsite and that Searle, Michael P et al ( 2008 ) state that the definition of the Main
Central Thrust “should not be limited to one thrust fault, but should be a broader fault
zone”.
Whilst accepting that the dam should be able to safely withstand a PGA of 0.547 g
POE do not believe that the probability of serious damage in a larger event is
negligible. Based on statistical evidence Douglas, Spannagle and Fell ( 1998 ) quote
an average return period of 28,571 years for the failure of concrete gravity dams, in
general. This seems optimistic for Tanahu dam in view of the risk of seismic
overstressing, GLOFs and landslide risks upstream. These risks do not, however,
seem to be such as to require any major changes to the design.
32
Outflow Floods ( GLOFs ) due to climate change also needs to be considered.
Proposals for consolidation and curtain grouting appear generally satisfactory but the
necessary quantities of grout may increase due to the presence of karstic dolomite at
the damsite. 4 No. galleries measuring 2.5 m wide x 3.0 m high should be able to
accommodate equipment for drilling for grouting and drainage.
Principal recommendations are as follows:
POE recommend the following;
a) Acceptance of the POE estimate for the probability of unacceptable seismic
cracking of the dam ( ie TR > 8,000 years )
b) Commissioning of a report on the potential for landslides and Glacial Lake
Outflows upstream ( including seismically triggered events ).
c) Preparation of a formal estimate of the return period of non-operation of the
spillway gates when needed to pass a flood. This could be made on the basis
of statistical evidence for similar installations elsewhere.
d) Preparation of a Quantitative Risk Assessment for the project as a whole. This
could be expanded to estimate the Risk posed by the dam in USD/year where:
Risk ( USD/year ) = Probability/year x Consequences ( USD )
e) The Consequences of dam failure should be relatively easily derived from the
results of the dambreak modelling.
f) Consideration of the possibility of satellite or other monitoring of potential
landslip areas upstream.
2.17 Engineering References

Annandale G.W, Morris G.L and Karki P, ( 2016 ),“Extending the Life of Reservoirs”,
IBRD/World Bank.
Douglas K, Spannagle M, and Fell R. ( 1998 ),“ Estimating the Probability of Failure of
Concrete and Masonry Gravity Dams” University of New South Wales.
Gosschalk E.M, Hinks J.L, Johnson F.S and Jarvis R.M, ( 1991 ), “Overcoming the
Build-up of Stresses, Cracking and Leakage in Mullardoch Dam, Scotland”, ICOLD
Conference, Vienna.
Herzog M.A.M ( 1980 ), “Failure Mechanisms in dams subject to earthquakes”, Water
Power and Dam Construction, August, pp 28 – 30.
Hinks J.L, “Behaviour of Dams in the Wenchan Earthquake” ( 2009 ), Issue 3 of
Hydropower and Dams.
ICOLD Bulletin 148, ( 2010 ) “Selecting Seismic parameters for Large Dams”, May.
ICOLD Bulletin 137, ( 2011 ) “Reservoirs and Seismicity”,
National Geographic Society, ( 2022 ), “Everest – Exploring the World’s tallest peak”.
Neville A.M, ( 1963 ), “Properties of Concrete”, Pitman’s Paperbacks.
33
Searle, Michael P et al, ( 2008 ), “Defining the Himalayan Main Central Thrust in
Nepal”,
Journal of the Geological Society, 165 (2): 523–534. March.
US Army Corps of Engineers ( 2003 ), Engineering and Design: Time-History Dynamic
Analysis of Concrete Hydraulic Structures (Engineer Manual EM 1110-2-6051), 22
Dec.
34
SECTION C - ENVIRONMENTAL SAFEGUARDS REPORT
3. SECTION C – ENVIRONMENTAL SAFEGUARDS REPORT
3.1 Introduction
This Section summarizes the activities carried out by the Environmental Expert of the
POE, and describes the findings and recommendations of the POE mission in relation
to the Environmental, Health and Safety (EHS) management capacity of THL and the
level of EHS compliance by the Project. In addition, this Section indicates the status
of preparation of four Project-wide Environmental Management Plans, and explains
the recommended mitigation approach to the impacts of the Project on fish species
and biodiversity.
Table 4 specifies the activities undertaken by the POE Environmental Expert and the
objectives pursued with them.
3.2 THL Environmental, Health and Safety (EHS) Management Capacity
In order to be effective in discharging their Environmental, Health and Safety (EHS)
responsibilities, executing agencies must have sufficient capacity in terms of staff and
resources, as well as tools and procedures for compliance and effects monitoring. This
subsection examines THL’s capacity to fulfil its EHS obligations under the Loan
Agreement signed with the Asian Development Bank (ADB) to implement the Tanahu
Hydropower Project (2013), in particular the enforcement of EHS clauses and
specifications of works contracts, the mitigation measures included in the Contractors’
and the Project’s Environmental Management Plans, national EHS regulations and
good international EHS practices.
In October 2015, THL established the Environmental and Social Management Unit
(ESMU) to deal specifically with the EHS and social safeguard issues related to the
Project. The ESMU staff comprise an Environmental Safeguards Officer
(Environmental Engineer), a Social Safeguards Officer (Sociologist), a
Communication Officer and a Civil Engineer. The staffing level in the original design
of the Unit did not specify a Civil Engineer but included, in addition to the Officers
mentioned, the Director of the ESMU and a Land Officer (THL, 2018, Request for
Proposals for Environmental and Social Management Service Provider, p. 18).2 The
Director position has been vacant for several months and apparently the Land Officer
was never appointed. The absence of the Unit Director is concerning, since this means
that the ESMU lacks a professional at a sufficiently high hierarchical level within the
organization to interact with other THL management units, the PSC and Contractors,
to liaise with environmental and other authorities, and to provide strategic direction
and guidance to the work of the Unit, all of which may undermine the overall
effectiveness of the ESMU, as well as the relevance and priority that EHS and social
2 According to the original requirements for the ESMU, the Unit should be based in the Project area and not in
its current location at the THL headquarters in Kathmandu (THL, 2018, Request for Proposals for
Environmental and Social Management Service Provider, p. 18). However, this arrangement does not seem
to impair the effectiveness of the ESMU, since the Unit’s staff visit the Project area regularly to coordinate with
the Environmental and Social Management Service Provider (ESMSP) the work related to safeguards
implementation, and oversee environmental and social performance on the Project and, in addition, the
ESMSP office is located in the Project area in Jhaputar, Tanahu. The location of the ESMU office in Kathmandu
may offer advantages in terms of access to key THL management and resources.
35
Table 4
Activities Carried Out by POE Environmental Expert during POE Mission and Objectives Pursued with Activities
DATES TIMES ACTIVITIES AND OBJECTIVES PARTICIPANTS

Friday, Night Arrival in Nepal POE Environmental
November 11 Expert.
Saturday, Morning Review of Environmental Management Plans and Monitoring Reports in preparation for POE POE Environmental
November 12 mission. Expert.
Afternoon Flight Kathmandu – Pokhara.
Review of Environmental Management Plans and Monitoring Reports in preparation for POE
mission.
Sunday, Morning Presentation by PSC on physical, financial and overall progress of Package 1 and Package 2 POE, THL
November 13 Contracts. management,
ESMSP and PSC.
Morning and Meeting with ESMSP on: POE Environmental
afternoon  Staffing levels, available resources, and procedures and tools at disposal of ESMSP to Expert and ESMSP.
conduct Environmental, Health and Safety (EHS) inspections and monitoring, and ensure
EHS compliance by Contractors.
 Level of EHS performance by Contractors.
 Challenges faced by ESMSP in carrying out responsibilities.
 Planning of site visits and meetings during POE mission.
Afternoon Visits to construction sites of Package 1 Contract and Package 2 Contract to ascertain POE Environmental
compliance with EHS requirements established in works contract, Contractors’ Environmental Expert and ESMSP.
Management Plans, Project’s EHS safeguards measures and national EHS legislation, as well as
with good international practices.
Monday, Morning and Site visit to Madi River to conduct a rapid visual survey of bio-physical and socio-economic POE Environmental
November 14 afternoon characteristics along River, with particular emphasis on 47 km section proposed as protection Expert, ESMU and
area to mitigate impacts of Project on fish biodiversity and population size of certain fish species, ESMSP.
and discuss main characteristics of proposed Madi River Protection Area.
Meeting with ESMU on EHS compliance by Contractors, and planning of site visits and meetings
during POE mission.
Afternoon Site visit to Package 1 Contract workshops and construction plants to ascertain compliance with POE Environmental
EHS requirements established in works contract, Contractors’ Environmental Management Plans, Expert, ESMU and
Project’s EHS safeguards measures and national EHS legislation, as well as with good ESMSP.
international practices.
Tuesday, Morning Site visits to Package 1 Contract labor camp, lab, first aid center, crushing plant and area of POE Environmental
November 15 implementation of erosion control measures, and to Package 2 Contract labor camp, workshops, Expert and ESMSP.
36
Table 4
Activities Carried Out by POE Environmental Expert during POE Mission and Objectives Pursued with Activities
DATES TIMES ACTIVITIES AND OBJECTIVES PARTICIPANTS

spoil disposal area and construction plants, in order to ascertain compliance with EHS
requirements established in works contract, Contractors’ Environmental Management Plans,
Project’s EHS safeguards measures and national EHS legislation, as well as with good
international practices.
Virtual meeting with ADB to discuss mitigation options explored in Fish Conservation ADB Senior
Management Plan in order to decide on final mitigation approach to impact of Project on Environmental
upstream migration of fish populations, fish biodiversity and population size, and to present Specialist, POE
preliminary findings on EHS compliance by Contractors resulting from POE Environmental Expert Environmental Expert,
site visits. THL management,
ESMU and ESMSP.
Afternoon Demonstration by ESMSP of catch and haul method in a stretch of Sange River, a tributary of POE Environmental
Madi River. This is a mitigation measure proposed in Fish Conservation Management Plan in Expert, ESMU and
addition to Madi River Protection Area to offset impact of Project on fish species. In addition, ESMSP.
micro-invertebrate sampling technique was also demonstrated in same river stretch.
Meeting to discuss plans for storage and disposal of spoils from Project. POE Environmental
Expert and PSC.
Wednesday, Morning and Site visit to Upper Seti River basin, focusing on future dam reservoir area, with stops at Bhimad, POE Environmental
November 16 afternoon Jalbire, Phadke Bridge, Phedikhola River and Badarkuna, as well as at Project Information Center Expert, ESMU and
for area. ESMSP.
Discussion with national Wildlife and Forest Experts on respective Management Plans.
Thursday, Morning Drafting of Environmental Section of Third Report of Panel of Experts. POE Environmental
November 17 Afternoon Flight Pokhara – Kathmandu. Expert.
Drafting of Environmental Section of Third Report of Panel of Experts.
Friday, Morning and Drafting of Environmental Section of Third Report of Panel of Experts. POE Environmental
November 18 afternoon Expert.
Night Departure from Nepal.
KEY: EHS: Environmental, Health and Safety; ESMSP: Environmental and Social Management Service Provider; ESMU:
Environmental and Social Management Unit; POE: Panel of Experts; PSC: Project Supervision Consultant; THL: Tanahu
Hydropower Limited.
37
safeguards issues must have as an integral component of the Project. The hiring of a
Civil Engineer represents an important technical reinforcement of the Unit, since this
strengthens its capacity to assess the operation and effectiveness of physical
mitigation measures implemented by Contractors, and to suggest technical
improvements to underperforming or failing physical mitigation measures.
The ESMU is ultimately responsible for the EHS and social performance of the Project,
having delegated authority in this area from THL (Ibid). It is charged with planning,
implementing, overseeing, enforcing and coordinating the management of EHS and
social measures and actions required to prevent and mitigate the EHS and social risks
and impacts posed by the Project. Since July 2019, the ESMU is supported in the
performance of its responsibilities by an Environmental and Social Management
Service Provider (ESMSP), a consortium formed by international and national
specialized consulting firms3 that works directly under the ESMU assisting in the
oversight of the implementation of environmental and social safeguards measures,
including compliance and effects monitoring.
Taking into consideration the shortcoming regarding the vacancy of the ESMU Director
noted above, and some limitations in field experience in environmental oversight as
well as in ESH enforcement effectiveness elaborated below, both the ESMU and
ESMSP largely have the capacity to undertake their compliance and effects monitoring
responsibilities. This assessment is based on: i) interviews conducted with the ESMU’s
Environmental Specialist and ESMSP managerial and technical staff; ii) the review of
Quarterly Environmental Safeguard Monitoring Reports produced by the ESMU and
the ESMSP; iii) an examination of compliance and effects monitoring tools employed
by the ESMU and the ESMSP (i.e., Minutes of Monthly Health, Safety and
Environment Inspection Review Meetings; Observations of Non-Compliance-ONC
Reports; ONC Letters; Weekly Environment Inspection Checklists; Quarterly Impact
Monitoring Reports for water, air, noise and spring discharge; and website “Tanahu
Hydropower Project - Survey System”); iv) site visits conducted at active construction
sites, construction plants, spoil disposal areas, workshops and workers’ camps; and
v) an examination of the levels of support staff (administration, drivers, etc.) and
resources (vehicles, computer and printing equipment, office space, communication
equipment, etc.) available to the ESMU and the ESMSP to perform their duties.
The ESMSP is fully staffed with the required key and non-key, national and
international specialists, as well as support personnel. The technical staffs of both the
ESMU and the ESMSP are qualified. The ESMP, which currently has a full time
Personal Safety cum Risk Management Officer, may need an additional such Officer
when the construction of the dam works starts, considering the health and safety risks
posed by this activity and in view of the fact that this Officer alone oversees health and
safety compliance for three construction packages, two of which have day and night
shifts.
The ESMSP as well as the ESMU have adequate resources such as budget, office
facilities and supplies, computer and communication equipment, and vehicles.
3 The consortium consists of ELC Electroconsult SPA, Italy, and GEOCE Consultants (P.) Ltd, Kathmandu.
38
The arrangement devised by the ESMSP to subcontract a private specialized
laboratory to undertake the quarterly sampling and analysis of the quality of water, air,
noise and spring discharge at selected monitoring sites is working satisfactorily.
Regarding the procedures and instruments for compliance monitoring, the ESMSP’s
two Environmental Officers and the Personal Safety cum Risk Management Officer
conduct daily, weekly and monthly inspections in their respective areas of expertise,
with each site inspected on average at least four times a week in all work fronts. The
daily inspections consist basically of walk throughs of construction sites, plants,
workshops and camps with the respective EHS staffs of the Contractor for Package 1
and the Contractor for Package 2, checking the status of previously detected non-
compliances, pointing out new issues if applicable and discussing appropriate
mitigation measures for the issues identified. The weekly inspections are more
thorough and systematic, employing an inspection checklist focusing on
environmental issues, although there is not such a checklist for health and safety
issues.
According to interviews conducted in the course of the POE mission, the monthly
inspections are supposed to involve staffs of ESMU, ESMSP, PSC and the respective
Contractor of the package inspected, all of whom should later participate in a meeting
to follow up on pending issues from the previous month’s inspection, and discuss the
findings of the inspection undertaken and actions to address them. However, a review
of a recent sample of attendance sheets of participants in monthly inspections shows
that apparently only staffs of the ESMSP and the corresponding Contractor for the
package inspected usually take part in this activity. Further, an examination of a recent
sample of minutes of monthly EHS meetings shows that staffs of ESMU, ESMSP and
the respective Contractor of the package inspected usually attend these meetings.
The fact that the PSC, whose supervision contract does not specify the hiring of an
Environmental Specialist, apparently does not take part usually in the monthly
inspections or the associated meetings means that there are missed opportunities for
the PSC to gain firsthand knowledge of noncompliance issues and their severity and
duration, and ways to address them. Given that the PSC acts on behalf of THL in the
management and supervision of the work of contractors, including quality control and
remediation of technical deficiencies and contractual non-compliances, its
participation in monthly EHS inspections and meetings is crucial.
An additional way in which the participation of the PSC in monthly EHS inspections
and meetings would contribute to enhancing enforcement of mitigation and
management measures is the sharing by the PSC with the ESMU and the ESMSP of
the approved method statements for the implementation of pertinent works and
measures that will be subject to inspections (e.g., slope stabilization measures,
location and technical specifications for spoil storage and disposal areas, technical
specifications for the exploitation of quarry and borrow areas, etc.). This would allow
the ESMU and the ESMSP to prepare better for the issues to look for and focus on
during site inspections.
Although THL, including its ESMU, and the PSC hold monthly meetings to evaluate
the progress of the Project, and issues in implementation and how to address them,
in which EHS compliance should be discussed, still the monthly joint EHS inspections
and meetings provide a forum to focus specifically on EHS issues and how to address
them. The need for a greater level of involvement of the PSC in EHS compliance
39
monitoring and enforcement is evidenced by some of the findings of the site visits
conducted during the POE mission. For instance, persistent non-compliances in the
Package 1 Contract with community safety and hazardous materials management
procedures and good practices, which could potentially have severe negative
consequences (please refer to Subsection 3.3.1.2 below for details), have been
unresolved for several weeks even though some of these issues have been previously
reported by the ESMU and the ESMSP. Further, a review of Tables 3-4 and 3-5 titled,
respectively, “ONC and Suggested Corrective Measures for Package 1” and “ONC
and Suggested Corrective Measures for Package 2” in THL’s Quarterly Environmental
Safeguard Monitoring Reports for April-June 2022 and July-September 2022,
evidences that the corrective measures for several non-compliances detected remain
“partially implemented” or “not implemented” for several weeks and thus are not
addressed in a timely manner.
The above findings call attention to the need to strengthen the EHS enforcement
system in place for the Project. According to the current system, EHS Officers make
note of Observations of Non-Compliance during their weekly inspections and also
prepare Non-Compliance Reports (NCRs) during more detailed inspections when
warranted. The latter are assigned a level from 1 to 3, depending on the significance
of the negative impact associated with the non-compliance, the time passed since its
detection and the urgency with which mitigation measures should be implemented.
The NCRs for the most serious violations that have the highest potential to cause
significant EHS impacts (NCRs Levels 2 and 3) are signed by a senior manager of the
ESMSP. The NCRs for all three levels identify the specific non-compliance that must
be addressed, specify mitigation measures and indicate a date for the implementation
of recommended measures. In the case of NCRs Level 3, which are the source of
significant negative impacts, the ESMSP sends a letter to the ESMU, which then
requests the Managing Director of THL to write a letter to the PSC requesting it to
formally inform the respective Contractor of the nonconformity, the measures to
address it and the deadlines to fix it. The persistency, level of risk and immediate
urgency to remediate the non-compliances noted above regarding community safety
and hazardous materials handling in the Package 1 Contract indicate that current EHS
enforcement mechanisms are not working with the necessary efficiency. In this sense,
THL management should consider requiring the PSC to employ more stringent
enforcement tools included in works contracts, such as the withholding of payments,
the imposition of fines and ultimately the issuance of stop work orders, for unrectified
continuous non-compliances that can lead to imminent significant negative impacts on
the biophysical environment and/or the safety and health of affected communities and
Project workers. Although it is clear that the main goal of all of the parties involved in
the Project is to ensure that it is implemented on time following best engineering and
EHS and social practices, the more stringent contractual enforcement measures may
be necessary to be implemented sparingly in the noted instances of utmost non-
compliance.
In addition to the above, the findings during the POE mission on community safety and
hazardous materials management in the Package 1 Contract also point to the need
for the institutional strengthening of the EHS staffs of the ESMU and the ESMSP,
composed mainly of dedicated and capable young professionals for whom, with the
exception of the ESMSP’s Personal Safety cum Risk Management Specialist, the
Project constitutes their first practical experience conducting EHS inspections and
enforcing EHS requirements for a major infrastructure undertaking. This need for
40
capacity development is further illustrated by a review of THL’s Quarterly
Environmental Safeguard Monitoring Reports for April-June 2022 and July-September
2022, which shows that in Table 3-4, titled “ONC and Suggested Corrective Measures
for Package 1”, the corrective measures for the non-compliance “storage of chemical
and other hazardous material must be safe” appear as “not implemented” for April and
May 2022, but beginning in June until September 2022 those measures are marked
as “implemented”, which is not the case as discussed in Subsection 3.3.1.2 below.
Further, regarding community safety, the presence of people walking through the
diversion tunnel during active construction works to go back and forth between their
communities and the markets in Damauli is not mentioned in either of the two
aforementioned reports. Thus, THL and the ESMSP should consider offering training
opportunities to their EHS staffs in the oversight, mitigation and enforcement of EHS
measures and requirements in large infrastructure projects, such as those offered
virtually by specialized institutions.
A good practice implemented by the ESMSP is the development of the website
“Tanahu Hydropower Project - Survey System” to centralize data in a largely
interactive digital format on relevant biophysical and socio-economic characteristics of
the Project area, including maps, and information on EHS and social compliance and
enforcement. This tool facilitates greatly the process of biophysical and socio-
economic data querying, as well as of accessing information on EHS and social
compliance and enforcement. Several of the fields in the databases of this digital
platform are being populated and updated. Figures 14 to 16 show sample screenshots
of the website.
Figure 14. Home page of website “Tanahu Hydropower Project - Survey System”
41
Figure 15. Screenshot of menu “Monitoring Observations” of website “Tanahu
Hydropower Project - Survey System”
Figure 16. Screenshot of menu “Tanahu Map” of website “Tanahu Hydropower Project
- Survey System”
42
3.3 EHS Compliance Monitoring
This subsection examines the level of observance by Contractors of the EHS
requirements and technical specifications contained in the Works Contracts, the
Project’s and the Contractors’ Environmental Management Plans, the Nepali EHS
regulatory framework and international EHS good practices. ESH conformance is
reviewed separately for each Contract Package, identifying both compliances and
non-compliances detected during the POE mission.
3.3.1 Package 1 Contract
Compliances Observed
 The workers’ camp and its facilities (workshops, yards, storage facilities,
laboratory, residences, kitchen, etc.) are physically separated from their
surroundings, and access to and egress from them are controlled by guards. In
addition, visitors are required to wear a helmet, a reflective vest and a mask.
 Adequate and separate sanitary facilities for female and male workers were
provided at the campsite.
 The kitchen and eating facilities at the camp were adequate and clean, and
potable water for drinking was provided.
 In general, workers wore appropriate Personal Protective Equipment (PPE) for

the types of tasks performed.
 In general, workers seemed to have adequate training for the types of tasks
undertaken. This is consistent with the information gathered through interviews
and the review of records during the POE mission, which indicated that workers
receive adequate on-the-job training as well as daily toolbox talks prior to the
start of the work day.
 Overall, housekeeping in workshops and yards was fair and needs

improvement.
 Although the crushing and batching plants were not operational during the site
visit, the staffs of the ESMU and the ESMSP remarked that these plants were
well maintained and operated, not generating visible fumes or dust. The
conveyor belts of the crusher have sprinklers to abate the generation of dust.
 No excessive blown away dust was observed on the roads used by construction
vehicles in the site visits undertaken during the POE mission. A water bowser
was seen watering the roads on two different occasions. The water bowser is
scheduled to spray the roads three times a day.
 Although there was no time during the POE mission to visit the spoil disposal
areas for Package 1, the Plans and Layouts for Disposal Areas prepared by the
Contractor were obtained and seemed technically sound.
43
Non-Compliances Observed
 Unfilled EHS position and limited presence of required Contractor’s staff: the
position of Environmental Officer has been vacant since the last week of August
2022 and therefore the Contractor does not have the required expertise among
its staff to undertake inspections specifically focused on environmental issues,
with the Safety Officer looking also at these issues to a limited extent. Some of
the findings below underline the urgent need for the hiring of this member of the
EHS staff.
 The main Contractor only has about 5 members of its key and non-key staffs
on site, including the Project Manager and the Safety Manager, with most of
the civil works being undertaken by local subcontractors. This may be an
underlying reason for some of the non-compliances noted below.
 Hazardous materials and wastes management: in contravention of contract

requirements and contrary to international good practice, the Contractor does
not have a system in place for the handling, storage, transport and disposal of
hazardous materials and wastes. This poses serious risks to the health and
safety of workers and the community, as well as the environment, and thus
warrants urgent attention. The main nonconformities observed in relation to
hazardous materials and wastes are the following:
 There are no Safety Data Sheets (SDSs), also called Material Safety
Data Sheets (MSDSs), for any of the chemicals stored at the camp or
used in the workshops. The SDSs provide specific essential information
for each chemical on: properties; physical, health and environmental
health hazards; protective measures; and safety precautions for
handling, storing, and transporting.
 Hazardous materials and wastes are not stored separately in an isolated
area of the camp in a locked room. Further, liquid hazardous materials
and wastes that may leak or spill are not stored in general in adequate
containers placed on an impervious surface and enclosed in a bunded
area with sufficient capacity to hold the volume of liquid that may spill.
Illustrative of this are: i) the storage in metal shelves within a storage
room of gasoline in plastic water bottles close to containers with diesel,
engine oil, brake fluid, engine coolant, transmission oil and even
chemicals (see Figures 16 and 17); and ii) the placement of barrels of
new motor oil in an open area without containment walls, mixed with
barrels of chemicals such as modified inorganic aluminum salt solution
and open containers of used engine oil (see Figures 18 and 19).
 A licensed hazardous waste management company has not been
contracted for the final disposal of this type of waste. Hazardous wastes
are inadequately stored at the camp as noted above and apparently
there is no plan for their final disposal.
 There are multiple unattended oil spills in different parts of the
workshops and there are no spill cleanup kits on site; in addition, since
the position of Environment Officer is vacant, no one on the Contractor’s
side knows how to clean up the spills and disposed of the contaminated
soil (see Figures 20 and 21).
44
 There is no grease trap or interceptor at the engine oil changing station,
which is located very close to the bank of the Seti River, posing the risk
of hydrocarbon pollution of this water body.
Figure 16. Plastic water bottles containing gasoline kept in storage room where other
hydrocarbons and chemicals are stored
Figure 17. Inadequate storage of hydrocarbons in storage room
45
Figure 18. Inadequate storage of barrels of motor oil together with chemicals, pipes
and tires
Figure 19. Inadequate storage of used motor oil
46
Figures 20 and 21. Evidence of unattended oil spills at refueling station and workshop
 Community safety and access: during the site visit to the Package 1 Contract
works, an elderly man was seen coming out of one of the diversion tunnels by
foot and then continuing walking towards his village. This person did not have
any reflective clothing and wore a hard hat provided by the Contractor. This
was observed in the middle of the afternoon, around 2:45, a time of high
circulation of construction trucks inside and in the proximity of the tunnels, as
well as of intense construction activity in the area (see Figures 22 and 23).
According to the Contractor’s Safety Manager, an average of 30 to 40 people
per day take this route by foot from the villages of Kundhi, Setang, Lokma and
Bajo to go mainly to the markets in Damauli and back to their communities. This
situation occurs daily, even though the Contractor provides a truck, also used
to transport workers, in the mornings and afternoons to drive villagers from and
to their communities. In separate interviews conducted on a different day in the
morning with the truck driver and villagers that were just getting off the truck to
go to the market, community members indicated that the pickup and drop-off
schedule is inconvenient and that therefore sometimes they do not wait for
transportation and just walk. The driver of the truck indicated that he transports
workers every day at 7:00 AM, 12:00 M and 6:00 PM, and villagers at
approximately 8:00 AM (from the villages) and 5:00 PM (back to the villages).
He said that he waits until there are at least 5 villagers before transporting them.
 The above transportation arrangement for the villagers is evidently not working,
as seen on the site visit and as explained by the community members
interviewed. Therefore, a solution must be proposed by the Contractor, which
may entail increasing the frequency of trips to two in the mornings from the
villages and two in the afternoons to the villages. The seriousness of the risk to
community safety that the situation described poses cannot be overstated and
a solution has to be found as a matter of urgency. In addition, the characteristics
of the platform on the back of the truck where villagers ride is not ideal, in
particular for the elderly and children, and thus the truck should be appropriately
retrofitted to facilitate accessing and exiting the platform; further, adequate
47
seating and grab handles for comfort and security should also be provided (see
Figures 24 and 25).
 The above issues relate to the obstruction by the Project of direct access by the
affected communities to other villages and the main settlements in the area
such as Damauli, an impact that has not been adequately addressed. In
discussions held during the POE mission, it became evident that there are no
clear plans on how to ensure access by affected communities to other villages
and settlements once the Project is operational, an issue that has been in
discussion since the time of preparation of the Environmental and Social Impact
Assessment for the Project but that has yet to be adequately solved.
Figures 22 and 23. Elderly person emerging from one of the diversion tunnels by foot
and walking in an active construction area
Figures 24 and 25. Inadequate truck used for transportation of villagers and villagers
getting ready to go to the market in Damauli after being dropped off by the truck
 Health services: in contravention of contractual obligations, there is no

ambulance on site and the First Aid Center is not staffed by a Medical Doctor
but by a Community Medical Officer and, in addition, a nurse has not been
hired. The Community Medical Officer degree is not equivalent to that of a
Medical Doctor, since it requires taking 18 months of coursework in college and
3 months of practical training. The medical supplies and equipment at the First
Aid Center seemed limited. These non-compliances have been noted
48
persistently in the Quarterly Environmental Safeguard Monitoring Reports
produced by the ESMU and the ESMSP, but have not been resolved.
 Solid waste management: there is no separation, reuse or recycling of domestic

solid wastes in the workers’ camp. This violation of contractual requirements
and specifications of the Contractor’s Environmental Management Plan has
been ongoing since the beginning of Project implementation. The ESMSP
prepared a solid waste management plan at the request of the Contractor, but
it is not being implemented. The solid wastes are collected by the municipality,
which does not have an adequate landfill and simply deposits them in an area
close to the Seti River.
 Topsoil management: contrary to environmental management specifications

