Guidelines for Biodiversity Assessment and Monitoring for Protected Areas

Guidelines for Biodiversity
Assessment and Monitoring
for Protected Areas

Authors

Graham Tucker (Ecological Solutions), Philip Bubb (UNEP-WCMC), Mireille de Heer
(UNEP-WCMC), Lera Miles (UNEP-WCMC), Anna Lawrence and Jeanette van Rijsoort
(Oxford University Environmental Change Institute), Siddhartha B. Bajracharya
(KMTNC), Ram Chandra Nepal (KMTNC), Roshan Sherchan (KMTNC) and Nawaraj

Chapagain (KMTNC).

TRUST FOR NATURE

:
iN
1 A

NEPAL. 1982

® NOLLVANSSNOD

UNEP etait

Digitized by the Internet Archive
in 2010 with funding from
UNEP-WCMC, Cambridge

http://www.archive.org/details/guidelinesforbio05gtuc

NANiue TF . ASH

Sue

Guidelines for
Biodiversity Assessment and
Monitoring for Protected Areas

Protected Area Monitoring Guidelines

Published by:

Copyright:

Citation:

Layout & Printed by:

Available from:

Contact:

The King Mahendra Trust for Nature Conservation, Nepal and
the UNEP-World Conservation Monitoring Centre, Cambridge, UK

e 3 (

Vv, y | p)
Vv

YOO) (

UNEP WCMC

TRUST FOR NATURE

an eg

\WW%/

NEPAL. 1982

@ KING MAHENDRA
© NolLwAuasNOD

© 2005 KMTNC and UNEP-WCMC

Reproduction of this publication for educational and other non-
commercial purpose is authorised without prior written permission from
the copyright holder provided the source is fully acknowledged.

Tucker, G., Bubb P., de Heer M., Miles L., Lawrence A., Bajracharya S.
B., Nepal R. C., Sherchan R., Chapagain N.R. 2005. Guidelines for
Biodiversity Assessment and Monitoring for Protected Areas. KMTNC,
Kathmandu, Nepal.

Graphic Plus, Nepal # 4267682

Information Unit
King Mahendra Trust for Nature Conservation
P O Box 3712 Kathmandu, Nepal

e-mail: info@kmtnc.org.np

The Guidelines are based on the practical experience of KMTNC and
UNEP-WCMC. We welcome comments and suggestions. Please contact

either info@kmtnc.org.np or info@unep-weme.org.
Cover Photo: Siddhartha Bajra Bajracharya

This publication has been produced with the support of the UK Darwin Initiative as part

of project 163/11/020 "Building capacity for biodiversity monitoring and assessment in
Nepal". The Darwin Initiative is a small grants programme that aims to promote
biodiversity conservation and sustainable use of resources in less developed countries.
The Initiative is funded and administered by the UK Department for Environment, Food
and Rural Affairs, (Defra).

http://www.darwin.gov.uk/index.htm

Protected Area Monitoring Guidelines

Table of Contents

Acknowledgements

Foreword

Foreword

Foreword

Office-in-Charge, UNEP World Conservation
Monitoring Centre

Vice-chairman, National Planning Commission
Member Secretary, King Mahendra Trust for

Nature Conservation

1. Introduction

1.1

12

Background to the guidelines

Use of the guidelines

2 Definitions and purpose of biodiversity assessment and
monitoring for protected area management

2.1

ED)
DS

Biodiversity assessment and monitoring as part of
protected area management planning

What is a biodiversity assessment?

What is monitoring?

3 Carrying out a biodiversity assessment for a protected area

3.1
32
3.3)
3.4
3h)
3.6

Defining the scope and approach of the assessment
Creating an enabling environment for participation

Data gathering and review

Biodiversity evaluation

Identification of constraints, opportunities and pressures

From the assessment to setting objectives

4. Developing a protected area biodiversity monitoring programme

4.1
4.2
4.3

Introduction
Determine what needs to be monitored

Review existing data

Protected Area MOMslorit1e | GUs Gel 11 C8 mm

111

Vi

4.4 Define the location and boundaries of the biodiversity
features to be monitored 43
4.5 Assess available monitoring resources and prepare budget 46
4.6 Determine the monitoring frequency 48
4.7 Select measurement methods 49
4.8 Establish the appropriate time to carry out surveys 58
4.9 Devise a sampling scheme 58
4.10 Devise data recording forms and document methods 65
4.11 Test methods 67
4.12 Review the monitoring programme in relation to available
long-term resources 68
4.13 Prepare a work plan 69
4.14 Carry out necessary training 73
4.15 Analyse data 1B
4.16 Report results to stakeholders 80
4.17 Review the monitoring programme 80
5. References 82
6. Acronyms 87
7. Glossary 88
8. Example protocols from the ACAP biodiversity programme 90
8.1 Introduction 90
8.2 ACAP Monitoring protocol for Snow Leopard 90
(Sa) ACAP Monitoring protocol for Himalayan Griffon and
other vultures oT
8.4 ACAP Monitoring protocol for broad-leaved forest habitat
quality 107
8.5 ACAP Monitoring protocol for broad-leaved forest birds 114
8.6 ACAP Monitoring protocol for remote sensing of habitat extent

and quality 119

ail aca) Orccied Area Oniorsnic GUiAehnes

ACKNOWLEDGEMENTS

We wish to thank the numerous people who contributed to these guidelines. We are especially
grateful to the staff of the KMTNC-Annapurna Conservation Area Project who took part in or
assisted with the various workshops and field training exercises, including:

Ajay Pandey — Conservation Officer, KMTNC-ACAP

Amar Gurung — Senior NRCA, KMTNC-ACAP, Lomanthang

Anil Bhattachan - CDA, KMTNC-ACAP, Jomsom

Arbin K.C.— NRCA, KMTNC-ACAP, Bhujung

Basu Dev Neupane - NRCA, KMTNC-ACAP, Lomanthang

Bhim Prasad Upadhya - NRCA, KMTNC-ACAP, Lwang

Bir Bahadur Thapa - NRCA, KMTNC-ACAP. Jomsom

Bishnu Prasad Paudel - NRCA, KMTNC-ACAP, Jomsom

Dhruba Laudari —- NRCA, KMTNC-ACAP, Ghandruk

Dom Bahadur Pun, JTA, KMTNC-ACAP, Ghandruk

Gehendra B. Gurung — Director, KMTNC-ACAP, Pokhara

Hira B. KC - NRCA, KMTNC-ACAP, Lomanthang

Kaji Ram Adhikari — NRCA, KMTNC-ACAP, Bhujung

Kamal Thapa—- NRCA, KMTNC-ACAP, Ghandruk

Krishna Gurung —- TDA, KMTNC-ACAP, Jomsom

Kriti Nath Paudel — Conservation Officer, KMTNC-ACAP, Lwang
Lizan Kumar Maskey — Conservation Officer, KMTNC-ACAP, Jomsom
Mani Prasad Gurung —- CEA, KMTNC-ACAP, Bhujung

Ms Hari Maya Gurung - GDA, KMTNC-ACAP, Lomanthang

Ms. Anu Kumari Lama — TD Officer, KMTNC-ACAP, Pokhara

Ms. Ganga Nakarmi — Gender Development Officer, KMTNC-ACAP, Pokhara
Ms. Min Maya Gurung —- GDA, KMTNC-ACAP, Manang

Ms. Shanti Gurung - GDA, KMTNC-ACAP, Lwang

Naresh Subedi — Conservation Officer, KMTNC-BCP

Professor Karan B. Shah —Tribhuvan University, Natural History Museum
Raj Kumar Gurung — NRCA, KMTNC-ACAP, Manang

Rajendra Suwal- Senior Ornithologist, Kathmandu

Ram Prakash Singh - NRCA, KMTNC-ACAP, Sikles

Rishiram Sudedi — Ranger, KMTNC-BCC, Sauraha

Sailendra Kumar Yadav — TDA, KMTNC-ACAP, Lomanthang

Shankar Chaudhary — Sr. Ranger, KMTNC-BCC, Sauraha

Sherjung Gurung — CEA, KMTNC-ACAP, Manang

Shree Krishna Neupane- NRCA, KMTNC-ACAP, Manang

Shreeram Ghimire — Senior Programme Assistant, KMTNC-BCP, Bardia
Sunil Marsani — CEA, KMTNC-ACAP, Sikles

Suresh Thapa — Senior NRCA, KMTNC-ACAP, Jomsom

Surya Bahadur Pandey — Conservation Officer, DNPWC

Top Bahadur Khatri, Project Manager, HMG/UNDP/PCP

The project was conceptualised and designed by Adrian Newton (University of Bournemouth,
formerly UNEP-WCMC) and Siddhartha Bajra Bajracharya (KMTNC). We are also grateful for
advice and information provided by other staff of UNEP-WCMC and KMTNC, including Tim
Inskipp, and by Carol Inskipp, Mark Eton (RSPB) and Peter Garson (Chair, WPA/BirdLife/SSC
Pheasant Specialist Group, c/o University of Newcastle). We also thank RPS and Scottish Natural
Heritage for permission to use figures from their Handbook of Biodiversity Methods: survey,
evaluation and monitoring (Hill et al. 2005).

Protected Area Monitoring Guidelines

Foreword by the Officer-in-Charge, UNEP-
World Conservation Monitoring Centre

Biodiversity plays a central role in our lives. We depend upon plants and animal species
for food, medicines and raw materials. The genetic resources contained within biodiversity
hold the basis of our continued existence. The services provided by biodiversity and
ecosystems helps to sustain our livelihoods and protect our health. And there is no doubt
that the beauty and variety of our living species greatly improve the quality of our lives.

There has been increasing global recognition of the importance of biodiversity. The 2002
World Summit on Sustainable Development reaffirmed the critical importance of
biodiversity in maintaining our wellbeing but also acknowledged that it was being lost at
an alarming rate. In response, it established a target of significant reduction by 2010 in the
rate of biodiversity loss, regarding this as one of the most important milestones in progress
towards a sustainable future. However, overall this recognition has yet to be transformed
into concrete action on the scale that will help us achieve the 2010.

The UNEP World Conservation Monitoring Centre (UNEP-WCMC) provides information
on the status of the world’s living resources, from plants and species to the ecosystems that
house them, in order to promote better informed decision-making and support sustainable
management of biodiversity. These Guidelines are a significant contribution to our goals,
as they help the managers of protected areas obtain and use the biodiversity information
necessary for their work. This information will help them know if their actions are being
effective, to prioritise their resources, and to promote their successes.

The Guidelines are based on the practical experience of KMTNC and UNEP-WCMC in
the development of a biodiversity monitoring system for the management of the Annapurna
Conservation Area. They are the result of a productive collaboration and we hope that they
will be of wide relevance for protected area managers throughout Nepal and beyond. I
would like to acknowledge the commitment of the King Mahendra Trust for Nature
Conservation to the success of this project and the support of the Darwin Initiative of the
UK Government, which have made these results possible. It is only through such
collaboration that we can use the expertise gained in actual practice to help implement the
policies expressed at national and international level and to bring about real change that is
vital to our future and that of our children.

Mr. Kaveh Zahedi

Officer-in-Charge

UNEP-World Conservation Monitoring Centre
Cambridge, UK

iv

Protected Area Monitoring Guidelines

Foreword by the Vice-Chairman, National
Planning Commission of Nepal

As we move on to the 10" Five Year Plan, a significant area of Nepal (18%) is in some
form of protected area — national parks, wildlife reserves, hunting reserve and conservation
areas. There is an extensive shift in protected area management approach over the period
from strict protection approach to community-based conservation approach. With the
changes in the management approach, protected areas management has been gradually
and successfully linked to local livelihood. This is a remarkabie achievement for Nepal
and a demonstration of importance accorded to biodiversity conservation in Nepal.

Until recently, however, it has been difficult to monitor biodiversity in protected areas in
an objective and scientifically rigorous way. Therefore, there is a growing realization on
need for blending biodiversity assessment and monitoring system within a piotected areas
management system to achieve sound and effective management of protected areas. The
present Guidelines to Biodiversity Assessment and Monitoring prepared through the
practical experience in Annapurna Conservation Area should help in effective monitoring
of biodiversity in protected areas. The Guidelines will be an important document for the
managers of protected areas in Nepal and else where.

I would like to congratulate the King Mahendra Trust for Nature Conservation (KMTNC)
and its partner organisation the UNEP- World Conservation Monitoring Centre for producing
the Guidelines document and extend my appreciation to those who contributed to this
document. I would also like to acknowledge the support of the Darwin Initiative of the UK
Government.

Shankar P. Sharma, PhD
Vice-Chairman

National Planning Commission
Singh Durbar, Kathmandu

EEO LCTE GLA ELEM OTL OTSTION G L650 111 CS an ar V

Foreword by the Member Secretary, King
Mahendra Trust for Nature Conservation,
Nepal

The King Mahendra Trust for Nature Conservation (KMTNC) in collaboration with the United
Nation’s Environment Programme (UNEP)-World Conservation Monitoring Centre (WCMC) has
developed ‘Guidelines for Biodiversity Assessment and Monitoring for Protected Areas’. With
the mission to conserve, manage and promote nature in all its diversity balancing human needs,
KMTNC has been actively working in biodiversity conservation in Nepal for more than two
decades. The document published in partnership with UNEP-WCMC is an initiative to contribute
in effective management of protected areas in Nepal.

With the establishment of protected areas since the last three decades, Nepal has made remarkable
achievements in conserving its rich biological diversity and cultural heritage. KMTNC, over the
years, with support from His Majesty’s Government, has developed a new and innovative concept
for protected area management effectively linking conservation with local livelihood. I hope the
guidelines will be a tangible tool for biodiversity assessment and monitoring in protected areas.

I praise the input of the team of KMTNC and UNEP-WCMC and wish their endeavour a success.
While it is difficult to name few out of a core team, I acknowledge the effort of Dr. Siddhartha
Bajra Bajracharya, Mr. Gehendra Gurung, Mr. Ram Chandra Nepal and Mr. Nawaraj Chapagain
for successfully designing and implementing the Darwin Initiative funded project.

I also take this opportunity to extend my appreciation to those who contributed to this document
and acknowledge the support of the Darwin Initiative of the UK Government.

Mr. Arup Rajouria

Member Secretary

King Mahendra Trust for Nature Conservation
Jawalakhel, Lalitpur

Nepal

vi Protected Area Monitoring Guidelines

Introduction

1.1 Background to the guidelines

These guidelines are a product of the project “Building capacity for biodiversity assessment
and monitoring in Nepal”. This was a joint project between the UNEP World Conservation
Monitoring Centre (UNEP-WCMC) and the King Mahendra Trust for Nature Conservation
(KMTNC), and was funded by the Darwin Initiative of the UK government from 2002 to
2005. The principal goal of the Darwin project was to strengthen the capacity of KMTNC
to include biodiversity information in management decisions of the Annapurna Conservation
Area Project (ACAP). These guidelines have been developed through the Darwin project’s
training courses and the field testing of the monitoring protocols by KMTNC staff.

The King Mahendra Trust for the Nature Conservation (K MTNC) was established in 1982
by a Legislative Act of the Parliament of Nepal, which mandated it as an autonomous, non-
profit and non-governmental organisation, to work in the field of nature conservation in
Nepal. KMTNC’s mission to promote, manage and conserve nature in all its diversity in
Nepal is supported by the following guiding principles: (a) always maintaining a balance
between human needs and the environment to guarantee long-term sustainability; (b) always
seeking maximum community participation in which locals are recognised both as principle
actors and beneficiaries; (c) always linking economic, environmental and ethical factors in
conservation activities; (d) always managing operations based on sound economic principles
and (e) always aiming for quality in all activities.

Geographically, KMTNC activities are spread from the tropical plains to the high Himalayan
regions, including Trans-Himalayan regions. The Annapurna Conservation Area (ACA) is one
of the major initiatives of KMTNC in the High Himalayan and the Trans-Himalayan regions.

Box 1.1. Annapurna Conservation Area and its importance

Annapurna Conservation Area Project, launched in 1986, is the largest undertaking of KMTNC,
and the first and largest Conservation Area in Nepal. ACA is located in the Mountain regions
of the west-central Nepal at latitude 28°50’N and longitude 83°57’E (Figure 1.1). ACA covers
an area of 7,629 sq. km. and is home to over 120,000 local people of different ethnic, cultural
and linguistic groups. ACA is rich in biodiversity and is a treasure house for 1,226 species of
plants, 38 species of orchids, 9 species of rhododendrons, 101 species of mammals, 474 species

of birds, 39 species of reptiles and 22 species of amphibians. It harbours rare and endangered
wildlife species such as the Snow Leopard, Musk Deer, Tibetan Argali, Impeyan Pheasant and
Tragopan Pheasant.

ACA is well known internationally and in Nepal for its beautiful mountains and a unique
ecology. The area is bounded to the north by the dry alpine deserts of Dolpo and Tibet, to the
(cont.)

Protected Area Monitoring Guidelines

Box 1.1. Annapurna Conservation Area and its importance (cont.)

west by the Dhaulagiri Himal, to the east by the Marshyangdi Valley and to the south by
valleys and foothills surrounding Pokhara. Some of the world’s highest snow peaks over 8,000
mand the world’s deepest valley of the Kali Gandaki river are in ACA. These extreme diversities
have made it Nepal’s most popular trekking destination with over 70,000 trekking tourists in
the year 2000, which is over 62 per cent of the total trekking tourists visiting Nepal.

ACA is a new model of protected area in Nepal where local communities are involved in
protected area management. KMTNC pioneered the ACA concept, realising that protected
areas cannot be isolated from the people living in and around them. The sustainable use of
local resources, particularly forest, remains integral both to the livelihoods of the local
communities and to the conservation of biodiversity and fragile environments. The local
community’s role as a partner in the management of a conservation area through a Conservation
Area Management Committee has been explicitly reflected in the Consevation Area
Management Regulations (CAMR). The regulations authorise Conservation Area Management
Committees to issue permits and collect revenues from the local community for allowing
fishing, forest resource utilisation, grazing and other resources utilisation. The Conservation
Area Management Committee (CAMC) is the main executive body constituted by the KMTNC-
ACAP to manage the conservation area. The villagers of every ward nominate nine of the 15
members. Committees exist in all the 55 Village Development Committees of ACA and under
these committees are several grassroots institutions, such as the forest management committees,
mother’s group, tourism management committees, electricity management committee, etc.
All these institutions are responsible for executing and linking their specific activities with
the conservation of natural resources.

The management of ACA is based on the participatory multi-land use protected area concept.
To balance global biodiversity conservation goals and local livelihood concerns, an integrated
conservation and development approach has been adopted. A Management Plan prepared in
1997 was based on eight management goals, with objectives, priority programmes and policies.
These management goals were: 1) to build and strengthen the institutional capacity of ACAP
through human resource development; 11) to develop a long term framework for conservation
of the natural resources in ACA; 111) to promote nature conservation through sustainable
development of tourism; iv) to enhance the status of women by according an equal role to
them in decision making processes in conservation and sustainable development; v) reduce
stress on critical resources primarily forests through wider use of micro hydro electricity and
other alternative programmes; v1) to promote community infrastructure development; vii) to
promote cultural heritage conservation; and viii) to carry out essential multi-disciplinary
management research to support conservation and development initiatives.

2 ————————————————————— Protected Area Monitoring Guidelines

SS ee (YM) 2A9SOy SIP

S19}SWO|4 00Z

(‘w'bs G/T)

(‘w'bs 005'T)
dN unJeg njexeW

(wy'bs S€0'Z)
yO ebuefueysuey

dN eujewuebes

‘IWdAN 40 SVAYV GALDSLOYd

V

(YH) eAuesey BuyunY

(wy'bs OTZ'T)

d

(Wy'bs pb)
dN undeatys
es

N bueybueq

(‘wy'bs €99'T
VO njseuel)|

OOl (YN) Wed jeuonen ie
(WO) Bay UOI}eAJasUuOD
SAI Pa}Da]0ld

Asepunog jeuoyeuse}u| [-~]

GN3931

(‘wy 'bs 7€6)
dN UeMAIYD |eAOY

(‘w'bs 896)
(‘woybs ¢ze’T) dN elpseg jeAoy
YH uejediouq

(‘wy bs 629'Z)

yo euindeuuy a

(w'bs 90T)
dN e1ey

ys

(‘w'bs ggc’e)
dN Opunsyolg ABys

‘jedan ul edly UOI}eAJaSUOD eUJNdeUUY au} JO UOIJeD0] AY] 1°} eunbi4

|
SJSJBWO|IY 008 0 008

g

(‘wy'bs S0¢)

UM BqUeUd BPNYS

&

(w'bs ¢7z)
dN peydey

at

Protected Area Monitoring Guidelines

Figure 1.2 Major habitat types in the Annapurna Conservation Area

y 4
=

oh

ham anthang

LEGEND

GM Conifer Forest
Hardwood Forest

MB Mixed Forest

7 0 7 14 Kilometers
= _ oe ee

MB Shrubland Contour lines
Grassland 1,000 m
_ Agriculture land 2,000 m
Waterbody “3,000 m
Barren Land 4,000 m
Gill Landslide ; ay 5,000 m

Protected Area Monitoring Guidelines

These guidelines have been developed with the management staff of the Annapurna Conservation
Area Project and are designed for their use. Whilst the examples refer to the management of a
mountain protected area, it is anticipated that much of the generic advice on establishing
assessment and monitoring programmes will be applicable to all types of protected areas.

Plate 1.1 A typical landscape of Annapurna Conservation Area

met

1.2 Use of the guidelines

Biodiversity assessment and monitoring in protected areas is normally, and most
appropriately, carried out as part of a management planning process. It is therefore suggested
that these guidelines are read and used in conjunction with appropriate guidance on
management planning within protected areas. However, in some protected areas,
management plans may not have been prepared, or may not deal with biodiversity issues
in detail (such as in the ACA up to now). Some advice is therefore given in Chapter 2 on
key aspects of management planning so that biodiversity assessments can be carried out
and monitoring programmes established in the absence of a detailed biodiversity
management plan.

An introduction to biodiversity assessments, with emphasis on participatory approaches is
given in Chapter 3, however, it is beyond the scope of these guidelines to give detailed
practical advice on this subject, and therefore the reader is directed to some recommended
references for further information.

Chapter 4 provides guidance on the key practical considerations and decisions involved in
establishing a protected area monitoring programme. This primarily focuses on issues
concerned with selecting field methods and survey sampling strategies that would be carried
out by trained staff or contractors. However, many of the key principles, such as those
associated with sampling, may also be applicable to participatory approaches (e.g. selection
of sample villages for holding interviews). The chapter includes some brief advice on
statistical analysis of monitoring data, but again this subject is too large to deal with in detail
here. Tables are therefore provided that give guidance on appropriate statistical tests for
various situations, and sources of further information, and statistical software. These should
enable readers to complete most forms of statistical analysis required for monitoring purposes.

5

Protected Area Monitoring Guidelines

2. Definitions and purpose of biodiversity
assessment and monitoring for
protected area management

2.1 Biodiversity assessment and monitoring as part of
protected area management planning

One of the principle reasons for creating National Parks and other types of protected area
is to conserve the special biodiversity values within them. Effective conservation of this
biodiversity normally requires management actions that are best carried out within some
form of management planning framework. This is most efficient if carried out as a continuous
process, where plans are prepared, implemented, reviewed and revised according to their
impacts as established by monitoring (see Figure 2.1).

Figure 2.1. A simplified management planning cycle

|
A Plan
aM management Ss \
Evaluate / \
and review _// )
| \ \
WV \
\ \ |
\ / :
A Monitor Implementation
activities and
impacts
ee ed y, y
— ——

Source: Adapted from Thomas & Middleton 2003

To develop an effective management plan requires an initial assessment of the status of
biodiversity, to set the priorities and objectives for management, and then ongoing
monitoring, to establish whether or not management actions are achieving their objectives.
Biodiversity assessments, therefore, normally form key components of protected area
management plans, from which monitoring strategies and programmes are identified and
implemented, as depicted in Figure 2.2. The process is essentially a circular one, with
periodic evaluations being carried out to assess progress in the implementation of actions
and the achievement of objectives. Over the long-term the monitoring data should also be
used to re-evaluate the most recent biodiversity assessment to ensure that decisions are
based on the best and most up-to-date information.

6

Protected Area Monitoring Guidelines

Figure 2.2. Biodiversity assessment and monitoring within a management

planning cycle

ASSESSMENT of situation

Vision of future

Monitoring plan
for objectives and

| Goal + objective setting | + | Goal + objective setting | setting Sotane

Work plan
| Implementation |

| Evaluation | MONITORING

Four key principles for effective management planning have been identified by Hockings
et al. (2001):

1. The plan should have conceptual rigour as a decision making framework. This
framework should provide:

a clear sense of a desired future for the area;

a set of strategies and actions for achieving this future;

clear guidance that can assist managers dealing with opportunities and eventualities
that arise during the life of the plan;

a basis for monitoring of plan implementation and progress towards the desired
future and adjustment of planning strategies and actions as required.

2. The plan should place the management of the area into a relevant environmental, social
and economic planning context. Where possible, planning decisions should be integrated
into this broader planning framework.

3. The content of the plan should be formulated within an adequate and relevant information
base and should place management issues within a broader context and in relation to
the desired future for the area: the needs and interests of any local and indigenous
communities and other stakeholders should have been considered within the plan.

4. The plan should provide a programme and prioritised set of actions for achieving the
desired future for the area.

Protected Area Monitoring Guidelines —_—AAA

Protected areas cannot remain in isolation from the communities and the economic activities
in and around protected area. As noted in point 3, plans should address the needs of local
communities and other stakeholders. The V" IUCN World Park Congress held in Durban,
South Africa has also emphasised on the rights of local communities in relation to natural
resources and biodiversity conservation. Participatory management approaches,’ such as
where "two or more social actors negotiate, define and guarantee amongst themselves a
fair sharing of the management functions, entitlements and responsibilities for a given
territory, area or set of natural resources" (Borrini-Feyerabend et al. 2000) should therefore
be used wherever appropriate. At the very least management planning should involve
adequate consultations with all stakeholders.

Management plans for protected areas are typically prepared following a logically ordered
sequence as summarised below (based on the IUCN Guidelines prepared by Thomas and
Middleton 2003).

1. Pre-planning phase (appointment of planning team, scoping of task, agreement
on the process to be followed).

Data gathering and review.
Evaluation of data and resource information.

Identification of constraints, opportunities and threats.

As eS)

Development of overall long-term vision for the protected area and specific
objectives.

6. | Development of options for achieving the vision and objectives (including zoning
if appropriate).

7. Preparation of a draft management plan.
8. Public consultation on the draft management plan.

9. Assessment of submissions, revisions of draft management plan, production of
final management plan and reports on consultation process.

10. Approval / endorsement of management plan.
11. Implementation of actions identified within the management plan.

12. Monitoring and evaluation of implementation and impacts of the
management plan.

13. Review and update of the management plan.

In practice some of these steps may be carried out iteratively (i.e. by going back and forth).
For example, collation of data on the protected area (Step 2) may be influenced by an evaluation
of what features are of particular value (Step 3). The proposed sequence also suggests that
consultations are carried out at Stage 8. However, we recommend that consultations start at
Stage | and should involve community participation at Stages 2, 3, 4, 5 and 6 to ensure that
the plan is adequately researched and that it deals with the interests of all stakeholders.

It is important to note that the proposed sequence is circular, such that Stage 13 is a review
and up-date of the management plan. This facilitates adaptive management. The key rationale
for adaptive management of biodiversity is the recognition that our knowledge of ecological

' Also known as co-management, collaborative, joint, mixed, multi-party or round-table management.

8 ———_— hh Protected Area Monitoring Guidelines

relationships is incomplete and, therefore, the management of natural resources is always
experimental. It therefore aims to improve our management effectiveness by studying the
impacts of implemented activities and learning from these. Adaptive management therefore
explicitly states objectives (and hypotheses on how they are to be achieved), monitoring
requirements and evaluation methods, and then adjusts and improves actions according to
the results obtained and lessons learnt. See BC Forest Service at http://www. for.gov.be.ca/
hfp/amhome/introgd/toc.htm and the website of Foundations of Success http://

www.fosonline.org/Resources.cfm for further guidance on adaptive management.

There are a large number of suggested contents, structures and formats for protected area
management plans (e.g. Ramsar Bureau 2002), but there is actually considerable similarity
amongst them. According to Thomas and Middleton (2003) the most commonly found
contents of a management plan include:

a Executive summary.

a Introduction (e.g. purpose and scope of plan, reason for designation of protected
area and authority for plan).

Description of the protected area.
Evaluation of the protected area.

Analysis of issues and problems.
Vision and objectives.

Zoning plan (if appropriate).

Management actions (list of agreed actions, identifying schedule of work,
responsibilities, priorities, costs and other required resources).

su Monitoring and review.

In Nepal a national framework for management plans has been agreed, which is summarised
in Box 2.2.

plans

Box 2.2. A national framework for management

1. Executive summary

2. Introduction and background Introduction, statement of significance, description of
protected area, legislation and policies, present practices

3. Rationale of a management plan Goals and objectives, vision and mission statement,
guiding policies, management approaches and logical

framework of outputs

4. Park management Management zones, management issues, strategies

and actions

5. Buffer zone management Management zones, management issues, strategies

and actions

6. Budget and action programmes Budget summary, park and buffer management
activities and budget schedule for 5 years

Protected Area Monitoring Guidelines —$<$<$< &$ —- i — —_— 9

2.2 What is a biodiversity assessment?

Biodiversity assessment is the first stage in the process of defining the biodiversity
management objectives for an area. Its purpose is to gather and assess the information
required to make decisions and recommendations for the future.

In the context of management planning for a protected area a biodiversity assessment involves
measuring or surveying what exists in the area and what is known about it, judging its value
and identifying the most important features (e.g. grasslands for livestock grazing, timber for
fuel and building materials, medicinal plants, water storage functions and habitats and species
of particular conservation concern). Assessments therefore need to involve a social component
that identifies biodiversity features of high socio-economic value, as well as features of high
aesthetic, cultural or intrinsic value. Assessments also typically include identification of the
principal factors affecting the important biodiversity features within the protected area
(e.g. the dependency of top level predators, such as a Snow Leopard Uncia uncia, on its
prey species, or the impacts of fuel wood collection on forest regeneration).

2.3. What is monitoring?

Monitoring is often thought of as a programme of repeated surveys or measurements,
usually by means of a standardised procedure. However, this is merely surveillance if
there is no predetermined objective or value that guides what the findings ought to be. For
example, daily measurements of rainfall are a type of surveillance. It is more appropriate
for protected area management needs to define monitoring more rigorously as:
"the collection and analysis of repeated observations or measurements to evaluate changes
in condition and progress toward meeting a management objective" (Elzinga et al. 2001).

Thus, in the context of protected area management needs, monitoring 1s carried out to determine
if biodiversity conservation, livelihood enhancement and other objectives are being met, such
as the maintenance of the existing area of a particular habitat or a specified number of a
particular species, or development of a community woodlot to reduce impact on natural forests.
It is better to think of monitoring in this more precise way, because it helps to ensure that
protected area monitoring programmes and their methods are focused on protected area
objectives, and so support their achievement. Thus, a protected area monitoring programme
has a specific purpose, tied to objectives that have already been defined.

Monitoring should not attempt to describe the general ecology of a site or measure things
that may merely be of interest. Unfortunately, monitoring schemes often resort to measuring
a wide variety of variables, which may or may not be related to the protected area objectives
and management questions that need to be addressed. As a result, time and money may be
spent collecting unnecessary data. Even worse, it may be found that key management
questions cannot be answered with the information obtained.

Nor should monitoring programmes be confused with research studies that are designed to
establish why something is happening (i.e. to test a hypothesis). Many of the field methods
and scientific principles of biodiversity assessment and monitoring can be used in research,
but their purpose is different. In particular, research may often need to be more detailed,
sensitive and scientifically rigorous than required for many monitoring purposes. For
example, it may be adequate to monitor vulture numbers by occasional counts of soaring
birds. This may establish if population trends are meeting conservation objectives, but will

10

Protected Area Monitoring Guidelines

not reveal the factors determining population size. To establish what influences population
size would require much more time consuming, difficult and costly studies. These would
probably not be necessary if, according to monitoring data, populations appear ‘healthy’
and conservation objectives are being met. However, such detailed research could be
triggered if monitoring data reveal a decline below a preset warning level (which should be
above the conservation objective population level).

In practice, monitoring data may sometimes be of use for research work. For example, if it
is necessary to measure livestock and vulture numbers within the same area, then it may be
possible to examine if vulture numbers are affected by livestock numbers (by correlation
analysis). However, such fortuitous use of monitoring data should not influence monitoring
designs. Instead monitoring and research requirements should be designed separately and
then if there is overlap between requirements (e.g. both need the same data on livestock
numbers) then they may be combined.

Plate 3.1 A female Cheer Pheasant

eae

Photo: Raju Acharya

Protected Area Monitoring Gedetines|$ _—__H_A_P_ $A A_aiaaiaa i 11

3. Carrying out a biodiversity assessment
for a protected area

A biodiversity assessment typically involves a number of key steps as outlined in Figure 3.1
(which are analogous to Stages 2-4 of the management planning process outlined in Section
2.1). The assessment may then lead on to the setting of broad goals, aims and objectives for
the protected area. These key steps are further described in the following sections.

Figure 3.1. Defining the scope & approach of the assessment

Define the scope and approach of the assessment (3.1)
i.e, purpose, area, stakeholders, information needs,
methods, responsibilities and work plan

Create enabling environment for
Participation (3.2), if necessary

Carry out new
Gather and review required data (3.3) baseline surveys if

necessary

v

Biodiversity evaluation (3.4)

Carry out overall biodiversity
evaluation of the protected area

Identify key biodiversity features including threatened species (e.g.
Cheer Pheasant), key habitats, ecological functions (e.g. grazing land) and
important resources (e.g. medicinal plants)

Identify
constraints, Identify overall vision and broad goals |
opportunities
and pressures
(3.5)

Assess available
monitoring
resources
(manpower,
equipment,

expertise, time)

Select biodiversity features and attributes of
each to be monitored (e.g. Cheer Pheasant
population size, forest area, forest tree diversity)

Set SMART objectives for each feature and
attribute (e.g. maintenance of > 50 Cheer
Pheasant)

Set SMART Objectives for

pressures and responses

(e.g. <2% timber extraction
per vear)

SMART means:
Specific,
Measurable,
Attainable,
Realistic and Time-
specific

li Develop monitoring programme (Chapter 4)

Note: figures in brackets refer to relevant text sections

Protected Area Monitoring Guidelines

3.1 Defining the scope and approach of the assessment

3.1.1 Defining the scope

A biodiversity assessment could potentially cover an enormous range of questions and
require a huge investment in data gathering and analysis. It is therefore necessary to define
the scope of the assessment. It should be focused on producing the information needed for
specific decision-making and planning purposes.

The stages in setting the scope and objectives of the assessment are:

1. Determine the purpose of the assessment, e.g. it may form part of the development of
a management plan (as described in Section 2.1) or it may have a separate specific
purpose. If it is for a management plan, then ensure that the scope and type of
management is well defined and understood, e.g. a conservation plan for the ACA, or
a Village Conservation Area Management Operational Plan.

2. Define the limits or boundaries of the area to be assessed. For example, the whole of
the ACA or sub-units of this, such as the village land under the responsibility of a
Conservation Area Management Committee (CAMC).

3. Determine who the stakeholders are in the assessment and what their information needs
are (see Box 3.1). In particular, identify the biodiversity and management issues and
priorities of the decision makers and other stakeholders. This will help establish which
questions the assessment will need to answer (e.g. which species are of highest conservation
importance, and which species are most important to local people and their livelihoods?).

4. Determine who may hold the information required to complete the assessment.

5. Select and agree the methods, responsibilities and work schedule for the assessment.

Box 3.1. Identification of stakeholders and their information needs
In the context of protected areas, stakeholders are likely to include:

mg Landowners, home owners and occupiers (e.g. tenant farmers), including those in adjacent
areas that may be affected by activities within the protected area.

Leaders of local communities (e.g. CAMCs and VDCs in ACA).

Businesses within the protected area (e.g. forestry, tourism, water supply), including
owners and employees.

Visitors to the area and those who organise such visits.
Researchers with sites or projects within the protected area.
Governmental, regional and local authority officials.

m Protected area authority staff.

A useful participatory process cannot begin until the stakeholders understand and respect
each others’ objectives and values. Usually an assessment facilitator will be needed to help
begin this process.

The information need of each stakeholder is likely to depend on their perception of whether
and why the area should be managed. For some, maintenance of livelihood will be most

(cont.)

Protected Area MOOS CLES, —$=—$——$———$—— $$$ 13

Box 3.1. Identification of stakeholders and their information needs (cont.)

important, for others, protection of culturally or spiritually important places, while others
may be motivated by a concern to protect threatened species for all humanity. Thus, each
stakeholder works with a set of assumptions, or values, about what is important, and it is
these that influence both decisions about what is important in the protected area, its management

objectives and evaluations of whether management has been successful or not. And it should
also be remembered that different value-laden needs can also exist within stakeholder groups,
including conservationists (Callicott et a/. 1999) and local communities (Salim ef al. 2001).
Facilitators, therefore, need to recognise what is important to each stakeholder, to help them
define their information needs.

3.1.2 The benefits of a participatory approach

As discussed in Chapter 1, protected area management planning should be undertaken
using participatory approaches wherever possible, and this also applies to biodiversity
assessments, whether part of a management plan or not. ‘Participatory’ in this context is
often understood to mean involvement of rural communities, but can also involve other
stakeholders such as students, policy makers, conservationists or volunteers. It can refer
to scientists and local people working together to assess biodiversity, so that they understand
each other’s perspectives better. Participatory monitoring is a powerful approach that can
improve the effectiveness of information gathering and help people understand the reasons
for certain management decisions. It is therefore increasingly being used to support
biodiversity conservation and management.