and international good practice, the Contractor has not collected, stored and
reused the nutrient rich layer of topsoil of the areas where earthworks have
been implemented.
 Crushing and batching plants: the inadequate treatment of wastewater coming

from these plants have been reported for several months in the Quarterly
Environmental Safeguard Monitoring Reports produced by the ESMU and the
ESMSP, as well as the External Environmental Monitoring Report prepared by
the External Monitoring Expert, but this nonconformity has not been resolved.
 Refueling station: the pump does not rest on a concrete slab and the refueling
of vehicles is carried out on bare soil rather than on a concrete platform. In
addition, there is no person in charge of the refueling station, as all drivers have
access to the pump, which has no lock.
 Fire extinguishers: there was an insufficient number of fire extinguishers at the

workshops, some of them were missing their service tags and the dates on
some service tags were illegible.
 Visitor log and talk on safety on site: contrary to international good practice,
there is no visitor log to register the presence of persons not affiliated with the
Contractor in the workers’ camp. In addition, visitors do not receive a brief talk
on how to stay safe on the Contractor’s premises.
3.3.2 Package 2 Contract
Compliances Observed
 The EHS staffing levels comply with contractual specifications.
 The First Aid Center is fully staffed with a medical doctor and two nurses. In
addition, the medical supplies and equipment available seemed adequate, and
there is an ambulance on site.
 The workers’ camp and its facilities (workshops, yards, storage facilities,
laboratory, residences, kitchen, etc.) are physically separated from their
surroundings, and access to and egress from them are controlled by guards. In
addition, visitors are required to wear a helmet, a reflective vest and a mask.
49
 Adequate and separate sanitary facilities for female and male workers were
provided at the campsite.
 The kitchen and eating facilities at the camp were adequate and clean, and
potable water for drinking was provided.
 In general, workers wore appropriate Personal Protective Equipment (PPE) for

the types of tasks performed.
 In general, workers seemed to have adequate training for the types of tasks
undertaken. This is consistent with the information gathered through interviews
and the review of records during the POE mission, which indicated that workers
receive adequate on-the-job training as well as daily toolbox talks prior to the
start of the work day.
 Overall, housekeeping in workshops and yards was satisfactory, and there

were no noticeable leaks or spills of hydrocarbons. In addition, there was a
sufficient number of properly maintained fire extinguishers at the workshops,
with adequately displayed up-to-date service tags.
 Although the crushing and batching plants were not operational during the site
visit, the staffs of the ESMU and the ESMSP remarked that these plants were
well maintained and operated, not generating visible fumes or dust. The
conveyor belts of the crusher have sprinklers to abate the generation of dust.
 No excessive blown away dust was observed on the roads used by construction
vehicles in the site visits undertaken during the POE mission. A water bowser
was seen watering the roads on two different occasions. The water bowser is
scheduled to spray the roads three times a day.
 Although available time during the POE mission only allowed a quick drive by
Disposal Area D, no major issues were apparent. The Plans and Layouts for
Disposal Areas prepared by the Contractor were obtained and seemed
technically sound.
Non-Compliances Observed
 Solid waste management: there is no separation, reuse or recycling of domestic

solid wastes in the workers’ camp. This violation of contractual requirements
and specifications of the Contractor’s Environmental Management Plan has
been ongoing since the beginning of Project implementation. The ESMSP
prepared a solid waste management plan at the request of the Contractor, but
it is not being implemented. The solid wastes are collected by the municipality,
which does not have an adequate landfill and simply deposits them in an area
close to the Seti River.
 Topsoil management: contrary to environmental management specification and

international good practice, the Contractor has not collected, stored and reused
the nutrient rich layer of topsoil of the areas where earthworks have been
implemented.
50
 Visitor log and talk on safety on site: contrary to international good practice,
there is no visitor log to register the presence of persons not affiliated with the
Contractor in the worker’s camp. In addition, visitors do not receive a brief talk
on how to stay safe on the Contractor’s premises.
3.4 Status of Preparation of Project-Wide Environmental Management
Plans, and Recommended Mitigation Approach to Project Impacts on
Fish Species and Biodiversity
After receiving final comments from ADB and the POE, the ESMU and the ESMSP
were in the process of developing the final versions of the following three
Environmental Management Plans for the Project: the Consolidated Environmental
Management Plan, the Forest and Watershed Management Plan, and the Wildlife
Conservation and Awareness Plan. According to the ESMSP, the first Plan would be
submitted by November 18, 2022 and the latter two Plans by the end of November
2022.
Regarding the Fish Conservation Management Plan, the POE mission afforded the
opportunity to hold a virtual meeting with the participation of the ADB Senior
Environmental Specialist, the POE Environmental Expert, THL senior management,
and the staffs of the ESMU and the ESMSP, in order to discuss and reach agreement
on a recommendation on the most effective mitigation approach to the barrier effect
created by the dam to the upstream migration of long-ranging species, as well as the
impacts of degradation of fish habitats, and decrease of ichthyofauna diversity and
populations of some species.
The meeting analyzed the technical feasibility, design and implementation complexity,
cost and likely effectiveness of the following four mitigation measures identified in the
Environmental Addendum for the Project, dated 2012 (NEA and THL, Volume 2a,
Environmental Management Plan: Dam/Reservoir Component, pp. 27-33), and further
developed and assessed in the Fish Conservation Management Plan prepared by the
ESMU and the ESMSP, dated 2022: a fish pass, a fish hatchery, a catch and haul
system, and the protection of a section of the Madi River. The analysis was based on
the findings of the literature and the experience of the participants with the
implementation of these measures in different parts of the word, including South Asia.
The Environmental Addendum recommended an examination of technical options for
the implementation of a fish pass, restricted to a technical feasibility study (p. 33). The
discussion during the meeting considered the following aspects: i) there is a very high
level of technical difficulty involved in the design and implementation of this structure,
which combines a conventional pool pass-type fish pass and a fish lift, and the current
design is preliminary, requiring further studies and adjustments; ii) it is the measure
with the highest cost, which will likely increase in view of the need for adjustments to
the present design; iii) the effectiveness of this measure is not guaranteed but, rather,
its chances of success seem very low because of the technical difficulties involved in
the design (for instance, the proposed fish lift system can be operated only with the
reservoir at full supply level and is functional exclusively for upstream migration,
among other issues); and iv) the results are mixed in different parts of the world,
including South Asia, regarding the effectiveness of fish ladders/lifts/passes, and there
is no experience in the country with the installation of this type of structures for high
51
dams such as Tanahu. The conclusion of the discussion was that this option does not
seem viable.
The fish hatchery option presents several challenges to the genetic and biological
integrity of fish species (e.g., inbreeding would lead to the loss of genetic diversity, not
all species of interest can be bred in a hatchery, such as Bagarius sp, etc.), its
implementation and maintenance costs are very high, and it involves the renovation
of the Kali Gandaki Hatchery, which was built to mitigate the impacts of the Kali
Gandaki Project, but its structures are seriously deteriorated, presents operational
problems and has not produced fish since 2019. Based on this, this option was not
considered optimal.
The catch and haul system was considered relatively easy to design and implement,
reasonably costed and with good chances of success.
The Madi River Protection Area was the measure assessed as having the highest
potential to mitigate Project impacts on ichthyofauna diversity and population sizes in
the context of the Seti River watershed. The discussion made clear that this option
involves, among other requirements for its design and implementation, a high degree
of community and institutional stakeholder consultations, defining the appropriate legal
protection figure for the area, devising the institutional arrangement for the
management of the area, and conducting biophysical and socio-economic studies for
the delineation of the protection area. In this sense, this measure is complex and
costly, with a medium-term implementation horizon.
The discussion concluded that the best mitigation approach would be to combine the
implementation of the Madi River Protection Area with the catch and haul system, with
all meeting participants agreeing on this, including the senior management of THL. It
was further agreed that the ESMU and the ESMSP would update the Fish
Conservation Management Plan with a detailed analysis of the viability and likely
effectiveness of the four measures examined and the justification of the measures
supported. It was indicated that ADB’s Safeguards Division must give its No-Objection
to the mitigation approach identified before it can be implemented as part of the Fish
Conservation Management Plan.
52
SECTION D – SOCIAL ISSUES
4. SECTION D – SOCIAL ISSUES
4.1 Summary of key findings and recommendations
Key findings which are summarized in Section 1.4 are repeated here together with
recommendations relating to issues that have a bearing on compliance with social
safeguards for the THP. Other findings and recommendations are included in the
Social Report that, in the opinion of the POE social expert, can contribute to enhancing
the work to address the needs and priorities of directly and indirectly affected people.
The recommendations are numbered according to the relevant section of the Social
Report.
4.2 Affected households that are cultivating acquired land
A number of households continue to cultivate agricultural land that has been acquired
by THL, as per THL policy. It is important to assess the continuing vulnerability of these
households as candidates for the Livelihood Restoration Plan (LRP) and to ensure
that they understand clearly and in a timely manner when they will have to stop
cultivating these lands.
Recommendation 2.3/1: In the development of the LRP, the analysis and
identification of vulnerable households should include AHs continuing to
cultivate acquired land in the reservoir area who may be at risk of sustained or
increased food insufficiency when they must cease cultivating that land, to
ensure they will have adequate agricultural land and/or are transitioning to non-
agricultural livelihoods.
Recommendation 2.3/2: The external monitor should target a small sample of
AHs continuing to cultivate acquired land in the reservoir area, to identify and
assess the strategies these AHs are developing and implementing to ensure
their livelihoods when they must cease cultivating that land.
Recommendation 2.3/3: THL management and the ESMU should develop and
implement a comprehensive public communications program to advise AHs
continuing to cultivate acquired land in the reservoir area about the filling of the
reservoir. The key elements of the communications program should include: i)
a confirmed date for reservoir impoundment; ii) a launch of the communications
program at least two annual crop cycles ahead of the date of reservoir
impoundment; and iii) multiple forms of communication including, among
others, community radio, TV, notices posted in prominent public places, and
local people/groups such as municipal/ward officials, school
principals/teachers, School Management Committees and Parent Teacher
Associations, local NGOs, Mothers’ Groups, etc.
4.3 Non-titled households
Among households cultivating acquired land, there are a number that did not have title
to the land and, in some cases, were deemed ineligible for an allowance for non-titled
affected households; as above, they should be targeted in vulnerability assessments
for the LRP. Also, other non-titled households that may be affected by the proposed
buffer zone have initiated the process of registering their land; it is recommended that
53
THL support them in this process but without being directly involved in order to ensure
impartiality and transparency.
Recommendation 2.4/1: The actions recommended above with respect to
titled AHs continuing to cultivate affected land in the reservoir area should be
extended to include non-titled AHs, as targets for the LRP, in external
monitoring surveys and in THL’s communication strategies to advise people
about the reservoir impoundment.
Recommendation 2.4/2: THL should negotiate with a Nepali NGO with legal
expertise and/or with a lawyer/law firm to consult with non-titled households in
Wangtankitar, Kapardi and any other communities that may be affected by the
buffer zone and represent them before the District Land Commission and/or in
any other negotiations related to obtaining land titles. This should be done
immediately to expedite the opportunity for households to obtain title regardless
of whether their land is identified as affected in the buffer zone study.
THL should adopt this approach rather than intervening directly or through
municipal officials, as proposed. The process of assisting potentially affected
households to obtain land titles should be independent of the project proponent
and local political leaders; and it should comply fully with the principles of
impartiality, transparency, and accountability as well as free, prior, and informed
consent.
4.4 Community assets
While public resting places affected by the reservoir have been replaced, most other
affected community assets have not. Of particular concern to local communities are
the loss of cremation sites, access to foot paths, and to activities on both sides of the
river (reservoir). The POE social expert recommends that steps are taken to move
forward to finding solutions to these issues.
Recommendation 2.5/1: With respect to both cremation sites and cross-
reservoir access, the ESMSP should work in close collaboration with
community leaders, Community Forest User Groups (CFUGs) and residents of
villages and settlements along both sides of the reservoir to document and map
the preferences for i) the location of cremation sites and ii) “desire lines” that
reflect people’s movements along each side of the reservoir (river) and from
one side to the other, as well as the reasons for and frequencies of these
movements. In consulting communities, different groups should be consulted,
including different ethnic groups/castes, women and men, children/students,
the elderly, farmers, people with livestock, laborers, businesspeople, etc.
Recommendation 2.5/2: THL should expedite the preparation of an
operational plan for the dam, including the proposed annual flushing of the dam,
in order that concrete steps can be taken, in consultation with affected
communities, to develop plans and an implementation schedule for
replacement of cremation sites, foot paths and suspension bridges, as well as
other affected community assets.
4.5 Resettlement along transmission line
54
Some affected households along the transmission have received compensation and
work is ongoing to determine rates for other types of affected assets. However, at the
same time, the civil works contractor is actively working to construct tower foundations
and erect towers. Also, a portion of the transmission line passes through forests but
there has been no determination if Community Forest User Groups (CFUGs) are
affected. Therefore, the POE social expert recommends actions to expedite
resettlement activities.
Recommendation 3.1/1: THL communicate and consult directly with the AHs
that have not received compensation for permanently acquired private land to
negotiate mutually acceptable written agreements to file applications for
compensation and/or to expeditiously resolve obstacles to filing applications for
compensation. The objective is to enable THL to hand over tower foundation
sites to the P3 contractor while complying with resettlement policy that affected
people are fully compensated prior to civil works; at the same time, THL should
not exert undue pressure on AHs.
Recommendation 3.1/2: THL take steps to ensure that the following occurs as
expeditiously as possible: i) the CDCs reach decisions regarding compensation
rates for restricted land use and affected structures in the ROW, ii) AHs are
consulted and encouraged to file applications for compensation, and iii) AHs
are paid. The objective again is to facilitate access for the P3 contractor to
complete construction of the TL while complying with resettlement policy that
affected people are fully compensated and are not unduly pressured.
Recommendation 3.1/3: THL take steps to identify if the construction of the TL
affects forest resources managed by CFUGs and, if so, consult with the affected
CFUGs to determine the extent of impacts, if any, on their use of those
resources.
4.6 Grievance redress mechanism
The POE social expert made an extensive analysis of the THP Grievance Redress
Mechanism (GRM) and found a number of issues related to widely accepted principles
and best practice. The following is a summary of recommendations that are further
elaborated in the main Social Report.
Recommendation 5.4/1: Review and, as required, revise the scope and
structure of the THP GRM i) to encompass all stages of the development,
construction, and operation of both the HP and TL components; and ii) to
establish major categories and sub-categories of grievances and complaints.
Recommendation 5.4/2: THL should expand the ways that people can lodge
a grievance/complaint.
Recommendation 5.4/3: THL should assign a unique code to each registered
grievance and complainant.
Recommendation 5.4/4: THL should ensure that a grievance file (electronic
and/or paper) is created to track the grievance/complaint.
55
Recommendation 5.4/5: THL should ensure that the person(s) making a
grievance/ complaint are advised about the process to verify and assess their
claims and how they can track the progress of this process and the outcomes,
in person and/or through access to written records.
Recommendation 5.4/6: THL should review and, as necessary revise, the
process of grievance redress to mandate the LCFs in each municipality/RM to
be the decision-making body for Level 1 adjudication of grievances, particularly
for grievances related to the entitlements, compensation, livelihood restoration,
and other assistance for people directly affected by land acquisition and/or
commitments made in various THP environmental management plans.
Recommendation 5.4/7: THL support the establishment of LCFs in the
municipality, rural municipalities and the metropolitan city affected by the
transmission line to address grievances.
Recommendation 5.4/8: With respect to grievances related to construction
activities and those related to the operation of the hydropower dam, THL should
review and, as necessary, revise the Level 1 process of grievance redress.
Recommendation 5.4/9: Following review and revision of the GRM, THL
should take steps to i) provide adequate training for THL staff, members of
LCFs, contractors’ representatives, affected people and others involved in
aspects of the GRM, to ensure that they have the knowledge, attitudes and
practical skills to support an effective GRM; and ii) develop and disseminate
written materials (in local languages) and conduct community consultations at
the ward and/or village levels, to ensure that people understand clearly how
different types of grievances will be addressed and how they can lodge, make
representations and track grievances.
4.7 Project employment
The THP has met a basic target of employing at least 1,000 people as set out in the
Design and Monitoring Framework. However, the THP has not met targets for
employment of local people, i.e., people living in Tanahun District, or women. The
target that 30% of project employees are women may not be realistic, however, given
the high expectations people have about project-related employment, the POE social
expert recommends action on existing commitments and other initiatives to expand
employment opportunities.
Recommendation 6/1: THL and the Package 2 contractor take immediate
steps to develop rolling 6-month projections of the workforce requirements,
including clear information about the required qualifications and the duration
and other relevant conditions.
Recommendation 6/2: THL and THP civil works contractors collaborate to
publish notices and widely disseminate information through other channels
about available project-related employment, at least 1 month prior to the start
date of the work. Wide dissemination of information should include community
radio announcements, information provided to village-level community leaders,
youth clubs, savings groups, etc., as well as via PICs and LCFs.
56
Recommendation 6/3: THL review and, as required, revise and/or develop and
implement a strategy to assist affected people and other local people who are
interested in employment opportunities to secure employment. The scope of
the review should include i) how THL identifies AHs where members are
interested in employment, validates their qualifications, and collaborates with
civil works contractors to link them to available positions; and ii) how PIC staff
in each municipality/RM as well as at the THL site can be resourced and tasked
to assist interested people to complete application forms, prepare for interviews
with contractors, etc.
Recommendation 6/4: THL review and take steps to reinforce and enforce the
RIPP policy that one person per AH be considered for employment on the
project, in order to make project-related employment available to as many AHs
as possible.
4.8 4.8 Livelihood Restoration Plan and Community Development Plan
The Livelihood Restoration Plan (LRP) and Community Development Plan (CDP) are
in development or recently developed. The POE social expert has reviewed the
research that has gone into the development of the plans as well as the early
formulations of them; and has made suggestions rather than recommendations. The
following recommendations, however, reflect lessons learned from activities that have
been undertaken and/or the information received from informants in local
communities.
Recommendation 8.2/1: Support for income restoration or improvement in the
LRP should focus on a packaged approach that combines technical skills with
entrepreneurship skills. Alternately, the LRP approach may want to focus
primarily on entrepreneurship skills, working with affected people who already
have skills and/or are trying to operate a microenterprise, with the objective of
assisting them to be more successful and/or able to grow their existing income
generating activities.
Recommendation 8.2/2: The skills development training to date has focused
on several skills that are in demand among project-affected people, but the
training has largely overlooked the highest priority, namely agricultural training.
The LRP should explore options to expand agricultural and livestock skills
training and entrepreneurship support (“farming as a business”), including the
introduction of simple, low-cost improved technologies to address climate
change impacts and promote greater food security.
Recommendation 8.3/1: THL should consult with Shukla Secondary School
and the other secondary schools targeted for student scholarships and based
on feedback, review and, as necessary, revise the proposed scholarship
program to optimize the allocation of scholarship funds to benefit the academic
and social development of the students who are selected.
4.9 Emergency Preparedness Plan
At present, this plan addresses issues downstream of the dam. The SOE social expert
believes that it is important that it be amended to reflect safety issues associated with
57
dam operation and annual flushing of the reservoir, for people and communities in
vicinity of the reservoir.
Recommendation 9/1: THL amend the Emergency & Preparedness Plan and,
as relevant, the River Safety Management Plan to address risks associated
with annual changes in water levels in the reservoir. In addition to early warning
systems (EWS) that are channeled through government agencies at district
and/or ward levels, it is recommended that a community-based ESW “layer” be
created that, for example, engages a network of women and/or men with mobile
phones. Further, it is recommended that a public education program be
developed and delivered to a wide cross-section of local community people.
4.10 Methodology and scope of report (see also Section 1.4 )
The methodology for the POE review of social safeguards issues for the Tanahu
Hydropower Project (THP) has combined document review, meetings with key
personnel from THL and its consultants, and site visits throughout the hydropower
(HP) project area including informal discussions with members of affected households
and other community members.
Prior to arriving in Nepal, the POE social expert conducted an extensive home-office
review of project-related documents that had been provided in the intervening time
since the previous POE mission in December 2018, by THL and/or ADB, as well as
other documents that were accessed from the THL website. The review produced
extensive notes on the status of social safeguard and gender equality and social
inclusion (GESI) aspects of the THP, along with the identification of issues to query
for clarification, updating, etc., during the POE mission.
The POE social expert arrived in Nepal on 4 November 2022. In Kathmandu, meetings
were held with the THL Managing Director, members of the THL Environment and
Social Management Unit (ESMU) and Environmental and Social Management Service
Provider (ESMSP) teams, and the consultant for external monitoring of social
safeguards issues. These meetings were invaluable to, first, meet people who had not
been in place at the time of the previous POE mission; and to clarify some initial
questions about the progress being made in the implementation of the Resettlement
and Indigenous Peoples Plans (RIPPs) for the HP and transmission line (TL)
components of the THP, from the perspective of people involved in the implementation
and external monitoring of the resettlement process.
From 8 November through 16 November, the POE social expert was based at the THL
camp to conduct site visits in the project area. The majority of this time was spent in
site visits to communities to talk with members of project-affected households and
other community members and/or to see some of the community infrastructure
projects that THL has implemented. Throughout this field work, the POE social expert
was supported by staff of THL, ESMU and ESMSP, in particular, ESMSP’s Deputy
Team Leader, Social Expert, and GESI Expert.
The POE social expert has been apprised of the issues that have led two separate
groups of households to appeal directly to the Asian Development Bank (ADB)
Accountability Mechanism and the European Investment Bank (EIB) Accountability
Mechanism. This includes one group of 15 directly affected AHs in Paltyang, as well
58
a group of 12 Dalit households in Wangtankitar and 5 Magar households in Jalbire
that may be affected by the establishment of a buffer zone around the reservoir.
As negotiations are ongoing through the ADB and EIB processes, the POE social
report does not address the issues directly. However, in the course of field work, the
POE social expert did visit Wangtankitar to discuss other issues and, also, had
occasion to talk with residents of Paltyang and Jalbire, again on separate matters.
The POE social report briefly reviews the planning and status of implementation of key
elements of the social safeguards, along with POE social expert comments,
suggestions and, as relevant, recommendations.
4.11 Households affected by hydropower project
As of the end of Quarter 2 (Q2) 2022, a total of 558 directly affected households (AHs)
have been identified in the HP project area, with 93% having received compensation
and allowances to which they are entitled. The remaining 39 AHs that have not
received compensation include i) 2 AHs in the Beteni-Banchare area who have yet to
be paid;4 ii) 15 AHs in Paltyang who have appealed directly to the accountability
mechanisms of the ADB and the EIB and continue to negotiate their claims; iii) 12 AHs
that are in court to resolve family issues related to compensation claims; and iv) 10
households that THL has been unable to locate.
4.12 Buffer zone study
The full assessment of people affected by the HP component will not be known until
the proposed buffer zone study is completed to assess the potential for flooding
primarily in the upper portions of the reservoir. The issue of the buffer zone has been
raised by people in the HP project area since at least as early as the first half of 2018.5
The study has been delayed due to various circumstances, however the ESMSP
estimates that it will begin in December 2022 and will take six months. Following this,
it will be necessary to prepare an addendum to the 2018 Resettlement and Indigenous
Peoples Plan (RIPP) or a new RIPP, including a census and detailed measurement
survey of all affected people and a sample socio-economic survey.
ESMSP is developing a comprehensive consultation process with one or more
community meetings in each ward around the reservoir, depending on the size of the
ward. Target participants include local officials and representatives of potentially
affected households and/or ethnic groups/castes. Following an initial round of
meetings in December, subsequent rounds of meetings are planned in conjunction
with different stages of field work to inform communities of the presence of the study
team and what they will be doing in and around communities. The ESMSP has recently
recruited a Magar-speaking person who will assist in facilitating the consultations for
participants who are more comfortable communicating in that language.
4.13 POE comment
It is important that the proposed consultations engage and encourage potentially
affected people, including women, marginalized groups, and others to understand the
4
The reasons for this are unknown.
5
The 2018 Updated RIPP mentions the buffer zone study.
59
implications of the buffer zone and to participate actively, make comments, ask, and
receive responses to their questions.
Recommendation 2.2/1: To that end, the ESMSP should consider the
development of visual materials, e.g., based on maps, aerial photos and/or
interactive digital presentations, that will show people the study area and,
subsequently, where field work will happen. These materials should be displayed
at the locations where consultations are held, in a manner that allows people to see
them prior to the formal start of the consultation meeting; ESMSP and ESMU staff
should be available to help explain the information, as many people may not be
familiar with how to read maps, aerial photos, etc. At the conclusion of the buffer
zone study, a large-scale map and/or aerial photo of the final determination of the
boundaries of the buffer zone can be printed on waterproof material and posted in
each affected village/settlement, on village notice boards, on the outside of
buildings and/or at other locations that are publicly accessible.
4.14 AHs continuing to cultivate acquired land
THL has allowed households with acquired agricultural land in the reservoir area to
continue to cultivate the land until the impoundment of the reservoir. The post-
resettlement survey provides data on AHs that have exercised this option. In total, 218
AHs or 71% of those in the reservoir area have not moved and, of them, 196 AHs
(89%) are still cultivating agricultural land. Fewer than 5% of all AHs have used
compensation money to buy other agricultural land (ELC, 2022b); therefore, the AHs
still cultivating agricultural land in the reservoir area are likely continuing to cultivate
acquired land as well as the remaining portion of their landholding that was not
acquired.6
Among AHs with title to affected land in the reservoir area,7 more than three-quarters
(77%) have fewer than 10 ropani of agricultural land remaining; more than half of these
households have 0-5 ropani remaining.8 Further, nearly two-thirds (64%) of these
households have less than 6 months of food sufficiency, half of whom have less than
3 months of food sufficiency. To meet their food needs, these households rely on
remittances (33% HH), selling agricultural products and livestock (28% HH), and
employment, working as laborers or operating businesses (19% HH).
4.15 POE comments and recommendations
In conversations with affected people, it is clear that some understand that they will
have to stop cultivating acquired land when the reservoir is filled; however, they are
not all clear about when that will happen or what they will do to sustain their livelihoods
at that time. Some of the most vulnerable may be targeted by the Livelihood
Restoration Plan (LRP),9 however other initiatives may be required to ensure that AHs
take appropriate steps to avoid a significant drop in livelihoods and economic
wellbeing when the reservoir is filled.
Recommendation 2.3/1: In the development of the LRP, the analysis and
identification of vulnerable households should include AHs continuing to
6
Note: The data are not entirely clear on this matter.
7
The situation of non-titled AHs is discussed in the next section.
8
The average agricultural landholding in rural areas of Nepal is 13 ropani (0.68 hectare). Source: Timilsina, R.H. et al., 2019.
9
See Section 7, below, for further discussion of the LRP.
60
cultivate acquired land in the reservoir area who may be at risk of sustained or
increased food insufficiency when they must cease cultivating that land, to
ensure they will have adequate agricultural land and/or are transitioning to non-
agricultural livelihoods.
Recommendation 2.3/2: The external monitor should target a small sample of
AHs continuing to cultivate acquired land in the reservoir area, to identify and
assess the strategies these AHs are developing and implementing to ensure
their livelihoods when they must cease cultivating that land.
Recommendation 2.3/3: THL management and the ESMU should develop and
implement a comprehensive public communications program to advise AHs
continuing to cultivate acquired land in the reservoir area about the filling of the
reservoir. The key elements of the communications program should include: i)
a confirmed date for reservoir impoundment; ii) a launch of the communications
program at least two annual crop cycles ahead of the date of reservoir
impoundment; and iii) multiple forms of communication including, among
others, community radio, TV, notices posted in prominent public places, and
local people/groups such as municipal/ward officials, school
principals/teachers, School Management Committees and Parent Teacher
Associations, local NGOs, Mothers’ Groups, etc.
4.16 Hs without land title
The 2018 RIPP for the HP component (THL, 2018b) establishes entitlements for AHs
lacking title to agricultural land they are cultivating; these include i) a special allowance
for AHs who do not have other agricultural land10 and ii) THL will help them to obtain
title to the affected land.
As of end of Q2 2022, 55 AHs in the HP project area have been identified as
encroachers (Table S-1). The majority (35 AHs or 64% of all non-titled AHs) are in the
reservoir area, particularly in Bhimad Municipality (21 AHs or 60% of non-titled AHs in
the reservoir area).
Table S-1: Non-titled AHs, HP Component

Total Female Structures Displacement Business Non- titled
AHs AHs allowance loss allowance
allowance
Reservoir Bhimad 21 2 17 15 6 9
Rishing 4 0 2 2 0 1
Myagde 9 1 9 9 1 7
Vyas 1 0 1 0 0 0
Access road/THL camp Rishing 5 3 5 4 0 0
Beteni/Banchare Rishing 10 3 5 5 0 2
Vyas 5 0 5 5 0 1
Totals 55 9 44 40 7 19
Source: Compensation Payment Status September 2022
10
The allowance was NR 400,000 for AHs in the areas of the reservoir and the access road/THL camp; due to inflation, it was
increased to NR 425,000 in the Beteni/Banchare area where compensation was paid at a later date.
61
Nineteen of the 55 non-titled AHs (35%) received the special allowance because they
do not have other agricultural land. In the Beteni/Banchare area, four households are
identified as encroaching on public land which makes them ineligible to apply for land
title. However, there is no indication that THL has assisted other non-titled households
to register their land, to be eligible to receive compensation for acquired land. The
POE social expert interviewed a member of one non-titled AH in the reservoir area in
Bhimad whose family initiated a process to register their land; however, when that
process was delayed, the family felt that they were pressured by THL to accept
compensation only for affected assets on the land.
Among non-titled AHs, 41 AHs (75%) had not moved from acquired land as of 2021,
with 80% of them continuing to cultivate acquired land, as permitted by THL, including
22 non-titled AHs in the reservoir area.11 As with titled AHs, some of these AHs such
as the one described above realize they will have to stop cultivating that land when
the reservoir is filled, although others may not; many will have spent their
compensation money that they received as much as 8-10 years earlier, causing some
to have a significant drop in their economic wellbeing.
The data from the post-resettlement household survey combined with data on
entitlements and compensation provide further insight into who received the special
allowance and, as well, the AHs that are continuing to cultivate affected land; and,
more generally, the food security strategies of non-titled AHs (Table S-2). As context,
over 80% of non-titled AHs have very little food security with 45% having only 0-3
months sufficiency and 37% having 4-6 months sufficiency. The following are three
“case studies” of non-titled AHs; however, from available information, it is not possible
to explain what appear to be several divergences from policy.
Table S-2: Case studies of non-titled AHs