Advice on participatory assessment and monitoring in these guidelines draws on shared
experience from an internet conference (Lawrence 2002) and published case studies. It is
anew field, and much of the experience is from developing countries, where a participatory
approach is particularly appropriate, but the processes would be similar in other contexts.
Participation ranges from passive participation, where people are only told what is going
to happen (and responses are often ignored), to self-mobilisation, where people take
initiatives independent of external institutions (Pretty 1994). To date, most examples of
participatory biodiversity assessment and monitoring reach only the halfway point in this
range: people participate by providing labour so that data can be gathered more quickly
and cheaply. Interactive participation where people contribute to decisions in biodiversity
management, or self-mobilisation where they have full rights and responsibilities in
biodiversity management, are still very rare. The current management structure within
ACAP already enables interactive participation by local communities in management
decision making, and therefore this existing framework can be expanded to include
biodiversity assessment and monitoring as well.

Although the use of participatory approaches may complicate issues and be time consuming,
their disadvantages are exceeded by their benefits. Local people are valuable participants
in assessing and monitoring biodiversity, because:

1. They may have knowledge about wildlife, plants and resources derived from
generations of use. For example, knowledge of the medicinal properties of some
plants may be undocumented and known only to the local communities.

14 = §— Protected Area Monitoring Guidelines

2. Most monitoring systems within protected areas focus on protected species of wildlife
and plants. Monitoring local resource use is a neglected but crucial dimension in
planning sustainable harvesting by local people.

3. It is internationally acknowledged that involving local people in the planning and
management of biodiversity and resources can increase their awareness and motivation
for conservation. It can enhance an exchange of local and outside perceptions on the
relationship between biodiversity and use patterns, leading to feedback on how to
change unsustainable resource use practices.

4. The basis of decisions on biodiversity management, especially in protected areas, is often
unclear to local communities depending on those resources. The involvement of local
people in the gathering and analysis of biodiversity data will enable local communities to
understand why certain decisions are made. This is likely to be particularly helpful when
difficult choices need to be made that may appear to be detrimental to a local community’s
immediate interests (e.g. restrictions on resource usage to allow recovery).

5. Communication among stakeholders is often limited. Interactive participation by
various partners, including nearby communities and protected area staff can improve
relations (Fabricius and Burger 1997; Van Rijsoort and Zhang 2002), and resolve
conflict (Bliss et a/. 2001).

6. Particularly in developing countries, resources for biodiversity assessment are limited
- human capacity, money and time are all scarce (Danielsen et al. 2000). Biodiversity
monitoring and management systems should therefore be based on locally available
capacity and resources to be sustainable.

However, it is important to recognise that there may be significant practical constraints on
participation by local communities. Some communities within protected areas may be unaware
of the relevance of appropriate protected area management for them, and hence its potential
benefits. They may, therefore, have no incentive to participate in management planning. They
may also have constraints on their time and inputs, especially if they are living on a subsistence
or near-subsistence basis where all time is used on basic life-supporting activities.

It may therefore be necessary to precede participatory management planning, assessment or
monitoring initiatives with basic socio-economic development, and awareness activities to inform
communities of the potential benefits of participation. Training will also often be necessary. If
stakeholders are to play a full role in management planning they will need to understand its
aims, who the decision makers are and how the management planning system works.

3.2 Creating an enabling environment for participation

The time needed to facilitate a participatory process in biodiversity assessment and
monitoring must not be underestimated. The process may take much longer than a non-
participatory approach, but this investment is essential for building mutual understanding,
to obtain useful data, and to promote local empowerment.

Before entering into a participatory process of biodiversity assessment and monitoring, an
enabling environment is needed — 1.e., favourable policy and institutional factors. In
particular, decentralised decision making is required rather than top-down management
(as has, for example, often been typical of the forestry sector in many countries). But
where protected areas are strictly protected, the possibilities for interactive participation

15

Protected Area Monitoring Guidelines

by surrounding communities may be limited, since the benefits perceived by these
communities may not be high. In cases where the rules and regulations of the protected
area enable sustainable use of resources and even involvement in management of the
protected area, as in the ACA, incentives for local communities to participate in biodiversity
and resource management planning, conservation and monitoring will be higher.

People who are leading or facilitating assessments need to be aware of any obstacles
perceived by stakeholders before entering into the process, in order to address
misunderstandings or justified fears. For example, in Yunnan, China, villagers were initially
reluctant to join in, fearing that the monitoring process would lead to further restrictions in
their resource use. This fear appeared to be justified during the analysis phase, when most
of the proposed solutions involved banning resource use. More constructive solutions that
provided benefits for all stakeholders had to be thought of, including sustainable resource
use and enrichment planting (Van Rijsoort and Zhang 2002).

It is also important that facilitators recognise their privileged position as stakeholders who,
despite striving to leave bias and subjectivity on one side, will nevertheless have personal
objectives and motives for becoming involved. This will help facilitators to be more self-
aware and protect against undue bias.

3.3. Data gathering and review

3.3.1 Data requirements

The principal aim of data gathering is to prepare an overall description of the protected
area, including an inventory of the known biodiversity components that are present. This
should be carried out in partnership with stakeholders, by collating and reviewing all relevant
and available information on the protected area’s status, biophysical characters, human
use and biodiversity. Additional information may also need to be gathered from new field
surveys and analysis of remote sensing data.

This stage of an assessment may potentially be very time consuming, and could easily
become overwhelming, so it is important to focus attention on key information requirements
of the stakeholders that are directly relevant to the management planning process (see Box
3.1). It is often best to collect the minimum of information first and then identify other
requirements as other stages of the management plan progress. This helps to ensure that all
information collected is relevant and avoids wasting time whilst irrelevant descriptive
information is compiled.

Information requirements for a protected area management plan assessment typically
include:

= Location and boundaries, and appropriate administrative boundaries (mapped).

m Area.

m Status (e.g. international, national and local designations and IUCN protected area
category).

= Administration (e.g. with respect to protected area, forestry, water resources and
community management).

16

Protected Area Monitoring Guidelines

Land ownership and occupancy.
Infrastructure and services (e.g. roads, airports, telecommunications, power supplies).
Physical information (e.g. geology, soils, topography, climate, hydrology).

Land use (historical and current).

Cultural information (cultural values and interactions with landscape and biodiversity).
m Socio-economic status and trends, and relationship with the protected area and its features.
m Visitor numbers, interests and influences.

m Ecosystems and habitats, (including the types of ecosystem and habitat that are present,
with maps of their location, quantification of their current and past extent and
condition, and descriptions of their use and management by local communities).

m= Important flora and fauna, including protected species, threatened species, utilised
wild species, ecological keystone species and species of cultural importance. And
for each species information on:

© Quantity: population sizes, abundance, stock volume, basal area.

© Quality: importance, trends in abundance, productivity and viability.

© Location: distribution; relationship between place and cultural value.

e Value: use by humans (e.g. food, forage for livestock, materials, medical uses,

cultural uses), trends in uses, conservation importance, aesthetic values etc.

= Domestic livestock, pest species and introduced species (listed and their interactions
with native species and ecosystems described).

= Factors affecting habitats and important flora and fauna, including pressures and
management responses.

3.3.2 Information sources

Existing information

Some of this information may be obtainable from scientific books, papers and reports,
land-use and habitat maps, aerial photographs, satellite images, historical records and
unpublished data held by experts. However, it will normally be essential to supplement
such scientific data with information from local people, gained through participatory
involvement in the assessment process where possible.

Any information and documents on the reasons for the establishment of the protected area
and the definition of its boundaries will be a useful start. Records held by government
agencies and religious authorities of land tenure, population census, tax collection, and
agricultural, forestry and fisheries production are all valuable resources.

New surveys

In some cases it may be necessary to collect new data to prepare biodiversity assessments
that are adequate for management plans. However, incomplete information should not be
used as an excuse for delaying management planning. In many cases it will be possible to

17

Protected Area Monitoring Guidelines

complete management plans with available data and to include further surveys amongst
the agreed actions, the results of which then feed back into the plan, thus completing the
adaptive planning loop.

Where surveys are required, these may necessitate fieldwork to establish species presence,
quantify species populations, map habitats and assess habitat conditions. These may be
carried out using some of the methods described in Chapter 3 of these guidelines, or as
described in Sutherland (2000) and Hill e¢ al. (in press). Local people may be able to
undertake some surveys (with any necessary guidance and training), thereby further
enhancing the participatory process. If surveys must be carried out by outside experts then
suitable local people (e.g. with an interest in wildlife or resource use) should be invited to
take part, perhaps as trainees or apprentices so that local capacity is developed to carry out
future surveys.

Remote sensing data

Habitat surveys may be supported using remote sensing data and Geographical Information
Systems (GIS). A GIS is a spatially referenced database that allows multiple layers of data
to be created and displayed together as computerised maps. Data sources may include
satellite data, aerial survey, existing maps, field survey and expert knowledge. GIS enables
the standard formatting of all maps used, no matter what their source. For further background
information on GIS, see Longley er a/. 2001, and Burrough & McDonnell, 1998.

Both satellite scenes and aerial photographs are types of remotely sensed data. The main
advantages of satellite data are that large and inaccessible areas can be covered using a
standard approach, with a uniform level of detail and at relatively low cost. The use of such
data may therefore be particularly cost-effective for mountainous areas, such as the ACA,
with their extremely difficult terrain, with the caveat that remotely sensed data are not
useful for areas that are persistently obscured by cloud. Remotely sensed data may be
extremely valuable for monitoring as repeated surveys using identical techniques are possible
throughout the lifetime of the remote measurement system (an anticipated 15 years for the
MODIS instruments on board EOS satellites). The maps produced can be used alongside
other layers in a GIS for land cover mapping, modelling and planning. However, a specialised
set of skills and software are required to interpret raw remotely sensed data — it should not
be assumed that all GIS software and personnel can carry out image analysis.

Two sources of remotely sensed data are the LANDSAT and EOS satellites. The first
LANDSAT satellite was launched in 1972, and the most recent in 1999, followed by the
EOS satellite in 2000. LANDSAT covers the Earth in 18 days and has a pixel resolution of
30 m. The MODIS Terra sensor on the EOS satellite visits every location every | to 2 days,
and has a pixel resolution of 250 m. LANDSAT is therefore better for monitoring fine
resolution, small area or relatively slow changes (such as changes in forest cover), and
MODIS for coarser resolution, large area or relatively fast changes (such as fires).

Free LANDSAT and MODIS satellite images can be readily obtained from the internet. These
can be found for a particular area using the USGS ‘GloVis’ interactive map (http://
glovis.usgs.gov/). Once the path and row number of the image(s) of interest have been found,
the Earth Science Data Interface site (http://glcfapp.umiacs.umd.edu:8080/esdi/index.jsp) can
be used to seek free images. The main scene for the Annapurna area has path 142 and row 40.

18 _————— Protected Area Monitoring Guidelines

If free data are not available for the place or period of interest, try the Tropical Forest Information
Centre, which makes tropical LANDSAT data available at a reduced price (US$25-50 per scene).

Aerial photographs can be used to examine smaller areas at a finer spatial resolution, and
to support the interpretation of satellite data. Vegetation types are most effectively identified
in satellite images when the classification is ‘supervised’. This requires a training set, used
to inform the software about different land cover on the ground. Satellite data therefore are
of most use when brought together with existing maps or georeferenced aerial photographs.

Habitat maps may be used in the design of a stratified sampling regime for species or
community assessment. The habitat maps can also be used to model species’ distributions
or abundance. These models may either directly be used to estimate species occurrence or
to identify areas for sampling. A number of techniques and packages for species distribution
modelling are listed in Table 4.8 (Section 4.15).

A unique contribution that can be made by satellite data is information on habitat changes
prior to the commencement of the assessment and monitoring programme. A change analysis
(comparison between years) highlights areas where land cover change has occurred. These
may be areas in need of conservation attention, or conversely where ecosystem restoration
is in progress.

3.3.3 Data presentation

Presentation of the descriptive information in the management plan should be brief and
easily understandable, and should focus on the key results of the assessment. Full use should
be made of maps and diagrams, with attention to design so that the most important information
is easily visible. Information that is not directly required should either be referred to (with
full reference details) or placed in a separate technical appendix. Technical language and
jargon should be avoided where possible, and a glossary provided for all technical terms
that are used. Local names of animals and plants should be given as well as scientific names.

3.4 Biodiversity evaluation

3.4.1 The purpose of a biodiversity evaluation

In general terms biodiversity evaluation’ is the process of establishing the value (ideally
quantitatively) of biodiversity components. Evaluations may be carried out on various
components of biodiversity (i.e. from genetic variation within species, to individual species,
species assemblages, habitats, sites and biomes).

In the context of protected area management planning, the overall purpose of a biodiversity
evaluation is to establish why the protected area is important and of value to society (and
hence protected) and what are its key features (and their values) that need to be conserved
(i.e. protected and/or used sustainably). It is normally carried out to identify sites that
merit some form of protection, but is often repeated and updated as part of management
planning process (especially when the reasons for creating a protected area have been ill-
defined or poorly communicated).

? Biodiversity evaluation should not be confused with monitoring evaluation (or result evaluation or management
evaluation or similar terms), which is the measuring of progress with respect to management objectives.

Protected Area Monitoring GuidehieS —<$<=<====——<— —<$—$—————— 19

Biodiversity evaluation is fundamental to any management planning process because it
identifies the features (such as particular habitats, threatened species, and medicinal plants)
that should be the focus of management actions. Unless protected area values are
understood, there is a risk that inappropriate management may be undertaken, resulting in
a decline in the value of the protected area and its important features. The evaluation,
therefore, forms the basis upon which conservation objectives should be set, which
explicitly ensure that each feature is conserved appropriately (see Section 3.5).

Great care must, therefore, be taken in the evaluation of the biodiversity assessment data.
And it is particularly important that all stakeholders participate in the evaluation process,
to ensure that all biodiversity values are identified. "Unless the preparation of the
Management Plan addresses the concerns of the local people then their support will be
impossible to secure" (Thomas and Middleton 2003). Local people, if given the opportunity
to discuss findings, can also often provide interpretations and insights that otherwise may
have been missed if the results had been interpreted solely by staff and advisors (Steinmetz
2000). Moreover, the drafted solutions emerging from participatory data analysis will be
more practical and adjusted to the local conditions.

A biodiversity evaluation consists of two parts:

= Anevaluation of protected area as a whole, which places the site in the context of its
regional, national and international importance, and identifies its overall value and
importance to various interest groups.

m The identification of the key biodiversity features that must be protected and
conserved to maintain the importance of the protected area.

3.4.2 Carrying out an overall biodiversity evaluation of the protected area

An evaluation of the biodiversity of a protected area as a whole may take into account a wide
range of potential biodiversity values, including intrinsic and socio-cultural values (Daily
1997, Posey 2000), and more direct socio-economic benefits (Daily 1997), such as food,
building resources, medicines and waste decomposition etc (Spellerberg and Hardes 1992).

The assessment of the ecological values (many of which are used for protected area
selection) is often the focus of nature conservation management plans, and various criteria
have been developed for these evaluations. Although, no standard set of criteria have
emerged for this purpose, one set that have been frequently used are those developed by
Ratcliffe (1977), as listed below:

Primary criteria

Size (including the protected area and habitats, and population sizes for key species).
Diversity (of habitats and species).
Rarity (of habitats and species).

Naturalness (habitats that are least modified by human use tend to hold richer wildlife
communities, often including rare species, and have high aesthetic appeal and
scientific value).

mu Typicalness (how well the area represents ecosystems and habitats on a wider scale).

20

Protected Area Monitoring Guidelines

m = Fragility (how vulnerable or sensitive ecosystems, habitats and species are to human
induced changes).

Secondary criteria (i.e. not always used)

= Recorded history (sites that have been studied and valued for a long time tend to be
highly valued).

m= Potential value (i.e. the likelihood that appropriate management could restore or
enhance an area’s ecological value).

= Position in geographical or ecological unit (e.g. some ecosystems, habitats and species
may be of high functional importance).

m Intrinsic appeal (e.g. species with public appeal promote the cause of nature
conservation and act as flagship species).

These criteria were first used in the United Kingdom to identify sites of high conservation
value, which later formed the basis for identifying Sites of Special Scientific Interest (Nature
Conservancy Council 1989). These criteria have been widely adopted and adapted in the
UK for management planning evaluations, e.g. by the RSPB (Hirons ef al. 1995) and the
Countryside Council for Wales (CCW 1996). Although developed in the UK, they are easily
applicable to other situations and have been widely adapted for use in many other countries,
e.g. in the Ramsar Management Planning Guidelines (Ramsar Bureau 2002)

Evaluations are simply carried out by considering the properties of the protected area in
relation to each of the criteria in turn. This may then be summarised in a description of the
overall importance of the protected area, such as the example for the ACA in Box 3.2.

Box 3.2. A possible ecological evaluation summary statement for
the Annapurna Conservation Area

The ACA is a very large (7,629 km?) area of Himalayan mountain ecosystem, which overlaps
several biogeographical regions and holds a diverse range of habitats and species. The Kali
Gandaki Valley runs through the ACA and is a biogeographical divide in the Himalayan
mountain chain. The ACA therefore has species typical of the eastern and western Himalayas.
The Kali Gandaki valley is also an important north/south bird migration route. The ACA also
lies within a region of overlap between the Palaearctic and Indo-Malayan realms and so has
species of both realms. The ACA has a huge altitudinal range, from 790 to 8,091 m. It has a
great range of rainfall, with some of the driest regions in Nepal in the trans-Himalayan zone
and some of the wettest regions south of Annapurna in the Modi Khola valley. Although some
of the habitats may not be species rich, they hold important populations of several globally
threatened species, some of which are of exceptional intrinsic appeal (e.g. Snow Leopard).

However, many management plans fail to properly identify other values of the biodiversity
of a protected area, such as cultural, social and economic values (Thomas and Middleton
2003). This is unwise as it may undermine the potential support of local people for the
protected area and its management.

Protected Area Monitoring Guidelines. ————m 21

3.4.3 Identifying key biodiversity features of a protected area

Key biodiversity features (e.g. species, habitats, ecological functions) should include features
that are:

a of high nature conservation importance.
m socio-economic importance.

gs cultural importance.

The identification of key biodiversity features of high nature conservation importance
should firstly take into account broad international and national conservation objectives.
In terms of global objectives, there is broad agreement that the prevention of global
extinction should be the highest priority, and therefore the degree of threat
(i.e. risk of extinction) is of primary concern in setting priorities. This is reflected in
the production of IUCN Red Lists (see www.redlist.org) of species that are considered
to be at risk of global extinction according to various categories of threat. The risk
of extinction at national level is also the commonest basis for identifying national
species conservation priorities.

Rarity has often been considered to be one of the most important factors influencing the
risk of extinction of a species, and many Red Lists have focused on this. Rarity has also
often been used as a secondary criterion whereby, for example, a declining species is not
considered to be threatened unless it has also crossed a rarity threshold. However, rarity is
not a straightforward concept as there may be a variety of circumstances under which
species may be rare (Rabinowitz 1981). Species may have small (or large) total ranges,
occupy few (or many) habitat types, and may be scarce or abundant where they do occur.
It is evident that rarity includes both a spatial and a numerical dimension. For some species
some aspects of rarity may be an evolutionary property, as a result of their habitat specificity,
small natural range or low natural densities. On the other hand, small range or low densities
may be the result of human impact, which may be reversible.

Consideration of rarity in an evaluation requires data on the range or number of individuals
of a species (or habitats, communities, or abiotic features), not only within the protected
area but at wider scales. This is because important elements of rarity are scale dependent.
For example, a locally rare species may also be regionally or globally rare, which is of
greater importance than if it is only locally rare because it is at the edge of its range (e.g.
Himalayan Musk Deer Moschus chrysogaster, or Snow leopard Uncia uncial or Golden
eagle Aquila chrysaetos). Normally, increased importance to rarity should be given to
species that are rare at the global level. Some locally common species may also be of high
conservation importance if the species in question is rare at a global or wide geographical
scale (e.g. Blue Sheep Pseudois nayaur in the ACA).

Whichever criteria are used for evaluations, a hierarchical level of importance should be
established so that the highest priority for conservation/protection is given to species, or
habitats, that are globally threatened. But it is also necessary to take into account their
local status to assess the necessity for taking action at a local scale. This enables the principle
of "thinking globally and acting locally" to be put into practice. The highest priority should
be given to species and habitats that are both globally and locally threatened.

tN
tN

Protected Area Monitoring Guidelines

Assessments below global scales should be based on appropriate biogeographical
populations where possible. In practice though, assessments of populations are more often
based on national or regional populations for political and administrative reasons (e.g. as
in most national Red Lists).

Any evaluation of conservation priorities for a species or habitat should also take into
account the importance of the population or resource being considered. Thus, the evaluation
of a habitat area or a species’ population should consider two key independent factors:

m the risk of loss of the habitat or species population (i.e. its threat status).

m its biogeographical importance, i.e. the proportion of the biogeographical (or national/
regional) resource or population.

Thus, for example, a very high priority should be given to a species’ population that is
biogeographically important (i.e. it is a large proportion of the species’ entire population)
and at risk of extinction within the area in question. However, it is important to note that a
species population may be a high priority nationally because it is highly threatened nationally,
irrespective of its numbers in relation to international or global populations e.g. Himalayan
Musk Deer in Nepal. This is because the maintenance of a species’ range (and potential
genetic variation associated with this) can also be an important conservation aim after
prevention of complete extinction. On the other hand, a population of a species may be very
important because it is a large proportion of the biogeographical population, irrespective of
its conservation status, e.g. Spiny Babbler Turdoides nipalensis, a bird species which is not
uncommon in secondary scrub in the ACA, but is endemic to Nepal. In these circumstances
a country has a particular responsibility for the species and should at least monitor the status
of the species and guard against potential events that could impact on the population.

Taking into account the issues described in this section and 3.3.2, it is possible to suggest a
checklist (Table 3.1) of biodiversity features that should be considered as key features, or features
of exceptional value, according to the IUCN terminology (Thomas and Middleton 2003).

Table 3.1. Biodiversity components that should be considered to be key
biodiversity features within a protected area.

Globally threatened species (i.e. on IUCN Red Lists)

Significant populations of rare or otherwise nationally or regionally threatened species
Endemic species or other species that have very high proportions of their biogeographic or
global populations within the protected area or region

Rare or threatened habitats (globally, regionally or nationally)
Particularly good examples (e.g. large and highly natural) of characteristic habitats
| Features of high intrinsic ecological importance or interest

Features of high socio-economic importance (e.g. forest products, grazing lands or species
and habitats that attract tourists)

Features of high cultural importance (e.g. sacred areas or species) or intrinsic appeal

LEO} AD ANREPIN OG LN ONO GILL 23

In practice it is often impractical to list and set objectives for all the habitats and species
that may qualify as key features. Some key features may therefore be combined. For example
a number of important medicinal plants may be just listed as medicinal plants of temperate
broad-leaved forests.

It is good practice at this stage not to include habitats or species etc as key features if they
are only of indirect importance (e.g. as a habitat or prey species) for other listed key features.
Key features need to be important in their own right. This can be tested by considering if a
substantial decline in the feature in the protected area would be unacceptable if all other
features were maintained (irrespective of how unlikely this might be). Thus for example, a
decline in Blue Sheep populations in the ACA would be extremely undesirable, not just
because it is a prime prey species for Snow Leopard. Thus this qualifies as a key biodiversity
feature. In contrast, some other key prey species might not be listed as key features. This does
not mean that management actions and monitoring may not be undertaken for such prey
species, but these requirements would be identified at a later stage of the management planning.

It is also advisable to avoid listing species as key features at this stage if they only act as
indicators (see Box 3.3) of a feature’s value, for example, a tree species as an indicator of
diverse broad-leaved forests. In this case the feature would be "diverse broad-leaved forest". If
it is necessary to use indicator species to define conservation objectives for this, these should
be identified at the conservation objective setting stage. However, it may be useful to note
that a species may sometimes act as an indicator of one feature and be a key feature in
its own right.

For further information on biodiversity evaluations see Margules and Usher (1981), Smith
and Theberge (1986), Usher (1986), and Spellerberg (1992).
Key biodiversity features within the Annapurna Conservation Area

On the basis of the criteria listed in Table 3.1, it is possible to provide a preliminary list of
some habitats and species that should be considered to be key biodiversity features within
the ACA (see Table 3.2).

24 FFF Protected Area Monitoring Guidelines

Table 3.2. Some key biodiversity features within the Annapurna
Conservation Area

Key Feature Rationale for listing as key feature

Caragana, Lonicera, Carex and Cobressia
species grasslands of the trans-Himalayan
Rangelands (Mustang and Manang)

Characteristic habitat of Trans-Himalayan
region, of key importance for several

threatened species, high socio-economic
importance.

Singdi forest, Pasgaon Village Development
Committee, Lamjung

Rare forest type and biodiversity hotspot
(supports over 250 plant species, also
recognized by ACA Management Plan, 1997),
has also patch of rare Hill Sal (Shorea robusta)
forest.

High socio-economic importance

Taxus baccata forest, primarily in Manang and
Mustang (below Larjung)

High commercial demand
High socio-economic importance

Pure Birch Betula spp. forest in gorge near
Ngawal (northern side of Ngawal)

Rare forest type, only known pure forest of
Birch in ACA (KMTNC, 1994); socio-
culturally important

Hill Sal (Shorea robusta) forest, Madi, Khudi
and Rudi Basin

Sal is a typical high value tropical species, rare
in ACA, High socio-economic importance; a
suitable habitat for Clouded Leopard

Juniperus forest stands of Samar, Upper
Mustang

Last remaining area of forest in upper
Mustang, culturally and ecologically important

Rhododendron forest of Ghandruk-Ghorepani High aesthetic appeal, and one of the largest
pure such forest in Nepal, with National flower

(Rhododendron arboreum)

Forests in the intensive use zone High socio-economic importance: fuelwood,

timber and non-timber products

Major non-timber forest products: Daphne
spp., Swertia chiraita (Chiraito), Piper longum,
Arundinaria spp., Alliums spp., Swertia spp.
(found in upper Mustang), Aconitum spicatum
(Bikh), Cordyceps sinensis (in Manang),
Dactylorhiza hatagirea (Panch Aule),
Picrorhiza scrophulariiflora (Kutki), Paris
poliphyla (Satuwa)

High socio-economic importance and some
species protected in Nepal

Globally threatened: Endangered, protected in
Nepal, CITES I, species of very high intrinsic
appeal.

Snow Leopard Uncia uncia

Globally threatened: Vulnerable, protected in
Nepal, CITES I

Clouded Leopard Neofelis nebulosa

(Cont.)

Note: Global threats status is based on 2004 IUCN Red List of Threatened Species
(IUCN 2004), unless otherwise indicated.

Protected Area Monitoring Guidelines 25

Table 3.2. Some key biodiversity features within the Annapurna
Conservation Area (cont.)

Asiatic Wild Dog Canis alpinus Globally threatened: Endangered
Red Panda Ailurus fulgens Globally threatened: Endangered, protected in
Nepal, CITES I

Assamese Macaque Macaca assamensis Globally threatened: Vulnerable and protected
in Nepal
Asian Wild Ass Equus hemionus Globally threatened: Vulnerable

Globally Near Threatened, CITES I, protected
in Nepal

Himalayan Tahr Hemitragus jemlahicus Globally threatened: Vulnerable

Tibetan Antelope Pantholops hodgsonii Globally Threatened: Endangered, CITES I,
protected in Nepal

Argali Ovis ammon

Wood Snipe Gallinago nemoricola

Himalayan Musk Deer Moschus chrysogaster

Globally Threatened: Vulnerable, protected in
Nepal, CITES I

Globally threatened: Vulnerable (BirdLife
International 2003); species of high meadows

Slender-billed Vulture Gyps tenuirostris, Slender-billed and Oriental White-backed
Oriental White-backed Vulture Gyps Vultures are Globally threatened: Critical
bengalensis and Himalayan Griffon Vulture (BirdLife International 2003), Himalayan
Gyps himalayensis, Griffon is common, but may be declining.

Cheer Pheasant Catreus wallichii Cheer Pheasant is globally threatened:
Vulnerable (BirdLife International 2003) and

protected in Nepal

3.5 Identification of constraints, opportunities and pressures

3.5.1 The purpose of identifying constraints, opportunities and pressures

This stage of management planning aims to identify potential constraints, opportunities and
pressures that may influence the development of the overall vision for the protected area
and specific management objectives. For example, constraints may include the legal and
institutional framework for the protected area, and resource, communication and infrastructure
limitations. Opportunities might include the potential capacity for increasing revenue from
tourism, new technological developments (e.g. micro-hydropower) or activities that may
bring unintended biodiversity benefits (e.g. afforestation programmes for soil stabilisation).

The identification of pressures that may threaten biodiversity is particularly important and
the aim should be to comprehensively identify all the pressures on the key features of
biodiversity importance. Where possible, these pressures should be quantified. Risk
assessments should be carried out for features of high importance, which take into account
both the probability of impacts and the likely magnitude of those impacts.

26

Protected Area Monitoring Guidelines

It is useful to develop a simple framework of the interrelations between important biodiversity
features and their influencing factors and management actions. This helps to understand
what management actions need to be taken and what needs to be monitored. A widely used
framework which is particularly useful in such assessments is the Pressure-State-Response
framework’ (see Figure 3.2). This was designed to aid analysis of the causes of change in
the natural environment and the response measures of human society to these changes.

Figure 3.2. The Pressure — State — Response framework

Reduce / Impacts
manage
MONITOR

Plan and
implement

When applied to a protected area, the State component is an assessment of the current
state or condition of a biodiversity feature in the area, and of the changes that it has undergone
in the past. This might focus on, for example, the extent and condition of important forest
habitats within the protected area.

The Pressure component is an assessment of what factors may potentially impact on the
biodiversity feature that influences its state. Examples of such pressures on the state of
important forest habitats could be deforestation for firewood and timber, pollution, or hunting.

The Response component is an assessment of the policies, laws, practices, etc. that have
been created to manage and conserve the biodiversity feature and alleviate or regulate the
pressures on it. The designation as a protected area is one response in itself, whilst others
could be tree planting programmes, awareness activities and hunting regulations.

3 This framework may sometimes vary, for example as Driving Force — State — Response, or Driving Force —
Pressure — State — Response, or other more complex variations

Protected Area Monitoring Guidelines|$ _—_XKHA_AHP_A a ?YW_ 4]

3.5.2 Identifying pressures on key biodiversity features

This stage of the assessment requires a review of available relevant information. Such
information, will include studies of the protected area itself. Some of this may be published,
but much may also be learnt from consultations with local experts. Published scientific
studies of other protected areas and the knowledge gained by protected area managers and
scientists elsewhere may also be relevant and useful. However, participatory approaches to
assessments of pressures on biodiversity may be particularly valuable and often the only
source of information in many areas. Local people may be able to provide particularly
useful information, which may be unavailable from other sources, especially regarding
exploited biodiversity components (e.g. forest products, hunted species and rangelands).
However, it is important to be aware that some opinions relating to possible pressures on
important socio-economic assets, such as the impacts of predators on livestock populations,
may be misconceptions or exaggerated by some stakeholders.

Pressures on key biodiversity features within the Annapurna Conservation
Area

Using this framework we can summarise the state and pressure of some key features in the
ACA according to available information as indicated in Table 3.3.

Plate 3.2 Fuel wood collection from forest

28 ——_———ewwm[wv—Vn—e_fjke—eass—— Protected Area Monitoring Guidelines

Table 3.3. | The state of some key biodiversity features within the Annapurna

Conservation Area and pressures affecting them

State of feature

Key feature

Caragana, Lonicera, Carex and
Cobressia species grasslands of the
trans-Himalayan Rangelands
(Mustang and Manang)

Degraded in over-grazed
areas

Juniperus forest stands of Samar,
Upper Mustang

Very sparse trees

Rhododendron forest of Ghandruk-
Ghorepani

Major non-timber forest products Increasing stock in some
area whereas decreasing in
the areas where illegal
collectors harvest
unscientifically

Snow Leopard Uncia uncia Inadequate information on
population.

Lack of detailed infor-
mation on population.
Increasing compared to
1-2 decades ago according
to local people

Himalayan Musk Deer Moschus
chrysogaster

Little known. First
scientific records in 2001.

Rare Tibetan Plateau species —
Asian Wild Ass, Tibetan Antelope,
Argali

Potential pressures on
feature

Livestock grazing pressure,
closure of Tibetan rangelands for
grazing Nepalese livestock

Fuelwood collection (lucrative
source of fuelwood compared to
its alternative of Caragana)

Degrading in certain parts | Tourism with increased demands
for fuelwood and timber.

Illegal harvesting and higher
commercial demand for legal
harvesting

Retaliatory killing by herders and
poaching.

High commercial interest leading
to poaching and disturbances of
habitat.

Competition with livestock for
grazing. Possible hunting.

Wood Snipe Gallinago nemoricola | Rare, trends uncertain Uncertain, possible overgrazing.

Substantial declines in
Slender-billed Vulture and
White-rumped Vulture and
some evidence of smaller
declines in Himalayan
Griffon Vulture,

Slender-billed Vulture Gyps
tenuirostris, Oriental White-
backed Vulture Gyps bengalensis
and Himalayan Griffon Vulture
Gyps himalayensis,

Cheer Pheasant Catreus wallichii | Rare, trends uncertain

Possible poisoning from
Diclofenac, a non-steroidal anti-
inflammatory drug used for
livestock, as across much of the
Indian subcontinent (Oaks ef al.
2004, Shultz et al. 2004).

Livestock grazing, forest fires,
hunting.

It may be seen from this table that some pressures are common to more than one feature
(e.g. Over grazing may impact on some wild threatened species and livestock) and some
pressures may be related to the state of some features. For example, degradation of the
State of the trans-Himalayan grasslands may result in a pressure on Snow Leopards as a
result of indirect impacts on their prey.

Vee BLN LN ORO ISS 29

Pressures may therefore be interrelated and complex, as indicated in Figure 3.3, which
summarises the example of pressures that are thought to affect Snow Leopards in the
ACA, based on studies within the ACA (Oli, 1991; Jackson ef al. 1996; Thapa 2000), and
elsewhere (Jackson et. al 2001).

Figure 3.3 Anillustrative example of the pressures affecting Snow Leopards
in the ACA.

; Grassland / livestock
Hunting shrubland habitat
condition es

Prey numbers

Sheep)

Numbers of trekkers
/ climbers etc Snow Livestock

Leopard in predation by
the ACA Snow Leopard

Hunting of Snow Retaliatory killing
Leopard for fur / due to livestock
medicine predation

3.6 From the assessment to setting objectives

3.6.1 Vision, goals and objectives in protected area management

Once the biodiversity assessment has been conducted the results should be analysed and
communicated to the planners and managers of the area. The assessment forms the basis
for developing a long-term vision, broad goals and specific objectives within the
management planning process (see Figure 2.2 and 3.1). A vision statement aims to provide
a broad and inspirational! description of the desired future of the protected area. Its main

purpose is to provide a focus or direction for management objectives and, according to
Thomas and Middleton (2003), should:

m Describe the kind of protected area that the plan is seeking to achieve in the long
term. This will help people to understand what it is hoped the area will be like in the
future, the reasons for this, and the action needed to achieve the vision.

m Bea long-term statement that is unlikely to change significantly over time. It should

therefore provide continuity in the process of managing the protected area in a
sustainable way.

30

Protected Area Monitoring Guidelines

m= ~=Include environmental, recreational, cultural and social and economic aspects of
the protected area.

The vision should be developed in a participatory manner, ideally with all key stakeholders,
to ensure that all interests are adequately dealt with and that there is broad ownership of
the vision.

Broad goals (or aims) relate to particular features or functions of the protected area, and
indicate general principles and directions of change. For example, the KMTNC 1997 ACA
Management Plan lists the following three primary long-term goals:

m to conserve the natural resources of the ACA for the benefits of the present and future
generations;

™ to bring sustainable social and economic development to the local people; and

@ to develop tourism in such a way that it will have a minimum negative environmental impact.

Objectives are specific outcomes or targets that the management activities will be designed
to achieve. Objectives should be clear descriptions of a measurable standard to be achieved,
or a desired state, threshold value, amount of change, or trend that you are seeking to
establish. Such objectives are often referred to as being SMART, i.e.:

Specific: Objectives must be focused and precise so that all stakeholders have a consistent
understanding of what is planned. For example, an objective such as "to conserve threatened
wildlife" would mean many different things to different people. Thus objectives should
not be easy to misinterpret. It is easier to identify and plan required actions if objectives
are specific.

Measurable: It is vital to be able to clearly determine whether or not an objective has
been reached. This can be done if measurable units are used to define the objective. It may
be relatively straightforward to set measurable objectives for habitat quantity or individual
species (€.g. a specific population size), but it is difficult to measure some objectives, such
as those relating to habitat quality. Biodiversity indicators may therefore be selected to
enable measurable targets to be set (see Box 3.4). For example, forest habitat quality might
be indicated by the presence of particular tree species that are only ever found in good
quality forests (i.e. old growth native forests that have not been subject to significant
exploitation). A measurable definition of forest quality may then be defined by setting a
minimum frequency of occurrence or density of the indicators.