Access road
Three non-titled AHs have not moved and are still cultivating affected land. Two have agricultural land elsewhere.
The third, a sharecropper, has identified his other available land as “sharecropping”. None of these non-titled AHs
received the special allowance, including the sharecropper. They all identified unspecified labor as their strategy to ensure
food security.
Two additional female non-titled AHs have moved and there is no post-resettlement information for them.
Beteni/Banchare
In this area, there are 12 non-titled AHs in Rishing and three non-titled AHs in Vyas. Of these, nine AHs (60%)
have not moved and are still cultivating affected land. Six AHs have agricultural land elsewhere, while three AHs are
sharecroppers who identified other available land as “sharecropping”. Two sharecroppers received the special allowance,
but the others did not, including the third sharecropper.
Three non-titled AHs (20%) have not moved from affected structures, although they are no longer cultivating
affected land.
Among most non-titled AHs that have not moved, unspecified labor or employment are their food security
strategies. Two AHs appear to be employed as laborers on the THP and one woman is employed by THL.
Three other non-titled AHs (20%) have moved. Two are identified as having other agricultural land but there is no
post-resettlement information for the third.
Businesses and other AHs in Rishing Patan
In Rishing Patan in the reservoir area, there is a cluster of 12 non-titled AHs. Four of these AHs (33%) have
affected businesses, including one owned by a woman and three by men. They may also be cultivating trees or other crops
as three identified having other agricultural land and several were compensated for trees. Of these four AHs, two received
the special allowance; they included one AH who identified “sharecropping” as other agricultural land and the woman
business owner although she identified having other agricultural land.
Among the eight other AHs, five have not moved from affected land. Three AHs remaining on affected land are
still cultivating it, but two AHs are not. Three of these AHs received the special allowance, including two who identified
11
These data are taken from the post-resettlement household survey that interviewed 91% of 558 AHs.
62
Table S-2: Case studies of non-titled AHs
not having other agricultural land and one who identified “sharecropping” as other land. The other two AHs who have
other agricultural land did not receive the special allowance.
Three AHs (25%) have moved. There is no information about their other land holdings but all three received the
special allowance.
Most non-titled AHs with businesses identified employment (versus labor) as their food security strategy, including
one AH who is employed by a THP contractor. There is limited information on the food security strategies of other non-
titled AHs in this area, although several rely on remittances, and one earns money from rental property and the sale of
dairy products.
In the Dalit communities of Wangtankitar and Kapardi, non-titled households that may
be affected by a buffer zone around the reservoir have initiated the process with the
District Land Commission to obtain registered title to their lands.12 Magar households
in Jalbire who may also be affected by the buffer zone already have land titles. THL is
proposing to the Dalit households that they channel their title applications through THL
or Bhimad Municipality.
4.17 POE comment and recommendations
Particular attention should be paid to non-titled AHs who are continuing to cultivate
project-affected land, in particular those in the reservoir area, to ensure they receive
adequate information and assistance to transition to other agricultural land and/or non-
agricultural economic activities prior to the reservoir impoundment. Further, THL
should adopt an approach to assisting non-titled households who may be affected by
the buffer zone to acquire title to their landholdings, that is impartial and does not
pressure households.
Recommendation 2.4/1: The actions recommended above with respect to
titled AHs continuing to cultivate affected land in the reservoir area should be
extended to include non-titled AHs, as targets for the LRP, in external
monitoring surveys and in THL’s communication strategies to advise people
about the reservoir impoundment.
Recommendation 2.4/2: THL should negotiate with a Nepali NGO with legal
expertise and/or with a lawyer/law firm to consult with non-titled households in
Wangtankitar, Kapardi and any other communities that may be affected by the
buffer zone and represent them before the District Land Commission and/or in
any other negotiations related to obtaining land titles. This should be done
immediately to expedite the opportunity for households to obtain title regardless
of whether their land is identified as affected in the buffer zone study.
THL should adopt this approach rather than intervening directly or through
municipal officials, as proposed. The process of assisting potentially affected
households to obtain land titles should be independent of the project proponent
and local political leaders; and it should comply fully with the principles of
impartiality, transparency, and accountability as well as free, prior, and informed
consent.
4.18 Community assets
In the HP project area, a number of community assets will be affected, including six
public resting places (Thati), nine cremation sites (Ghat), two temples (one private and
12
Magar households in Jalbire who may also be affected by the buffer zone already have title to their potentially affected lands.
63
one used by the Magar community in Beteni), two water sources, seven suspension
bridges across the Seti River, and eleven foot trails (THL, 2018b). The resting places
have been replaced at sites agreed with communities. However, affected households
and communities remain concerned about the restoration of other communal assets.
In January 2022, the Bhimad Local Consultative Forum (LCF) was informed by THL
that three cremation sites would be restored following the completion of a cultural
impact assessment (CIA) that was projected to be completed within 2-3 months (THL,
2022a). The POE understands, however, that THL has put a hold on conducting the
CIA. Further, the POE understands, the buffer zone study as well as the THP
operational plan that includes annual flushing of the reservoir will have an impact on
how cremation sites can be re-established around the reservoir.
Throughout discussions the POE social expert had with THL and ESMSP staff and
people living in the HP project area, the issue was raised about the need to replace
affected suspension bridges and foot paths and/or ensure adequate means for local
people to continue to move along either side of the reservoir and/or across the
reservoir. These movements are essential to the daily lives of many people in the
reservoir area, most of whom currently walk across the river to reach agricultural
and/or grazing land, to maintain family relations, to access markets, as well as schools,
health care and other public services, etc. Several options were discussed, including
catenary suspension bridges,13 floating bridges and ferries and/or water buses that
connect multiple points along both sides of reservoir. However, a solution will also
depend on the THP operational plan.
Recommendation 2.5/1: With respect to both cremation sites and cross-
reservoir access, the ESMSP should work in close collaboration with
community leaders, Community Forest User Groups (CFUGs) and residents of
villages and settlements along both sides of the reservoir to document and map
the preferences for i) the location of cremation sites and ii) “desire lines” that
reflect people’s movements along each side of the reservoir (river) and from
one side to the other, as well as the reasons for and frequencies of these
movements. In consulting communities, different groups should be consulted,
including different ethnic groups/castes, women and men, children/students,
the elderly, farmers, people with livestock, laborers, businesspeople, etc.
Recommendation 2.5/2: THL should expedite the preparation of an
operational plan for the dam, including the proposed annual flushing of the dam,
in order that concrete steps can be taken, in consultation with affected
communities, to develop plans and an implementation schedule for
replacement of cremation sites, foot paths and suspension bridges, as well as
other affected community assets.
4.19 Transmission line - Resettlement entitlements and compensation
A 34-km 220-kV transmission line (TL) is being constructed from the THP powerhouse
in Tanahun District (Gandaki Province) to the Bharatpur Substation located in Chitwan
District (Bagmati Province). It traverses Vyas Municipality and four Rural Municipalities
(RMs) in Tanahun District and Bharatpur Metropolitan City (MC) in Chitwan District.
13
In 2020, the Baglung Parbat pedestrian bridge opened in Gandaki Province; at 567 m, it was then the world’s longest catenary
suspension bridge. Since then, two longer suspension bridges have been built in Italy (586 m) and Czech Republic (721 m).
64
As of the end of Q3 2022, the detailed cadastral survey was complete, foundation
works were complete for 48 towers of a total of 94 towers, and 25 towers had been
erected (Tractabel, 2022).
Initially, a total of 253 households were identified as affected by the acquisition of
private land and/or restrictions on land use within the 30-meter right-of-way (ROW) for
the TL14 (THL, 2020). THL has recently provided the POE social expert with an
updated assessment of land acquisition impacts involving a total of 353 households
(Table S-3).
Table S-3: Summary of TL land acquisition and resettlement impacts

Tanahu Chitwan Total
Anbu Khaireni RM
Vyas Municipality
Bharatpur MC
Bandipur RM
Devghat RM
Impacts
AHs
Agricultural land: Area (ha 0.65 0.36 0.08 0.10 0.21 67 1.5 ha
Permanently acquired for
tower foundations Number AHs 31 15 5 3 7 61
Agricultural land: Area (ha) 9.79 5.43 1.00 1.37 1.68 186 21.2 ha
Restricted land use in
ROW Number AHs 154 83 25 16 14 292
Structures - cowsheds 3 3 (560 sq.ft.)
Trees 36 trees
Source: DMS data provided by THL
Based on aerial photos, approximately 24 km (70%) of the TL passes through forest

land. At present, THL believes that the majority of forest land is state-owned; some
may be part of community-owned forests, but there is as yet no assessment of impacts
of the TL on Community Forest User Groups (CFUGs).15
4.20 Status of entitlements and compensation along TL
The Compensation Determination Committees (CDCs) in the two districts have set
compensation rates for the permanent acquisition of agricultural land and trees, as
well as allowances for vulnerable AHs and for lost fruit and crop production. The POE
social expert understands that by the end of 2022, the rates and relevant allowances
will be set to compensate restricted uses within the ROW and for affected structures.
As of November 2022, compensation for permanently acquired agricultural land and
affected trees has been paid to 51 AHs (84%). The remaining AHs with affected land
to be permanently acquired have not received compensation for several reasons: they
have not filed an application for compensation; their case has been referred to the
courts; etc.
14
Owners of land within the ROW will be required to remove any structures as well as trees/plants that exceed 3 meters in height.
They will be allowed to cultivate annual and perennial crops (that are less than 3 meters at maturity) and to graze livestock within
the ROW.
15
This information was communicated verbally by a THL representative.
65
The Package 3 (P3) contractor has already completed more than half of the tower
foundations and erected more than one-quarter of the towers, work that requires
access to private land that is to be permanently acquired and/or will be located in the
ROW. Further, it is unclear how the ROW will affect resources in forests managed by
CFUGs.
Recommendation 3.1/1: THL communicate and consult directly with the AHs
that have not received compensation for permanently acquired private land to
negotiate mutually acceptable written agreements to file applications for
compensation and/or to expeditiously resolve obstacles to filing applications for
compensation. The objective is to enable THL to hand over tower foundation
sites to the P3 contractor while complying with resettlement policy that affected
people are fully compensated prior to civil works; at the same time, THL should
not exert undue pressure on AHs.
Recommendation 3.1/2: THL take steps to ensure that the following occurs as
expeditiously as possible: i) the CDCs reach decisions regarding compensation
rates for restricted land use and affected structures in the ROW, ii) AHs are
consulted and encouraged to file applications for compensation, and iii) AHs
are paid. The objective again is to facilitate access for the P3 contractor to
complete construction of the TL while complying with resettlement policy that
affected people are fully compensated and are not unduly pressured.
Recommendation 3.1/3: THL take steps to identify if the construction of the TL
affects forest resources managed by CFUGs and, if so, consult with the affected
CFUGs to determine the extent of impacts, if any, on their use of those
resources.
4.22 Grievance redress mechanism
This section looks at the project-specific grievance redress mechanism (GRM) as
proposed and implemented for the THP. It is based on discussions the POE social
expert had with staff of THL, the ESMSP, and the Project Supervision Consultant
(PSC); and reviews of the THP social and environmental safeguard documents (THL,
2012a; THL, 2012b; THL 2012c; THL, 2018a; THL, 2018b; THL, 2020a), various THP
SSMRs, and GRMs in other Nepali infrastructure projects.16
To provide a context for POE comments, a first sub-section briefly outlines the
objectives and principles of an effective GRM. An assessment of the THP GRM
database and the provisions for GRM as set out in THP social and environmental
safeguard documents is followed by an evaluation and recommendations.
4.23 Objectives and principles of an effective GRM
An objective of a GRM is to provide a predictable, transparent, and credible process
by which directly affected people can seek redress to ensure that their rights to
compensation and other benefits are respected and fulfilled; and directly and indirectly
affected people and communities can express concerns, obtain information, influence
16
See References – Social Report for the other Nepali infrastructure project documents.
66
decisions, and achieve remedies for a range of social and environmental issues that
may arise at different stages of project development, implementation, and operation.
Equally, an objective of a GRM is to avoid conflicts and reduce risks for the project
proponent. Early identification and attention to grievances reduce the risks that
complaints will escalate into judicial action or public controversy that, in turn, may
result in lengthy delays and associated cost overruns in project implementation.
A well-functioning GRM is a means to ensure fair, effective, and lasting outcomes to
social and environmental issues and to build trust and productive relationships
between local communities and the project proponent. An effective GRM embodies a
number of principles, including (Agarwal, 2013; CAO, 2008):
a) Impartiality: the design and procedures of the GRM should eliminate bias in
favor of any party; and should be open to addressing concerns arising from
perceptions and misperceptions as well as those that are based in fact.
b) Accessibility: the GRM should be accessible to individuals, groups of individuals
and/or institutions/organizations via multiple local locations and modes of
access and with procedures that are understandable to people, without
discrimination based on sex, ethnicity/caste, age, education, disability, etc.
c) Proportionality: the GRM should address and resolve grievances and disputes
at the lowest possible level and in the shortest possible delay, commensurate
with the objective of achieving fair, effective, and sustainable outcomes.
d) Cultural and social inclusivity: the GRM should be designed and implemented
in collaboration with the community, to ensure that it reflects and responds to
the needs and priorities of diverse cultural and social groups. For example,
culturally appropriate forms of negotiation and conflict resolution; gender, age
and/or ethnic/caste dimensions of cultural systems of authority, social rank, and
status; the intersectionality of cultural, social, and gender identities and
relations; cultural ways of understanding social and environmental issues; etc.
e) Transparency and accountability: the GRM criteria and procedures for
assessing and resolving grievances should be transparent, including
opportunities for complainants to participate in and be heard during the process;
accountability requires full disclosure to complainants of the outcomes and the
reasons for them and public-facing disclosure commensurate with any need to
protect the privacy of individuals or groups.
4.24 Analysis of grievances
Since 2014, there have been a total of 171 registered grievances, however many are
registered by groups that range from 2 to 29 individuals and/or households each; in
total, it is estimated there have been 350-400 complainants.17 In addition to individuals
and households, some grievances have been registered by institutions. It should be
noted that these grievances relate to the HP project area. While compensation
17
The number is imprecise because the number of households in all groups is not identified.
67
payments have been initiated for AHs along the TL, to date there is no information
about registered grievances from people in affected communities.
Table S-4 is THL’s analysis of the types of the registered 171 grievances, with the
status of addressing and resolving them: “in process” refers to assessments underway
to verify the validity of the grievance and “ongoing” means that redress actions are
being implemented.
Table S-4: THL analysis of registered grievances by type and status, HP project area
On- In Not
Category Completed Total
going Process Initiated
Compensation for Land 26 2 6 3 37
Compensation for Assets 12 4 3 0 19
Livelihood Restoration (i.e., allowance for
54 0 0 3 57
affected crops/trees)
Replacement of Impacted Structures 15 0 1 1 17
Community Infrastructure Damaged by
10 0 1 1 12
Construction Activities
Water, Sanitation & Hygiene 1 0 0 0 1
Environmental Damage/Protection 9 0 3 4 16
Community Health & Safety 4 0 2 2 8
Community Complains to ADB/EIB 0 2 0 0 2
Group Grievances to Change the Alignment of
1 0 0 0 1
Project Access Road (PAR) in Vyas-05
Group Grievances Mixed to Demands– 35 Points 0 1 0 0 1
Total 132 9 16 14 171
Distribution (%) 77.2 5.3 9.4 8.2
Source: ESMSP
A rapid analysis of the GRM database indicates that 40% of the registered grievances
come from individuals or groups of households who are identified by their AH code as
directly affected people, however, slightly more than 60% of registered grievances
deal with land acquisition and compensation issues. Further, 60% of registered
grievances come from people who are not identified as directly affected and/or not
identified as to their status with respect to land acquisition impacts; and it is not
possible to verify the total number of complainants, i.e., the numbers of people in
grievances registered by groups, or the number of unique complainants.
Among the registered grievances related to land acquisition entitlement and
compensation issues:
a) The largest number come from people living in the reservoir area (50%); 27%
come from the Beteni/Banchare area, 10% from the area of the access road
and THL camp and 13% are listed as “other” although it is not clear why.
b) By ethnicity/caste, the distribution is 65% Janajati, 4% Dalit, and 31% “other”
that is assumed to include, among others, Bhramin and Chhetri.
68
c) By sex of the person recorded as registering the grievance, men outnumber
women by a factor of more than four.
d) The completion rate for grievance redress for issues specifically related to
entitlements and compensation is 87%, with only 5% of registered grievances
deemed ineligible. Grievance redress that is “in progress” or “ongoing” is 8% of
registered grievances related to entitlements and compensation. However, it is
not clear what “completed” means.
A second category accounting for 11% of registered grievances concerns problems
related to construction activities such as blasting that produces noise and/or damages
structures and truck traffic, dust, and pedestrian safety along roads. Most grievances
are registered by people living in areas close to the construction works for the dam
and the powerhouse. Where the information has been recorded, 38% of the
grievances are registered by women. Of all the construction-related grievances, 21%
have been disallowed (i.e., “not initiated”), primarily because they related to issues
within the mandate of the National Electricity Authority (NEA).
A third category of grievances deal with environmental management issues and
account, overall, for 9% of registered grievances. They are evenly divided between
grievances registered by households and by institutions. Household grievances tend
to focus on drainage management and embankment protection; the majority (63%)
are currently being investigated or have been disallowed with the comment that the
issue may be addressed during project construction works. Institutional grievances
include the need for a conservation plan for 200 households located in the upper
portion of the reservoir area in Bhimad, landslide protection for an irrigation scheme in
Rishing Patan, a gabion wall to protect the community and community forest land in
Belbas and, downstream of the dam, protection against scouring at the junction of the
Madi and Seti rivers; most are indicated as completed or in the process of evaluation.
Slightly more than 20% of registered grievances have come from groups of
households. The strategy of group grievances, as reported to the POE social expert,
is the solidarity and confidence of working together combined with a perception of
being more effective as a group in achieving the desired outcomes. Several of the
grievances registered by groups are those made directly to the ADB Accountability
Mechanism and EIB Complaints Mechanism by Magar and Dalit communities.
4.25 POE Comments
Overall as a tool, the GRM database does not fully provide the scope and precision of
information necessary to enable complainants as well as THL to track and monitor the
process of grievance redress, or to ensure full transparency and accountability of the
GRM. The following observations will be addressed in recommendations in Section
5.4, below.
 The GRM database is incomplete, with missing, inconsistent and/or imprecise

data. As examples, a) the transmission line should be added to project
components; b) “other” under ethnicity/caste should be clarified and/or
expanded to identify specific other groups; c) the designation of sex should be
clear as to whether this refers to the land title holder, the head of household
and/or the person filing the grievance; d) grievances registered by groups
should clearly and consistently identify the number of people in the group, as
69
well as the number of discrete grievances; e) “Lokma residents” and “Kapardi
residents” are listed as institutions.
 The grievance code assigned to each registered grievance appears to be linked

to the municipality and village of the complainant; further, when a group
registers a grievance, there is a single grievance code. The POE social expert
believes this is insufficient, and that a) the grievance code should also include
reference to the AH code, where relevant, to facilitate the ability of AHs to track
action on their grievances; b) in cases where the complainants are not AHs, a
different personal identification should be incorporated into the grievance code;
and c) members of groups registering grievances should each be given a
unique grievance code, particularly when different members of the group have
different grievances, e.g., about the amount of compensation received. Note: If
there is an overlap at village level between AHs affected by the HP and TL
components, e.g., in Vyas Municipality, it may be necessary to incorporate that
distinction in a grievance code.
 The system of classifying the “type of complaint” requires revision to promote

greater consistency in how grievances are categorized and to clarify the
analysis of grievances. Major categories should be established and used
consistently; for example, grievances related to a) land acquisition entitlements
and compensation grievances, b) construction-related grievances, c)
environmental management/ protection grievances, d) post-construction
grievances (e.g., related to the operation of the dam/ reservoir).
Within each of these categories, sub-categories should also be established. For
example, land acquisition entitlements and compensation grievances may be
sub-divided by grievances related to entitlements, the accuracy or amount of
compensation and/or allowances, and activities related to the livelihood
restoration plan, the community development plan and/or the gender equality
and social inclusion plan. Construction-related grievances may be sub-divided
by short-term impacts, e.g., dust, noise, pedestrian safety, versus grievances
that require technical assessment, e.g., damage to structures. Environmental
management grievances may be sub-divided according to the different
management plans for the THP. The information under “description” in the GRM
database can be used to provide relevant further information about specific
grievances.
 The information on the status of grievances is insufficient, particularly as it

relates to the “completed” category. It is important to understand the nature of
the decisions made. For example, was a household correct that their land had
not been properly measured, or not; was it confirmed they received their full
compensation or were they awarded additional compensation? Similarly, were
households that complained about damage to their houses due to the project
correct or not? What happened when people complained about dust or the
safety of pedestrians?
4.26 Overview of THP Grievance Redress Mechanism
The following is a brief description and, by way of comments, assessment of the
project-specific GRM for the THP as presented in the various social and environmental
70
safeguard documents cited above and discussions with ESMSP and PSC personnel.
The GRM comprises several levels, with some time-bound schedules.
Level 1: At the “local” level, there appear to be several approaches proposed to initially
address complaints and grievances from AHs and other stakeholders:
a) Public Information Centre (PIC): The PIC will be the first level to address
grievances (THL, 2012a; THL, 2018a; discussions with ESMSP): Full-time PIC
staff hired by THL receive and attempt to resolve complaints and grievances by
providing information to AHs and other stakeholders.
b) Local Consultative Forum (LCF): The LCF will be the first level to address
grievances (THL, 2018b): As noted previously, the LCF is constituted of elected
municipal and ward officials, local government staff, representatives of directly
affected people; and supported by ESMU and PIC staff.
c) Land Acquisition and Rehabilitation Cell: The Cell will be “solely responsible
to be in close contact with all project-affected people and public”, and it will
resolve issues at the field level within 7 days (THL, 2020a). The Cell will also
work “in coordination with the construction contractor”.
d) Civil works contractor: The PSC reported receiving information from the
Package 1 (P1) contractor’s Health and Safety Manager that indicates that
some people living in the vicinity of construction activities are addressing their
complaints directly to the contractor, as often as several times per week with
recent complaints focusing on the nuisances and damages related the
contractor’s open surface blasting activities.
As per a follow-up assessment report received from the PSC (ELC, 2022a), the
approach to addressing one set of complaints about blasting activities involved
the ESMSP, ESMU and the P1 Health, Safety and Environment (HSE), with the
following steps:
i) in mid-July 2022,18 community members filed a written complaint about
noise, dust, house cracks caused by blasting, and restricted mobility due
to project activities;
ii) ESMSP visited the community in early September 2022 to update the
situation; and over several days in early November 2022, the ESMSP,
ESMU, and P1 HSE returned to the community and
iii) conducted tests to observe and/or measure noise and dust (both
exceeding permissible levels), vibration cracks (found in 4 residential
structures), several road cracks that raise the risk of landslides and the
functioning of the blasting siren system (unsatisfactory); and
iv) consulted with community members (including several women, as
shown in pictures). The recommended actions included informing the
community one day prior ahead and at the time of blasting activities,
using hand mikes and electric sirens; continued practice of providing
safety helmets and “use of vehicle for mobility of pedestrians”; and
continued regular monitoring of noise, dust emissions and “possible
structural damage due to project activities”.
18
Nepali date: 2079-03-31
71
4.27 POE Comments
With respect to the existing modalities of an initial level of grievance response and
redress.
 There is a lack of consistency in the approach to the initial step in addressing

grievances and complaints within and across the different components of the
THP.
 The PIC approach is being employed in the HP project area, as described by

the ESMSP to the POE social expert. It is understood that when a
complaint/grievance is presented to a PIC, it must be in written form and
accompanied by supporting documents and must be addressed within 7 days
of receipt. As necessary, the PIC will refer the matter to the ESMSP to conduct
relevant social and environmental assessments and/or to the ESMU to go
onsite to inspect, verify the claim and make arrangements for implementing
remedies.
 A LFC has been established in each of the four municipalities/RMs in the HP

project area. However, as noted previously, the POE social expert was unable
to identify any instances in which a grievance or complaint has been referred
to and/or adjudicated by the LCF.
 Also as previously noted, no PIC or LCF has been specifically established in

the jurisdictions traversed by the transmission line.
 With respect to some construction-related grievances, there has been

coordination between civil works contractors and the ESMSP and ESMU to
verify and assess claims. That said, the reporting does not indicate or explain
the following: the delay – from July to November – in addressing the complaints;
or if the owners of the damaged houses received any cash and/or in-kind
compensation; if measures are being taken to prevent landslides caused by
road cracks, that may inflict damage on private property/structures; or if
remedial measures are being taken to address the excessive noise and dust
problems, such as limiting the times of day of blasting activities and watering
roads.
 To ensure transparency and accountability in the GRM process, it is important

that each grievance is registered, recorded, and processed properly; that
complainants have opportunities to make representations in person or through
their chosen representatives regarding their grievances/complaints; and that
they are promptly and fully informed in writing of the receipt of their grievance,
the results of any technical assessments and the final decision, including
reasons for the decision.
 The construction-related impacts cited above is an example of a community

consultation conducted as part of the grievance redress process. However,
nowhere in the THP social safeguard documents is there any mention of these
matters. Examples from other infrastructure projects in Nepal suggest the
following19: complainant(s) are informed within 3 working days to acknowledge
19
See References – Social Report, in particular ADB, 2021.
72
receipt of the grievance/complaint, provide the unique registration number, and
set up a meeting to discuss the grievance; at the meeting, the complainant(s)
may come with representatives of their choice and there is an option for
interpreters, as required; minutes of any meetings are prepared and signed by
both parties and appended to the complainant’s file; and the complainant is
promptly informed with a decision is made including the reasons for the
decision. Communications with complainants should be in writing, preferably in
the language of choice of the complainant, although other means may be
required with complainants who are not literate, etc.
 THP social safeguards documents for the HP and TL components indicate that
grievances at the local level must be addressed within 7 days of receipt.
However, it is clear from the above reporting on construction-related impacts,
that it took several months to conduct the field work and testing to assess the
validity and scope of the complaints.
Level 2: As per a review of the THP social and environmental safeguards documents,
there is consensus that if a resolution to a grievance is not achieved at the local level,
it is referred to the THL ESMU. The complainant is informed of this and, if not already
done, must provide a written complaint with appropriate documents to substantiate the
complaint. The ESMU is responsible for registering the complaint, making the
necessary inquiries and verifications, and resolving grievances “through ongoing
interaction with the affected persons” (THL, 2018a).
The ESMU Chief and the Social Safeguards Specialist and Public Relations Officer of
the ESMSP are identified as having key roles (THL, 2020a) and as required, the ESMU
may seek assistance from the supervision consultant’s safeguard specialist (THL,
2018b). The 2018 Consolidated EMP also suggests a role for NGOs at this state of
grievance redress (THL, 2018a).
Social safeguards documents for the HP component indicate that if the ESMU is
unable to reach agreement with the complainant, “the issue will be discussed in the
LCFs”. The 2018 Consolidated EMP indicates that the ESMU has 7 days to address
complaints/grievances at this stage; the various resettlement plans for the HP and TL
components indicate 15 days.
4.28 POE comments
During this mission, no information was received about grievances that had advanced
to this second stage. The grievance register does not clearly provide this information
and there was no information forthcoming from informants. Therefore, the following
comments relate to the activities as proposed in project documents.
 The referral of grievances that are not resolved at the local level to the project
proponent, in this case THL’s ESMU, is consistent with other recent
infrastructure projects in Nepal.20
 However, the “trigger” for moving a grievance to level 2 is not always clear in
the various THP social and environmental safeguards documents.
Resettlement plans clearly state that “if the issue is not addressed within 7 days
20
See References – Social Report.
73
of written application to the satisfaction of the affected person”, the grievance
moves to level 2, but the 2018 Consolidated EMP is less clear.
 The referral of Level 2 cases to the LCF when there is no agreement between
the ESMU and the complainant is consistent with some other recent
infrastructure projects in Nepal (e.g., NEA, 2016). However, as with level 1
grievances, the POE social expert was unable to identify any cases of
grievances being referred to the LCFs established for the THP.
 The POE social expert was unable to identify any cases of grievances where
NGOs have been involved in representing AHs and/or negotiating resolutions,
with the exception of the grievances that members of three communities have
filed directly with the ADB and EIB formal complaint mechanisms.
 There is a lack of consistency between THP social and environmental

safeguard documents regarding the time frame for this level of the GRM. The
various resettlement plans indicate a 15-day window while the 2018
Consolidated EMP identifies 7 days.
 The POE comments related to the registration, recording and communications

about grievances at level 1 apply equally to level 2.
Level 3: There is consensus across THP social and environmental safeguards
documents that anyone who has filed a grievance or complaint that has risen to level
2 can, if they are not satisfied with the outcome of the GRM process, appeal the
decision.
4.29 POE comments
There is, however, a lack of consistency for the appeal process, in particular:
 THP social safeguards documents identify appeal to the CDC at the district
level; the CDC is a statutory body mandated under the Land Acquisition Act
(LAA) 1977. Its membership includes the Chief District Officer (CDO) as CDC
Chair, the Chief of the Land Revenue Office, and representatives of the District
Development Committee (DDC), relevant Municipalities/RMs, and the
Executing Agency, that is, THL. The initial resettlement plan for the HP
component also identifies representatives of affected people as members of the
CDC (THL, 2012c).
 The 2018 Consolidated EMP stipulates appeal to a Grievance Redress