Achievable: It must be possible to achieve the objectives within the protected area. For
example, it would not be possible to ensure the maintenance of a declining migratory bird
population (as it might be declining due to impacts outside the protected area). But it
would be appropriate to set an objective for its conservation within the protected area (e.g.
related to maintaining breeding success).

Realistic: Objectives should not be aspirational, such as to reverse all previous forest
loss. This might be a suitable long-term vision, but a more suitable objective for a
management plan might be to reverse 20% of previous loss over the 10 year period of the
management plan.

EF OIE CIE ARCLAMOTILONETS GUIGE NCS 31

Time-specific: It is important to set a time period for reaching the objective, to help
prioritise and plan actions. For maintenance objectives where ongoing actions are required,
the objective should state the period over which the objective will apply.

The rationale for objectives should be clearly explained within management plans. Some
form of priority category should also be given to each objective, so that decisions can be
made without re-evaluations if resource or time limitations require some objectives to be
dropped. Secondary objectives may also be included in case resources increase unexpectedly.
Objectives should be set for the biodiversity features themselves (i.e. their state) the factors
that influence them (i.e. pressures) and for the management activities that may influence
these pressures (i.e. the responses). Each of these objectives should then be monitored
(see Figure 4.1).

3.6.2 Setting objectives for the state of key biodiversity features

The setting of objectives for a biodiversity feature (or group of features) involves the
definition of the following six components:

1. Selection of the key biodiversity features within the protected area that will be the
focus of management and monitoring (see 3.4.3 above)

2. Definition of the geographic location and extent of the key features. In many cases it
may also be appropriate to sub-divide the features and set different conservation
objectives for different areas. For example it may be desirable to set different levels
of desired forest cover for different valleys and altitudes.

3. Identification of attributes that define the desired condition of the key features, e.g.
dominant species composition or species richness for habitats, or breeding population
size or breeding success for species. Further examples of attributes that may define
the condition of biodiversity features are given below in Box 3.3. For example, an
objective might be to maintain broad-leaved forest quality as defined by threshold
levels of its attributes of tree species richness, frequency of occurrence of particular
keystone tree species, frequency of occurrence of mature trees, shrub density and %
tree regeneration for specific forest types.

This stage also needs to take into account the monitoring resources that are available
and the ease of monitoring particular attributes of a feature. For example, establishing
the population density of Snow Leopards in the ACA would be extremely time
consuming, difficult and costly. It may, therefore, be necessary to set the objectives
for this species in relation to an index of relative abundance (e.g. frequency of
occurrence) rather than absolute numbers.

4. Establishment of the action to be taken with respect to each feature and its attributes,
e.g. increase, maintain, or decrease.

5. Setting of measurable standards for the state or degree of change for each feature’s
attributes (e.g. maintain >50% tree cover)

6. Setting of a time frame for the objective, e.g. increase rhododendron forest cover by
20% by 2020.

32

Protected Area Monitoring Guidelines

Box 3.3. Attributes of habitats and species that may be used to define
conservation objectives (adapted from Hill et a/. in press)

HABITAT ATTRIBUTES

Quantity:
wm Area

Quality — physical attributes:
g Geological (e.g. presence of bare rock or soil depth)
mg Water (e.g. presence of open water or depth of water table)

Quality — composition:
mg Community species richness/diversity
m Typical/keystone/indicator species
@ Presence/absence
@ Frequency
@ Number/density
@ Cover
Biomass

Quality — structure:
m Inter-habitat (landscape) scale (e.g. fragmentation, habitat mosaics)
gw Intra-habitat scale
gm Macro scale

@ Horizontal (e.g. plant community mosaics)

e@ Vertical (e.g. ground-, shrub- and tree-layer topography)
mg Micro scale

e@ Horizontal (e.g. patches of short and tall vegetation)

e Vertical (e.g. within-layer topography)

Quality — dynamics:
a Inter-habit
mw Succession
gm Reproduction/regeneration
gm Cyclic change/patch dynamics

Quality — function:
mg Physical/biochemical (e.g. soil stabilisation, carbon sinks, water storage)
mg Ecosystem (e.g. net producer)

SPECIES ATTRIBUTES

Quantity:
Presence/absence

Range

Population size
Frequency of occurrence
Number/density

Cover

Quality: Population dynamics
@ Recruitment rate

g Survival rate

gm Emigration rate

mg Immigration rate

Quality: Population structure
m Age
gm Sex ratio
mg Fragmentation/isolation
m Genetic diversity

Quality: Habitat requirenients (see above)

Protected Area Monitoring Guidelines. —_—$RA $A <—$—-_—-—$— << 33

Box 3.4. Biodiversity indicators

Biodiversity indicators are measures of biological or other features of the environment that
reflect to some degree the state of an ecosystem, habitat or other components of biodiversity.
Such indicators aim to fulfil three basic functions:

Simplification, i.e. to provide a simplified measure of a complex feature.

Quantification, i.e. to enable a numerical measurement to be made of a subjective property,
such as habitat quality.

Communication, i.e. to help understand the condition of a feature.

It is preferable to use a limited number of indicators, so that key conclusions are apparent.
The challenge is to strike a balance - the number of indicators should be small to minimise
monitoring requirements and to keep the main messages clear, but equally the issues must not
be oversimplified.

The presence of a particular species is often used as an indicator, e.g. of habitat quality, or
species richness. Such species indicators should preferably be:
m= widespread and relatively common in the habitat.

easy to identify, observe and census.

|
gu well understood with respect to its ecology and interactions with land-uses;
a

able to respond rapidly to environmental changes, so that they can provide early warning
of detrimental impacts;

representative of the habitat requirements and ecology of a large number of species;

of high intrinsic or popular appeal so that they can help motivate action (e.g. ‘flagship’
species).

The selection process of indicators should include a test to verify that it clearly reflects the
changes in the ecosystem for which it was chosen as an indicator.

It is, however, important to note that in many protected areas where resources and capacities
are limited and threats to biodiversity are high it will be necessary to set simple objectives
that can be monitored easily. Instead of spending scarce resources on quantifying the
amount of change it may be sufficient to know if there is a positive or negative trend, why
biodiversity is changing, and what are the local perceptions of the causes of change, in
order to formulate management decisions.

As an example, some SMART objectives for the state of the Singdi forest (Pasgaon VDC,
Lamjung District) key biodiversity feature in the ACA (see Table 3.2) might simply relate
to the following four forest attributes: forest area, tree species richness, tree age diversity
and the presence of indicator bird species (associated with diverse, native old growth and
little impacted forest). Some objectives relating to these attributes might be:

au SF1. Restore the areas of Singdi forest encroached by shifting cultivation to increase
forest extent by 25% in relation to 1995 coverage, by 2015.

mu SF2. Maintain the diversity of native tree species in Singdi forest, such that mean
native tree species richness is maintained above 50 species per ha, and non-native
tree species account for no more than 5% of cover in mature forests within the
ACA, for the next 50 years.

34

Protected Area Monitoring Guidelines

SF3. Maintain the natural forest of Singdi forest by ensuring that less than 10% of
trees are pole class (10-29.9 cm diameter over bark) and more than 30 % of trees
are above 30 cm diameter over bark over the next 20 years.

SF4. Maintain the wildlife quality of Singdi forest, by ensuring that at least 20% of
characteristic native old-growth forest indicator bird species (as listed in the
management plan) are present, over the next 20 years.

As an example of objectives for a species, some SMART objectives for the state of
the Snow Leopard population in the ACA might cover the attributes of range,
relative population density and breeding success. Some objectives for these
attributes might be:

SLI. Maintain the presence of Snow Leopards in each primary sample block in
each of the species’ key areas in the ACA.

SL2. Maintain the relative index of abundance in each of the four key areas in the
ACA, over the next 10 years.

SL3. Ensure successful breeding (rearing of at least one cub) at least once every
three years in each key area.

Some other SMART objectives for certain attributes of the state of some biodiversity
features in the ACA region (see Table 3.2) might be:

G1. Maintain the area of utilisable grassland (i.e. that can support over | livestock
unit per hectare over three months per year) in the Caragana, Lonicera, Carex and
Cobressia species grasslands of the trans-Himalayan Rangelands (Mustang and
Manang) within 10% of baseline levels as determined by 2005 satellite image
analysis, until at least 2015.

CP1. Maintain at least two populations of Cheer Pheasants, with at least 50 calling
males in each, in the Southern Annapurna Region over the next 10 years.

RF1. Increase Rhododendron forest cover in the Ghandruck-Ghorepani region of
the ACA by 10% (according to 2005 baseline surveys) by 2015.

IF1. Increase the forest biomass of the intensive forest use zone for fuelwood use
in the Ghandruk district by 25%, against 2004 baseline data by 2015.

3.6.3 Pressure and response objectives

Once the overall objectives have been set for the state of each biodiversity feature, then
objectives should be set for the pressures that affect it and the responses or actions that
may be needed to control the pressures. These would be typically developed as part of the
management planning process, with the response objectives guiding the development of a
detailed workplan.

Some hypothetical objectives for biodiversity pressures and responses with respect to some
biodiversity features in the ACA region are indicated in Table 3.4.

Protected Area Monitoring Guidelines ——_—<—a ——M— —_—_ — — — — — — — — — — — — — — 35

Table 3.4. Some example pressure and response objectives for some key
biodiversity features within the Annapurna Conservation Area

Pressures on
feature

Key Feature
(see above for
objectives)

Fuel wood collec-
tion

Singdi forest

Timber collection

Snow
Leopard
Uncia uncia

Disturbance by
trekkers

Decreased prey
numbers (e.g. Blue
Sheep)

Retaliatory killing
by herders

Hunting for skins
and traditional
medicines

36

Pressure objectives

FW1. Limit fuel wood
collection to less than
3000 kg per annum per
household; review in 2010

T1. No felling of native
timber within defined
reserve zone until 2020

V1. Maintain less than
0.1 trekkers per 100 km?
on average per month in
sensitive parts of key
areas; review in 2008.

BS1. Maintain Blue
Sheep densities of > 73 /
10 km square.

SLP1. No persecution of
the species in any area.

SLP2. No hunting of the
species in any area

Response objectives (note
some may address more
than one pressure objective)

Carry out forest conservation
awareness workshop in each
village within Skm of Singdi
forest

Build micro-hydro electric station
to supply power to 200 households
in Singdi village, by 2010

Build paraffin fuel store in each
village within 2km and provide
half price paraffin stoves to
villagers, by 2008.

Increase tree planting in
surrounding intensive use zone
to increase forest cover in
district by 10% by 2020

Identify sensitive areas in each
key area where special permits
are required for trekking /
climbing and limit permits to
<200 per year; review in 2008.

Establish livestock free core zones
in key areas; review in 2005.

Distribute posters indicating
protection from hunting of Blue
Sheep and other protected
species to all > 2,000 m villages
in ACA by 2006.

Set up livestock predation
compensation / insurance
schemes in each village in each
key area by 2010.

Preventive measures such as
improved corrals set up in each
village in each key area by 2010.
Ensure a Snow Leopard Conser-
vation Committee is established
in each key area by 2006, and
in other priority areas by 2008.

Ensure at least two Snow
Leopard conservation awaren-
ess packages organised/provi-
ded in each key area by 2007.

Protected Area Monitoring Guidelines

4. Developing a protected area biodiversity
monitoring programme

2 PY

4.1 Introduction

A protected area biodiversity monitoring programme‘ requires a great deal of time and
effort to develop and implement. It is therefore important to ensure that it is undertaken
efficiently, whether or not it involves participatory approaches or field surveys by protected
area staff, contractors or others. This requires careful planning, but this is time well spent
as monitoring programmes may last many years and poor decisions may not become
apparent for a long time; when it is then usually too late to do anything about.

Monitoring programmes should, therefore, be carefully planned to ensure that they are
effective (i.e. provide adequate answers to the questions that they were set up to address)
and efficient (i.e. collect the required data with as little effort and cost as possible). Many
of the most common pitfalls associated with monitoring, as summarised in Table 4.1, can
be avoided by careful planning.

Table 4.1. Some common monitoring pitfalls and ways of avoiding them

Common failures in monitoring Means of avoiding it
programmes

Data are collected that are of no use, and/or key | Focus on determining if conservation objectives
questions cannot be answered. are being achieved, and identify data requirements
and analytical methods carefully.

Poor design leads to inconclusive results. Plan carefully and test with pilot surveys and
analysis.

Multiple observers differ in field skills and use | Document precisely the methods to be used and
inconsistent methods. ensure observers have been trained in using the
methods, have practiced them and follow them.

Methods are changed during the monitoring | Plan and test the methods and then stick to them.
programme, and thus survey results cannot be | Ifchanges are absolutely essential document them
compared. and if possible calibrate effects of changes.

Inappropriate methods are chosen for habitats or | Know your habitat and species, and if necessary
species. carry out pilot field tests.
(Cont.)

+ 4 protected area biodiversity monitoring programme is the whole series of related surveys
and analysis that aims to establish whether or not all the specific biodiversity objectives
for the area are being achieved.

Protected Area Monitoring Guidelines

Table 4.1. Some common monitoring pitfalls and ways of avoiding them (Cont.)

A few large samples are taken rather than many
small ones, so natural variation cannot be
accurately measured.

Natural fluctuations in populations and habitat
conditions obscure management changes.

Locations of permanent sample sites are not
properly marked and recorded so that the same
areas are not revisited.

Biased results are obtained because samples are
purposefully taken in good/interesting areas of
habitat or favoured sites for species etc.

Spurious results may be obtained by comparing
data using the same methods in different habitats
because the results are incorrectly assumed to be
directly comparable.

Results are assumed to be completely accurate and
biases are ignored (e.g. birds counted are assumed
to be all the birds present).

The data cannot be analysed statistically.

Data are lost, either physically or within
organisations.

Data are not analysed because the biologists lack
the skills to do so.

Managers, other staff and external stakeholders
do not use monitoring results because the data are
not useful or not trusted.

Monitoring programmes are not sustained and
results remain unused due to loss of institutional
support.

Use many small samples in preference to a few
larger ones, and if possible carry out pilot surveys
to estimate variation and calculate the required
number of samples to detect significant changes.

Ensure that sampling is sufficient in terms of
numbers and frequency and carried out using
appropriate methods.

Carefully mark sample locations with more than
one marker, designed to last a sufficient length of
time. Photograph the site, take GPS coordinates
and map the location.

Take random or systematic samples (see
Section 4.9).

The efficiency of methods may vary between
habitats, so take care with inter-habitat
comparisons or measure and calibrate habitat
effects.

Identify and document biases, and standardise
methods to maintain consistent biases.

Ensure that planning includes identification of the
statistical methods to be used and their data
requirements.

Ensure data are copied as soon as possible,
maintain records of datasets and their locations,
with full documentation of methods used to collect
them. Inform others of the data.

Ensure that the methods of analysis are identified
at the planning stage and that suitably qualified
staff are available and responsible for analysis of
the results.

Spread ownership of the programme by involving
managers and other stakeholders in planning of
monitoring.

Ensure that there are adequate long-term resources
for the whole programme, and use and increase
the local capacity for participatory monitoring.
Share results with a broad community to increase
possible sources of funding.

Good planning of a monitoring programme involves consideration of a number of key
issues, which are outlined in the diagram in Figure 4.1. Although the diagram suggests that
the planning is a step by step process, in practice it requires iterative decision making,
because decisions may have to be revised according to the outcome of later decisions. In
other words, some decisions may have to be reconsidered at later stages of the planning
process. For example, it may be necessary to reconsider the frequency of monitoring of
some features, once the overall programme has been costed.

38 —_—————— Protected Area Monitoring Guidelines

Figure 4.1. Important steps in planning a biodiversity monitoring programme

Determine what needs to be monitored (4.2) State and pressure
objectives for key
biodiversity features
(Chapter 3)
FOR EACH OBJECTIVE
(Le. feature and attribute)

Review existing survey / monitoring data (4.3)

Define location of features to be monitored, and target and
sample population (4.4)
Assess available Det 3 onan (46 Assess risk to each
monitoring etermine monitoring frequency (4.6) feahetand
resources (4.5) anticipated rate of
Select measurement method (4.7)

Determine best time to use method (4.8)

Devise a sampling scheme where necessary (4.9)

inherent change

Revise if
necessary

Devise recording forms and document methods (4.10)

Test methods and sampling strategy with pilot surveys or
existing data (4.11)

FOR THE ENTIRE PROGRAMME

Review feasibility in relation to resources (4.12)

Prepare a work-plan (4.13)

Carry out training if necessary (4.14)

bitline

Carry out monitoring and analyse data (4.15)
Review
i objectives and

monitoring
Report to stakeholders and act on results programme
(4.16) (4.17)

Note: Figures in brackets refer to relevant text sections

Protected Area Monitoring Guidelines 39

Plate 4.1 Cheer Pheasant observation

Photo: Raju Acharya

Each step outlined in Figure 4.1 requires consideration of a number of key decisions (which
are further described in Sections 4.2 — 4.17 below. These decisions will require careful
consideration of your monitoring aims, available information on the biodiversity features
to be monitored, the physical nature of the areas to be monitored (e.g. size, topography,
climate, accessibility) and available resources (time, funding, expertise).

Some recommended sources of further information on biodiversity monitoring are listed
in Box 4.1.

Box 4.1. Recommended references for further information on
biodiversity monitoring strategies and methods

See reference list for full citation details.

Doak and Pollock. Statistical / monitoring tools for the design and analysis of conservation
monitoring data http://www.biology.ucsc.edu/people/doaklab/natconserv/index.html

Elzinga et al. (2001). Monitoring plant and animal populations.

Fancy. Monitoring Natural Resources in our National Parks. http://www.nature.nps.gov/im/
monitor/

Feinsinger (2001). Designing field studies for biodiversity conservation.

Goldsmith (1991). Monitoring for conservation and ecology.

Hill ef al. (in press). A species and habitat survey, evaluation and monitoring handbook.

Krebs (1999). Ecological methodology 2nd ed.

Southwood (1978). Ecological methods.

Spellerberg (1991). Monitoring ecological change.

Sutherland (1996). Ecological census techniques.

40

Protected Area Monitoring Guidelines

4.2 Determine what needs to be monitored

4.2.1 Monitor objectives for the key biodiversity features

Clearly and explicitly defining the purpose of the monitoring programme is probably
the most important step. Failure to do so may result in data being collected that are of
little value, or of important data requirements being overlooked.

In these guidelines for monitoring biodiversity in protected areas, we recommend that
monitoring should focus on establishing whether or not the conservation objectives for the
area are being achieved (see Section 3.6). Thus the first stage in developing a monitoring
programme is to ensure that you have clearly defined conservation objectives. Key features
of biodiversity importance in the protected area should be identified and these should be the
focus of management actions and monitoring. Each key feature should have clear (SMART)
conservation objectives set for it. Ideally these will have been already established during the
production of a management plan or similar document (as described in Section 2.1).