Committee (GRC) that is chaired by the Project Director (i.e., THL), with other
members from relevant Municipalities/RMs, affected people, NGOs, and the
“local area committee” (THL, 2018a).
 All social and environmental safeguards documents identify a 15-day time

frame for this stage of the GRM, with the exception of the RIPP for the TL
component that identifies a 30-day time frame (THL, 2020a).
i) Appeal to the Ministry of Home Affairs: If within the prescribed time frame the
CDC is unable to resolve social and/or environmental grievance issues, the LAA
provides that people and communities in the project area may appeal to the
Ministry of Home Affairs. This is acknowledged in the resettlement documents for
74
the HP component (THL, 2012c; THL, 2018b). In the case of the TL component,
the resettlement plan identifies appeal to the district court (THL, 2020a).
ii) Access to GRM free of cost: The ADB safeguards policy (ADB, 2009) stipulates
that affected people may seek redress through a project-specific GRM at no cost
and without retribution. This provision is explicitly included in the resettlement plans
for the HP component (THL, 2012c; THL, 2018b), but is missing from the RIPP for
the TL and the 2018 Consolidated EMP (THL, 2020a; THL, 2018a).
iii) Reporting on the GRM is included in the SSMRs that have been prepared on a
semi-annual or quarterly basis, with varying degrees of detail.
4.30 Evaluation of THP Grievance Redress Mechanism and
recommendations
The GRM for the THP has encompassed grievances related to three major categories
of complaints. The process of addressing and resolving minor grievances at the local
level via a contractor’s Health and Safety Manager is a good example of
proportionality. Nonetheless, aspects of the structure, procedures, and resources of
the THP GRM do not fully comply with key principles of an effective GRM, namely
objectiveness and independence, accessibility, responsiveness, proportionality, and
participation.
In evaluating the THP GRM, the approach taken is to assess how it is currently
functioning, what may be needed to ensure that it continues to provide appropriate
redress throughout all phases of the development and operation of the THP, and how
it can establish a model for future THL projects including the Lower Seti (Tanahu)
Hydropower Project.
4.31 Scope and structure of the THP GRM
The scope of the THP GRM should encompass all stages of project development and
operation; and the structure should facilitate comprehensive and consistent reporting
and analysis of a range of social and environmental management issues.
Recommendation 5.4/1: Review and, as required, revise the scope and
structure of the THP GRM i) to encompass all stages of the development,
construction, and operation of both the HP and TL components; and ii) to
establish major categories and sub-categories of grievances and complaints.
As part of the review and revision of the THP GRM, i) clarify and/or expand the
information reported in the GRM database including, among others, information
on ethnicity/caste and sex of complainant(s); ii) multiple types of grievances
registered by groups of people should, to the extent feasible, be reported in the
GRM database and other records (see below) in a manner that clarifies the
distinguishes the types of grievances and the number of people involved; and
iii) clarify and expand the information reported in the GRM database regarding
the status of the grievance/complaint, in particular to specify the decisions
and/or actions taken to resolve the grievance/complaint and why (see also,
recommendation below regarding the grievance/complaint file).
75
4.32 Procedures for lodging, registering, tracking, and reporting grievances
The procedures for lodging, registering, tracking, and reporting grievances should be
accessible, socially, and culturally acceptable, and transparent and accountable. In
the review of the GRM, the POE social expert was not able to corroborate all aspects
of how it functions presently. The following recommendations are made to increase
the accessibility, transparency, and accountability of the GRM and to ensure that it is
socially and culturally acceptable; they reflect good practice in handling grievances
and are offered to the extent that measures are not already in place for the THP.
Recommendation 5.4/2: THL should expand the ways that people can lodge
a grievance/complaint.
In addition to having to travel to a PIC, THL should consider ways that people
can lodge a grievance/complaint with a person in their village/settlement
designated by THL to collect the necessary information; the types of
grievances/complaints that could be lodged through the THL website, by phone
at numbers provided, by submitting a note in a suggestion box and/or by
sending a letter; and the types of grievances/complaints that can be lodged
directly with civil works contractors.
Recommendation 5.4/3: THL should assign a unique code to each registered
grievance and complainant.
The code should i) enable complainant(s) to track the progress of adjudication
and decisions regarding the grievance/complaint and ii) facilitate tracking and
analysis of grievances/complaints by major categories as well as municipality
and village. In cases of groups registering grievances, unique codes should be
assigned to discrete grievances and the individual(s) involved.
Recommendation 5.4/4: THL should ensure that a grievance file (electronic
and/or paper) is created to track the grievance/complaint.
The file should be linked to the unique code and should contain initial
information collected from the complainant including copies of any supporting
documents provided by the complainant. The file should be updated with
relevant information at each stage of assessment, verification, adjudication, etc.
of the grievance/ complaint, with entries by THL staff signed and dated. When
a decision is reached regarding the grievance/complaint, a full description of
the decision and the reasons supporting it should be entered into the file, with
THL staff signing and dating the information; this information should be included
in the GRM database under the heading “status”.
Recommendation 5.4/5: THL should ensure that the person(s) making a
grievance/ complaint are advised about the process to verify and assess their
claims and how they can track the progress of this process and the outcomes,
in person and/or through access to written records.
To this end, within three days of a grievance being lodged, THL should assess
its eligibility under the GRM and communicate this information in person and/or
in writing to the complainant(s); if the grievance is eligible, the complainant(s)
should be provided with the unique code(s) as well as information about when
76
and how to make representations and/or obtain information about the status of
the grievance/complaint, in person and/or in writing.
4.33 Grievance redress process
The GRM process adopted by THL has delegated responsibilities to ESMSP and
ESMU to conduct, as required, social and environmental assessments, field
inspections, etc. to verify and determine the nature of actions required to redress
grievances; and, as evidenced by the construction-related grievances described
above (Section 5.3), to consult with and provide information to complainants and other
community members. These are important elements of an effective GRM.
Nonetheless, the POE social expert is concerned there is a lack of impartiality,
transparency, and accountability in a grievance redress process that, at the local level,
is managed and controlled by the project proponent; and, by not integrating the LCFs
and, in some instances, civil works contractors into the process, does not fully reflect
the principles of being socially and culturally appropriate or proportional.
Recommendation 5.4/6: THL should review and, as necessary revise, the
process of grievance redress to mandate the LCFs in each municipality/RM to
be the decision-making body for Level 1 adjudication of grievances, particularly
for grievances related to the entitlements, compensation, livelihood restoration,
and other assistance for people directly affected by land acquisition and/or
commitments made in various THP environmental management plans.
In this capacity, the LCFs would convene public meetings at regular intervals
(e.g., every two weeks or less frequently as warranted), to which complainants
and/or their representatives would be invited and offered an opportunity to
make presentations and/or ask questions. The LCFs would continue to receive
technical support and advice from the ESMSP and ESMU; and the PICs would
continue to be the liaison between people in the communities and the project.
However, the LCF as a more independent and impartial community-based body
with representation of affected people as well as local leaders and experts
would be responsible for final decisions regarding redress of these rights-based
grievances.
Recommendation 5.4/7: THL support the establishment of LCFs in the
municipality, rural municipalities and the metropolitan city affected by the
transmission line to address grievances.
Recommendation 5.4/8: With respect to grievances related to construction
activities and those related to the operation of the hydropower dam, THL should
review and, as necessary, revise the Level 1 process of grievance redress. In
the case of construction activities, it is recommended that THL consult with civil
works contractors to agree on the types of grievances that can best be
addressed to and resolved directly by contractors versus THL, along with how
THL and contractors can collaborate on technical assessments, etc. In the case
of grievances related to the operation of the dam, it is recommended that THL
consider extending the mandate of the LCFs and/or constituting a new
consultative forum that draws on local government officials, relevant
district/provincial government departments (e.g., Department of Forests, etc.),
and representatives of community organizations and NGOs.
77
Recommendation 5.4/9: Following review and revision of the GRM, THL
should take steps to i) provide adequate training for THL staff, members of
LCFs, contractors’ representatives, affected people and others involved in
aspects of the GRM, to ensure that they have the knowledge, attitudes and
practical skills to support an effective GRM; and ii) develop and disseminate
written materials (in local languages) and conduct community consultations at
the ward and/or village levels, to ensure that people understand clearly how
different types of grievances will be addressed and how they can lodge, make
representations and track grievances.
4.34 Project employment
The entitlements for directly affected people stipulate that one person from each
household will “…as far as possible receive priority for temporary employment at
project-related construction works, based on their capability, skills, and qualifications.
Specific skill training for them may be provided….” (THL, 2018b). This reflects a
commitment made in the 2017 21-point agreement that “affected people will be
provided employment opportunities as per their capacity and skill on a priority basis”
and “the project will recommend (to contractors) to mobilize capable locals in
construction works.” (THL, 2018b).
The target for local employment on the project as set in the Design and Monitoring
Framework (DMF) and the initial Gender Equality and Social Inclusion Action Plan
(GESIAP) (ADB, 2013) is 1,000 people, of whom 30% are women. In the following
analysis, local people are considered to be people living in Tanahun District.
As of the end of Q3 2022, a total of 1,230 people were employed on different
components of the project, of which 58 were women (4.5%). The contractors for the
three construction packages were employing a total of 1,164 people, of whom 299
(26%) were local people; 18 women accounted for 6% of local people (Table S-5)
.
Table S-5: Analysis of employment for THP construction packages, September 2022
Women Men Total
Tanahun Other Foreign Tanahun Other Foreign Tanahun Other Foreign
District Districts District Districts District Districts
Package 1 6 18 0 94 446 56 100 464 56
Package 2 12 0 1 186 303 39 198 303 40
Package 3 0 0 0 1 1 1 1 1 1
Totals 18 18 1 281 750 96 299 768 97
Source: Tractabel, 2022. PSC Quarterly Report, No. 29, July, August, and September 2022
Data for the end of Q2 2022 indicate that 298 people from Tanahun District were
employed across all components of the project, including 41 people or 14% from
directly affected households (Table S-6).
Table S-6: Analysis of AH employment on THP, June 2022

Tanahun District Members of directly affected households (AHs)
AHs Non-AHs Women Men Skilled Unskilled Full time Part time
Package 1 4 35 0 4 4 0 2 2
Package 2 22 177 5 17 22 0 21 1
Package 3 0 5 0 0 0 0 0 0
78
Table S-6: Analysis of AH employment on THP, June 2022
Tanahun District Members of directly affected households (AHs)
AHs Non-AHs Women Men Skilled Unskilled Full time Part time
THL (site) 12 16 2 10 10 2 12 0
PSC 2 18 1 1 2 0 2 0
ESMSP 1 6 1 0 1 0 1 0
Totals 41 257 9 32 39 2 38 3
Source: THL, 2022. Social Safeguard Monitoring Report, April-June 2022
In 2022, THL adopted a hiring policy that prioritizes gender inclusive employment of
directly affected people (THL, 2022g); and committed to the communities of
Wangtankitar and Jalbire to share 6-month plans projecting contractors’ human
resources requirements for skilled and unskilled labor (THL, 2022b). The focus of
these initiatives has been to increase the representation of directly affected people
and women in the workforce of the Package 2 contractor, as the construction of a
concrete dam has the potential to create a significant number of temporary unskilled
and semi-skilled jobs, for example, to construct formwork for the dam. However,
according to the ESMSP, THL and/or the contractor have not yet developed any six-
month workforce projections.
Based on the above analysis and a review of social safeguards documents including
Social Safeguards Monitoring Reports (SSMRs), the POE social expert has the
following comments:
 Despite monthly variations in staffing, the THP has recently been employing
more than 1,000 people. Nonetheless, a target to employ local people defined
as people living in Tanahun District is not being met; further, people from AHs
represent only 4% of total employment.21
 The target that women constitute of 30% of project employment is also not
being met; but may be an unreasonable target for a major hydropower project,
at least for employment with civil works contractors. A more nuanced target in
the GESIAP would have been better.
 The THL hiring policy (THL, 2022g) stipulates that civil works contractors
publish notice of required manpower at least 15 days prior to the
commencement of work. However, if there are 6-month projections of workforce
requirements, it should be possible to publish notices earlier; this would assist
interested people to make arrangements within their households to ensure that
other economic activities such as crop cultivation are not compromised by work
on the THP.
 The RIPP policy that one member of an affected household be considered for
employment on the project has not been respected. As reported in the SSMR
for July-December 2020 (THL, 2020d), there were several instances in which
up to four members of a directly affected household were employed in different
capacities on the project.
21
Based on a total project employment of 1,049 reported in the SSMR for Q2 2022.
79
Persistent feedback from directly affected people and other informants that the POE
social expert met related to unmet expectations that the THP would provide
employment for local people. This may be the result of a combination of factors
including lack of knowledge and, therefore, unrealistic expectations in local
communities about the number, type, and duration of employment opportunities for
unskilled and semi-skilled labor; lack of clarity and communications by THL to inform
local communities about opportunities and how to access them; and lack of clear
requirements and targets in bidding documents and civil works contracts regarding
local employment.
Recommendation 6/1: THL and the Package 2 contractor take immediate
steps to develop rolling 6-month projections of the workforce requirements,
including clear information about the required qualifications and the duration
and other relevant conditions.
Recommendation 6/2: THL and THP civil works contractors collaborate to

publish notices and widely disseminate information through other channels
about available project-related employment, at least 1 month prior to the start
date of the work. Wide dissemination of information should include community
radio announcements, information provided to village-level community leaders,
youth clubs, savings groups, etc., as well as via PICs and LCFs.
Recommendation 6/3: THL review and, as required, revise and/or develop and
implement a strategy to assist affected people and other local people who are
interested in employment opportunities to secure employment. The scope of
the review should include i) how THL identifies AHs where members are
interested in employment, validates their qualifications, and collaborates with
civil works contractors to link them to available positions; and ii) how PIC staff
in each municipality/RM as well as at the THL site can be resourced and tasked
to assist interested people to complete application forms, prepare for interviews
with contractors, etc.
Recommendation 6/4: THL review and take steps to reinforce and enforce the
RIPP policy that one person per AH be considered for employment on the
project, in order to make project-related employment available to as many AHs
as possible.
4.36 Livelihood Restoration Plan
The ESMSP in currently in the process of finalizing a Livelihood Restoration Plan
(LRP) as a component of THL’s commitment to assist directly affected people and
households to restore if not improve their livelihoods and incomes. In order to establish
priorities for people to target via the LRP, ESMSP has undertaken a comprehensive
household survey that interviewed 91% of AHs, conducted focus group discussions
(FGDs) with groups of affected people, and collected primary data from district level
stakeholders.
Three in ten AHs have moved since receiving their compensation, to locations
elsewhere in Tanahun District (52%), to Chitwan District (22%), and elsewhere in
80
Nepal. The 338 AHs that remain in the HP project area are concentrated in two areas,
around the project construction sites and in Bhimad Municipality (ELC, 2022d).
The objectives of the proposed LRP are i) to design appropriate, tailor-made livelihood
restoration activities that target the needs of the most vulnerable AHs still living in the
project area; ii) correctly match household economic activities that are land-, wage-
and/or enterprise-based through the provision of financial literacy and entrepreneurial
skills training as well as technical skills training; and map and prioritize support for
public services and facilities that have the most bearing on strengthening AH
household economic activities (ELC, 2022d). Factors in assessing the most vulnerable
AHs include project impacts such as the extent of land loss and physical displacement;
and socio-demographic factors including gender, age, disability, income, and food
sufficiency (ELC, 2022d).
The process of formulating the LRP is ongoing, however, several significant findings
of the baseline survey are noteworthy, in and of themselves and for their relevance to
other aspects of the THP social safeguard initiatives. They include (ELC, 2022d):
i) The reasons that AHs have not moved include the fact that the loss of agricultural
or other land has not significantly changed their livelihoods, or because they cannot
afford to migrate.
ii) One-quarter of the remaining AHs are headed by women, however, in the Magar
community, the proportion of female-headed affected households increases to
39%. The current overall rate is twice the proportion of 12% female-headed
households in the project area in 2018 (THL, 2018b); the principal reason cited is
the death of a spouse (ELC, 2022d).
iii) The training demands of men and women are similar in that agriculture is the top
priority as indicated by 44% of men and 45% of women. For men, the 2nd, 3rd, and
4th most important types of training are driving (18%), computer skills (15%), and
wiring and plumbing (12%). For women, the corresponding priorities are tailoring
(19%), beautician (15%), and computer skills (11%).
iv) The most common use of compensation money has been to upgrade existing or
build new houses (44% of all AHs; 40% of FHHs). Only 6% of all AHs and 4% of
AHs headed by women used compensation money to buy agricultural land.
v) Slightly more than half of AHs (55%) have post-resettlement living standards that
are similar to their living standards prior to land acquisition, based on their income
and the proportion of their landholding that was acquired. Three in ten AHs (29%)
are better off and the remainder (16%) are worse off, compared with their pre-land
acquisition living standards.
vi) Measured in terms of food sufficiency, however, the proportion of AHs with a
severe lack food sufficiency (<3 months) has increased from 22% to 37%. The
proportion of AHs with moderate food sufficiency (7-9 months) has increased from
5% to 11%, however the proportion of AHs who are food sufficient (10-12 months)
has dropped from 35% to 15%.
vii) In communities along the transmission line, unemployment was identified as a
concern by 55% of the people interviewed. More than six in ten people are counting
81
on the compensation package to improve their situation, and 22% expect to have
employment on the project.
4.37 POE comment
The ESMSP has adopted an evidence-based approach to the development of the
LRP. The POE social expert is aware, for example, that there is ongoing ground-
truthing of the data revealed through the household survey to ensure that the AHs who
remain the most vulnerable despite entitlements and compensation are the targets of
LRP initiatives. Further, ESMSP is establishing contacts with government services and
other resources in the THP project area to identify potential linkages for LRP initiatives.
However, at this stage, no information was available regarding the budget and other
resources that have been allocated to implement the LRP.
4.38 Community Development Plan
Prior to the development of the Community Development Plan (CDP), THL committed
just under US$750,000 for projects in response to community priorities;22 these
included about US$730,000 (97%) for different types of infrastructure, as well as funds
for skills development programs (2.5%) and school scholarships (0.5%). An analysis
of the distribution of these funds (Figure S-1) shows that the majority of funds were
invested in communities in the vicinity of the access road, THL camp site and the
powerhouse that were impacted by early THP construction activities (THL, 2022f);
disproportionately less was invested in Myagde RM compared to the proportion of
directly affected people.
Figure S-1: Distribution of CDP projects in HP project area
CDP Projects VS AH S
Vyas Rishing Myagde Bhimad
CDP Projects, 22% CDP Projects, 29%

AHs, 12%
AHs, 27%
CDP Projects, 5%
AHs, 41%
AHs, 20%
CDP Projects, 44%
Source; THL, 2022. Social Safeguard Monitoring Report, July-September 2022
The budget presented in the CDP document is 1.3 million USD. Most of these funds
(60%) are allocated for a range of investments in education, health, WASH and GESI;
22
Two school construction projects included in lists of THL pre-CDP commitments have also been included in the CDP budget.
To avoid double-counting, they are not included in the analysis of pre-CDP commitments.
82
they are distributed among affected communities in the HP project area including 30%
in Myagde RM, 26% in Bhimad Municipality, 23% in Rishing RM and 21% in Vyas
Municipality (ELC, 2022c). The remaining 40% are contingency funds that are
identified for infrastructure and other investments in communities in the reservoir area
(67%) and along the transmission line (33%).
4.39 Community infrastructure
THL has been highly responsive to requests (also referred to as “demands”) from
communities in the HP project area for funding to develop a range of infrastructure;
the allocation of funds included river protection works (54%), roads and drainage
(24%), a variety of community buildings and facilities (11%), drinking water systems
(8%), education and health facilities (2%), and contribution to the development of an
irrigation scheme (1%).
4.40 Skills development training
Several skills development training initiatives have been funded by THL in
communities in the HP project area (Table S-7). Initially the training focused on men
and skills presumed to offer employment opportunities on the project. Subsequently,
more attention was paid to women’s training priorities.
Table S-7: Skills development training

Training Municipality/RM Participants
Women Men Total
Plumbing Myagde, Vyas, 0 20 20
Rishing, Bhimad
House wiring Myagde, Vyas, 0 40 40
Rishing, Bhimad
Tailoring Myagde, Vyas, 28 0 28
Rishing, Bhimad
Beautician Myagde, Vyas, 21 0 21
Rishing, Bhimad
Dhaka weaving Myagde 2 0 2
Goat raising Rishing (Dhobla) 9 12 21
Off-season vegetable Rishing (Malinga) 26 2 27
Totals 86 74 160
Note: As these data come from various sources, the accuracy of the numbers of women and men involved may require updating.
Source: ESMSP Presentation – POE Mission – November 7, 2022
4.41 POE comments and recommendation

In general, the training was short-term (several days to several weeks) and focused
on basic skills. As a result, as reported in SSMRs and to the POE social expert, few
of the people who participated have been able to use their acquired skills to secure
employment or start a business.
In brief, the skills development training has been unsuccessful for several reasons.
First, the basic skills that participants learned are not sufficient to make them
83
competitive on the job market; that is, they do not have the full range of knowledge
and skills that employers are looking for. There are also risks of over saturating market
demand in small rural communities when large numbers of people receive the same
skills training; this may have been a factor for people participating in the house wiring
training. Further, none of training programs provided participants with any business
development skills such as financial literacy and other entrepreneurship skills.
The following recommendations encompass an approach that is being adopted in the
development of the LRP, based discussions that the POE social expert has had with
the ESMSP team.
Recommendation 8.2/1: Support for income restoration or improvement in the
LRP should focus on a packaged approach that combines technical skills with
entrepreneurship skills. Alternately, the LRP approach may want to focus
primarily on entrepreneurship skills, working with affected people who already
have skills and/or are trying to operate a microenterprise, with the objective of
assisting them to be more successful and/or able to grow their existing income
generating activities.
Recommendation 8.2/2: The skills development training to date has focused
on several skills that are in demand among project-affected people, but the
training has largely overlooked the highest priority, namely agricultural training.
The LRP should explore options to expand agricultural and livestock skills
training and entrepreneurship support (“farming as a business”), including the
introduction of simple, low-cost improved technologies to address climate
change impacts and promote greater food security.
4.42 Education
The proposed CDP identifies 15 priority schools in the municipalities/TMs in the HP
project area including the Shree Deepak Secondary School and 14 primary schools;
11 of the primary schools also include a pre-school Child Development Center (CDC).
Approximately one-quarter of the CDP budget is allocated for education, with more
than three-quarters (77%) going to the two municipalities of Bhimad and Vyas; 20% is
allocated to Rishing RM and 3% to Myagde RM.
Nearly three-quarters of the education budget is to be used for construction projects.
These projects include new buildings at two secondary schools and renovations to a
primary school. In addition, two secondary schools and 13 primary schools will get
new toilets with access ramps for disabled students, as well as other repairs; and one
secondary school and six of the primary schools will get new storage and filtration
equipment for drinking water.
Table S-8 CDP Education

% Education Schools benefiting
budget Primary / CDC* Secondary
New buildings 73 1 2
Construction (toilets, ramp access, repairs) 5 14 3
Drinking water storage/filtration equipment 1 5 3
Furniture 6 6 3
Digital equipment 5 4 2
84
Table S-8 CDP Education
% Education Schools benefiting
budget Primary / CDC* Secondary
Learning materials 4 10 2
Sports materials 2 1 2
Warm clothing for students 1 5 1
Training (parents; teachers) 1 14 2
Scholarships 2 3
** Eleven of the 14 priority primary schools also include a pre-school Child Development Center (CDC); however, it is not
possible to confirm whether CDP funds are planned for pre-school and/or primary levels.
Source: ELC, 2022c
Other funds will be spent on furniture, learning materials, sports materials, and
children’s clothing; and three secondary and two primary schools will each get ten
computers as well as related digital equipment. Three secondary schools will receive
funds to support scholarships for 30 students at each of the schools. Parents of
students at one secondary school will receive awareness training; and teachers at one
secondary school and 14 primary schools will receive training on teaching methods,
gender-based violence and other matters.
The proposed education program responds to a number of priorities that have been
identified by education stakeholders as well as community members who participated
in the GESI FGDs. The provision of toilets including access for disabled students is an
important addition to the physical plant of schools. While the CDP does not specify,
assuming these are separate toilets for boys and girls, the provision of toilets at the
secondary level is an important factor for adolescent girls to decide to attend and/or
remain in school. Similarly, the provision of hand washing and drinking water facilities
are important improvements to the school environment, to promote personal hygiene
and reduce the risk of disease spread.
An initial scholarship program designed to fund ten students at the Shukla Secondary
School was introduced prior to the formulation of the CDP; the CDP is proposing funds
for 100 students at this school and two other secondary schools. In discussions with
the principal of the Shukla Secondary School, he indicated that he was able to use the
funds provided by THL to fund 12 students, all of whom benefited academically and
socially. However, THL allocated funds for one year and did not renew them. The POE
social expert concurs with the opinion of the school principal that the educational and
social benefits of the scholarship increase when sponsored students are able to
continue to the end of their secondary education.
Recommendation 8.3/1: THL should consult with Shukla Secondary School
and the other secondary schools targeted for student scholarships and based
on feedback, review and, as necessary, revise the proposed scholarship
program to optimize the allocation of scholarship funds to benefit the academic
and social development of the students who are selected.
4.44 Health
The CDP allocates nearly US$250,000 for health-related initiatives, with 78% to be
spent on repairs for all existing health facilities and construction of two birthing centers.
85
The remainder of funding includes new equipment for the birthing centers and existing
health facilities (15%); first-aid kits for a range of health, education, and community
facilities (3%): training for Female Community Health Volunteers (1%); and awareness
raising on sexual and reproductive health, HIV/AIDs, and health and hygiene (3%).
4.45 Water and Sanitation
In response to community demand, THL invested approximately US$50,000 in the
development of water supply systems in several communities in Rishing and Myagde
RMs. In a community in Myagde RM that the POE social expert visited, the benefits
for the 150 households are very evident. Women no longer spend up to one hour at
least once a day to haul water from the source below the village; they now have taps
outside their houses. Metered taps enable the community to adopt a fee structure that
is affordable for normal usage but that encourages people not to waste water.
4.46 POE comments
The CDP identifies all villages as priorities for improvements to water supply, and the
85% of the CDP budget for WASH is allocated for three additional projects. However,
it is unclear which communities and how, overall, how communities were selected for
these improvements.
4.47 Gender Equality and Social Inclusion Action Plan ( GESIAP )
The information reported about directly affected people and other groups in
communities throughout the HP project area demonstrate that women and diverse
ethnic groups/castes factor significantly into how people have been adversely affected
and/or are beneficiaries of THL initiatives.23 Throughout these communities, the
majority of people are Magar, other Janajati or Dalit. On average, women head 25%
of directly affected households, although in the dominant Magar community, women
head nearly 39% of households. Women account for 23% of people who hold title to
project-affected land, a rate that is more than twice the provincial rate of 11% of women
who are sole owners of land (UN Women, 2021).
In the pre-loan preparation of the THP, a Gender Equality and Social Inclusion Action
Plan (GESIAP) was prepared (ADB, 2013). Subsequently, in 2021, ESMSP conducted
20 GESI-focused focus group discussions (FGDs) in communities throughout the HP
project area, to talk with groups of women and/or men, youth, children, social leaders,
and others representing the different ethnic/caste groups (ELC, 2022e). This research
resulted in recommendations based on stakeholder priorities that encompass i) social
norms and practices, ii) overcoming education barriers, iii) young women, men,
marginalized groups and challenges, iv) overcoming health related challenges, v)
gender based violence and safety, vi) addressing human rights and good governance,
vii) economic empowerment and opportunities, and viii) preservation and protection of
cultural, religious, archaeological, and other elements.
The CDP allocates 7% of the budget to GESI initiatives that include orientation and
awareness raising activities (60% of the GESI budget) and a social media campaign
(13%); health camps (21%); and initiatives to train savings groups and social
mobilizers and review social safeguard policies (6%). The evaluation of proposed
23
To date, there are insufficient data regarding directly and indirectly affected people in the TL project area.
86
GESI activities is made against this background, with comments and suggestions from
the POE social expert rather than specific recommendations.
4.48 Orientation and awareness raising
The scope of the proposed activities to provide orientation and awareness raising
includes more than 2,400 participants representing THL, government officials, local
social/political leaders; marginalized men, women, people with disabilities, and
members of the LGBTQI community; students, parents, teachers, and members of
school management committees; out-of-school youth; and members of indigenous
communities. The 1-day or ½-day sessions for small groups of participants cover a
range of topics including basic gender, indigenous, and social inclusion rights; gender-
based violence (GBV) and the importance of engaging men and boys in preventing
physical and/or sexual violence against women, girls, and children; and risks
associated with social media and cyber abuse as well as drug abuse.
THL and ESMSP have already conducted GBV orientations with a total of 228 school
students, 61% of whom have been girls. One group included students at the Shulka
Secondary School. According to the principal, all the participants from his school were
girls.
A first GESI orientation session was recently conducted with 24 staff of THL and its
consultants (21 men, 3 women), facilitated by the ESMSP GESI specialist who has
extensive experience, and very well received as reported by participants.
4.49 POE comments
The POE social expert has concerns. It is essential that girls learn about their rights
related to GBV, however this violence will not stop unless and until boys and men stop
perpetrating it. Further, the efficacy of short-term, one-off orientation and awareness
raising to effect long-term, sustainable behavior change is limited. Basic awareness of
GESI and indigenous rights and the importance of preventing GBV and other forms of
abuse is essential. However, experience has shown that a more sustained and
comprehensive approach is necessary to address the social and cultural norms that
are often the root cause of discrimination and violence and to effect changes in
people’s attitudes and behaviors. Such initiatives may be beyond the scope, time limits
and resources of a hydropower project.
Therefore, the POE social expert would have preferred that this significant portion of
the GESI budget were re-allocated to other priorities that have been identified by the
GESI FGDs; and that these activities were designed to serve as “platforms” to
introduce and stimulate discussion about the key issues that are addressed by the
proposed orientation and awareness raising activities.
4.50 Social media campaign
The POE social expert understands from the CDP documentation that this initiative
proposes to work with 200 people from youth and child clubs in the project area and
to provide them with 1-day training on the development of social media campaigns to
address child rights and the prevention of sexual abuse of children, as well as on the
importance of democratic rights and participation in local governance;24 the training
24
Note: It is not clear in the CDP whether the proposal is 200 people for a 1-day training or 100 people for a 2-day training.
87
will be designed by a social media expert and facilitated by, among others, sports
persons, social media influencers and other “male champions”.
4.51 POE comments
The POE social expert finds this initiative valuable in that it targets building skills of
young people and empowering them to become influencers among their peers. It
addresses a number of recommendations of FGD participants, by strengthening child
and youth clubs, engaging with existing and local organizations, and using a medium
that has been appropriated by young people to address social issues at an age where
there is potential for transformative change. That said, the sustainability of this initiative
may be enhanced if there are resources that can be identified within the scope of the
THP or from other sources to provide ongoing mentoring for the groups, to support
them to leverage the training and continue to develop effective social media
programming and content.
4.52 Health camp
Just over 20% of the GESI budget in the CDP is allocated to health camps focused on
sexual and reproductive health and child health; the detailed budget identifies with
three 2-day events with a total of 1,500 participants. In fact, THL reports that a health
camp was recently held with a total of 4,000 participants, 63% of whom were women
(ELC, 2022f). The POE social expert did not obtain any further information about this
health camp or the impact on the budget allocation for future health camps.
4.53 POE comments
The reluctance of many people to seek health care may be a good justification for THL
to bring clinical services closer to the local communities. It is not clear from the CDP
whether there are any linkages between the health camps and other, health-related
initiatives of the CDP. If those linkages do not already exist, the POE social expert
believes it would be important, for example, to engage Community Female Health
Volunteers to participate in the health camp and, thereby, obtain “on-the-job” training
and support from the medical personnel engaged for the camp. The health camps
would also be opportunities to establish linkages with other awareness raising
activities on sexual and reproductive health, HIV/AIDs, etc., proposed in the health
section of the CDP; and, to provide information on available referral services for
women and girls who have experienced or are at risk of Gender Based Violence
(GBV).
4.54 Training for self-help groups
A nominal amount ( 2,300 USD ) has been allocated to establish and train self-help
groups on credit and savings activities. However, there is no further information in the
text of the CDP or the detailed budget.
4.55 POE comments
It appears that self-help and/or savings-led microfinance groups have been
established in many of the communities in the HP project area, although the available
survey and FGD information is limited about the scope and status of these activities.
The amount allocated in the CDP and the time available within the context of the THP
are insufficient to establish and train groups. Based on experience, this is a process
88
that requires intensive training over a 2- to 4-month period, followed with up to 18
months of regular mentoring, to ensure that groups are able to self-govern, manage
finances, and establish sustainable savings and lending practices.
However, existing self-help and/or savings-led microfinance groups are excellent
“platforms” for a wide range of awareness raising, training and other support activities.
For example, trainings on family nutrition and hygiene, financial literacy, and
entrepreneurship have been successfully layered into bi-weekly and/or monthly
meetings of these groups; as many of these groups are exclusively or largely women,
training on sexual and reproductive health is another option. The benefits of the
savings/lending activities in combination with other substantive training include
increased confidence and self-esteem among the women, as well as increased
knowledge and skills. The POE social expert suggests exploring options to use these
existing “platforms” in addition and/or instead of the 1-day group orientation and
awareness raising sessions.
4.56 Social mobilizer training
The GESI budget includes training for social mobilizer(s). The ESMSP has indicated
that it has proposed the recruitment of four social mobilizers, although the POE social
expert is not clear about their role.
The 2013 GESIAP identified the recruitment of 18 women mobilizers to ensure that
women could access the GRM. This did not happen and may have, in the opinion of
the POE social expert, contributed to the lack of accessibility and transparency of the
GRM as discussed previously.
As noted previously, the ESMSP has recently recruited a Magar-speaking person to
help facilitate consultations with people in this community. Given the ongoing focus on
the LRP and the CDP, it is the opinion of the POE social expert that social mobilizers
should include women and members of ethnic groups/castes in the project area; and
their mandate should include encouraging and facilitating underrepresented people
including women and marginalized people to participate in consultations and project-
sponsored initiatives.
4.57 Social policy on GESI and Gender Based Violence – POE Comment
It is unclear from the available information what the scope and objective of this initiative
area. However, as a result of a summary review made by the POE social expert, it
appears that THL corporate policy does not explicitly address GESI and/or GBV
issues. Although protections exist in Nepal’s Constitution and labor and other laws,
there are no explicit provisions in THL policy for non-discrimination in recruitment and
promotion, based on sex, age or ethnicity/caste; no provisions for equal remuneration
for work of equal value; no prohibition of sexual harassment or violence in the
workplace; etc.
89
4.58 Economic empowerment
The GESI portion of the CDP does not include any initiatives related to economic
empowerment opportunities in communities in the HP or TL project areas, beyond the
skills development training that has already been provided.25
4.59 POE comment
The absence of economic empowerment opportunities in the CDP is a significant
missed opportunity. Over 40% of women and men in directly affected households have
identified agricultural training as their first priority, along with other skills development
training. Agriculture is mentioned in several social safeguard documents as a principal
livelihood throughout the project area. Significant effort has gone into research to
identify the cultural and economic value of a wide range of non-timber forest products,
but there is a lack of a similar cultural and gender analysis of roles in subsistence
and/or commercial agriculture and animal husbandry.
The POE social expert is aware that ESMSP is considering an initiative to establish a
collection center for marketing of agricultural products; there have also been initiatives
to establish linkages with the local chapter of the Federation of Women Entrepreneurs
Association. In discussions with women from communities in the project area, the POE
social expert learned that it is for women to work together to assist one another in their
economic activities; women’s collectives are often very effective to support their
economic empowerment by addressing constraints related to other demands on their
time, social norms that curtail women’s activities or mobility, etc.
The THP has directly affected the economic conditions of households affected by land
acquisition and the focus on this in the LRP is appropriate. At the same time, the THP
has altered economic conditions throughout local communities – creating
opportunities and curtailing others. It would therefore be very appropriate to allocate
more resources in the CDP to this area, with a focus on the agricultural and non-
agricultural aspirations of local women and young people.
4.60 General comments about GESI
Throughout the THP social safeguard documents dating from the project preparation
stage in 2012 through current research and proposals for livelihood restoration and
community development, the social analysis has been limited in several important
ways. Socio-economic and baseline surveys have largely limited gender and cultural
analyses to statistical data. Very importantly, throughout resettlement planning and
implementation, the unit of analysis is the household. The lack of more nuanced
analyses tends to contribute to strategies that do not adequately protect the rights of,
respond to the needs of or assist groups of people within households and communities
who are affected by the project.
Social safeguard documents should, as a matter of course, include comprehensive
gender and cultural analyses that encompass the intersectional nature of people’s
identities based not only on sex or ethnicity/caste but also on age, disability, sexual
orientation, location (rural or urban), etc. The resettlement impacts on directly affected
people vary within households based on roles, responsibilities and decision-making
authorities defined by relations based on sex and age. Gender relations for example
25
See Section 8.2, above.
90
how household decisions are made vary across ethnic groups/castes in the project
area. Impacts at the community level will also vary depending on the age group,
ethnicity/caste, wealth ranking, etc.
4.61 2013 GESIAP
Most of the actions identified in the 2013 GESIAP have been incorporated into
initiatives undertaken and/or proposed by THL in the planning and implementation of
the THP. One initiative not reported elsewhere is the provision of separate toilets for
women and men employed by the project. Table S-9 summarizes the situation as of
the end of Q3 2022.
Table S-9: Provision of separate toilets for women and men

Staffing Toilets
Component Women Men Women Men Total
Package 1 - office 2 57 1 32 33
Package 1 - labor 22 539 2 28 30
Package 2 13 528 4 80 84
Package 3 – office 0 1 1 2 3
Package 3 - labor 0 0 0 0 0
THL 13 56 2 3 5
PSC 5 38 2 3 5
ESMSP 3 7 1 2 3
Source: Tractabel, 2022. PSC Quarterly Report, No. 29, July, August, and September 2022
The data for civil works contractors has been validated. In the case of Package 3, the
staffing data reflect the fact that at the time they were collected, the contractor did not
have people working in the field. The numbers of toilets provided by Package 1 and 2
contractors are assumed to reflect future workforce levels.
4.62 “Other” CDP funding
Approximately 15,000 USD is allocated in the CDP budget for water supply and road/
foot bridge initiatives in Wangtankitar and Paltyang, two of the three communities that
have referred grievances to the ADB and EIB accountability mechanisms. In fact,
these investments have already been accounted for in the THL commitments made in
response to demands from local communities.
4.63 Contingency provisions
The CDP allocates 39% of the total budget as contingency, with two-thirds allocated
for additional work in the reservoir area and one-third in communities along the TL.
The available information suggests that these funds will largely be used for various
infrastructure investments, although there are no details.
4.64 Dam Safety
The THP River Safety Management Plan and the Emergency Preparedness Plan both
focus on issues downstream of the dam. However, the proposed operation of the dam,
specifically the proposed annual flushing of water in the dam, creates safety issues for
people and communities around the reservoir. The POE social expert has discussed
91
with ESMSP staff responsible for these plans and requested amendments that reflect
risks related to the changes in water levels in the reservoir.
Recommendation 9/1: THL amend the Emergency Preparedness Plan and, as
relevant, the River Safety Management Plan to address risks associated with
annual changes in water levels in the reservoir. In addition to early warning
systems (EWS) that are channeled through government agencies at district and/or
ward levels, it is recommended that a community-based ESW “layer” be created
that, for example, engages a network of women and/or men with mobile phones.
Further, it is recommended that a public education program be developed and
delivered to a wide cross-section of local community people, for example, through
primary and secondary schools, at health posts and clinics, in conjunction with
Mothers Groups and other community-based groups and organizations.
4.65 Rural electrification scheme
The ADB loan includes a rural electrification (RE) scheme implemented by the
National Energy Agency (NEA), to deliver electricity to areas of Tanahun District, most
of which are where households are directly affected by the HP or the TL.26 The work
to implement the RE scheme occurred from 2017 through mid-2022. As a result of the
scheme, total of approximately 6,000 households (or 21,960 people27) have access to
electricity, and the majority are connecting to electricity.28 This is facilitated by
Government of Nepal subsidy programs to defray connection and service costs.29
There are no data available for the RE scheme on the distribution of connected
households by the sex of household head, but recent research in Nepal suggests that
in grid-electrified areas, 95% of female-headed households and 92% of male-headed
households are connected (World Bank, 2019).
The achievements of the RE scheme while important do not fully align with the targets
identified in the DMF) or the GESIAP developed as part of the loan preparation (ADB,
2013).
i) The target of 17,636 households connected to electricity, cited in both the DMF
and the GESIAP, reflects the total population the Village Development Committees
(VDCs) being targeted by the RE scheme. In 2010, in the period prior to the
implementation of the RE scheme, 73% of households in the Western Rural Hills
of Nepal that would include Tanahun District, had access to electricity (Government
of Nepal, 2011). In the following decade, the electrification rate in Tanahun District
increased to more than 90%.30
ii) of the total population of the selected set a target for the RE scheme for 17,636
households including 20,000 women to be connected by 2020. The GESIAP also
identified 17,636 households but set targets that 10% of the households (1,700)
are female-headed households and women account for 50% of 100,000 people
26
The extension of the electricity grid occurred in Vyas and Bhimad Municipalities and the Rural Municipalities (RMs) of
Aabookhaireni, Bandipur, Devghat, Ghiring, Myagde, and Rishing.
27
According to the 2021 Census, the average household size in Gandaki Province is 3.66 persons.
28
Personal communication, NEA
29
In 2021, the Government instituted a policy to provide electricity to poor households free of charge for the first 20 kwh of
consumption per month although households will have to pay a monthly services charge that was NR 30 at the time the policy
was instituted
(https://thenepalweekly.com/2021/08/25/7231/). However, no corroborating information was accessed via an internet search
regarding connection subsidies.
30
https://myrepublica.nagariknetwork.com/news/over-86-pc-households-electrified-through-national-grid/ (accessed 13
November 2022)
92
connected through the RE scheme. The GESIAP set further a further target that
1,000 people connected through the RE scheme, 60% of whom are women and/or
socially excluded people (SEP), be given access to micro-finance, financial and
administration skills training, and market assistance to support micro-enterprise
development.
It is clear from THL newsletters that access to electricity via the RE scheme is a matter
of interest to households affected by the THP. The January 2022 edition of the
newsletter mentions a meeting of the LCF in Bhimad attended by THL’s Managing
Director, at which it was agreed to facilitate and coordinate with NEA regarding rural
electrification of project-affected communities. However, to date, no data are available
to indicate the number of AHs who are direct beneficiaries of the RE scheme.
4.66 References
ADB, 2009. Safeguard Policy Statement
ADB, 2013. Nepal: Tanahu Hydropower Project: Report and Recommendations of the
President to the Board of Directors
ADB, 2017. Nepal: Power Transmission and Distribution Efficiency Enhancement
Project – Resettlement Plan
ADB, 2021. Nepal: Electricity Grid Modernization Project – Additional Financing.
Resettlement and Indigenous Peoples Plan
ADB, 2022. Nepal: Tanahu Hydropower Project, External Social Monitoring Report
No-2 (December 2021 – June 2022)
Agarwal, S. 2013. Effective Grievance Redress Mechanisms. World Bank: Social
Development Department
Compliance Advisor/Ombudsman (CAO), 2008. A Guide to Designing and
Implementing Grievance Mechanisms for Development Project. International Finance
Corporation: The Office of the Compliance Advisor/Ombudsman
ELC, 2022a. Follow-up Assessment and Community Consultation Report of Lokma
Village of Rishing Rural Municipality 1, Tanahu
ELC, 2022b. Community Development Plan, Baseline Survey Report
ELC, 2022c. Community Development Plan
ELC, 2022d. Livelihood Restoration Plan (2nd Initial Draft)
ELC, 2022e. Assessment Report on Focus Group Discussion (FGD)
ELC, 2022f. Presentation on Social Safeguard Monitoring, POE Mission, 7 November
2022.
Government of Nepal, 2011. Nepal Living Standards Survey, 2010/11
Government of Nepal, Ministry of Industry, Commerce and Supply (MoICS), Ministry
of Agriculture and Livestock Development (MoALD), 2020. Strategic Road
93
Connectivity and Trade Improvement Project (SRCTIP)-Trade Facilitation
Component: Resettlement Policy Framework (RPF)
Government of Nepal, Ministry of Water Supply, 2022. Nepal Water Governance and
Infrastructure Project (NWGIP): Draft Stakeholder Engagement Plan
Nepal Electricity Authority, 2016. Resettlement Action Plan of Samundratar 132 kV
Substations
THL, 2012a. Indigenous Peoples Planning Framework, Tanahu Hydropower Project
THL, 2012b. Resettlement Framework, Tanahu Hydropower Project
THL, 2012c. Resettlement and Indigenous Peoples Plan (Draft), Tanahu Hydropower
Project
THL, 2017. Social Safeguard Monitoring Report (January 2017 to June 2017), Tanahu
Hydropower Project
THL, 2018a. Consolidated Environmental Management Plan, Tanahu Hydropower
Project
THL, 2018b. NEP: Tanahu Hydropower Project, Resettlement & Indigenous Peoples
Plan
THL, 2018c. Social Safeguard Monitoring Report, Semestral Report, January 2018
THL, 2018d. Social Safeguard Monitoring Report, Semestral Report, July 2018
THL, 2019a. Social Safeguard Monitoring Report, Semestral Report, January 2019
THL, 2019b. Social Safeguard Monitoring Report, Semestral Report, July 2019
THL, 2020a. Resettlement and Indigenous Peoples Plan, Tanahu Hydropower Project
– 220 kV Transmission Line Component
THL, 2020b. Environmental and Social Safeguard Monitoring Report, Semestral
Report, 2020
THL, 2020c. Social Safeguard Monitoring Report, January-June 2020, Semi Annual
Report, June 2020
THL, 2020d. Social Safeguard Monitoring Report, July-December 2020, Semi Annual
Report
THL, 2021a. External Social Monitoring Report, January 2021
THL, 2021b. External Social Monitoring Report No-2 (January 2021 - November 2021)
THL, 2021c. Social Monitoring Report, Quarterly Report, April 2021
THL, 2021d. Social Monitoring Report, Quarterly Report, July 2021
THL, 2021e. Social Monitoring Report, Quarterly Report, October 2021
THL, 2022a. Newsletter. 078/079-65, January 2022
94
THL, 2022b. Agreed Actions: THL Commitments for Dalit and Magar Complaint
Members, February 2022
THL, 2022c. Social Monitoring Report, Quarterly Report, January 2022
THL, 2022d. Social Monitoring Report, Quarterly Report, April 2022
THL, 2022e. Social Safeguard Monitoring Report, April-June 2022, Quarterly Report
THL, 2022f. Social Safeguard Monitoring Report, July-September 2022, Quarterly
Report
THL, 2022g HR Hiring Procedure
Timilsina, R.H. et al., 2019. Agricultural Land Use in Nepal: Prospects and Impacts on
Food Security. Journal of Agriculture and Forestry University (2019), Vol 3: 1-9
Tractebel, 2022. PSC Quarterly Report No. 29, July, August, and September 2022
UN Women, 2021. Factsheet on Women, Nepal, Province 4 – Gandaki
(https://un.org.np/sites/default/files/doc_publication/2021-
01/RevisedWomenFactsheet-Province4_1.pdf)
World Bank, 2019. Nepal – Beyond Connections: Energy Access Diagnostic Report
ACKNOWLEDGEMENTS
5. Acknowledgements
The POE wish to acknowledge the help given by Mr. Kiran Kumar Shresta and his
staff with special mention of Mr. Shyamji Bhandari and Dr. Sudhir Rajure ( Dept of
Mines and Geology ). Thanks also need to be expressed all the other local experts
who have helped POE with their work and also to Mr. William B. Dobbs of Tractebel
International who gave full and generous cooperation to the panel.
95
APPENDIX 1: LINK TO PAPER ABOUT CHAMOLI MUDFLOW
6. APPENDIX 1: LINK TO PAPER ABOUT CHAMOLI MUDFLOW
Link to paper about Chamoli 27 Mm3 debris/mudflow in Uttarakhand Province, India
on 7 February, 2021.
Chamoli Disaster -
Reynolds et al.pdf
96
APPENDIX 2: EXTRACT FROM “EARTHQUAKE ENGINEERING FOR DAMS AND
RESERVOIRS”
7. APPENDIX 2: EXTRACT FROM “EARTHQUAKE ENGINEERING FOR DAMS AND
RESERVOIRS”
Chapters 17 and 18 of “Earthquake Engineering for Dams and Reservoirs”. Now with
the publishers for publication in summer 2023.
Chapter17:Earthquake-triggered landslides
John M. Reynolds
17.1 Introduction
Earthquake triggered landslides are a very important phenomenon particularly in high
mountainous regions such as the Andes and the Himalayas, which are particularly
prone to earthquakes. Many lives have been lost and much infrastructure destroyed.
Hydrological events, following earthquakes, also pose a great problem and a fully
integrated approach is advised.
Lake Sarez in Tajikistan was created in 1911 when a strong earthquake triggered a
massive landslide that, in turn, became a 500 m high dam on the Murgab River, now
called the Usoi dam. Water level is steadily rising behind the dam which, if overtopped,
could release 17 km3 of water that could affect 5 million people downstream. Other
examples and case histories are given in this chapter illustrating the great threat that
earthquake-triggered landslides can pose.
Although not caused by an earthquake there was a 350 Mm3 rockslide into the Vaiont
reservoir in northern Italy that caused waves of water more than 250 m high to overtop
the 262 m high arch dam. Although the dam itself did not fail total fatalities downstream
were estimated at 2,056, resulting in Europe’s deadliest landslide in Europe in
recorded history. The event caused tremors that were felt as far away as Vienna and
Brussels.
Earthquake-triggered landslides (EQTLs), also referred to as co-seismic landslides,
are mass movements triggered within seconds to minutes of strong ground shaking
(Fan et al., 2019). The slope instabilities so generated can include small, shallow, soil
failures and rock falls; deep slumps and landslides; snow, ice, and rock avalanches;
as well as rapidly moving devastating rock avalanches, landslides, and sturzstrom
events. For the sake of clarity, a sturzstrom (rock avalanche) is defined (Hsü, 1975)
as a flow of very rapidly moving debris derived from the disintegration of a collapsed
rock mass of very large size. The speed of such a rock avalanche can exceed 100
km/h and involve volumes greater than 1 million m³ of material.
A single large magnitude earthquake (e.g., M>7) can generate tens of thousands of
co-seismic landslides, many of which may be large enough to form landslide dams
that block river valleys impounding large volumes (multi-million m³) of river water in a
matter of hours. Such landslide dams frequently fail within hours to days of their
formation, releasing large volumes of water as Landslide Dam Outburst Floods
(LDOFs). These can be devastating to downstream riparian communities and
infrastructure.
97
Earthquakes make also reactivate ancient landslides and slope instabilities, leading to
sequences of parasitic and subordinate landslides, commonly exacerbated by heavy
rainfall, for years after the main earthquake sequence.
Earthquakes may generate small rock falls and rock avalanches that in themselves
would be inconsequential if it were not for the presence of vulnerable infrastructure.
Even a single dislodged boulder can cause millions of US dollars’ worth of damage or
rupture key arterial transport routes.
In this chapter, a brief overview of the preconditioning of mountain environments and
of the occurrence and characteristics of earthquake-triggered landslides is provided.
These are illustrated by case histories from both well-studied earthquake events as
well as by previously unreported examples that demonstrate the range of impacts
arising from perhaps small physical events with disproportionate consequences.
This all goes to demonstrate that, when considering possible consequences of seismic
triggering on a landscape, a more holistic approach should be taken. This is
exemplified by the development of Disaster Risk Management (DRM) Action Plans
that incorporate the recently developed protocols associated with Integrated
Geohazard Assessments (IGAs).
17.2 Earthquake-induced chains of geological hazards
For a comprehensive review of earthquake-triggered chains of geological hazards,
readers are directed to Fan et al. (2019) and references therein. It is important to
recognise that a major earthquake can induce virtually simultaneous geomorphic
responses over very large areas (many thousands of km²). If this coincides with, for
example, a monsoon, then the slopes will be significantly affected by the amount of
precipitation and may exacerbate both the scale of the initial mass movement and the
resulting run-out of material with their own sequential consequences as will be
discussed below. Where subordinate processes run out into converging river
tributaries the magnitude of the developing event may be magnified multiple times.
This is known as a cascading multi-phase event, where the magnitude of the impacts
of the accumulated processes are many times the sum of those of each constituent
process.
What might start out as relatively small-scale processes in many separate locations
may end up coalescing to form a much larger and more impactful event downstream.
Consequently, it is important to consider possible timelines of events ranging from:
how a given landscape is preconditioned by prior processes over geological time to
the present; the co-seismic events themselves in the immediate time around the
triggering earthquake (minutes to days); through to the ongoing consequences in the
years and even decades and longer after a significant earthquake. It is then important
to consider how the post-seismic landscape might be modified and how its revised
conditioning might influence the subsequent behaviour of that environment in terms of
susceptibility to future, perhaps lower magnitude, earthquakes, and in relation to
prevailing and future changes in impactful processes such as from climate change.
Major earthquakes and meteorological processes (monsoon, atmospheric river
downpours; typhoons and hurricanes) can impact very large areas virtually
simultaneously. These are shown as items 1 and 2 in Table 17.1. When such
significant meteorological processes occur at or around the same time as a major
earthquake, the consequences arising from the latter can be exacerbated. Physical
damage can occur from the initial earthquake sequence, from the co-seismically
98
triggered mass movements, and then from the subordinate processes as indicated in
Table 17.1.
Table 17.1: Consequential mass movement types (modified from Reynolds, 2014)
Event trigger Examples of mass Flood-type events

movement types
Sturzstrom, landslides,
Landslide Dam Outburst
1 Earthquake snow/ice/rock avalanches,
Floods
rock falls, debris flows
Cloudburst or Landslide Dam and