4.2.2 Monitor states, pressures and responses

We have also recommended that SMART objectives should be set for regulating pressures
on key biodiversity features, and for management responses to pressures which result in
changes in the state of features (in accordance with the Pressure — State — Response
framework described in Section 3.5). Thus in addition to establishing if conservation
objectives are being met (i.e. establishing the state of the key feature) monitoring should
also establish if objectives relating to pressures and responses are being achieved. For
example, a monitoring programme for Snow Leopard might include:

State monitoring

m& Presence of Snow Leopard throughout the ACA (from reported sightings).

m Relative abundance of Snow Leopard in selected key areas within the ACA.

m Successful breeding in key areas (i.e. presence of juvenile animals in the population).
Pressure Monitoring

m Livestock abundance and seasonality in grasslands in Snow Leopard range within the ACA.

m Abundance of Blue Sheep and other natural prey species in Snow Leopard range
within the ACA.

= Condition of high altitude grassland / shrublands in key areas for Snow Leopard.

= Numbers of herders within key areas for Snow Leopard.

m Disturbance by trekkers / climbers etc in key areas for Snow Leopard.

= Cases of killing of Snow Leopards (persecution / retaliatory killing etc).
Response monitoring (illustrative actions)

= Number of anti-poaching patrols carried out in high risk areas

= Proportion of communities with established Snow Leopard Conservation Sub-
committees

= Number of meetings held with village communities, to raise awareness of threats and
to advise on livestock protection measures that avoid the need to kill Snow Leopards

EX OLECKEHAT EAN OTESLO NETIC GLEBE 197 CS NTT TTT 41

m Success in rangeland restoration measures (livestock exclusion)

mw Extent of wildlife conservation awareness camps and inclusion of primarily herders and
farmers who are intrinsically dependent on forest and grazing land (awareness creation)

= Number of legal actions initiated by the KMTNC-ACAP as per the rules and regulations
(law enforcement aspect)

a Inclusion of wildlife conservation as a subject in the school CE curriculum

When considering pressure and response monitoring requirements it may be found that
some monitoring activities can be combined with those needed for other key features. For
example, grassland/shrublands is a key biodiversity feature in its own right, and therefore
monitoring of the state of grasslands meets the requirement to monitor pressures on Snow
Leopards (through impacts on Blue Sheep and other important prey).

These guidelines focus on biodiversity monitoring, and therefore the monitoring of abiotic
pressures and management actions (i.e. responses) is not within its scope. The rest of this
chapter provides detailed guidance on how to plan and undertake monitoring of objectives
that relate to biodiversity components.

4.3 Review existing data

It is important to establish whether monitoring has previously been undertaken, including
the features and attributes covered, the methods used, the time-scale and frequency over
which it took place, and whether or not it is ongoing. This may require careful investigation
as it is not unusual for the results of monitoring studies to be forgotten, especially where
they are unpublished and collected by an external organisation, or where personnel have
since moved on.

Data from previous monitoring programmes or ad hoc surveys should be used in the
assessment process (described in Chapter 3) to identify key features to be monitored, if
these have not been identified in a management plan, and to help assess the appropriateness
of potential survey methods and sampling strategies.

Where possible and appropriate, monitoring should build on existing data collection
programmes (e.g. government forest inventories), institutional arrangements (e.g. ACAP
regional structure and CMACs) and local community activities (e.g. forest patrols,
sheparding). But previous monitoring programmes should not be simply repeated without
careful consideration of their suitability, as they may have been established to meet different
objectives. However, where existing programmes are likely to contribute to current objectives
they should be continued and developed if necessary. Where appropriate, existing
methodologies should be followed to maintain the validity of long-term datasets. It may
also be useful to use existing fixed marker systems or permanent quadrats.

If prior surveys have not been carried out and the identity and condition of biodiversity
features of importance is in doubt, then it may be necessary to carry out a baseline survey
before a detailed monitoring programme can be planned. It is necessary to establish the
baseline condition of features so that appropriate conservation objectives for them can be
set and any subsequent changes in them detected.

42

Protected Area Monitoring Guidelines

4.4 Define the location and boundaries of the biodiversity
features to be monitored

It is vital to clearly define the geographical area or species population that you wish to
monitor and draw conclusions on. This requires consideration and definition of four types
of population that you may be interested in: biological population, target population, sampled
population and statistical population (Elzinga er al. 2001).

A biological population is "a group of organisms of the same species, present in one
place at one time". For example, all the plants of a particular wetland species that occur at
an isolated wetland might be appropriately defined as a biological population. But in many
cases biological populations are difficult to define in practice because the "place" occupied
by a species is often not clearly definable. This is particularly the case for mobile species
or species that have a wide capacity for dispersal, so that there is much emigration and
immigration between apparently geographically separate populations.

Where feasible it is best to monitor on a biological population basis. This means that
different biological populations are monitored (or analysed) separately and whole
populations are monitored where possible. In practice this is often difficult to achieve,
especially for large or mobile populations, as protected areas often cover only a small part
of the population of such species. However, within the ACA it might be possible to conclude
that the main west-east Annapurna mountain chain divides the area into two broad areas
which may support distinct and separate populations of many species. Monitoring may
therefore need to consider the population status of each of these separately.

In many species biological populations will extend beyond the ACA boundary. Where
such extensions are small then monitoring should cover the entire biological population
rather than be constrained by the administrative boundary of the ACA. This enables more
ecologically meaningful conclusions to be drawn from the monitoring. For example, one
might imagine a situation where a bear population is found within a valley, half of which
occurs within the ACA. Monitoring of the ACA section only might reveal that bear numbers
have declined. But monitoring of the whole valley might indicate they have merely moved
to another section of the valley outside the ACA. Although the loss of bears from the ACA
may be undesirable for some reasons, the conclusion that the bears have moved is much
less serious than the erroneous conclusion that they have declined.

In practise monitoring normally focuses on part of the biological population, typically that
which is within the protected area. Or sometimes we may only be interested in monitoring
part of the population within the protected area, if for example, part of the area is being
managed specifically for a habitat or species, although they may occur elsewhere in the
protected area. The population that we are interested in can be called the target population
(see Figure 4.2a) and the area in which it is found is the target area.

It is sometimes possible to monitor a whole feature across a target area. For example it
may be possible to reliably measure the full extent of a habitat feature by aerial photography
or carry out a complete census of a localised and conspicuous species. But more often it
will be necessary or more efficient to monitor a feature by assessing samples. The area
over which samples may be drawn then defines the sample area and sample population.

2

45

Protected Area Monitoring Guidelines

The statistical population is the entire set of observations across all the samples, from
which statistical inferences are made.

Ideally the sample population should be the same as the target population, but in many
cases the sample population will be different due to practical constraints. When this occurs
it is vital to recognise the difference between your target population and your sampled
population so you know the limitations of your data. It is only possible to draw valid
statistical conclusions about your sampled population.

When the target population has irregular boundaries, it may be most practical to redefine a
new sample population by fitting a regular shaped polygon over the bulk of the target area.
This newly defined area, referred to as a macroplot, becomes the sample population.
Macroplots are relatively large areas, with sampling units such as quadrats, lines or points
(see Section 4.7) located within them. They facilitate the positioning of sampling units.
Macroplots are usually permanently defined to ensure that the same area is sampled on
each sampling occasion.

Sample populations may also differ from target populations because areas cannot be sampled
for practical reasons, because for example they are too steep or wet or inaccessible for
other reasons (see Figure 4.2).

If a target population covers a large area then it may not be feasible to sample the whole
area because of time and resource constraints. Sampling over a large area will usually
result in widely scattered sample locations, which will entail considerable amounts of
travel time. This is particularly a problem in areas such as the ACA where there is no
means of transport, paths are limited and much of the terrain is extremely steep and
hazardous. It is therefore necessary in such circumstances to restrict the sampling area to a
smaller subset of the target population. This can be accomplished by placing a random
sample of macroplots (primary plots) within the target population. Further sampling (i.e.
with secondary plots) is then carried out within the primary plots (leading to a two-stage
sampling design as described further in Section 5.5). Because the macroplots are randomly
selected from within the entire target area, the sampled population and the target population
are the same, and therefore statistical conclusions apply to the entire target population.

However, if the target population is very large and difficult to sample, then it may be
necessary to restrict sampling to a few selected key areas. For example, it is not practical
to attempt sampling of Snow Leopard occurrence over its entire target population (i.e. the
entire extent of suitable habitat for the species within the ACA). Although the species is
probably widespread, in many parts of its range within the ACA it may only occur irregularly
and at very low densities. It is therefore only practical to identity and monitor (by field
surveys) a few key areas where Snow Leopards are known to occur. Samples are then
taken within each key area. This is again two-stage sampling, but in this case the sampled
population is only each key area. Statistical inferences may only be validly drawn for each
key area. Values from the key areas should never be averaged because they are not randomly
drawn samples; key areas are selected (normally subjectively) with a particular intent (e.g.
to study areas of known importance for a species).

44 —S— tected Area Monitoring Guidelines

Because statistical inferences can only be made to the key areas that are actually sampled,
it is important to set conservation objectives that are specific to each key area, rather than
the target area as a whole (unless other means exist for monitoring the whole target area).
Where necessary it is equally important to clarify within management plans that actions
may be taken according to the results of the monitoring in key areas, despite the fact that
they are not necessarily representative of the situation across the whole target population.

Figure 4.2. Example illustrations of target populations, sample populations
and the use of key areas
Key:

thick black line = protected area boundary and limit of monitoring area;
thin black line = river;

grey = forest;

white = grassland / shrubland;

hatched = inaccessible areas (steep).

stippled grey = high altitude rock and ice;

Q0 9001!

hatched = inaccessible areas (steep).

sampled population

4.a. Forest bird monitoring. The target population 1s all the solid grey (forest). The sampled
population is the grey area excluding the hatched grey areas (inaccessible land). Survey
plots are randomly distributed (see Section 4.9) over the sampled population (white squares).

. . 78 A 4
Protected Area Monitoring Guidelines —<<—<—@— — — — ._ +5

4.b. Snow Leopard monitoring in key areas. The target population is the white area (:.e.
suitable grassland / shrubland habitat). The three key areas (dotted lines) excluding the hatched
black and white areas (inaccessible land) are three separate sampled populations. Survey
plots are distributed by restricted random sampling (see Section 4.9) in each key area.

4.5 Assess available monitoring resources and prepare budget

It is particularly important to take into account available resources (e.g. staff, time, expertise,
transport, funding) when planning a monitoring programme. Many scientific monitoring
activities are time-consuming, expensive and require experienced personnel. Careful
consideration therefore needs to be given to what is achievable when setting monitoring
objectives, especially in the long-term. Monitoring will be of no value if it cannot be
repeated. Some of these problems can be alleviated by the careful targeting and efficient
design of monitoring programmes, as suggested in Box 4.2.

Box 4.2. Approaches to minimising costs of monitoring in protected areas

The key aim of monitoring in protected areas is to establish whether or not conservation
objectives are being met. Even where monitoring resources are severely limited, this may be
achieved by taking the following actions:

Prioritise monitoring for features that are of highest conservation interest and at highest
risk (based on a risk analysis and estimate of the probability of each possible impact
and its likely magnitude).

Set simple conservation objectives where possible and restrict monitoring to what is
required to test if these objectives are being met (e.g. if the aim is to maintain the
presence of a species do not use more time consuming methods that estimate densities).

Monitor pressures (see Section 3.5) more frequently than the state of features, as these
are often easier to measure and can provide early warning of potential problems. But do
not restrict monitoring to pressures. Carry out occasional state monitoring, and use
pressure monitoring to trigger further state monitoring if pressures increase.

cont.)

46

Protected Area Monitoring Guidelines

Box 4.2. Approaches to minimising costs of monitoring in protected areas
(cont.)

Focus on key areas (see Section 4.4) where it is not possible to cover all areas effectively,
but realise that this restricts one’s ability to draw conclusions about the protected area as
a whole.

Only monitor as frequently as necessary (see Section 4.6, i.e. in accordance with expected
rates of change or risks). For example, there is little to be gained from monitoring forest
condition at less than 5 year intervals (unless pressure monitoring or casual visits indicate
a sudden event e.g. fire).

Use the most cost-effective methods, but be aware that these need not necessarily be the

most simple. For example, remote-sensing may be the most cost-effective method for
monitoring habitat extent. Consider participatory approaches where appropriate, but
ensure that they really are cost-effective (as they may take a lot of time to set up and run
over the long-term) and provide the data that you need consistently and reliably.

Be clever and use efficient sampling strategies. For example consider multi-stage
sampling to cut down on travel time between sites (see Section 4.9). And use permanent
sample sites where feasible. Although these may be time-consuming to establish they
are much more efficient in detecting changes in the long run provided that a sufficient
number are established and located randomly.

Use a phased approach to develop monitoring when major project funding is available.

Participatory monitoring methods may also be more sustainable and provide other
advantages, such as increased ownership of results (e.g. see Danielsen et a/. 2000, Danielsen
et al. 2003). However, although participatory biodiversity monitoring can be cheaper than
more conventional scientific monitoring, this is not always so. Focussing monitoring on
pressures, or carrying out threat reduction assessments (Salafsky and Margoluis 1999) can
also reduce costs and training requirements etc, but the monitoring of pressures should
never replace monitoring the state of key features. Pressure monitoring may, however, be
carried out, more frequently than state monitoring, and then used as a possible trigger for
more intensive and frequent state monitoring if pressures appear to increase and thus warn
of possible impacts (see Section 4.6).

A preliminary budget should be defined at this stage of the process since there is no point
developing programmes without the funds to implement it. The items of the budget should at
least include costs of external data (e.g. aerial photographs), staff time, equipment, local
meetings, transportation, stationery, and other operational costs. Funds for publicity and
dissemination are important as well. Staff and local participants may need to be trained in
field methods, record keeping and data analysis, depending on which methods are to be used.

The budget may need to include payment to participating stakeholders, particularly villagers.
For local communities, especially in the case of poor farmers, being involved in biodiversity
monitoring is extra work which takes time and money. It is fair to offer a fee to take
account of these costs borne by local people, keeping in mind that this will be temporary.
After developing the rest of the methodology, i.e. after determining how to monitor each
objective and how to document and disseminate the results, the budget should be reviewed
and finalised. If a relatively large amount of short-term funding is available, e.g. as a three-
year development project, then this can be used to develop monitoring through a phased
approach:

Protected Area Monitoring Guidelines ——<— @— @ i i_i—-—-—— — §m—m—m 4

m= Use the project funding to do baseline assessments, test monitoring methods and
carry out training to develop capacity.

m Validate and calibrate simple methods against more detailed methods during the
project phase.

mw = Inthe longer-term, adapt monitoring according to increased knowledge, and rates of
environmental change and risk (e.g. use simple methods if these have been shown to
provide reliable results, and reduce quantitative monitoring of features that are of
low risk of sudden change).

4.6 Determine the monitoring frequency

The frequency at which monitoring is carried out is a key factor affecting the cost of
monitoring. One should therefore avoid monitoring more frequently than is necessary.
The likely rate of change in each feature as a result of natural events and management
interventions is of key importance in deciding how often monitoring visits should be carried
out. Thus, for example, it may be appropriate to visit forests at five-year intervals because
major changes in such habitats are normally very slow in the absence of disturbance. But
bird populations may vary considerably from year to year, so surveys may need to be
carried out annually if resources allow.

But unexpected events may affect the biodiversity features. Monitoring programmes should
therefore incorporate sufficient flexibility to deal with unforeseen, potentially rapid and
catastrophic events (e.g. storms and fires). Additional very basic inspections may be needed
to detect such events and then additional monitoring can be designed to establish the
condition of a site.

A general procedure for determining the appropriate frequency of monitoring for a particular
feature is:

1 Select an interval consistent with:

m the likely rate of change of the feature in relation to intrinsic natural rates of
change and known management influences, taking into account the precision with
which that change can be measured;

m™ any reporting requirements (e.g. to project donors); and

mw the availability of funds for monitoring.

2 Aim to make a detailed assessment of the attribute at the required interval (e.g. for a
protected forest area, aerial photography may be required at intervals of 10 years).

3 Assess the risk of change from external factors.

Unless very frequent monitoring is required, aim to make a basic inspection of the features
and their attributes more frequently for signs of abrupt change (e.g. for forests, a basic
inspection at intervals of 3 years may be appropriate).

48 SS . TOL CIE Are MOTO ning GIG hIES

4.7 Select measurement methods

Choosing the most appropriate method for measuring each key feature and its attributes,
or pressures On it, is another critical step in planning a monitoring programme. Essentially
one should aim to use the most cost-effective method that provides an adequate assessment
of whether or not the conservation objective for the feature (i.e. its state) or the management
objective (i.e. relating to pressures on it) are being met. Very often the most cost-effective
method may be the simplest, but this is not always the case. Well planned and implemented
scientific studies may in the long-term provide better value for money than very simple
subjective methods that may produce results of little value.

The choice of methods needs to take into account some key questions, as outlined in Figure
4.3 and described below (much of which is based on Hill et al. in press). From this brief
discussion of potential measurement methods it is clear that there are a number of important
issues to be considered. However, it is difficult to make the best choices without practical
experience and data, and therefore as discussed in Section 4.11 it is highly recommended
that methods are tested in some preliminary field trials before a final selection is made.

Plate 4.2 Snow Leopard habitat in ACA, Nepal

Photo: Kamal Thapa

PEPOTERICCUA REINO TIEL ORI GLEE C LY Cm aa 49

Figure 4.3. Selection of methods for measuring attributes of each feature

For each feature’s attribute
to be monitored
Consider the next

Consider the most cost-effective method TOS COST:
effective method

Is it likely to have an unacceptable environmental or
socio-economic impact?

Is it able to provide a type of measurement
consistent with the objective?

Is it able to detect appropriate degrees of change?

Does the bias matter for
monitoring purposes if it is
consistent?

Can the bias be controlled
or measured?

Establish best time to use method, devise sampling strategy, devise recording
procedures, test and document.

50 —— nn POC ANCA OOF GUIGENTES:

4.7.1 Ensure that the method will not have unacceptable environmental or
socio-economic impacts

Unfortunately, there are many documented cases where research and monitoring
programmes have merely measured and recorded the damage caused by their own activities.
Therefore, great care should be taken to ensure that the methods chosen will not cause any
damage, and the following precautions should be observed:

m= Do not use destructive sampling methods unless absolutely necessary.

mw Ensure that important wildlife, natural resources, livestock and crops are not damaged
during field surveys (e.g. by trampling).

m Minimise disturbance to sensitive species.
m Do not use vehicles on sensitive habitats unless impacts can be avoided.

= Position fixed sampling locations sensitively and avoid or minimise damage during
their establishment.

a Ensure any fixed or temporary structures that are left unattended do not pose a risk
to wildlife, livestock or local people.

m Avoid excessive re-visiting of sites and sampling locations.