Landslides, snow/ice/rock
exceptionally Glacial Lake Outburst
2 avalanches, rock falls, debris
prolonged heavy Floods; river and surface-
flows, mud flows
rainfall water floods
A summary and analyses of reports of earthquake-triggered (a) landslides and (b)

snow and ice avalanches have been provided by, for example, Keefer (1984, 1994,
2002), and by Perla (1980), Perla and Martinelli (1978), Podolskiy et al. (2010), and
Schweizer et al. (2021), respectively.
17.3 Preconditioning of mountain environments
The impacts of sudden-onset events, such as a major earthquake, may be
exacerbated by preconditioning over long time periods before hand (Reynolds, 2016).
Many of the processes involved have inter-dependency. Commonly, risk is considered
monolithically and detrimentally through, for example, separate analysis of geological,
seismic, and hydrological risks, where each element is considered in isolation. It is
only by considering the interplay between these processes that a more realistic
understanding of physical processes and their likely impacts on infrastructure and
communities can be achieved (Reynolds, 2015). The collective consequence of these
past events is the preconditioning of the current landscape. When an external trigger
occurs, such as an earthquake, new, and often larger processes, can be initiated. By
way of example, consider the situation in northern Pakistan in the Karakoram. The
mountains there have been formed by millions of years of plate tectonic and orogenic
processes through which the rocks and associated structures have formed and
through which these processes are still at play.
Uplift and erosion are ongoing as they have been for millions of years. It is thought
that the area went through a major glaciation about 20,000 years ago and the many
glaciers, which are now such a feature of the region, developed to achieve their
maximum glacial extents. Valleys were eroded by, and filled by, major glaciers. High
mountain slopes over-steepened and rock falls occurred frequently as they do today.
Glaciers began to retreat and, in so doing, de-buttressed the mountain flanks resulting
in increased slope instability and more rock avalanches and mega rock slope failures
that often contain more than 1 km³ and in rare cases more than 5 km³ in volume
(Hewitt, 2015).
Most of these massive events formed dams across the river valleys, the largest known
being that associated with the pre-historic mega rock slope failure at Nomal on the
99
Hunza River that created a dam 1,100 m high (Hewitt, 2011). Many smaller but
similarly formed dams led to impoundment of large volumes of water that developed
reservoirs tens of kilometres long, leading to deposition of extensive stretches of thick
(>100 m) lacustrine deposits now stranded hundreds of metres above the current river
valley floor. Whether such large collapses of whole mountain flanks were triggered by
major earthquakes is not known. Major earthquake sequences can destabilise hill
slopes for years afterwards that can be re-activated in response to intense rainfall,
effects of climate change, and/or subsequent smaller magnitude earthquakes leading
to many more landslides and mudslides that often claim many lives each year (ADPC
and Partners, 2010).
The major landslide dam that occurred in 2010 blocked the Hunza River in northern
Pakistan forming Attabad Lake, which inundated the valley forming an 18-km long
reservoir that submerged several villages and a substantial length of the Karakoram
Highway. The hill slopes were known to be unstable, with major fractures having
developed years before the eventual failure. It is thought that this landslide was both
preconditioned by long-acting physical processes but particularly by the two local
Astore earthquakes of October 2002. When the landslide occurred, there was no
earthquake or monsoon rain to trigger it; the slope clearly had reached a point of failure
under gravity.
17.4 Consequences of earthquake-triggered landslides
17.4.1 Landslide dams and Landslide Dam Outburst Floods
A detailed review of earthquake-triggered landslides and dams has been
provided by Fan et al. (2019), as well as by Fan et al. (2020), both of which
provide much more detail than can be provided here. There have been several
reviews of earthquake-triggered landslide dams (Costa and Schuster, 1988;
Evans et al., 2009, Hewitt et al., 2008, Korup, 2002, Korup and Wang, 2015) but
few systematic analyses. Undoubtedly, and as stated by Fan et al. (2019) more
scientific and practical insights can be gained from studying earthquake-triggered
landslide dams to better inform risk assessments and emergency responses.
The formation of landslide dams following earthquakes is common and so too
are the Landslide Dam Outburst Floods (LDOFs) than can occur when a
landslide dam breaches. Such floods can propagate for many kilometres
downstream. Furthermore, such dam impoundments can have slow catastrophic
effects upstream as impounded water levels rise for hours to many days after the
dam formation, submerging key infrastructure and communities, and displacing
local inhabitants. Catastrophic failure of such dams can contribute to
redistribution of sediment downstream for many years after the seismic event.
Landslide dams largely comprise soil, rock debris, and/or fragmented rock
materials. In high mountain glacierised environments, some landslide dams may
incorporate glacier ice entrained during the original mass movement if an
overlying glacier was involved.
LDOFs may entrain material from the landslide dam itself, backwater sediments,
and rework fluvial deposits downstream. They can also incorporate eroded
valley flank material to form fast flowing mixtures with high sediment-water ratios,
such as hyper-concentrated flows or even debris flows (Korup and Wang, 2015).
Given the practical implications of such sediment fluxes, there is a lack of
predictive frameworks for estimating the effects of single earthquakes on the
100
fluvial system. This even more so for earthquake sequences and cascading
hazard chains triggered by earthquake activity compounded by hydro-
meteorological events.
Longevity of a landslide dam refers to the whole process, from dam formation to
failure (Shen et al., 2020). There have been three main studies undertaken of
landslide dam longevity by Costa and Schuster (1988) [73], Peng and Zhang
(2012) [204], and Shen et al. (2020) [352], with the number of landslide dams in
their respective analyses given in square brackets. All three studies produce
broadly the same longevity profile. Around 7.4% lasted ≤1 hr and ~30% survived
≤1 day, which are important when considering emergency responses to an
ensuing LDOF. The proportion lasting longer than 1 year is less than 16%. In
some cases, once a landslide dam has settled geotechnically and survived the
early weeks to months, it may survive for many years. The difficulty is telling
which landslide dams are more likely to fail quickly from those that might survive
long term.
Attempts have been made to derive models to predict the failure probability of a
landslide dam (e.g., Dong et al., 2011). Using a limited dataset of ten landslide
dams, following the 1999 Chi-Chi earthquake (Taiwan), 2008 Wenchuan
earthquake (China), and 2009 Typhoon Morakot (Taiwan) they devised a logistic
regression model to test landslide dam failure. However, they found that the
failure probability assessments are materially affected if the flow rate of the
dammed river is also taken into consideration. It remains extremely difficult to
predict if and when a given landslide dam will fail; ideally each case needs to be
taken on its own merits and any assessment made based on detailed site
investigations and analysis. Indeed, it is foolish to think that an historic landslide
regarded as being ‘stable’ because of its longevity will remain so in perpetuity.
An example of the failure of an ancient landslide dam (more than 13-15 ka old)
has been provided by Hermanns et al. (2004) for Las Conchas valley in NW
Argentina. Multiple rock avalanches impounded two lakes with surface areas of
~8 km² and 600 km². Both landslide dams remained stable during a strong
earthquake. It was from subsequent rockfalls into a lake that produced a
tsunami-like wave that destroyed its dam. They found that, instead of direct
effects of strong seismicity on landslide dams, landsliding into a landslide-
dammed lake, abrupt hydrological changes, and climate changes related to
enhanced run-off are processes that can result in the failure of ‘quasi-stable’
landslide dams. This is an important finding in that it emphasises the complex
interaction between hydro-meteorological and geological processes in
influencing the effective stability of a landslide dam. These factors need to be
included in any landslide dam assessment.
17.4.2 Landslides on to glaciers
A summary overview of rock avalanches onto glaciers has been provided by
Deline et al. (2015). Sudden additional mass onto the surface of a glacier can
increase the insulation afforded by increased debris thickness overlying the
glacier. It can also lead to increased flow rates of the glacier tongue leading to
avalanche-induced short-term glacier advances. Furthermore, rock avalanches
onto glaciers can have extended run-out distances by the low sub-slide friction
on the glacier surface, and incorporation of glacial ice into the transported debris
101
reducing its overall viscosity (e.g., Haeberli et al., 2004). Another example of a
long run-out rockslide onto a glacier has been provided by Jiskoot (2011), at Tsar
Mountain, Canadian Rocky Mountains. The rock avalanche had a source
volume of 1.6 million m³, a deposited volume of 2-4 million m², was <1.3 km wide,
and maximum length 1.3 km, and covered an area of 1.35 km² and smothered
30% of the cirque glacier onto which it collapsed. In this case, its run-out was
contained entirely within the footprint of the underlying glacier.
17.4.3 Landslides into lakes
On 22 April 2002 a rock avalanche occurred immediately to the south-west of Laguna
Safuna Alta, in the Cordillera Blanca, Peru. A detailed analysis of this event was
carried out by Hubbard et al. (2005). Field mapping indicated that the avalanche
deposited 8 to 20 million m³ of rock onto the surface of the frontal region of Glaciar
Pucajirca and into the lake (Figure 17.1). Repeated bathymetric surveying indicated
that ~5 million m³ of this material was deposited directly into the lake. The immediate
effect of this event was to create a displacement wave that gained in height as it
travelled along the lake basin, overtopping the impounding moraine at the lake’s
northern end. To achieve overtopping, the maximum wave height was estimated to
be >100 m, possibly as much as 110 m, in part going airborne. This, and subsequent
multiple seiche waves, caused extensive erosion of both the proximal and distal faces
of the impounding terminal moraine.
Figure 17.1 Landslide that collapsed onto Glacier Pucajirca and into Safuna Alta,
Cordillera Blanca, Peru (Hubbard et al, 2005 ) causing displacement waves that
overtopped the moraine dam.
Further deep gullying of the distal face of this moraine resulted from the supply of
pressurised water to the face via a relief overflow tunnel constructed in 1978 following
the devastating 1971 Huascarán earthquake. It can be argued that the debuttressing
102
of the valley flanks above Glaciar Pucajirca caused by widespread deglaciation,
thawing permafrost at high altitude, coupled with destabilisation from the 1971
earthquake, led to the critical conditions at this slope.
Two-dimensional, steady-state analysis of the stability of the post-avalanche moraine
rampart indicates that its proximal face remains susceptible to major large-scale
rotational failure. The backwall area adjacent to the rock avalanche source area
remains highly fractured and is prone to further collapse.
17.5 Post-seismic relaxation
The consequences of especially large magnitude earthquakes are thought to last for
many years after the principal event, influencing subsequent entire neo-tectonic
zones, down to individual slope behaviour and susceptibility to particularly rainfall-
induced movement and later re-activations.
At a regional scale, it has been found that, following major stress release during a
large magnitude earthquake, the crust and upper mantle regions affected adjust to
account for that stress reduction (Klein et al., 2016). They found that, for example,
following the 2010 Mw 8.8 Maule earthquake, Chile, a step-change occurred in the
ratio between the cumulative post-seismic displacement after almost 5 yr. The
respective co-seismic displacement, for stations located directly in front of the rupture
zone, increases with distance from the rupture zone. Similar trends were observed
after both the Aceh and Tohoku earthquakes (Trubienko et al., 2014), indicating
consubstantial properties of the mantle, regardless of local particularities. Indeed,
major earthquakes, such as the Mw≥~9.0 26 December 2004 Sumatra-Andaman
(Boxing Day) earthquake, given their magnitudes that rupture hundreds of kilometres
of major crustal fault systems, are expected to be followed by vigorous viscoelastic
relaxation involving both the upper and lower mantle (Pollitz et al., 2006). Rock healing
as a relaxation process has been discussed by Schneider et al. (2017).
It is only relatively recently that attention has turned to post-seismic relaxation at a
local scale, i.e., in relation to individual valleys and specific mountain slopes (e.g.,
Parker et al., 2015). They investigated the 1929 Buller and 1968 Inangahua
earthquakes that occurred in northwest South Island, New Zealand. They found that
the behaviour of slopes affected by the first event were noticeably more affected by
the second event than by the first. Thus, the damage legacy of large earthquakes may
persist in parts of the landscape for much longer than observed sub-decadal periods
of post-seismic landslide activity and sediment evacuation. Consequently, a lack of
knowledge of the damage state of hillslopes in a landscape potentially represents an
important source of uncertainty when assessing landslide susceptibility. However, this
assessment was based purely on seismic responses. What is evident is that hill slopes
affected by previous earthquake activity are also preconditioned to behave differently
in response to other external stimuli, such as intense monsoonal rainfall (cf. Section
17.3).
The 2015 Mw 7.8 Gorkha earthquake sequence in Nepal triggered more than 20,000
landslides across a wide region of Western and Central Nepal. Rosser et al. (2021)
investigated how landsliding has evolved since the 2015 seismic events. They
undertook a time series landslide mapping to see how landslides evolved over time
since the earthquakes. They found a significant step-up in the level of landslides
following the earthquakes but noted that, in following monsoons, the activity reduced,
as did the areas of landslides, although new landslides also occurred. This
103
earthquake-induced increase in landslide activity is well known. What has been
unclear previously is that the occurrence of a major earthquake can lead to an increase
in overall landside activity and susceptibility to other external non-seismic processes,
such as intense rainfall, albeit with a reducing number or post-seismic relaxation over
several years.
By way of illustration of this possible post-seismic relaxation and preconditioning, the
case of the Upper Bhote Koshi valley in north Central Nepal is presented. The Bhote
Koshi emanates from Tibet (ARC), where the river is called the Poiqu. The valley
contains an economically and politically important road (the Araniko Highway) that
links Tibet (ARC) and Nepal. The valley also houses several important hydropower
schemes, both constructed and operational as well as in construction. Geologically, it
has been postulated (Reynolds, 2018, 2019) that the valley has been formed by a N-
S normal fault system that cuts across the regional geological fabric associated with
the various E-W striking thrust faults (e.g., the Main Central Thrust). The sides of the
valley are formed variously of steep fault planes and associated conjugate structures
that have formed unstable wedge features that are prone to collapse. Parts of both
sides of the valley comprise ancient Massive Rock Slope Failures that regularly
contribute subordinate shallow landslides and many debris flows and rock falls that
often result in fatalities.
One such ancient Massive Rock Slope Failure (MRSF) is about 2 km wide and 3.5 km
from backscarp to toe. It comprises several discrete flow units, the central one of
which is extremely active and the steeper front across the width of the toe is subject
to multiple shallow subordinate landslides virtually every monsoon. It survived the
2015 Gorkha earthquake sequence although many buildings locally were badly
affected by both shaking and seismically triggered rock falls. Buildings on the toe of
the MRSF fared generally well through the earthquakes. However, in subsequent
years, during each monsoon, parts of the central flow unit have become reactivated.
The lowermost third of the central flow unit slipped by more than 5-10 m vertically in
2016, leaving a prominent sinuous backscarp. The middle third activated during the
2017 monsoon and dropped about 3-5 m at its upper end. Further shallow landslides
occurred across the toe of the wider MRSF during the 2018 monsoon. There is no
satellite imagery freely available for 2020, but in 2021, the steeper toe slope of the
central flow unit and part of the toe of the adjacent flow unit to the south that activated
in 2018 failed. This badly damaged the Araniko Highway that had been rebuilt by the
Chinese in the first part of 2019 but which by October 2019 was showing significant
vertical and horizontal displacements (of the order of 0.5 m) after just 4 months. There
had been no such landsliding activity in the years 1985-2015 for which images are
freely available, only since the 2015 earthquakes, suggesting that parts of the MRSF
had become co-seismically re-activated. Perhaps the various landsliding activity from
2016-date is a measure of the post-seismic slope relaxation and increased
susceptibility of the MRSF to slope movement, exacerbated by the heavy monsoon
rains each year.
While there are obvious areas of slope failures on the MRSF described, elsewhere on
its surface deformation and distortion were continuing to occur. For example, one
building escaped unscathed from the 2015 earthquakes but was badly misshapen by
the 2016 monsoon slope movements.
17.6 Geographical examples
17.6.1 Huascarán, Peru, 1971
104
The Cordillera Blanca in the Peruvian Andes (Figure 17.2) consists of
granodioritic peaks up to 6,500 m with deeply dissected U-shaped valleys
draining into the Callejón de Huaylas, a fertile valley corridor through which the
Rio Santa flows northwards and ultimately towards the Pacific Ocean. The
region is seismically very active, both with many earth tremors associated with
the Cordillera Blanca Fault (e.g., over 150 tremors with magnitudes up to Richter
3.2 were recorded in 34 days by Deverchère et al., 1989) as well as major quakes
associated with the subduction of the Nazca Plate in the Peru-Chile Trench. It
was the occurrence of four major earthquakes with magnitudes from 5 to 7.9 on
31st May 1970 with an epicentre some 25 km west of Chimbote (Figure 17.2) that
resulted in the loss of around 80,000 lives with countless more made homeless.
Of these deaths, about 20,000 were due to the catastrophic rock/ice avalanche
sturzstrom from Huascarán Norte, details of which were provided by Plafker and
Erickson (1978).
105
Figure 17.2. Map of the Cordillera Blanca and Cordillera Negra in northern Peru
with the epicentre for the May 1971 Huascarán earthquake (redrawn from
Reynolds, 1992)
The steep face of Huascarán Norte, with its faulted carapace of metamorphic
rocks over a core of granodiorite, towered over the foothills of the Cordillera
106
Blanca. The summit ice cap topped what was an inherently friable mountain. In
1962, part of the summit pyramid broke away in a rock avalanche that swept
down the adjacent river valley and destroyed the town of Ranrahirca with the loss
of 4,000 lives. This event left the summit weakened.
The short swarm of earthquakes in May 1970 loosened the toe of the main face
which in turn caused the more massive failure of the western part of the
mountain. Part of the debris fell into Laguna Llanganuco while the majority fell
westwards into the main valley. The debris fell from the face of the mountain
about 800 m onto a glacier below from which it incorporated water from surface
snow and a small glacial lake. The flow split as it crossed a major fault. The
total volume of mobile material has been estimated at about 180 million m³, much
of which travelled at speeds greater than 270 km/hr. Some boulders with weights
of around 60 tonnes were hurled through the air over for 2-3 km at speeds of
>1,000 km/hr. One resulting impact crater was 33 m long by 15 m wide and 8 m
deep. Associated with the sturzstrom were an air blast and mud spatter that
were reported to have been capable of shredding skin and disembowelling
victims.
The sturzstrom sped down-valley, ricocheting from side to side like a bobsled
down a sinuous chute. Debris rose to 80 m higher on the outside of the bends
than on the inside. When the debris reached Cerra de Aira, it split into two lobes,
the larger one of which followed the route of the 1962 sturzstrom over the town
of Ranrahirca. The other over-topped a 150-m high ridge in a wave of debris 300
m wide and 80 m high, which was reported by survivors to have looked just like
a huge ocean wave breaking onto a beach. The tragedy was that that the debris
flow impacted onto the central plaza of the town of Yungay, its population swelled
by market-day visitors.
107
17.3 Schematic to illustrate the various components of the sturzstrom from
Huascarán Norte that led to the destruction of Yungay town with the loss of
thousands of lives (redrawn from Reynolds, 1992). LL = Lago Llanganuco;
towns: Y = Yungay; R = Ranrahirca; M = Matacoto.
17.6.2 Wenchuan, China, 2008
The 12 May 2008, Great Sichuan Earthquake, also called the Wenchuan
Earthquake, occurred at 14:28 local time, in Sichuan Province, China. The
108
earthquake magnitudes were Mw = 7.9 (USGS), Ms = 8.0 (Chinese Earthquake
Administration). The epicentre (Figure 3.5 in Chapter 3) was 80 km west-
northwest of Chengdu, the capital city of Sichuan province, at a focal depth of
14-19 km (Fan et al., 2012a). Around 87,225 people were killed or reported
missing with 374,643 injured. At least 15 million people were evacuated from
their homes and more than 5 million were left homeless.
The earthquake occurred on the NE-trending Longmen Shan thrust fault zone,
which itself comprises three sub-parallel faults, namely the Wenchuan-Maowen
Fault (WMF), Yingxiu-Beichan Fault (YBF), and the Pengguan Fault (PF). The
rupture initiated near Yingxiu City and propagated unilaterally towards the
northeast, generating a 240-km long rupture along the YBF and a 72-km rupture
along the PF (Fan et al., 2012a). The earthquake resulted in >60,000 co-seismic
slope failures with 828 landslide dams identified of which 501 blocked their rivers
entirely, while the remainder caused only partial damming or channel diversion
(Fan et al., 2012b).
The Wenchuan earthquake epicentre was very close to the Niujuan Valley the
upstream and northern part of which is located on the hanging wall of the
seismogenic YBF thrust fault and many co-seismic landslides occurred in the
area (Fangqiang et al., 2010). The largest, known as the Niujuan landslide,
occurred at the headwall of the Lianhuaxingou Valley. The landslide moved
down to the headwall foot where it entered the river channel and travelled rapidly
for nearly 2 km to the confluence with the Niujuan Valley. The original slide mass
(estimated to have a volume of 3-4 million m³) fractured during its movement and
transformed into a debris avalanche. At the confluence with the Niujuan Valley,
the flow was obstructed by the right bank slope where it turned downstream and
accumulated in the riverbed forming a 30-m high dam. This formed a lake with
a volume of ~0.11 million m³. The significance of this event was its proximity to
the upstream end of the Zipingpu Reservoir, the largest on the Minjiang River,
with a capacity of 1.12 billion m³.
The Tangjiashan landslide dam (Figure 17.4), in the upstream section of the Jian
River, was the largest and most dangerous landslide dam triggered by the
Wenchuan earthquake (Xu et al., 2009). The dam crest extended ~600 m across
and 800 m along the valley; its height varied from 82 m to 124 m and was
estimated to have a volume of 20.4 million m³. The newly formed impounded
lake had a volume of up to 300 million m³. As seven smaller landslide dams had
formed downstream, the breaching of the Tangjiashan dam could have induced
the failure of several of these, posing a severe threat to more than 1 million
people.
Given the seriousness of the situation, the Chinese Government decided to build
an artificial spillway by excavation and blasting. After 7 days and 6 nights of
continuous working, a 475-m long, 12-13-m deep channel had been formed.
Once the water level had risen sufficiently the lake began to flow through the
spillway to start draining the lake that, at that time, had a volume of 247 million
m³. As the flow started to erode the channel the discharge rate increased to a
maximum of ~6,500 m³/s four days after the drainage started. The lake level fell
by 28.8 m and its volume reduced to 86 million m³, at which point the emergency
109
was declared over and residents who had evacuated were permitted to return
home.
The Tangjiashan landslide affected slopes that had been gravitationally
deformed long before the earthquake (Chigira et al., 2010; Pánek and Klimeš,
2016).
Figure 17.4 Tangjiashan landslide dam (Photo by Mr Li Gang)

17.6.3 Nepal in April and May 2015
On 25 April 2015, a magnitude 7.8 earthquake followed by a series of M>6
aftershocks occurred affecting much of central Nepal (Figure 17.5) (Collins and
Jibson, 2015). A magnitude 7.3 event struck on 12 May 2015. These events
have become known as the ‘Gorkha earthquake sequence’, with the two largest
events known separately as the ‘Gorkha’ and ‘Dolakha’ earthquakes,
respectively, referring to the locations of the two epicentres. About 9,000 people
were killed and 16,800 injured during the onset of the sequence with a further
100+ killed and 1,900 injured in the May event. Some 2.8 million people were
displaced with more than 8 million people affected by the seismic events and
their aftermath. Several hundred thousand people remained in temporary
accommodation for years after the earthquakes.
110
Figure 17.5 Map showing the location of the epicentres of the two large
magnitude earthquakes as part of the Gorkha earthquake sequence (25th April,
Gorkha earthquake; 12th May, Dolakha earthquake) and the rectangular thrust
slab involved in the seismicity. The direction and amount of annual
foreshortening is also indicated.
The April 2015 Gorkha earthquake occurred because of thrust faulting along the
main interface between the subducting Indian plate and the overriding Eurasian
plate to the north. Fault rupture propagated south-eastward with maximum slip
of 4-6 m beneath the Kathmandu Valley. Maximum fault slip of about 3 m
occurred south of the Dolakha epicentre.
Strong motion sensor-derived PGA values were more than 0.5 g although
localized areas probably experienced PGA values exceeding 1 g. For the
Dolakha earthquake, lower values (>0.2 g) were noted but some areas are likely
to have had PGA values up to 0.83 g (Collins and Jibson, 2015). In the two
months following the main shock, there was 1 aftershock with M>7 (the Dolakha
event), 2 aftershocks in the M6-7 range, 22 in the M5-6 range, and 39 in the M4-
5 range.
Initial mapping, based on satellite data, identified 4,312 earthquake-induced
landslides (co-seismic and post-seismic) (Kargel et al., 2015). High landslide
densities were identified between the epicentres for three major earthquakes in
Nepal, namely, those in 1833, and the Gorkha and Dolakha events of 25 April
and 12 May 2015. This suggests the possibility of a long-term preconditioning
from historic quakes. Subsequent analysis indicated that more than 20,000
landslides occurred across 14 districts in Western and Central Nepal (Rosser et
al., 2021).
One of the most tragic of the co-seismic landslides that occurred in 2015 was
that in Langtang Valley. The area is one of Nepal’s major trekking regions and
for its accessibility, it has been the focus of much glaciological, hydrological, and
meteorological research. The magnitude of this landslide focussed significant
111
international attention and thus it is one of the best documented events arising
from the 2015 earthquake sequence. The valley underwent moderate shaking
(up to ~0.26 g above Langtang village). Co-seismic snow and ice avalanches
and rockfalls with their large concurrent air blasts led to the destruction of
Langtang village with the loss of 350 people.
The Langtang landslide deposits covered an area of 7.61 million m² (Kargel et
al., 2015). The primary co-seismic event was a combined ice-snow avalanche
that started near 7,000 m. Subsequently, rockfall material was entrained with ice
and snow and descended a low-gradient part of the glacier near ~4,500 m. The
rock-ice mass then became airborne as it fell off a cliff below 4,500 m. After the
debris reached the riverbed at ~3,250 m, it ran up the opposing slope ~200 m.
The air blasts propagated 400 m up the mountain across the valley. From the
point of impact on the valley floor, devastation extended ~1 km up- and down-
valley.
In Langtang village, landslide winds flattened what had survived the initial slide,
including some buildings constructed of stone slabs; a small forest was also
completely levelled, indicating wind speeds of >322 km/h.
A second large post-mainshock mass movement happened near Langtang
village sometime before 25 May 2015, the source of which may have been a rock
detachment from the summit ridge of Langtang-Lirung, ~6,700 m elevation. At
least one other large post-seismic landslide took place in the valley between 8-
10 May 2015.
While it was stated initially that there had been no co-seismic ground rupture,
there is photographic evidence of the occurrence of hill crown splitting. This was
observed in the Bhote Koshi Valley to the east of Langtang. The pre- and post-
earthquake images show the effects of the earthquake on a distinguishable hill-
top feature that was also videoed during a helicopter flight as part of the 2015
USGS field investigations (Collins and Jibson, 2015). The earthquakes clearly
had disturbed the upper vegetated soil layer into rhomboidal slabs and, further
down the slope near the prominent feature (arrowed), fractures around the hill
crown suggesting extensional movement over the hilltop.
17.7 Other consequences of earthquake-triggered landslides for hydropower
17.7.1 Earthquake-triggered rock avalanches
In addition to the well-known effects of earthquake shaking on buildings, which
are often ignored or omitted in pre-construction risk assessments for hydropower
stations, is the impact of other seismically induced processes, such as
associated more with slope instabilities. It is often thought that only large-scale
events cause major damage. This is not the case as will be demonstrated by the
following couple of examples.
In high mountain environments with steep valley flanks, many of which are de-
buttressed following deglaciation, seismically induced landslides and rock
avalanches are very common. When rock avalanches occur, the norm is to
consider that the run-out is across the ground surface. What can happen is a
combination of ground-hugging landslides coupled with a rock avalanche in
which isolated boulders can travel considerable distances through the air and
112
provide aerial strikes on properties and infrastructure. All it takes is an airborne
boulder of sufficient mass and momentum to strike at a critical point in an
engineered system to wreak significant physical damage but even greater
financial losses.
During the devastating Gorkha earthquake sequence in Nepal in 2015, complex
multi-phase slope failures occurred that involved seemingly relatively minor
individual rock avalanches. The Upper Bhote Koshi valley was particularly badly
affected by the earthquake and seismically induced landslides. One community
built along the Araniko Highway connecting Nepal and China exemplifies the
problem. Landslides and rock avalanches punctuated the road causing
significant damage to the road itself. Buildings were badly shaken and damaged
by the earthquake sequence. Some suffered from aerial bombardment from
large boulders that bounced down the slopes and hit buildings below. One
example is of a building badly damaged by earthquake shaking, hit from the side
by a co-seismic landslide, with the coup-de-grace being delivered by an aerial
rock bomb that hit almost vertically onto the roof of the building, penetrating
through two storeys.
A similar fate occurred at the Upper Bhote Koshi Hydropower Plant powerhouse
during the same earthquake sequence. When the power station was being
designed, there were two choices for the penstock pipe leading to the
powerhouse: (a) leave it exposed (the slightly cheaper option), or (b) cover it to
protect it from boulders. The cheaper option was taken, and the penstock lay
exposed. A ~1-m diameter boulder from a relatively minor rock avalanche from
the steep mountain flank above the powerhouse landed on and ruptured the
penstock pipe. The contained water, being at high pressure, discharged with
force, eroding the penstock’s engineered channel sides flooding the powerhouse
and filling it with debris. All the electro-mechanical equipment inside the
powerhouse was destroyed resulting in cessation of power generation and a
substantial physical cost and loss of revenue.
In a related case along the same valley, the Chinese had constructed a single-
span bridge across the Bhote Koshi for the Araniko Highway. During the Gorkha
earthquake sequence a single large airborne boulder from another relatively
minor rock avalanche hit the road above the central bridge span and destroyed
it. The Araniko Highway was effectively cut, making vehicular travel impossible.
It took months for a Bailey bridge to be constructed across the river to replace
the destroyed structure and restore communications along the Araniko Highway.
The lesson from these examples is to consider a wider portfolio of geomorphic
slope responses to seismic activity other than just earthquake shaking when
undertaking risk assessments for key infrastructure. Even critically located but
small processes can have disproportionate practical consequences.
17.7.2 Earthquake-triggered landslides at critical hydrological nodal
points
In a high mountain region, such as in the Himalayas, river catchments so critical
for the supply of water to downstream populations should be carefully assessed
for a wide range of physical hazards. What are often neglected or omitted in
hydropower development risk assessments are the regional-scale neo-tectonic
and geological processes that can trigger catastrophic geomorphic responses at
113
critical nodes within a river system. The Himalayas are often referred to as the
‘water tower of Asia’ on which ~1.2 billion people depend to various degrees.
Hydropower hydrological risk assessments, conducted separately from any
seismic risk assessments, concern themselves just with hydrological water flows.
Hydrological run-off models use the watershed areas of a catchment coupled
with river discharge measurements to predict the likely river flows, and hence the
hydropower potential. Seismic risk assessments concern themselves with
whether engineered structures associated with the hydropower station
(headworks, powerhouse, etc.) will cope with a prescribed Maximum Credible
Earthquake. The two types of risk assessments are done independently of each
other.
What these independent risk assessments fail to recognise is the roles that neo-
tectonics and seismic activity play in both shaping and affecting the drainage
within an overall catchment.
An example of the impact that neo-tectonism and seismicity can place on
affecting the hydrological run off from a catchment is that of Kashmir.
Of critical significance to the drainage behaviour from the Kashmir Basin is the
seismic activity and surface dislocations associated with faults associated with
the Kashmir Basin Fault where the Jhelum River drains through the western
mountain range near Barambula (Shah, 2016). Fluvial flooding in the Kashmir
Basin is largely influenced by the active nature of the NE-dipping Kashmir Basin
Fault, which has dragged the whole of the basin downward. Ongoing
deformation has tilted the basin toward the south-east. He raises concern that
future movement on the Kashmir Basin Fault could lead to major rock slope
instability that could dam the Jhelum River again.
Such a river damming event has occurred historically. Bilham and Bali (2014)
identified the occurrence of multiple landslide events that had occurred between
Barambula and Gantamula in response to a major earthquake reported to have
occurred in 883 AD. This led to the formation of a lake 1,600 km² in area that
submerged much of the Kashmir Basin. The landslide dam was breached
through engineering works and the lake drained. The Jhelum River was
subsequently canalised after this event in the late ninth century to aid drainage.
The physiography of the area has been pre-conditioned by the geology and
structure such that there is an ongoing subsidence of the Kashmir Basin coupled
with sporadic seismic events. This can result in changes to the hydrological
gradient, including in places, drainage reversal. These seismic events have in
the past and may again in the future lead to major slope instability in key locations
that have led to blockage of the Jhelum River, especially near Barambula in
Indian-controlled Kashmir. When extreme meteorological events occur, such as
in September 2014 (Ray et al., 2015), the subsided portions of the Kashmir
Basin, which are very vulnerable to flooding disasters (Shah, 2016), can undergo
major inundation. If such events become more commonplace and/or more
extreme associated with climate change, coupled with ongoing subsidence of the
Kashmir Basin through tectonic processes, drainage from the Kashmir Basin
through the Jhelum River may be compromised. There is a clear association
between long-term tectonic processes, geological structures, contemporary
seismic activity, and major slope instabilities preconditioning the geomorphic
behaviour of the mountain environment (Reynolds, 2016, 2017) within the overall
114
Jhelum Basin upstream of the Karot HPP, including between the HPP site and
Muzaffarabad. This behaviour can also be strongly influenced by meteorological
events the magnitude and frequency of which may be affected by changing
climate.
Impoundment of lakes through the interaction of major geological faults and
probable landslide damming is also evident in the Upper Indus Basin on the
Tibetan Plateau. In Ladakh, northern India, the blockage of the Pangong Tso
outlet to the Shyok River (Tangtse River), which has cut-off the sub-basin from
the Upper Indus watershed, is thought to have happened possibly in the last 2 to
10 ka (Dortch et al., 2011; Phartiyal et al., 2015) but the cause for it is less known.
Some authors have proposed that the Tangtse River has been dammed by flood
deposits resulted from the catastrophic partial drainage of the Pangong Tso into
the Tangtse Valley, which may have been caused by seismic activity on the
Karakorum Fault (Dortch et al., 2011). Some others have suggested that the
blockage of the Pangong Tso outlet to the Tsangtse River was due to a step in
the bedrock resulting from tectonic movement on the Angmong (normal) Fault
system (Van Buer et al., 2015). Alternatively, seismic activity coupled with
increasing slope gradients through tectonic uplift may have resulted in faster
rates of denudation and greater sediment supply rates from the adjacent
mountains. This higher rate of sediment influx could have progressively choked
off the Tangtse River drainage from Pangong Tso.
Similarly, there is morphological evidence of large palaeo-lakes having formed
in the Pumqu Basin (Tibetan Plateau, China) (Chen et al., 2016; Reynolds,
2019). It is postulated that individual lakes formed from sub-basins being
dammed by landslides at critical nodal points. These are thought to have
breached episodically, lowering the lake levels in stages. The critical nodal
points all occur at the intersection between the faulted margins around the Ana
Prime Dome and more east-west trending structures aligned with the main
central thrust sequences. At each node there is evidence of possible landslide
activity with the southernmost postulated landslide dam having attained a height
of 400 m above the valley floor. Landslide dam outburst floods emanating from
such large lakes through steep incised river valleys would have led to very long
run-out distances, potentially depositing coarse debris in the Bay of Bengal.
Elsewhere on the Tibetan Plateau several endorheic lakes are present, whose
water levels are dropping as evidenced by stranded former shorelines, e.g., Cuo
Muzhelin, Cuochuolong, and Peikucuo (all in the Pumqu Basin, China), to name
but three. It is plausible that these sub-basins have been progressively switched
off by reversals in the hydrologic gradient with most water loss occurring through
evapo-transpiration perhaps coupled with decreasing rates of precipitation.
17.8 The role of Integrated Geohazard Assessments
What has become extremely clear over recent years (Reynolds, 2016, 2017, 2019,
2022a,b; Reynolds et al., 2018) is the linkage between a range of physical processes
and the formation of natural disasters. The ranges of process types and resultant
disasters are far broader than those usually investigated in hydropower projects
through Geotechnical Baseline Reports (GBR), Hydrological Risk Assessments, and
Seismic Risk Assessments, etc. Given the large financial losses associated with
hydropower facilities destroyed or badly damaged by natural disasters over recent
years (Reynolds, 2015), it is clear that the hydropower sector needs to adopt methods
115
by which hydropower infrastructure becomes less vulnerable to more extreme physical
processes, such as seismically induced landslides and Landslide Dam Outflow Floods
(LDOFs). Since 2016, a protocol has been developed, referred to as an Integrated
Geohazard Assessment (IGA), that can build upon and extend beyond pre-existing
risk assessments. Importantly, the process provides a mechanism through which the
physical processes and major slope instability events within an upstream catchment
as well as immediately downstream of a proposed/existing hydropower facility can be
determined. The relationship between various timescales and concurrent but different
types of physical processes was identified as discussed by Reynolds (2016, 2017).
This inter-relationship and follow-through from this are indicated in Table 17.2, where
the different but typically disparate risk analyses undertaken in major infrastructure
development can be considered collectively and more holistically within an Integrated
Geohazard Assessment. The interplay between geological causes of hydrological
events (e.g., LDOFs) and hydrological triggers to geological events (e.g., rain-induced
landslides) is illustrated by the two sets of double-headed arrows in Figure 17.6.
Table 17.2: Association of HEP project studies with physical process types
(Reynolds, 2016, 2017)
116
Geophy
Physical sical & Geotech Geomor Glaciologi Climate Meteorolo
Processes Geologi nical phic cal Change gical
cal
Existing
Seismic GLOF
discrete Geotechnical Hydrological Risk
Risk Assessm
study Baseline Report Assessment
Analysis ent
types
Recomme
nded
Integrated Geohazard Assessment
additional
study
Figure 17.6 Integrated Geohazard Assessment (Reynolds, 2015)

The Integrated Geohazard Assessment approach is congruent with and has been
adopted in recent guidelines published by The World Bank (Climate Change
Resilience and Disaster Risk Management, Mott MacDonald, 2017; Dam Safety in
Nepal, Hatch Consultants, 2017). In addition to existing risk assessments, the IGA
method includes ‘Event Mapping’ and ‘Process Identification’, as described in detail
by Reynolds, 2017, 2019, 2022a,b; and in Reynolds et al., 2018). In essence it brings
together and builds on the various baseline reports and adds more detailed and
integrated assessments of seismic generation of landslides, landslide susceptibility
mapping, neo-tectonic and seismicity assessments, regional geological appraisals
including lineament studies, critical nodal points where landslides have occurred/may
occur, glacial and permafrost hazard assessment, palaeo-lake identification, climate
change analyses, and detailed hydrological modelling. This also provides a framework
to identify physical features that should be the subject to future monitoring and
investigation, and ultimately can feed into consideration of any mitigation measures
117
that may be required and could be feasible. Such a comprehensive analysis needs to
be catchment wide and not restricted by political boundaries in the form of a
transboundary geohazard assessment (Reynolds, 2019).
17.9 Conclusions
Moderate to large magnitude earthquakes can trigger tens of thousands of co-seismic
landslides across large regions simultaneously. Of these a significant proportion can
result in landslide dams that block the river valleys into which they fall. Of these
landslide dams, less than ~70% are likely to fail in less than one month from formation,
with 7.4% failing within one hour, and 48% surviving for less than a week. Landslide
dam outburst floods can wreak devastation downstream for tens of kilometres. Such
floods can result in over-steepened valley flanks that become more prone to slope
failure and can contribute significant volumes of additional sediment into river systems
for years after the initial earthquake.
Seismically triggered landslides that collapse into lakes create large seiche waves that
can overtop their dams (natural or engineered) and can lead to major dam breaches
with the attendant consequences.
Major earthquake events and sequences of significant aftershocks can result in a step-
change increase in the number of landslides that occur within an affected area. The
number of post-seismic landslides does tend to reduce over a period of several years
after the initial earthquake in a form of post-seismic relaxation. However, the physical
effects of such earthquakes and of the slope instabilities they cause effectively
precondition the landscape to subsequent events.
When major earthquakes trigger landslides, if the event coincides with periods of
heavy rain, such as during the monsoon is the Himalayas, what starts out as a
geological set of processes can morph and cascade into a series of floods to form a
major hydrological disaster. There is a transposition between geological causes of
hydrological disasters (e.g., landslide dam outburst floods), and hydrological causes
of geological disasters (e.g., rain-induced landslides). Consequently, when
considering any infrastructure project, it is strongly recommended that an Integrated
Geohazard Assessment is undertaken during which both geological and hydrological
process are reviewed for an entire upstream catchment that can act in concert within
the same catchment simultaneously, magnifying the scale of the consequences.
17.10 References
ADPC and Partners. (2010). “Nepal Hazard Risk Assessment”. Asian Disaster
Preparedness Center (ADPC), Norwegian Geotechnical Institute (NGI) and
Centre for International Studies and Cooperation (CECI), Bangkok, Thailand.
Bilham, R. and Bali, B.S. (2014). “A ninth century earthquake-induced landslide and
flood in the Kashmir Valley, and earthquake damage to Kashmir’s Medieval
temples”. Bull. Earthquake Eng., 12: 79-109.
Chen, Y., Aitchison, J.C., Zong, Y., and Li, S-H. (2016). “OSL dating of past lake levels
for a large dammed lake in southern Tibet and determination of possible controls
on lake evolution”. Earth Surface Processes and Landforms;
doi:10.1002/esp.3907.
118
Chigira, M., Wu, X., Inokuchi, T., and Wang, G. (2010). “Landslides induced by the
2008 Wenchuan earthquake, Sichuan, China”. Geomorphology, 118: 225-238.
Collins, B.D. and Jibson, R.W. (2015). “Assessment of existing and potential landslide
hazards resulting from the April 25th Gorkha Earthquake Sequence”. U.S.
Geological Survey Open-File Report 2015-1142, 50 pp;
https://doi.org/10.3133.ofr20151142.
Costa, J.E., and Schuster, R.L. (1988). “The formation and failure of natural dams”.
Geological Society of America Bulletin, 100(7): 1054-1068.
https://doi.org/10.1130/ 0016-7606(1988)100<1054:TFAFON>2.3.CO;2.
Deline, P., Hewitt, K., Reznichenko, N., and Shugar, D. (2015) “Rock avalanches onto
glaciers”. In: Davies, T. (ed.) Landslide hazards, risks, and disasters. Elsevier,
263-319.
Deverchère, J., Dorbath, C., and Dorbath, L. (1989). “Extension related to high
topography. Results from a microearthquake survey in the Andes of Peru and
tectonic implications”. Geophysical Journal International, 98(2): 281-292.
Dong, J-J., Tung, Y-H., Chen, C-C., Liao, J-J., and Pan, Y-W. (2011). “Logistic
regression model for predicting the failure probability of a landslide dam”.
Engineering Geology, 117: 52-61, https://doi:10.1016/j.enggeo.2010. 10.004.
Dortch J.M., Owen, L.A., Caffee, M.W., and Kamp, U. (2011). “Catastrophic partial
drainage of Pangong Tso, northern India and Tibet”. Geomorphology, 125: 109-
121.
Evans, S.G., Roberts, N.J., Ischuk, A., Delaney, K.B., Morozova, G.S., and Tutubalina,
O. (2009). “Landslides triggered by the 1949 Khait earthquake, Tajikistan, and
associated loss of life”. Engineering Geology, 109(3-4): 195-212.
https://doi.org/10.1016/j. enggeo.2009.08.007.
Fan, X., Dufresne, A., and 13 co-authors. (2020). “The formation and impact of
landslide dams – State of the art”. Earth-Science Reviews, 203, 103116, Error!
Hyperlink reference not valid..
Fan, X., van Western, C.J., Korup, O., Gorum, T., Xu, Q., Dai, F., Huang, R., and
Wang, G. (2012a). “Transient water and sediment storage of the decaying
landslide dams induced by the 2008 Wenchuan earthquake, China”.
Geomorphology, https://doi.org/10.1016/j.geomorph.2012.05.003.
Fan, X., van Western, C.J, Xu, Q., Gorum, T. and Dai, F. (2012b). “Analysis of
landslide dams induced by the 2008 Wenchuan earthquake”. Journal of Asian
Earth Sciences, 57: 25-37, http://dx.doi.org/10.1016/j.jseaes.2012.06.002.
Fan, X., Scaringi, G., Korup, O., et al. (2019). “Earthquake-induced chains of geologic
hazards: Patterns, mechanisms, and impacts”. Reviews of Geophysics, 57: 421-
503. https://doi.org/10.1029/2018RG000626.
Fangqiang, W., Chernomorets, S., Aistov, K., Petrakov, D., Tutubalina, O.,
Pengcheng, S., Yuhong, J., Aisong, X., and Petrasov, A. (2010). A Seismically
Triggered Landslide in the Niujuan Valley near the Epicenter of the 2008
Wenchuan Earthquake. Journal of Earth Science, 21(6): 901-909.Haeberli, W.,
Huggel, C., Kääb, A., Zgraggen-Oswald, S., Polkvoj, A., Galushkin, I., Zotikov, I.,
and Osokin, N. (2004). “The Loka-Karmadon rock/ice slide of 20 September 2002:
119
an extraordinary event of historical dimensions in North Ossetia, Russian
Caucasus”. Journal of Glaciology, 50(171): 533-546.
Haeberli, W., Huggel, C., Kääb, A., Zgraggen-Oswald, S., Polkvoj, A., Galushkin, I.,
Zotikov, I., and Osokin, N. (2004). “The Loka-Karmadon rock/ice slide of 20
September 2002: an extraordinary event of historical dimensions in North Ossetia,
Russian Caucasus”. Journal of Glaciology, 50(171): 533-546.
Hatch Consultants. (2017). “Dam Safety in Nepal”. Hatch Ltd, Ontario, Canada, for
The World Bank, Washington DC, USA.
Hermanns, R.L., Niedermann, S., Ivy-Ochs, S., and Kubik, P.W. (2004). “Rock
avalanching into a landslide-dammed lake causing multiple dam failure in Las
Conchas valley (NW Argentina) - evidence from surface exposure dating and
stratigraphic analyses”. Landslides, 1: 113-122. DOI 10.1007/s10346-004-0013-
5.
Hewitt, K. (2011). “Large, topographical-constrained rockslide complexes in the
Karakoram Himalaya, Northern Pakistan”. Proceedings of the 2nd World Landslide
Forum – 3-7 October 2011, Rome, Italy, 9 pp.
Hewitt, K. (2015). “Glaciers of the Karakoram Himalaya - Glacial environments,
processes, hazards and resources”. Springer, 363 pp.
Hewitt, K., Clague, J.J., and Orwin, J.F. (2008). “Legacies of catastrophic rock slope
failures in mountain landscapes”. Earth-Science Reviews, 87(1-2): 1-38.
https://doi.org /10.1016/j.earscirev.2007.10.002
Hsü, K.J. (1975). “Catastrophic Debris Streams (Sturzstroms) generated by rockfalls”.
Geological Society of America Bulletin, 86: 129-140.
Hubbard, B, Heald, A., Reynolds, J.M., Quincey, D., Richardson, S.D., Zapata, M.,
Santillan, N. and Hambrey, M.J. (2005). “Impact of a rock avalanche on a
moraine-dammed proglacial lake: Laguna Safuna Alta, Cordillera Blanca, Peru”.
Earth Surface Processes and Landforms, 30: 1251-1264.
Jiskoot, H. (2011).” Long-runout rockslide on glacier at Tsar Mountain, Canadian
Rocky Mountains: potential triggers, seismic and glaciological implications”. Earth
Surface Processes and Landforms, 36:203-2016; DOI:10.1002/esp.2037.
Kargel, J.S., Leonard, G.J., Shugar, D.H., Haritashya, U.K., Bevington, A., Fielding,
E.J., Fujita, K., Geertsema, M., and Miles, E.S. (2015). “Geomorphic and geologic
controls of geohazards induced by Nepal’s 2015 Gorkha earthquake”. SIAS
Faculty Publications, 342; https://digitalcommons.tacoma.uw.edu/ias_pub/342.
Keefer, D. K. (1984). “Landslides caused by earthquakes”. Geological Society of
America Bulletin, 95(4): 406.
Keefer, D. K. (1994). “The importance of earthquake-induced landslides to long-term
slope erosion and slope-failure hazards in seismically active regions”. In:
Geomorphology and Natural Hazards, Elsevier: Amsterdam, Netherlands, 265-
284.
Keefer, D. K. (2002). “Investigating landslides caused by earthquakes - A historical
review”. Surveys in Geophysics, 23(6): 473-510.
https://doi.org/10.1023/A:1021274710840.
120
Klein, E., Fleitout, L., Vigny, C., and J.D. Garaud, J.D. (2016). “Afterslip and
viscoelastic relaxation model inferred from the large-scale post-seismic
deformation following the 2010 Mw 8.8 Maule earthquake (Chile)”. Geophys. J.
Int., 205: 1455-1472; doi: 10.1093/gji/ggw086.
Korup, O. (2002). “Recent research on landslide dams - A literature review with special
attention to New Zealand”. Progress in Physical Geography, 26(2): 206-235.
https:// doi.org/10.1191/0309133302pp333ra.
Korup, O. and Wang, G. (2015). “Multiple landslide-damming episodes”. In: Davies, T.
(ed.) Landslide hazards, risks, and disasters. Elsevier, 241-261.
Mott MacDonald. (2017). “Hydropower Sector Climate Resilience Guidelines” Mott
Macdonald, Cambridge, UK, for The World Bank, Washington DC, USA.
Pánek, T., and Klimeš, J. (2016). “Temporal behaviour of deep-seated gravitational
slope deformations: A review”. Earth-Science Reviews,156: 14-38;
http://dx.doi.org /10.1016/j.earscirev.2016.02.007.
Parker, R.N., Hancox, G.T., Petley, D.N., Massey, C.I., Densmore, A.L, and Rosser,
N.J. (2015). “Spatial distributions of earthquake-induced landslides and hillslope
preconditioning in the northwest South Island, New Zealand”. Earth Surf. Dynam.,
3: 501-525; doi:10.5194/esurf-3-501-2015.
Peng, M., and Zhang, L.M. (2012). “Breaching parameters of landslide dams”.
Landslides, 9:13-31.
Perla, R.I. (1980). “Avalanche release, motion, and impact”. In: Colbeck, S.C. (ed.)
Dynamics of snow and ice masses. Academic Press, New York, USA, 397-462.
Perla, R.I. and Martinelli, M. (1978). “Avalanche Handbook”. U.S. Department of
Agriculture, Forest Service, Agriculture Handbook 489, US Government Printing
Office, Washington, D.C., USA.
Phartiyal B., Singh R., and Kothyari, G.Ch. (2015). “Late-Quaternary geomorphic
scenario due to changing depositional regimes in the Tangtse Valley, Trans-
Himalaya, NW India”. Palaeogeogr. Palaeoclimatol. Palaeoecol., 422: 11-24.
Plafker, G., and Ericksen, G.E. (1978). “Nevados Huascarán avalanches, Peru”. In:
Voight, B. (ed.). Rockslides and Avalanches. 1: Natural Phenomena. Elsevier,
Amsterdam, 277-314.
Podolskiy, E.A., Nishimura, K., Abe, O., and Chernous, P.A. (2010a). “Earthquake-
induced snow avalanches: I. Historical case studies”. Journal of Glaciology,
56(197): 431-446. https://doi.org/10.3189/002214310792447815.
Pollitz, F.F., Bürgmann, R., and Banerjee, P. (2006). “Post-seismic relaxation
following the great 2004 Sumatra-Andaman earthquake on a compressible self-
gravitating Earth”. Geophys. J. Int., 67: 397-420; doi:10.1111/j.1365-
246X.2006.03018.x.
Ray, K., Bhan, S.C. and Bandopadhyay, B.K. (2015). “The catastrophe over Jammu
and Kashmir in September 2014: a meteorological observational analysis”.
Current Science, 109(3): 580-591.
Reynolds, J.M. (1992). “The identification and mitigation of glacier-related hazards:
examples from the Cordillera Blanca, Peru”. In: McCall, G.J.H., Laming, D.C.J.
and Scott, S. (eds), Geohazards, London, Chapman & Hall, pp. 143-157.
121
Reynolds, J.M. (2014). “Natural disaster preparedness for hydropower projects in high
mountain environments”. HYDRO 2014 Conference Proceedings, Cernobbio,
Italy, 13th-15th October 2014.
Reynolds, J.M. (2015). “On the need to integrate Disaster Risk Management within
the hydropower sector”. HYDRO 2015 Conference Proceedings, Bordeaux,
France, 26th-28th October 2015. Panel presentation.
Reynolds, J.M. (2016). “Climate Change Resilience and Disaster Risk Management
in the hydropower sector in high mountain environments”. HYDRO 2016
Conference Proceedings, Montreux, Switzerland, 10th-12th October 2016.
Reynolds, J.M. 2017. “Integrated Geohazard Assessment as part of Climate Change
Resilience and Disaster Risk Management in the hydropower sector in high
mountain environments”. HYDRO 2017 Conference Proceedings, Seville, Spain,
9th-11th October 2017.
Reynolds, J.M. (2018). “Integrated Geohazard Assessments in high mountain
environments: examples from the Hindu Kush-Karakoram-Himalayan Region”.
ASIA 2018, Conference Proceedings, Da Nang, Vietnam, 13th-15th March 2018.
Reynolds, J.M. (2019). “On the need for transboundary management of geohazards”.
HYDRO 2019, Conference Proceedings, Porto, Portugal, 14th-16th October 2019.
Reynolds, J.M. (2022a). “Integrated geohazard assessments to increase the resilience
of hydropower schemes to future natural disasters in the Himalayan Region”.
HYDRO 2022 Conference Proceedings, Strasbourg, France, 25th-27th April 2022.
Reynolds, J.M. (2022b). “Multi-hazard risk in the hydropower sector - breaking the
bottlenecks”. Koshi Basin Knowledge Hub, The World Bank, Washington D.C.,
USA, & ICIMOD, Kathmandu, Nepal, 2nd May 2022.
Reynolds, J.M., Donnelly, C.R., Rutherford, J., Taylor, S., Karki, P., and Sulpya, R.M.
(2018). “Accounting for natural disasters in dam design and dam safety”.
Canadian Dam Association 2018 Annual Conference, Québec, QC, Canada, 15th-
17th October 2018.
Rosser, N., Kincey, M., Oven, K., Densmore, A., Robinson, T., Pujara, D.S., Shrestha,
R. Smutny, J., Gurung, K., Lama, S., and Dhital, M.R. (2021). “Changing
significance of landslide Hazard and risk after the 2015 Mw 7.8 Gorkha, Nepal
Earthquake”. Progress in Disaster Science, 10;
https://doi.org/10.1016/j.pdisas.2021.100159.
Schweizer, J., Bartelt, P., and van Herwijnen, A. (2021). “Snow avalanches”. In:
Haeberli, W., and Whiteman, C. (eds). Snow and ice-related hazards, risks, and
disasters, pp. 377-416; https://doi.org/10.1016/B978-0-12-817129-5.00001-9.
Shah, A. (2016). “The KASHMIR Basin Fault and its influence on fluvial flooding in
the Kashmir Basin, NW Himalaya”. In: Wessel, G.R. and Greenberg, J.K. (eds)
Geoscience for the Public Good and Global Development: Toward a Sustainable
Future. The Geological Society of America Special Paper 520.
Shen, D., Shi, Z., Peng, M., Zhang, L., and Jiang, M. (2020). “Longevity analysis of
landslide dams”. Landslides, 17:1797-1821, https://doi.org/10.1007/s10346-020-
01386-7.
122
Snieder, R., Sens-Schönfelder, C., and Wu, R. (2017). “The time dependence of rock
healing as a universal relaxation process, a tutorial”. Geophysical Journal
International, 208(1): 1-9; https://doi.org/10.1093/gji/ggw377.
Trubienko, O., Garaud, J.-D., and Fleitout, L. (2014). “Models of post-seismic
deformation after mega earthquakes: the role of various rheological and
geometrical parameters of the subduction zone”, Solid Earth Discuss., 6(1): 427-
466.
Van Buer N.J., Jagoutz O., Upadhyay R., and Guillong, M. (2015). “Mid-crustal
detachment beneath western Tibet exhumed where conjugate Karakoram and
Longmu-Gozha Co faults intersect”. Earth Planet. Sci. Lett., 413: 144-157.
Xu, Q., Fan, X-M, Huang, R-Q., and Van Western, C. (2009). “Landslide dams
triggered by the Wenchuan Earthquake, Sichuan Province, southwest China”.
Bull. Eng. Geol. Environ., 68: 373-386; https://doi.org/10.1007/s10064-009-0214-
1.
Chapter18: Rockfalls
Jonathan L. Hinks and John M. Reynolds
18.1 General
A major feature of the magnitude 7.9 Wenchuan earthquake of 12 May, 2008 in China
was the rockfalls which obliterated many kilometres of main roads, damaged power
plants and destroyed transmission lines. The rockfalls would have made access to
many dams impossible for long periods of time. Even a single dislodged boulder can
cause millions of US dollars’ worth of damage or rupture key arterial transport routes.
8. Figure 18.1. Typical rockfall during the Wenchuan earthquake in China on 12 May 2008.
123
Figure 18.2 Road obliterated by rockfalls in Wenchuan earthquake
As noted, rockfalls seriously damaged hydropower stations in the Wenchuan
earthquake. This was particularly the case at the Shapai hydropower scheme where
the RCC dam itself performed well. In this case an underground powerhouse would
have been the preferred solution as this would have eliminated the rockfall hazard
completely.
There is clearly a potential for boulders, some as big as houses, to cause direct
damage to various types of dam and high frequency pulses measured at the Zipingpu
CFRD are thought to have been caused in this way. Direct impingement on the facings
of such dams would doubtless be very damaging.
It is possible to deliberately dislodge potentially dangerous boulders in preparation for
road schemes and Swiss engineers, among others, regularly build reinforced concrete
shelters to protect particularly vulnerable sections of roads and railways In the seismic
design of structures and hydro-mechanical and electro-mechanical equipment ground
shaking has been and is still the main concern as practically all seismic codes,
guidelines and recommendations deal with this hazard. However, since the 2008
Wenchuan earthquake it has been realized that, in mountainous regions, the rockfall
and landslide hazards can be more devastating than ground shaking. Earthquakes
that have triggered thousands of mass movements are the 1990 Manjil earthquake in
Iran, the 1999 Chi-Chi earthquake in Taiwan, the 2008 Wenchuan earthquake in
China, the 2015 Gorkha earthquake in Nepal and others. These four earthquakes have
also caused damage to storage dams and powerhouse equipment etc. It has been
observed that, in the seismic design of dams, appurtenant structures and equipment
the rockfall hazard has often been ignored or greatly underestimated.
On the morning (local time) of Sunday 7 February 2021 a major avalanche occurred
in the Chamoli District, Uttarakhand, northern India (Figure 18.3 inset) in the environs
of the Nanda Devi National Park, a UNESCO World Heritage Site in the outer Garhwal
Himalayas in Uttarakhand state, India (Reynolds, 2021). As summarised by Shugar
et al. (2021), at 4:51 UTC (10:21 Indian Standard Time [IST], ~27 million m³ of rock
(80%) and ice (20%) detached from the steep northern flank of Ronti Peak (6,029 m)
from an elevation of ~5,500 m asl and fell ~1,800 m down the steep face into the Ronti
Gad valley floor. The large avalanche, better referred to as a ‘sturzstrom’ (Hsű, 1975),
was partly airborne during its descent, fragmented further on impact below and
pulverised much of the rock mass. This is thought to have generated a great deal of
124
heat (Hanisch and Schulze, 2012) that melted ice avalanche content (~5.4 million m³;
equivalent to ~5 million m³ of water) and the remnants of previous ice avalanches from
the northern flanks of Ronti Peak that had been deposited along the floor of Ronti Gad
valley.
This contributed a large volume of water to the debris mix (discussed further below).
The impact also created a huge dust cloud that smothered the lower flanks of Ronti
Gad, as can be seen in Figure 18.3, and flowed down valley. The resulting avalanche
run out transported some rock blocks up to 20 m diameter and chunks of ice of ~1 m
diameter and flowed at speed along Ronti Gad (blue arrow) and blocked the
Rishiganga Valley (yellow star) and formed a backwater impoundment. It destroyed
the under-construction 13.2 MW Rishiganga HPP, engulfed the tunnels, trapping and
killing many workers inside. It progressed to demolish the road bridge at Raini (brown
arrow), depositing much of its coarse material by this point. Upon flowing into the
Dhauliganga Valley it morphed into more of a fast-moving water-rich debris/mud flow
(light brown arrow) that overwhelmed the headworks at the 520 MW Tapovan HPP
largely with mud with some ice blocks and continued downstream. As the sturzstrom
moved along Ronti Gad valley it incorporated any flow within the river, as it also did
when it joined both Rishiganga and Dhauliganga rivers, increasing its water content.
Figure 18.3 3-D perspective across the area affected by the 7 February 2021
sturzstrom viewed towards the south with (inset) the location of the disaster in Chamoli
District, Uttarakhand, northern India. Google Earth image dated 10 February 2021
(Reynolds et al., 2021)
Despite oft-cited local and international media reports, including early statements from
official government sources, stating that the event was caused by a “glacier burst”, a
“Glacial Lake Outburst Flood”, or a breach of a temporary landslide dam, none was
true. There are no glacial lakes within the Ronti Gad catchment and the impact zone
of the sturzstrom in Ronti Gad was ~1.5 km downstream of the snout of the nearest
debris-covered glacier, which was not involved in the event in any way.
125
Taking advantage of a previously unseen high resolution (0.75 m) Jilin-1 satellite
image dated 20 September 2020, coupled with undertaking an analysis of 2-m
resolution Digital Elevation Models, obtained from WorldView-1 and WorldView-2
images processed following methodology proposed by (Shean et al., 2016), available
for 2015 (Bushan et al., 2021) and 10-11 February 2021 (Shean et al., 2021), Reynolds
et al. (2021) were able to determine further perspectives of the source area.
The structural geological and lithological boundaries of the area local to the Chamoli
sturzstrom of 7 February 2021 were tentatively mapped by analysing a combination of
digital elevation models (DEMs), published literature, and open-source map data, as
well as the recent 0.75 m resolution Jilin-1 satellite image. Partially identifiable
features include a series of northeast-southwest and east-west trending thrust faults,
lineaments, and joint and fracture sets (Figure 18.4). Some of these features display
an apparent radial geometric orientation, appearing to slightly change in direction, like
a fan-like shape. Despite surface mapping using the aforementioned data types, the
presence of many blind structures with no obvious surface expression are expected;
however, due to the continual reworking of the landscape by erosion, deposition, and
meteorological processes they remain unmapped. a fan-like shape. Despite surface
mapping using the aforementioned data types, the presence of many blind structures
with no obvious surface expression are expected; however, due to the continual
reworking of the landscape by erosion, deposition, and meteorological processes they
remain unmapped.
126
Figure 18.4 Interpreted indicative photo-lineament trends associated with the 7
February 2021 sturzstrom source area (blue arrow; Figure 18.3) in the small redsquare
marked on a high-resolution satellite image dated 26 September 2020. (Reynolds et
al., 2021)
The main features evident in Figure 18.4 are a northeast-southwest (dashed yellow
lines) series of postulated faults, an east-west set of structures (brown lines) with some
that are slightly oblique to this trend, but which are apparently related. There is also
a northwest-southeast trend (white dashed lines) in the northeastern part of the image.
These structures appear to terminate in the Ronti Gad Valley. Distinctive
geomorphology is apparent along the west facing Ronti Gad Valley, especially in the
northern part of the image, which might be more commonly associated with more
massively structured lithologies, such as those listed by Rana et al. (2021) (e.g.,
conglomerates and quartzites). This is strongly suggestive that there is a marked
difference in lithologies between the western and eastern side of the valley.
Furthermore, the distinct change in orientation of structures either side of the valley
(white vs yellow dashed lines) indicates the probability that the valley exploits a major
northwest-southeast/north-south trending fault (purple dashed line). It was postulated
127
by Reynolds et al. (2021) that this might be a more westerly location for the South
Tibetan Detachment. The area within the small red square is the primary sturzstrom
source area shown in Figure 18.3.
Another major feature shown in Figure 18.4 is an arcuate set of fractures facing north,
that are cross-cut by other fault/joint sets to form a blocky step-like feature (within the
large red rectangle).
Evidence of large deposits of what look like landslide deposits like those from the 2021
sturzstrom but from much earlier events, of the order of ~200 m thick is present on the
Jilin-1 September 2020 image as well as on various Google Earth scenes. One
location is indicated on Figure 18.4.
Previously unseen high-resolution Jilin-1 satellite imagery of the area before the event
coupled with that available on Google Earth afterwards yield remarkably clear
evidence of the avalanche source area and surroundings. This permitted a detailed
structural geological assessment of the mountain flanks in which the source area was
located from which a clearer picture of the constraints on the rock avalanche that
occurred has been developed, as discussed above. There appears to be a clear link
between the geological and environmental preconditioning of this mountain
(geological structures; accumulation and ablation of glacier ice and snow; permafrost),
coupled with the residual ice avalanche debris from previous ice avalanche events.
Warming temperatures in these regions are having an increasing impact by increased
thawing of permafrost and subsequently destabilisation of very steep mountain flanks.
By considering all these aspects a much better understanding of the controls at work
in this environment has been developed. The paper by Reynolds et al. (2021) also
revealed that the rock and ice avalanche in February 2021 is unlikely to be the last
that occurs from this source area.
References
Bhushan, S., and Shean, D. (2021). 2021 Chamoli disaster pre-event 2-m DEM
composite: September 2015. Zenodo; doi:10.5281/zenodo.44554647; 2021.
Hanisch, J. and Schulze, O. (2012). The genesis of Huascarán-like sturzstroms – a
thermo-dynamic approach. Proceedings of 11th International Symposium on
Landslides, Banff, Canada, pp. 901-906.
Hsű, K.J. (1975). Catastrophic debris streams (sturzstroms) generated by rockfalls.
Geological Society of America Bulletin, 86:129-140.
Rana, N., Sharma, S., Sundriyal, Y., Kaushik, S., Pradhan, S., Tiwari, G., Khan, F.,
Sati, S.P., and Juyal, N.A. (2021). A preliminary assessment of the 7th February 2021
flashflood in the lower Dhauli Ganga valley, Central Himalaya, India. Journal of Earth
System Science, 130(78), https://doi.org/10.1007/s12040-021-01608-z.
Reynolds, J.M. (2021). Statement on the disaster in Chamoli District, Uttarakhand,
northern India, on 7 February 2021. The International Journal of Hydropower & Dams,
28(2):8, March 2021, also more detail at https://www.hydropower-dams.com /7032-2/.
Reynolds, J.M., Diaz-Moreno, A., and Noble, R. ( 2021 ). Forensic review of the
Chamoli Disaster, Northern India, February 2021, and implications for the hydropower
sector. International Journal of Hydropower & Dams, 28(5):31-38.
Shugar, D.H. and 52 co-authors. (2021). A massive rock and ice avalanche caused
the 2021 disaster at Chamoli, Indian Himalaya, Science, 10.1126/science.abh4455.
128
Shean, D., Alexandrov, O., Moratto, Z. M., Smith, B.E., Joughin, I.R., Porter, C., and
Morin, P. ( 2016 ), An automated, open-source pipeline for mass production of digital
elevation models (DEMs) from very-high-resolution commercial stereo satellite
imagery. ISPRS J. Photogramm. Remote Sens. 116:101-117; 2016.
Shean, D., Bhushan, S., Berthier, E., Deschamps-Berger, C., Gascoin, S., and Knuth,
F. ( 2021 ).Chamoli Disaster post-event 2-m DEM composite (February 10-11,
2021) and difference map (v1.0). Zenodo. doi:10.5281/zenodo.4558692; 2021.
129
APPENDIX 3: 2002 PAPER ON qra FOR HYDRAULIC GATES
9. APPENDIX 3: 2002 PAPER ON qra FOR HYDRAULIC GATES
Simplified Quantitative Risk Assessment for Hydraulic Gates Published in
International Hydropower and Dams, Issue 2, 2021.
Simplified Quantitative Risk Assessment for Hydraulic Gates
J.L.HINKS HR Wallingford, United Kingdom
R.DIGBY KGAL Consulting Engineers, United Kingdom
P.J.BAUR Independent Consulting Hydrologist, United Kingdom
R.S.STILWELL Independent Consulting Civil Engineer, Canada
SYNOPSIS
In many dam projects, the mechanical and electrical systems/components associated
with the operation of spillway and low-level outlet gates can be regarded as “safety
critical”. This is due to the fact that failure to operate gates as and when required can
have a detrimental effect on the ability of the dam to pass flood flows. It should also
be noted that gates and their operating mechanisms are particularly vulnerable during
extreme events such as violent electrical storms, hurricanes and earthquakes. This
paper will focus on spillway gates.
When reviewing risk, detailed Failure Modes and Effects Analysis ( FMEA), relying on
fault tree investigations, is a time consuming and expensive activity. It requires
geotechnical field investigations when design data are unavailable. Quantitative Risk
Assessment (QRA) is a shortened version of FMEA appropriate when many dams
must be assessed for a single owner. It relies on review of dams by expert engineers,
using their familiarization with dam safety industry data. The approach must be used
with caution and is not recommended for large dams in developed countries [ Rigbey,
20191 ].
The procedure can be applied to a portfolio of dams and failure modes (eg
geological/geotechnical, seismic, hydrological, etc ) to estimate probabilities of
different outcomes leading to economic analyses of remedial options, as exemplified
here with particular reference to Valdesia dam in the Dominican Republic.
GENERAL RELIABILITY PRINCIPLES FOR SPILLWAY GATES
A gated spillway should, ideally, have a target reliability (i.e. the probability of failure
on demand ( PFD ) of the order of 10-4 per year [ Lewin, 20022 ]. However, this is not
easy to achieve in practice without providing several levels of operating system
redundancy.
Note that the levels of redundancy required should be clearly identified through the
structured use of the hazard identification (HAZID) or hazardous operation (HAZOP)
process, perhaps even involving failure modes and effects analysis (FMEA). Ultimately
a Fault Tree Analysis (FTA) can be used to provide a full quantitative estimate of the
PFD, but as mentioned above, that is beyond the remit of this paper).
Following the completion of the HAZID/HAZOP/FMEA process elements as required,
redundancy in the operating system will most likely to be found in several key areas:
130
 operator access to a site prior to a major flood or hurricane.
 power supply
 operating mechanism
 operating controls
 operating system sensors and switches
 spillway gate structural design and maintenance [ USBR3].
It is worth noting that many dams have a relatively high potential rate of failure in
modest floods ( eg. 100 years return period ) when combined with the probability that
all gates may fail to open. In terms of putting reliability into a more practical context,
the following example is worthy of note.
At the Tavera dam in the Dominican Republic reliability ( failure on demand ) of the
spillway gates was estimated to be of the order of 3 x 10-3 per year. This relates to
failure on demand not dam failure ). A higher figure will apply at Valdesia which is
referred to below.
The experience of the authors is that such low reliabilities are not unusual. The gates
at Tavera failed to open in Hurricane David in 1979 because of a power outage and
failure of the standby generator. Despite heroic efforts at manual opening by soldiers,
water level rose to within 2 m of the crest of the dam. Later there was no mains power
in Tropical Storm Hortensia ( 1996 ) or in Hurricane Georges ( 1998 ) because the
electricity grid is generally shut down during hurricanes.
STANDBY GENERATORS
In the context of having power available to operate gates when required, standby
generators are extremely important. They do, however, require reliable starting
arrangements and should be run regularly under load. This can often prove
problematic in practice as it may not be permissible to operate the gates at the time or
in the manner required (often due to water discharge limitations) or due to the lack of
a dedicated load-bank against which to run the generator. The need to carry out
regular maintenance and testing of the generator cannot be over-stated.
131
Figure 1. Standby generator at Tavera.
Mobile generators or power packs may be provided in addition although these are of
limited usefulness if they are shared between projects.
Figure 2. Power pack at Sirikit Dam in Thailand