4.7.2 Choose the most cost-effective method that provides the appropriate
type of measurement

One of the key questions when planning monitoring is whether to use subjective assessments
(e.g. villagers are asked if fuel wood collection has increased, decreased or remained the
same over the last year), or objective measurements (such as by counting the amount of
fuel wood logs in a sample of local houses over the year in question). Subjective assessments
are generally easier and, therefore, more readily applicable to participatory approaches. As
a result they may be cheap and a lot of data (or samples) can then be collected per unit cost.
However, many studies have shown that people vary greatly in their subjective assessments.
Subjective assessments are also usually more biased than objective measurements, and
very importantly, such biases may vary between occasions. One must also be aware that
such biases may result from vested interests or subconscious wishes. There are many cases
where local resource users (e.g. fishermen) have claimed that resources are not being
overexploited, despite overwhelming scientific evidence to the contrary.

One must therefore carefully weigh up the risks of relying on subjective assessments, and
they should not be used by themselves when impacts on an important feature could be
rapid, substantial and irreversible. Subjective assessments may, however, be particularly
useful if validated, and even better, calibrated against objective methods. They may also be
useful as a simple early warning system of impacts on features that are in good condition
and at low risk of rapid change.

It should also be remembered that the use of objective methods does not preclude
participation by local communities. The scientific principles described in these guidelines
can be applied to participatory approaches, and we should not underestimate the abilities
of local people to record detailed and complex data, though the analysis and generation of
useful results may require external support. On the other hand, highly technical and labour
intensive scientific methods are more often than not unsustainable in the long-term.

Protected Area Monitoring Gedeli€S: —<$<_<_<_—$_—$—$———————————————— 51

Another key consideration is whether to use quantitative, semi-quantitative or qualitative
methods.

= Qualitative data: e.g. single assessments of presence / absence of a feature in a site or
area with no indication of abundance. These data can be combined to create frequency
datasets, which are semi-quantitative data.

= Semi-quantitative data: e.g. data that can be analysed mathematically, but do not
provide absolute measurements of the attribute. Relative abundance is an example of
this type of data. For example, with relative abundance data one may say that species
A is twice as abundant now compared to 10 years ago, but one cannot know how
abundant the species is or was.

= Quantitative data: e.g. direct counts, estimates or indices of absolute abundance, area etc.

In practice it is often necessary to use simple qualitative methods in protected area
management monitoring, where resources and capacities are limited and threats to
biodiversity are high. However, this is not necessarily a significant problem as qualitative
data can normally be obtained more quickly and cheaply than semi-quantitative data, and
especially quantitative data. The ability to take more samples than quantitative methods
may then overcome the greater variation in measurements resulting in greater precision
overall. In other words it may be better to take a lot of simple measurements than a few
reliable estimates. For example, with the same amount of limited time and resources it
may be possible to assess the population status of Snow Leopards in the ACA by measuring
its presence or absence over approximately 30 (thirty) 4-km* sites, or by reliably measuring
its relative abundance in just 3 (three) 4-km?. In this case it would be much wiser to
monitor the status of the population by the simpler and qualitative wider scale presence/
absence method, combining the data to produce a semi-quantitative frequency of occurrence.

Another consideration is whether to use direct or indirect measure of the attribute. A direct
measure involves making measurements of the attribute itself (e.g. counting the number of
livestock present). An indirect (surrogate) measure involves measuring a related variable,
which is used to infer the status of the attribute being monitored (e.g. counting dung as an
index of the number of livestock present). Such measurements are described as an index
(in this case of population size). An index of population size is also obtained from direct
sampling of a subset of a total population. For example, male pheasants can be counted by
their calls, but this does not offer an index of total population size, since one cannot be sure
of the numbers of females.

Finally it must be remembered that the selected method must produce a measurement that
is consistent with the objective for each feature and its attributes. For example, the objective
for species composition in a forest stand might be “70 - 90 % of trees of Quercus species”.
This could be measured by simply calculating the frequency of Quercus species encountered
along a number of randomly located line transects (see Table 4.3). However, if the objective
specifies a required density of Quercus trees then quantitative estimates of Quercus density
on continuous scales of measurement are required using a different method. It is therefore
important to consider the potential methods and costs of measuring the achievement of an
objective when setting that objective (see Section 3.6). Thus if a method cannot be found
that can adequately measure achievement of the objective with the resources (time, money,
expertise) that are available, then it may be necessary to modify the objective.

52

Protected Area Monitoring Guidelines

4.7.3 Use a method that measures attributes across an appropriate range
of conditions

It is essential that the method is suitable for the range of conditions over which it is to be used.
For example, a forest bird survey method must be appropriate for use across the range of
forest types that are being monitored, e.g. from open forest to dense forest.

Within the ACA it is particularly important that methods can be applied across a wide range of
terrains and altitudes. Methods that cannot be used on steep slopes will be of very little use!

4.7.4 Use methods that measure appropriate degrees of change

Time may be wasted if a method is used that is very precise when only large changes need to be
detected. For example, it is not necessary to use complicated methods for measuring tree height
to the nearest centimetre if the objective relates to ensuring 20% of trees are above 10m height.

4.7.5 Use a method with acceptable and controllable bias

What is bias?

Bias is a systematic source of error that results in under- or over-estimation of the attribute
being measured. For example a survey methodology may result in half of the individuals of a
species being routinely overlooked, and as a result population estimates from the survey will
always be half of the true value. Thus, bias causes estimates to be inaccurate. Methods free of
bias are said to be accurate, but completely bias free methods are always impossible to obtain.

Sources of bias
Bias may arise from several sources in a study, including:

Observer

m Incorrect identification of species.

Failing to detect and count all individuals of a particular species being monitored.
m Different observers may record identical observations in dissimilar ways.

m Differences in expertise between observers.
a

Variation in observer effort (e.g. speed of assessment).

Location

m Studying a species only where it is common introduces bias; if areas where it is rare
have been ignored the full dispersion of the species will not be understood.

= Using a small subjectively selected sample area when the site being studied is not
homogenous will introduce bias.
Habitat differences
m Variation in the detectability of species in different habitats.

@ Inabilities to access some habitats or areas.

Species differences
@ Variation in detectability between species.

m= Some species may be more easily identifiable than others.

Protected Area Monitoring Guides 5)

Time related sources

u_—_ The time of year (or day) when a survey is carried out can affect the results.

Weather

= Weather may affect the activity of animals and therefore their detectability (e.g. many
birds stop singing in wet weather, and vultures and other raptors will not fly in bad
weather when there are no thermals to enable soaring)

a Bad weather can reduce an observer’s ability to see and hear (e.g. it is difficult to
hear singing birds in windy weather)

a __ Inclement weather affects observers’ concentration, as well as variation between different
observers’ capacity for working under difficult conditions can introduce a bias.

Dealing with bias

Awareness of such potential sources of bias when planning a monitoring programme can
help avoid future analytical problems. Hill et a/. (in press) suggest that there are three
ways of combating bias:

1. Anticipated sources of bias can be reduced or controlled by:

a _ Using the same methods, observers and analysis etc., across years and sites. Ensuring
that procedures are well documented helps to maintain consistency (see Section 4.10).
If recording effort cannot be kept constant, the next best thing is to measure it, that
way any observer bias can be more easily assessed.

a Checking that methodological assumptions are valid for the habitat or species you
wish to study and for the period of time over which it is to be studied.

a Recording relevant weather conditions when surveying. Agree and record beforehand
under which weather conditions work should be postponed.

Agreeing and recording definitions (e.g. sample size, type, population unit etc.) beforehand.

Calibrating observers against each other before and during monitoring. Introduce a
system for verifying the data (perhaps by using a person unconnected with the study).

2. With careful design it is possible to avoid the problem by confining comparisons to
results that have the same bias.

3. It may be possible, though difficult, to measure the bias. Measuring bias can be done
only if the true value can be occasionally ascertained, which is normally unachievable.
A separate experiment may be helpful - for example, one could compare the results
obtained by different observers measuring the same population.

If the bias adversely affects the monitoring and cannot be adequately measured, controlled
for or reduced, then an alternative method should be used. If it is not possible to find one
method that provides an apparently unbiased estimate, use a number of different methods
and compare the results, or change the objectives to match what is achievable.

4.7.6 Methods for measuring habitat and species attributes

The recommended uses and advantages and disadvantages of some methods for measuring
attributes of habitats, vegetation and plants, and animals are outlined in Tables 4.2 — 4.4
below. See the references listed in Box 4.1 for further guidance on the use of the methods
and how to carry them out in practice.

54

Protected Area Monitoring Guidelines

Table 4.2

Method and
main uses

Satellite remote
sensing

Measurement
of habitat
extent & major
changes in
composition

Aerial
photography
remote sensing

Measurement
of habitat
extent and
broad changes
in composition

Fixed point
photography

Records broad
changes in
habitat
structure

Protected Area Monitoring Guidelines. A OCS

Advantages and disadvantages of different habitat measurement

methods

Advantages

A large area can be covered from a
single data source. Repeated samples
can be expected for the same area for
the lifetime of the satellite.

Satellite derived maps can be used to help
design a stratified sampling programme.

Data can be used to identify landscape
scale changes.

Archived data may be accessed to
provide a historical baseline prior to the
initiation of the monitoring programme.

Very useful for mountainous areas such
as the ACA where the terrain can limit
field surveys.

Provides a relatively quick assessment
of extent of broad habitat types and
broad changes in these.

Historical trends can be examined using
past photographs.

Allows quicker and more accurate
mapping than by ground survey.

Records a wide range of attributes of a
habitat and it is not necessary to
anticipate the changes that are likely to
take place. Therefore this method may
detect unexpected changes which are
unmeasured by other methods.

Quick and simple, and provides a visual
picture of change with time.

Better than aerial photographs for steep
slopes.

Disadvantages

The lifetime of the satellite may be
more limited than the lifetime of the
monitoring project, leading to potential
difficulties in comparing outputs from
original and replacement sensors.

Satellite scenes are likely to be
incomplete when clouds are present.

Error rates in habitat classification may
be unacceptable.

Range of expertise required is
broadened from ecology into GIS and
remote sensing

Good photographs are required for
accurate analysis.

Habitats cannot be classified in as much
detail as with ground surveys.

Some habitats can be hard to
distinguish on photographs,
necessitating field checking of results.

Area will be underestimated for slopes
unless three co-ordinates are used to
digitise maps. High altitude areas will
be overestimated relative to low ones.

Atmospheric/geometric corrections may
be more complex than with satellite
monitoring, as the aircraft is not in a
stable orbit.

Generally only gives broad indications
of change, which cannot easily be
quantified or tested by objective
statistical methods.

55

Table 4.3.

Advantages and disadvantages of different methods for

measuring vegetation and plant attributes

Method and
main uses

Quadrats

Measurement of
cover, density,
biomass and
frequency of
small shrubs,
herbs, grasses,
and lower
plants, and
associated
variables, e.g.
height.

Line transects

Measurement of
cover and
frequency of
trees, shrubs,
herbs and
grasses, and
associated
variables, e.g.
height.

Plot-less
techniques

Measurement of
density of tree
density in
forests, and
associated
variables, e.g.
mean girth.

56

Advantages

A very widely used and documented
method.

Cover values provide good descriptions
of the contribution that each species
makes to vegetation communities.

Frequency estimates are easier to
collect and more reliable than cover
values, but require more samples and
are affected by quadrat size.

Often simpler to use than quadrats,
especially in sparse or tall vegetation.
Easier to search thoroughly than
quadrats of the same area and quicker to
record.

Line point transects are useful for
measuring changes in total vegetation
cover, but accuracy depends on the
length of line and number of points
used per line.

Generally much faster method to
employ than quadrats or transects.

Little equipment is required.

Disadvantages

Recording all species present can be
time-consuming.

Inaccuracies in species identification
occur for difficult species.

Estimation of cover can vary
significantly between recorders.

Cover scales (e.g. Domin) are non-
linear and therefore values, even from
randomly located quadrats, can only be
analysed using less powerful non-
parametric tests.

Transects are often not suitable for
measuring cover of individual species
where plants are closely intermingled
and vegetation type boundaries are not
distinct.

Long transect lines produce under-
estimates of species cover when points
are widely spaced. However, estimates
of total cover are unaffected by length
of line

If species are at very low density it may
take a long time to locate the nearest
individual.

When surveying areas with high species
diversity, the time taken when
measuring separate species will
increase.

The method contains inherent bias due
to the non-random selection of trees.

Protected Area Monitoring Guidelines

Table 4.4.

Advantages and disadvantages of different methods for

measuring animal population attributes

Method and
main uses

Recording
incidental
observations

Establishment
of presence of
species, crude
estimation of
range if
sufficient
records.

Timed searches

Establishment
of presence of
species in a
sample area,
and estimation
of relative
abundance.

Point counts

Measurement of
relative
abundance (or
density) of
highly visible

or vocal species
(usually birds).

Transects

Measurement of
relative
abundance (or
density) of a
variety of birds,
mammals and
other animals.

Advantages

Collects records on rare or rarely
encountered species that cannot in
practice be monitored by other means.

Simple method which can involve all
staff and local communities.

Relatively simple efficient method.

Useful for scarce species or species that
are difficult to detect.

More practical than transects in difficult
terrain.

More practical than transects in areas
where vegetation is dense.

Less prone to causing disturbance than
transects.

Easier to locate than transects.

Efficiency can be increased by multi-
stage sampling.

If observation distances are recorded
Distance* analysis can be used to
estimate densities.

Relatively efficient compared to point-
counts especially where birds are at low
densities.

If observation distances are recorded
Distance analysis can be used to
estimate densities.

Good for open habitats.

Disadvantages

Observation and recording effort varies,
and therefore data cannot be easily
analysed for trends etc, although staff
activities data can be assessed in
relation to effort if properly recorded.

Reliability of records variable and
difficult to assess.

Variable biases.

Normaliy only crude indices of
abundance.

Likely to be biased according to
observer ability and effort.

Detection biases cannot be easily
controlled for.

Less efficient than transects,
particularly for less detectable species.

More observations are by song / call
than by transects and therefore
considerable experience is required.

Distance analysis more prone to bias if
used than transects.

Often difficult to randomly allocate
routes.

Difficult in dense vegetation and steep
terrain.

Movement may scare shy species.
Strip transect densities may be
unreliable.

5 See Buckland et al. (2001) for details of Distance sampling techniques, but these are normally too
complicated and time consuming for routine monitoring purposes.

UEROTECIE MALE GI COTILONIIIC GLELA CH LCS ET aN

57

4.8 Establish the appropriate time to carry out surveys

The appropriate time for carrying out monitoring will vary according to the feature and
attribute being measured. For example, most breeding bird monitoring is best carried out
early in the breeding season when singing is at its greatest, whilst Snow Leopard surveys
are best carried from January to April when they are mating and marking their home ranges
most intensively.

The time of day is also often a vitally important factor to take into account, for example
pheasants generally only call for a short period around dawn, and therefore surveys carried
out an hour or so later will be ineffective. In contrast, surveys of soaring vultures must be
carried out later in the day when the sun is up and thermals have been created. Knowing
the activity patterns of your study animals is therefore important, and it is essential that if
there are strong diurnal activity patterns that survey times are standardised with respect to
these.

It is particularly important that repeat surveys in subsequent years are carried out at the
same time of year each year, unless seasonal cycles are being investigated, and at the same
time of day. Serious bias may occur if surveys are carried out at different times.

4.9 Devise a sampling scheme

4.9.1 Decide if sampling is needed

In some situations it may be possible to make a complete assessment of the whole feature
within your target area. For example it may be possible to reliably measure the full extent
of a habitat feature by aerial photography. Or it may be possible to carry out a complete
census of a rare species by counting all individuals if the species is easily detectable and
highly localised (e.g. vultures breeding at cliff nesting sites). However, care should be
taken, as you may be mistaken in believing that you have detected all occurrences of the
species. This is particularly important when dealing with mobile species, because a decline
in your monitored population may simply be due to their relocation to another site that you
are unaware of. Thus a decline in a vulture breeding colony could be merely because they
have moved to a new nesting site.

In practice, it is seldom possible, or even necessary, to establish the total population size of
a species. Unless species are very rare, very conspicuous, and very localised, total counts
will probably be too time-consuming and will produce biased results.

Generally, it is most efficient to assess samples of the feature and to extrapolate from the
observations made in each sample to the whole feature (or that part covered by the sampling
area). For the inferences that one draws about the whole to be valid, sampling must follow
certain principles:

= Samples must be representative of the site.

= More than one sampling unit per habitat is required. This is known as replication.

Sampling enables the estimation of an attribute’s value for a whole site, and also estimates
the inherent uncertainty in this value due to having only studied part of the site

58

Protected Area Monitoring Guidelines

(or population). For example, the area of a monitoring unit (e.g. a hill) covered by a
Rhododendron may be estimated by calculating the mean area of Rhododendron in a sample
of 10 m x 10 m plots and multiplying this figure by the size of the site in square metres.
The uncertainty in this estimate can be measured by the standard deviation of the estimate,
or by confidence intervals (see Elzinga 2001) or other references listed in Box 4.1 for
details, and Section 4.15 for advice on statistical analysis).

When designing a new monitoring programme it is advisable to design your sampling
strategy to your specific needs. The design of a sampling strategy is a particularly important
stage in the development of an effective monitoring programme and should, therefore, be
carried out carefully according to the key steps summarised in Figure 4.4. Sampling should
be designed for each feature’s attribute, taking into account the method being used to
measure it, the inherent properties (e.g. relative density) and variability of the attribute (if
known), the required precision of measurement, and the time and costs of sampling. The
design of a sampling strategy will also need to take into account the size of the sampling
unit being used (e.g. a 2 m by 2 m quadrat). This will in turn depend on the species or
habitat being sampled, the type of measurements being made and the method used for
sampling.

Once the required sampling has been identified for each feature and attribute, then these
can be combined to create an overall sampling programme, including combined data
collection where appropriate. For example one might use the same quadrat samples to
collect information on bare soil, vegetation height, vegetation species richness and the
presence of selected species.

Plate 4.3 Himalayan Tahr in the Sagarmatha National Park, Nepal

Weis)

oa 5

Photo: Som Ale

59

Protected Area Monitoring Guidelines

Figure 4.4 Designing a sampling scheme

For each
feature and
attribute to be
monitored

Carry outa
preliminary
survey

Are the features well known, are data
available on mean values and variation
in the attribute?

Decide if sample locations
should be permanent or not
Is it feasible to take more than
Use 30 samples and is distribution Locate samples
random information required? systematically

sampling

Are there known or likely variations in the
attribute across the site? Use stratified
sampling

Will travel time between
sample units be high? Consider cluster or
multi-stage sampling

Calculate minimum sample
sizes required for detecting
important changes

Calculate cost of sampling
Is the sampling programme feasible given
other monitoring requirements? Reconsider sampling
strategy
Obtain more
Document methods and sampling resources
strategy and train personnel

Protected Area Monitoring Guidelines

60

4.9.2 Decide if sample locations should be permanent or temporary?

Advantages of permanent plots

Permanent sample locations can provide a good approach for reducing variability when
temporal changes are to be monitored. Therefore, they increase the statistical power of the
monitoring, which means that fewer samples are needed to obtain a desired level of precision
and hence detect an important change. If permanent plots tend to change in similar ways,
then any changes documented are more likely to be real than due to random variation
between samples. For example, if mean species richness over 20 temporary randomly located
quadrats is 20 in one year and 10 in a subsequent survey, this may not be due to a real
change in overall species richness, but could instead be caused by the chance location of
quadrats in the first survey in richer parts of the site. However, if the repeated observations
were made at the same locations and the locations are representative of the site, we can be
more certain that species richness is declining on the site as a whole.