CABLING
Separate cabling (power and control) to gates is advised. Armoured cables with PVC
coating should generally be used and be routed through ducts with adequate slack to
avoid damage in earthquakes.
Power and control cables should ideally be run in separate ducts, but where this is not
possible, cable spacing should be consistent with the relevant electro-magnetic
compatibility (EMC) directives to prevent cross talk and signal interference.
OPERATIONAL CONSIDERATIONS
Where staff need to get to a dam to operate the gates the following points need to be
borne in mind:
1. Availability of qualified and trained staff on call, including senior management
for approval/direction, as per the requirements of the operating manual
2. Dedicated, secure phone lines are operational
3. Availability of a vehicle that is operational, with sufficient gasoline and
charged battery
4. Access roads are passable and not obstructed by washout, landslide, fallen
trees or road subsidence
5. Bridge crossings, ( e.g. of the tailrace ), are not flooded
6. Access to the control building with appropriate keys
7. In case of power failure, an emergency power supply with fuel and charged
battery is ready to go
8. Gate operating equipment works
9. Gate raising is attempted one at a time, not all together
132
Procedures for gate opening can be quite complicated and this can affect the time
required to open the gates ( two hours at Valdesia- see below ). This is why there is a
need for staff to be adequately trained in the use of the gates using primary,
secondary, tertiary and any other backup systems on a regular, scheduled basis to
ensure familiarity with all modes of operation and to prove that these backups are
working and available.
STATISTICAL APPROACH
In the reference “Hydraulic Gates and Valves” [ Lewin, 20014 ] Table 12.1 lists 30 major
failures of spillway gate installations, noting that the list is not comprehensive. In Table
12.3 he reports a survey [ Lagerholm , 19965 ] of 362 radial gates in Sweden having a
fault frequency of 6.5% per gate per year. This figure presumably includes for quite
minor faults which may affect only one or two spillway gates at a particular dam.
If it is assumed that, say, 1% of the above failures have serious implications for the
safety of the dam, this equates to an overall M&E annual probability of dam failure of
650 x 10-6 for a dam with the spillway controlled by a number of radial gates. This
figure is of the same order of magnitude as the independently derived figure of 500 x
10-6 quoted in Table 1 below for the Valdesia dam where all five gates were lost in
Hurricane David in 1979.
It should be noted that while the above figures relate to radial gates, values for vertical
lift gates of the sliding or roller type will be different.
More worldwide data on the reliability of spillway gates would be very useful as would
specific data on reliability in typhoons and hurricanes, since it is under these conditions
that the need to open the gates is likely to coincide with the lowest reliability.#
CASE HISTORY - VALDESIA DAM
The 78 m high Valdesia hollow gravity dam in the Dominican Republic was completed
in 1976. It provides an interesting case history culminating in economic analyses of
measures to improve the safety of the structure.
Figure 3. Valdesia Dam, Dominican Republic.
133
Spillway Gates
The five spillway gates can pass 47 % of the Probable Maximum Flood. Prior to a
study in 2003 [ Halcrow/Cor Ingenieria6 ] consideration had been given to the
construction of an ungated auxiliary spillway to allow passage of the PMF with all gates
open, or the 10,000 year flood with all gates closed. The estimated cost of this
proposal was USD 36.2 million (2019 prices) which was considered prohibitive. A
possible alternative of a fuse plug spillway is not discussed in this paper.
There were some particular problems with the spillway gates at Valdesia, not least
because they were each 24.0 m long but only 5.0 m high. They were, therefore, prone
to twisting and jamming when being operated. Large quantities of floating debris are
also thought to have played a part. In addition, the original gate design may not have
had adequate nappe aeration leading to larger than originally designed hydraulic
loading if overtopped by a large flood.
Each gate was operated by two hoist units, one on either side of the gate. Unlike at
many dams, the two hoist units were not mechanically interconnected by a shaft.
Cross-synchronisation was, instead, effected by an electro-mechanical system
whereby gate end positions were transmitted by electrical impulse to an integrating
unit which displayed lead or lag on a dial. When either lead or lag exceeded a pre-set
value a limit switch was activated and the hoist or lower motion stopped. It had then
to be corrected to bring the two ends into alignment. It was not a comforting thought
that this might well have to be done under hurricane conditions.
Figure 4. Spillway gate at Valdesia Dam.

Seismicity
The most likely source of an earthquake damaging the dam is thought to be the
Enriquillo-Plantain Garden Fault, although some sources show the fault terminating
about 60 km west of Valdesia.
A seismic hazard study [ De Mets and Wiggins-Grandison, 20077 ] predicted a
maximum earthquake magnitude of 7.2 on the E-PG Fault ( although the earthquake
of 7 June 1692 which destroyed Port Royal in Jamaica, and which was on the same
134
fault system, is stated by McCann [ 20048 ] to have had Moment Magnitude of 7.5 ).
Assuming a focal depth of 15 km and an epicentral distance of 10 km, the mean Peak
Ground Acceleration ( PGA ) returned by 6 commonly used attenuation formulae was
0.60 g (see below) for a return period of 27,500 years. This method, assuming an
event of moment magnitude 7.2, is thought to be the more appropriate for extreme
events near the damsite.
Hinks and Gosschalk [19939 ] imply that, based on experience at Sefid Rud,
Hsingfengkiang, Koyna and elsewhere, a PGA of perhaps 0.6 to 0.8 g would be
needed to destroy a large buttress or gravity dam31. The gates near to the crest of
Valdesia dam could be subjected to high seismic accelerations, hydrodynamic loads
and possibly loads arising from seismic seiches in the reservoir. Risks associated with
earthquake damage were estimated and included in the estimates of annual failure
probabilities given in Table 1.
Hansen K.D. and Nuss L.K. [ September201110 ] give updated details of large concrete
dams that have been subjected to Peak Horizontal Ground Accelerations ( PHGAs )
greater than 0.3 g. With the exception of the Shih Kang weir in Taiwan in 1999 none
of these dams suffered major failure.
Table 1. Estimated probabilities of dam failure by Failure Mode
Cause of Failure Annual Probability of Failure
Hydrological 100 x 10-6 ( 10,000 year flood

)
Geological/Geotechnical [ 14 x 10-6 ( see Ref 11 )

Douglas et al 11]
Mechanical & Electrical 500 x 10-6 ( 5 % probability of

non operation of gates in 100 yr
flood )
Seismic 36 x 10-6 ( see text )
Adjustment for Monitoring 34 x 10-6 ( lack of

piezometers )
TOTAL 684 x 10-6
It will be clear that the primary cause of dam or gate failure is likely to be the occurrence
of a hurricane with very high winds and intense rainfall. Opening of the spillway gates
in advance of such an event is strongly advised where possible. Alternatively raising
the gates at the beginning of the wet season could be considered although this would
involve generation losses.
In the experience of the authors, operational factors can lead to overtopping of the
gates (if they are not opened) or flooding downstream (if they are opened excessively
31
The only large concrete dam to have failed in an earthquake is the Shih Kang weir in Taiwan which failed in 1999. The PHGA
was about 0.51g but failure was caused by a 7 m vertical fault movement beneath the dam.
135
). In fact, for whatever reason, the gates at Valdesia were not opened in Hurricane
David on 30 August, 1979 and collapsed as a result of being overtopped to a significant
depth, flooding the 54 MW power station downstream.
Remedial Options
Two principal options for improving the reliability of gate operation at Valdesia were
considered. The first involved cross connecting the gate hoists using torque shafts.
This would have required extensive modification to the hoist units and to the gate
structures and would have been costly.
Alternatively one could employ a number of measures including use of variable speed
motors and programmable logical control (PLC) based on proportional, integral,
derivative (PID) control. Such a system is not costly but requires backup devices which
can work on battery power.
In addition to the above the following were also suggested:
 The gates were supplied by a single cable looped in and out of each gate.
This was not considered satisfactory because failure in any section of the
feeder cable would have caused the failure of electrical supply to all the
gates. Separate cables were recommended for each gate.
 Replacement of the simple PVC insulated cables with PVC wrapped with
steel wire armour which is, in turn, PVC coated. This protects the cables
from inadvertent damage and is valuable under earthquake conditions
when sections of concrete can fall on the cables or when the cables may
be subjected to shear forces.
 All floor mounted control panels should be secured against tilting during an
earthquake.
 Adequate slack should be provided at terminations to allow for movements
during earthquakes.
 Cables should be run on saddles in trunking.
 Standby generators should be run regularly under load.
Implementation of remedial measures leads to a reduction in risk of failure (Table 2).
Table 2. Effect of remedial measures on annual probabilities of dam failure
Cause of Failure Annual Probability of Failure
M&E Improvements M&E

Improvements
only
plus Auxiliary
Spillway
Hydrological 100 x 10-6 0 x 10-6
Geological/Geotechnical 14 x 10-6 14 x 10-6
Mechanical & Electrical 250 x 10-6 2.5 x 10-6
Seismic 36 x 10-6 36 x 10-6
136
Adjustment for 34 x 10-6 34 x 10-6
Monitoring
TOTAL 434 x 10-6 86.5 x 10-6
To avoid double counting for the M & E improvements plus auxiliary spillway the
probability of hydrological failure is taken as zero. The risk of a flood with a return
period greater than 10,000 years combined with the residual probability of M&E failure
is estimated at only 2.5 x 10-6 / year.
Geological and Geotechnical risk for concrete structures is impossible to estimate
when a record of the dam design is not available. A preliminary stability analysis must
be done in that case, using foundation uplift and shear strength data. If not available,
resort can be made to Stone and Webster [ 199212 ] for typical shear strength and uplift
distribution parameters based on test data.
The raft of M&E improvements recommended above was estimated to cost USD
240,800 (2019 prices).
Economics
The annual probability of failure when multiplied by the estimated cost of failure ( Table
3 ) gives the theoretical annual premium payable to insure against dam failure in a
perfect market (Table 4). This is USD 281,800 per annum at 2019 prices (Table 4).
Table 3. Estimated cost of dam failure
Cost Element Estimated Cost (USD

million)
Replace Valdesia dam and power 236.4

station
Repair Las Barías dam downstream 47.6
Loss of Life (see below) 33.8
Loss of water supply to Santo 12.0

Domingo
Loss of housing 0.8
Agriculture and infrastructure 81.4
TOTAL 412.0
An important, but largely unexpected element, was the high cost of suddenly depriving
the city of Santo Domingo, with its 4 million inhabitants, of 40 % of its water supply [
Edelmann, 201913 ] . The cost of bringing in emergency supplies would be significant
and there might be outbreaks of water- borne diseases with accompanying fatalities.
137
The Value to Prevent a Fatality (VPF) can be calculated by taking the VPF used by
the Highways Agency in UK (USD 2.19 million) and scaling it according to the GDP
per capita in the Dominican Republic (USD 8,629) compared to that in the UK (USD
39,532). This yields a VPF of USD 0.48 million although a lower figure (USD 0.1
million) was actually used after consultation with the Government of the Dominican
Republic.
Multiplying the costs of failure by the relevant probabilities of failure the figures in Table
4 were obtained:
Table 4. Insurance premiums payable in a perfect market
Condition Failure Cost Annual Theoretical

(USD m) Failure Premium
Probability USD/year
Present 412.0 684 x 10-6 281,800
With M&E 412.0 434 x 10-6 178,800

improvements
With auxiliary 412.0 86 x 10-6 35,400

spillway plus
M&E
improvements
The figures in Tables 2 and 4, referring to M&E improvements only, assume that the
probability of dam failure, because of M & E failure, will be reduced by 50 % by the
implementation of the proposed measures. For M&E improvements only a capital
expenditure of only USD 240,800 will save USD 103,000 per annum in theoretical
premium, making it a very attractive investment as well as having the potential to save
human life.
We may summarize as follows:
(1) The theoretical insurance premium is presently USD 281,800 per annum.
(2) This can be reduced by USD 103,000 per annum for a one time capital cost of
USD 240,800 (i.e. by improving the reliability of gate operation). Assuming an
economic life of 25 years and a discount rate of 5% the reduction in premium
would be capitalised at USD 1.45 million ( Benefit/Cost = 6.02 ).
(3) The economic case for more extensive M & E improvements ( eg the
introduction of torque shafts ) would be worth studying, but has not so far been
investigated in detail.
(4) The construction of a new auxiliary spillway would reduce the annual premium
by a further USD 143,400 per annum but the cost of doing so (USD 36.2 million)
would be prohibitive. Assuming an economic life of 25 years and a discount
rate of 5% the reduction in premium would be capitalised at only USD 2.0
million.( Benefit/Cost = 0.05 ).
CONCLUSIONS
138
When assessing the reliability of spillway gates on dams, it is crucial to understand the
basic requirement for adequate levels of redundancy to be provided in terms of power
supply, operating mechanism, operating controls, operating system sensors/switches
and the spillway gate structural design.
Methods of achieving the required level of reliability therefore rely heavily on
conservative engineering design and ensuring that the redundant systems provided
will, indeed, cover all foreseeable failure modes identified using the
HAZID/HAZOP/FMEA processes.
The return period of failure of a large concrete dam with a community downstream
would normally be expected to be at least 10,000 years which is close to the
statistically derived figure of 7,143 years reported by Douglas et al [ 1998 11 ] for
concrete gravity dams over 5 years old and commissioned between 1930 and 1992
(they do not quote figures for buttress or hollow gravity dams). ICOLD Bulletin 188 (
2019 14 ) is also relevant.
The exercise described here obtains a present return period for failure of Valdesia
dam of just 1,460 years ( the return period for collapse of the spillway gates will be
much lower ). The present low value takes account of:
 Hurricanes, with extreme rainfalls and windspeeds, of which there are
reported to have been 11 in the last 80 years
 High risk of seismicity
 Low reliability of spillway gates
 Heavy siltation in the reservoir ( increasing the load on the dam )
 High uplift pressures beneath the structure
 Lack of monitoring ( particularly piezometers ).
With modest investment, in improving spillway gate operation, the return period for
failure of the dam could be increased to about 2,300 years. This is still a very low
figure for a major dam and ways should probably be sought to achieve further
increases in the overall return period for failure. These could include drawing down the
reservoir whenever a hurricane is forecast, more comprehensive improvements to the
spillway gate operation system ( such as the introduction of torque shafts ),
improvements to dam instrumentation and measures to prevent blockage of the power
intake trashracks.
Simplified Quantitative Risk Assessments will rarely be adequate for large dams in
developed countries [ see Rigbey, 20191 ] where much more expensive analyses,
based on formal fault trees, are likely to be required. In less developed countries, and
where ranking is needed of measures at a portfolio of dams, the methodology
described in this paper, however, encourages the practitioner to make use of all readily
available information in a structured way.
In the longer term the systematic gathering of international data on the operational
reliability of spillway gates is desirable to improve the results at dams with gated
spillways.
Acknowledgements
The authors would like to express their very sincere thanks to the Instituto Nacional
de Recursos Hidráulicos ( INDRHI ) of the Dominican Republic and, in particular, to
139
Ing. Olmedo Caba Romano ( Executive Director ) and Ing José Raúl Pérez ( Planning
Manager ) for their assistance and for allowing the publication of this paper, which
does not necessarily express the views of INDRHI.
References
1. Rigbey S. “Looking beyond Oroville: Where to from here ?”, Canadian
Dam Association Bulletin, Winter 2019.
2. Lewin J. Personal Communication, 2002.
3. US Bureau of Reclamation (USBR) Mid-Pacific Region Sacramento, California
American River, Document No. TC 555 W45, Forensic Report on Spillway Gate
3 Failure Folsom Dam,1996
4. Lewin J, “Hydraulic Gates and Valves”, Second Edition, Thomas Telford, 2001.
5. Lagerholm, S. “Safety and Reliability of Spillway Gates”, ICOLD Symposium,
Repair and Upgrading of Dams, Stockholm, June 1996.
6. Halcrow en Asociación con Cor Ingeneria SA, “Estudio de la Vulnerabilidad de
las Presas Grandes en la República Dominicana, Informe Final”, Enero 2003.
7. DeMets, C.; Wiggins-Grandison W. (2007). "Deformation of Jamaica and
motion of the Gonâve microplate from GPS and seismic data" (PDF).
Geophysical Journal International. 168 (1): 362–378.
8. McCann W.R. Estimating the threat of tsunamigenic earthquakes and
earthquake induced-landslide tsunami in the Caribbean. In Mercado-Izary A
and Liu P ( Ed ) Caribbean tsunami hazard. Proceedings of the NSF Caribbean
Tsunami Workshop. World Scientific Publishing Co pp 43 – 65., 2004.
9. Hinks, J.L. and Gosschalk, E.M, “Dams and Earthquakes – A Review”, Dam
Engineering, February, 1993.
10. Hanson K.D and Nuss L.K “Lessons learned from the Earthquake Performance
of Concrete Dams”, International Water Power and Dam Construction,
September 2011.
11. Douglas K, Spannagle M and Fell R. “ Estimating the Probability of Failure of
Concrete and Masonry Gravity Dams”, University of New South Wales, 1998.
12. Stone and Webster, “Uplift Pressure, Shear Strengths and Tensile Strengths
for Stability Analysis of Concrete Gravity Dams”, Report EPRI TR- 100345,
1992.
13. Edelman D.J, “Managing the Urban Environment of Santo Domingo, the
Dominican Republic” . Current Urban Studies, March 2019.
14. ICOLD Bulletin 188 ( Preprint ) “Dam Failures – Statistical Analysis”, 2019.
140

013 Esmr en - 0

Uploaded by

Copyright:

Available Formats

013 Esmr en - 0

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

013 Esmr en - 0

Uploaded by

Copyright:

Available Formats

Environmental and Social Monitoring Report

Project Number: 43281-013

Nepal: Tanahu Hydropower Project – Panel of

Tanahu Hydropower Project

Third Report of Panel of Experts

Jonathan Hinks Susan Novak Jose M. Cabral

SECTION B – DAM SAFETY REPORT................................................................... 15

SECTION C - ENVIRONMENTAL SAFEGUARDS REPORT ................................. 35

SECTION D – SOCIAL ISSUES .............................................................................. 53

DDC District Development Committee

NFTP Non-timber forest product(s)

 Jonathan Hinks, Dam Safety Expert

 Susan Novak, Social Development Expert

 José Cabral, Environmental Expert

 The hiring of a Civil Engineer represents an important technical reinforcement

 An additional way in which the participation of the PSC in monthly EHS

 Regarding community safety, an average of 30 to 40 people per day walk

 The discussion concluded that the best mitigation approach would be to

Figure 3. Upper catchment of River Seti

Return Period ( Years ) Peak Inflow ( m3/s )

2. PSC’s Response as to Sediments

A sediment flushing study has been conducted in the preparational phase of

 Thereafter, power production is discontinued, and the water lowering outlets

At the present stage, reservoir operation is envisaged as intended in earlier

Contractor POE POE POE

The contractor’s calculations assume a Demand-capacity Ratio of 1.5 as previously

Block 0.8 g 0.75 g 0.7 g

Table 3. Manjil Earthquake

Note: Stress exceedance areas exceeding 15 % at 0.75 g are highlighted

2.8.4 Return Period for dam failure due to seismic cracking

 Hsingfengkiang buttress dam in China, 1962

 Koyna gravity dam in India, 1967

 Pacoima arch dam in USA, 1994

2.9 Other Issues

2.9.2 Spillway Gates

2.11 Power Intake

In the absence of a detailed engineering geological survey, this impression

 Construction joint monitoring

 Pendulums and Inverted Pendulums ( with dataloggers )

 Piezometers and pressure gauges

 Monitoring of crest movements

 Seismographs on rock and on the crest of the dam

2.17 Engineering References

DATES TIMES ACTIVITIES AND OBJECTIVES PARTICIPANTS

DATES TIMES ACTIVITIES AND OBJECTIVES PARTICIPANTS

 In general, workers wore appropriate Personal Protective Equipment (PPE) for

 Overall, housekeeping in workshops and yards was fair and needs

 Hazardous materials and wastes management: in contravention of contract

Figure 17. Inadequate storage of hydrocarbons in storage room

Figure 19. Inadequate storage of used motor oil

 Health services: in contravention of contractual obligations, there is no

 Solid waste management: there is no separation, reuse or recycling of domestic

 Topsoil management: contrary to environmental management specifications

 Crushing and batching plants: the inadequate treatment of wastewater coming

 Fire extinguishers: there was an insufficient number of fire extinguishers at the

 The EHS staffing levels comply with contractual specifications.

 In general, workers wore appropriate Personal Protective Equipment (PPE) for

 Overall, housekeeping in workshops and yards was satisfactory, and there

 Solid waste management: there is no separation, reuse or recycling of domestic