The usefulness of permanent plots varies depending on the degree of correlation between
two successive measurements. Permanent sampling will be most advantageous for
monitoring when there is a high degree of correlation between sampling-unit values between
two periods. This is most likely to occur with long-lived plants (e.g. trees, shrubs, some
perennials and lichens) and large territorial mammals, and least likely with erratic, transient
or mobile populations such as some annual plants, small mammals and insects.

See Elzinga (2001) or other sources of statistical information listed in Tables 4.7 and 4.8
for advice on analysis of permanent plot data.

Disadvantages of permanent plots

There are some significant disadvantages with permanent plots that should be considered.
Most importantly, marking and relocating permanent sample locations can be difficult and
time-consuming. This may offset any advantage from additional precision if observations
from non-permanent samples can be obtained much more quickly.

Repeatedly surveying the same locations may alter or damage the attribute being monitored
or its surroundings, e.g. by trampling. Apart from the potential unacceptability of such
damage, this may cause the samples to become unrepresentative of the site as a whole.
However, this is more likely to be a problem for intensive scientific studies where frequent
measurements are necessary, rather than routine protected area monitoring.

If the use of permanent sampling results in very few samples being taken, then additional
practical problems may result. If there are only a few plots, then these may become
unrepresentative of the whole study area (assuming that they were representative initially)
as a result of chance events with a different effect on the plot to that on the rest of the area.
Such events may also have permanent or long-lasting effects, as successive changes at one
point tend to be correlated. Therefore, any recorded changes will not reflect the true pattern
of change over the area. This difficulty (termed autocorrelation) can be overcome by
recording a second set of samples at the end of the first monitoring period, which are used
to estimate changes in the second period and so on. Samples A are enumerated on the first
survey occasion, samples A and B on the second, samples B and C on the third and so on.

ol

Protected Area Monitoring Guidelines

Permanent sample locations may also be effectively lost due to unforeseeable events such
as permanent or long-term flooding of part of the site, or the growth of trees over long time
periods. Human encroachment may also lead to loss of samples, particularly in developing
countries with expanding populations and agricultural landuse. For example this may lead
to the loss of forest plots as forests are cleared for housing or slash and burn cultivation.
This problem can be alleviated by recording ‘spare’ samples, though this may also reduce
the advantage of the approach compared with temporary sampling.

4.9.3 Choose an appropriate means of distributing samples

According to Elzinga et al. (2001), there are three requirements that must be met with
respect to the positioning of sampling units in the sample population:

1. Some type of random, unbiased sampling method must be used;

2. The sampling units must be positioned to achieve good interspersion throughout the
populations; and

3. The sampling units must be independent of each other.

Of particular importance is the random selection of sampling units. If this is not done, then
you cannot make any statistical inferences from your results. Selection of samples by
judgement (or preferential sampling) should be strictly avoided.

The advantages and disadvantages of some means of sampling are described below in
Table 4.5 and illustrations of different sample designs are given in Figure 4.5.

Table 4.5. Summary of the advantages and disadvantages of different

approaches to sample distribution

Method and main uses Disadvantages

Simple random sampling

Requires minimum knowledge of a population
in advance.

Easy to analyse data and compute errors.

Selecting sample units is quicker and easier than
for other random designs.

Systematic sampling (regular)

If the population or attribute is ordered with
respect to some pertinent variable such as an
altitudinal or moisture gradient, there is a
stratification effect, which reduces variability
compared with random sampling.

Calculation of sample positions and location in
the field is relatively easy and efficient.

Collection of sample observations can be time-
consuming.

Often larger errors for a given sample size than
with systematic sampling.

Estimates will be less precise on heterogeneous
sites than with stratified sampling.

Travel time between sample units can be high.

If sampling interval is correlated with a periodic
feature in the habitat, bias may be introduced.

Strictly speaking, statistical tests are not valid,

though in practice conclusions are unlikely to be
affected if there are more than 30 potential sample

(Cont.)

62 —_— ee _ Protected Area Monitoring Guidelines

Table 4.5.

Summary of the advantages and disadvantages of different

approaches to sample distribution (cont.)

Provides an efficient means of mapping
distribution and calculating abundance at the same
time.

Stratified random

Ensures that all the main habitat types present on
a site will be sampled (if these are used to define
strata).

Characteristics of each stratum can be measured
and comparisons between them can be made.

Greater precision is obtained for each stratum and
for overall mean estimates if strata are
homogeneous.

Restricted random sampling

Like systematic sampling it results in good
interspersion of samples, but is more robust when
the number of possible samples is few.

Like stratified random sampling in that the plot
is subdivided (but in this case with arbitrary
divisions and 1 sample in each). Results in more
precise estimates than simple random sampling.

Multi-level (plots within plots)

Can reduce sampling times thus increasing
efficiency.

Can avoid loss of precision caused by combining
individual sampling units which occurs with
cluster sampling.

Useful for sites which are heterogeneous at small
spatial scales

locations. (e.g. a 1m? sample could be located in
36 locations in a sampling grid of 6m x 6m).

If strata have not been identified prior to
monitoring, preparation can be time-consuming.

The most appropriate stratification for a site at one
time may have changed when repeat surveys are
carried out. Monitoring efficiency may therefore
also change.

Sampling units can occur side by side (like other
forms of random sampling), and therefore
systematic sampling is normally better if the
potential number of systematic samples is large
enough (i.e. more than 30).

Larger errors are obtained if secondary sample
plots within each primary plot are highly
correlated.

LSB) CHT LA EON EET Dp LECT O ) ———eeeee 63

Figure 4.5 Different types of sampling strategy

(a) Random sampling (b) Stratified random sampling

(a) Random sampling: samples taken randomly from the whole sampling area (b) Stratified
random sampling: sample area divided into strata and random samples taken in each stratum
(c) Systematic sampling: samples taken at regular intervals (d) Restricted random sampling:
one sample randomly located in each arbitorily defined sub-section of the sampling area
and (e) Mutli-level sampling: major units (large squares) chosen randomly and minor
units (x) sampled randomly from each, major units may also be trasects.

64 J — (ProlecteddArca Moniorns Guuelines

4.9.4 Estimate the number of samples that will be required to reliably
establish if objectives are being met

A key principle of sampling is that with increased sample size our uncertainty decreases
regarding how closely the estimated population value reflects the true population value.
Thus, we would expect that as more samples are taken the closer the estimated mean will
be to the true value. Unfortunately, the greater the sample size the greater the amount of
survey time required. Additional time is required to take the measurement at each sample
location and to move between sample locations, and the latter time may be particularly
substantial in large protected areas with difficult terrain (as in the ACA). Furthermore our
precision (i.e. the closeness of the sample measurements to each other) in estimating the
mean only increases slowly once we go beyond a few samples. Typically, precision increases
only in proportion to the square root of the sample size. Hence, to double the precision
obtained from ten sample units requires another 30 units.

So a balance is needed between limiting time in sampling and ensuring the estimate is
adequate. But defining what is ‘adequate’ sampling is not easy because it depends on the
reliability of information required, which in turn depends on the importance, objectives
and condition of the feature being monitored. For example, if a key feature is very important
and its population is believed to be close to the limit of what is viable, then a good, or
precise, population estimate is likely to be required. In other situations, only a quick check
may be needed to confirm that an objective is being met.

Because increasing the number of replicated samples increases the cost of monitoring, it is
very useful to carry out pilot surveys. These pilots aim to assess the distribution and
abundance of the species or habitat attributes being monitored, so that the amount of variation
in each can be approximately calculated. This can help in designing the sampling system
and establishing the number of samples required to achieve a desired level of precision or
to detect a given level of change.

In general, measurements should be taken from at least five plots before any generalisations
can be made about a population or habitat within the sample area. Even this low level of
replication can improve the confidence with which the results can be regarded. For frequency
data, it has been shown that with less than 50 samples, only very large changes are likely to
be statistically significant, and 100 samples has been suggested as a minimum. Bonham
(1989) suggests that for most purposes 25 quadrats randomly and temporarily located on
25 randomly and temporarily located transects should give satisfactory results within a
homogenous plant community.

There are now sophisticated means of establishing the number of samples required to
detect desired changes, using Power Analysis. This is beyond the scope of this guidance
but further information on these approaches can be found in Thomas & Krebs (1997),
though more up to date information can be found on the Internet.

4.10 Devise data recording forms and document methods

4.10.1 Design field data recording forms

Once the method and sampling design has been chosen, field data sheets should be designed
and tested. Specially designed forms encourage consistency and reduce unnecessary writing.

65

Protected Area Monitoring Guidelines

Forms are easy to read and help ensure that all necessary data are collected and
not forgotten. See Elzinga ef al. (2001) for suggestions for standard information to
include in forms.

It is vital that all relevant sections of survey forms are completed at the time of the survey,
and checked immediately after. Do not leave form filling for later, as information may be
forgotten or entries in field notebooks misinterpreted.

Where lots of data are being recorded relatively quickly, it may be advantageous to type
the data directly into a hand-held data-logger, which may include a GPS, thus providing
accurate spatial reference data for each record. However, such data loggers can be heavy
and expensive. If a data logger is used, a database structure should be written which
prompts the observer to enter the appropriate record. The advantage of this method 1s that
a large data set can be downloaded directly to a computer via a cable.

4.10.2 Document field methods as monitoring protocols

It is essential that the monitoring methods are constant between surveys. Therefore, before
the first survey is carried out, a monitoring protocol should be written describing
in detail the methods to be used, so that everyone understands what is required, and
the methods are kept consistent between observers. A monitoring protocol should be
prepared for each feature to be monitored and should document each of the issues
as listed in Box 4.3.

Example protocols for remote sensing of habitat extent, forest quality, forest
bird assemblages, vultures and Snow Leopards are provided in section 8.2 — 8.6
of these guidelines.

Box 4.3. Format and headings for a monitoring protocol

Protocol author, version and date
Monitoring objectives

Reasons for monitoring:
Users of the monitoring data / conclusions:
Conservation objectives for the key feature:
Location of the feature, monitoring population / area and sub-units:
Frequency of measurement:
Measurement Method
Observation / data types:
Method:
Timing of observations:
Potential causes of bias and rules for standardization:
Sampling scheme
@ Complete census or sample survey:
@ Temporary or permanent sample location:

@ Method for sample location: (cont.)

66

Protected Area Monitoring Guidelines

Box 4.3. Format and headings for a monitoring protocol (cont.)

@ Number of samples:
Monitoring requirements & organisation
Personnel responsible and time required:
@ Experience / training necessary:
Licence and access permission requirements:
Equipment required:
Data recording and storage
Data analysis procedures
Reporting format and procedures:
Costs: capital (equipment) and annual recurrent (including staff time and travel etc)
Health and safety

g@ Any particular risks with carrying out the fieldwork, and requirements for any special equipment 01
measures to be taken to reduce risks.

References

Fieldwork can be dangerous and so before carrying out any such work a risk assessment
should be carried out. Careful consideration must then be given to identifying safety
precautions that reduce any identified risks to acceptable levels. All identified safety
precautions should then be strictly followed.

Each protocol should then be followed as closely as possible in all subsequent surveys.
However, if deviations from the protocol are necessary, then these should be recorded.
Monitoring reports should ensure that the methods are written out in full and the original
monitoring protocol placed in an appendix. Deviations from the protocol should be reported
and the implications for the results and interpretation of the monitoring discussed.

Even when monitoring methods are very carefully documented, surveys should be repeated
by the same observers as much as possible. There is no doubt that the accuracy of
interpretation is considerably enhanced when one observer repeats the surveys over a long
period of time. If this can’t be done, then another option is to use many observers and to
randomly allocate sampling to them, so that systematic biases are avoided.

4.11 Test methods

It is highly recommended that you test your proposed monitoring methods and sampling
strategy once you have drafted your protocols. Data from such pilot surveys enable observers
to become familiar with the practicalities of using the method in relation to the terrain, the
physical structure of the habitat and the behaviour of the study species. It may also provide
an initial assessment of how close biodiversity features are to their conservation objectives
(which may influence the effort needed to monitor them fully) and an estimate of the
degree of variation present in each feature’s attributes. This information is invaluable for
finalising the optimal type of sampling, the distribution of samples, the required level of
precision and the number of samples required to achieve this. Monitoring protocols
should then be revised according to the findings and any methodological changes
that need to be made.

Protected Area Monitoring Guidelines). —————. 67

Testing of monitoring methods should also include trials of the statistical analysis of the
data (see Section 4.15 for selection of methods). These trials would ideally be with data
from pilot surveys, but even the use of invented data is better than no testing of the statistical
methods selected. The statistical method for analysing the data from monitoring should
always be determined before the data is collected, as this will influence the survey design.
If surveys are to be carried out by different observers then it is important to check the
repeatability of the method where feasible. This can be tested by having one observer
repeat a survey immediately after another observer, or by the same observer conducting
duplicate counts. The results of such tests may indicate differences in the ability of surveyors,
such as in species identification, which might then be overcome by training, etc. The results
of method assessments can also be incorporated into statistical tests. Confidence limits
and variance can be calculated on the variation in total counts or mean values in order to
separate variation caused by observer bias from all other variation. If major discrepancies
are found between two of these calibrating surveys, the underlying cause should be identified
and corrected if possible.

Although pilot trials may be time-consuming, they save time and resources in the long
term, especially in a large area such as the ACA where monitoring is likely to require
considerable travel between sites. In such cases, sampling needs to be as efficient as possible.

4.12 Review the monitoring programme in relation to
available long-term resources

Once all the monitoring requirements have been identified, and methods and sampling
approaches devised to meet them, then the whole cost of the programme and staff time
requirements should be assessed. The assessment should take a long-term view of the
requirements for monitoring and available resources, including likely year-to-year variations
in monitoring needs and budgets. A poor monitoring design is one in which the monitoring
effort changes from year to year, or monitoring is dropped because of a lack of resources.
It is a common mistake for new monitoring programmes to be too ambitious.

It is therefore vital that the overall long-term requirements for all the proposed monitoring
is reviewed in relation to available resources (including funding, equipment, staff time,
staff expertise and the capacity for local participation) before finalising the programme
and preparing a detailed monitoring plan. The review must take into account training needs
for staff and other personnel (e.g. contracted surveyors or local participants). As a minimum,
it is essential that all monitoring personnel are familiar with the habitat, study species and
survey methods required. The correct identification of target species may require specialist
personnel even if the methods themselves are straightforward. If the monitoring involves
several people, they should all be trained to a minimum standard and recording techniques
should be standardised, e.g. as part of a pilot study.

A careful evaluation of the equipment needed should also be undertaken during the review
of the monitoring programme. All equipment needed for the monitoring programme should
be available for its duration. If equipment is to be purchased, especially if this is expensive,
detailed advice on its suitability for the monitoring plan should first be obtained. If it is
found that the resources needed for a full monitoring programme exceed those available,
then it will be necessary to seek more funds, and/or trim the monitoring programme in the

68 SS Protected Area Monitoring Guidelines

least damaging way, e.g. by targeting monitoring as suggested in Box 4.2. However, it
should be remembered that cutting back on monitoring may be a false economy, as
monitoring may enable early management intervention which avoids very costly damage.

4.13 Prepare a work plan

4.13.1 Agree responsibilities for the monitoring programme

A complex monitoring programme requires careful coordination, to allow enable integrated
data analysis and reporting. For example, key decisions need to be made on:

m who enters the data and is responsible for quality control

who will manage the data

who holds mastercopies — how is this related to versions of the database elsewhere?
who has access to the final information

who will analyse the data

who will report on the results of the monitoring to users and managers

who owns the data

@ who has rights of use
It is therefore recommended that a monitoring manager / coordinator is appointed who has
overall responsibility for deciding on these issues and coordinating the implementation of
the overall programme. Key interactions between the coordinator and other members of
staff, other organisations and stakeholders should also be identified, together with
information flows. This can be usefully documented as a monitoring programme
coordination system (as for example indicated in Figure 4.6).

irs c
RADA B O GIMALLDS:. 69

Figure 4.6. The coordination and data flow system for ACAP biodiversity
monitoring

Local people

Village Development
Committee

Conservation Area
Management
Committees

(CAMCs) Visitors

Rangers,
KMTNC-ACAP Conservation

fee Officers

Scientists (Univ.
Govmt, foreign)

Scientific

Monitoring
Coordinators

KMTNC-ACAP

community MONITORING
COORDINATOR

a Remote
District and national KMTNC-ACAP sensing data
Government HQ staff providers

agencies

International
conservation
monitoring agencies

Key: Arrows indicate directions of information flow. Dashed lines indicate coordination guidance, analysed
data and reports. Solid lines indicate primary data. ACAP staff are indicated in bold type.

70 —_—. $——. —————$ ————— —— Ptec ted Area Monitoring Guidelines

4.13.2 Organise data collation and management

Where feasible and appropriate monitoring data should be analysed and reported locally to
increase local ownership of the monitoring programme. However, it will normally also be
necessary to collate data for larger scale analysis and reporting, including for the protected
area as a whole. Coordination will therefore be required to transfer data from the field, via
a local office to a central database. Key coordination issues to be addressed will include:

H how will the data be transferred?

M whois responsible for data collation and transfer?
@ in what format should the data be sent?
|

what time schedule will be used, etc.?

Data management procedures should be documented and clearly communicated with
everybody involved, making sure that responsibilities are clear.

When data arrive for central processing and storage there should be aa initial quality check
before any analysis is undertaken. This should look for common errors such as:

H missing data (1.e. gaps on data recording forms)
M recording errors (e.g. impossible or unlikely observations)

@ unreadable data entries

Data locations should also be checked by plotting the stated coordinates of the sampling
sites on a map (e.g. using a GIS), so that incorrect coordinates might be apparent from a
visual inspection of this map. Sometimes mistakes can best be detected by an expert,
assessing the plausibility of results. Any detected errors and inconsistencies should be
queried immediately with the original data providers (preferably the field workers
themselves), who should keep a copy of any revised data themselves. Such quality control
procedures will be the basis for ensuring scientifically sound and reliable outcomes of the
monitoring programme.

Requirements for data storage should be addressed when the design of field forms and the
design of the central database are being done. This will allow for an easy transfer from
data recorded on paper to any electronic system, and help prevent mistakes. The use of
standardized and widely recognised recording forms and coding systems (see Section 5.4)
will greatly assist sharing data with others.

4.13.3 Develop and agree a monitoring workplan

A monitoring workplan should be developed, which summarises all the actions in the
monitoring programme. There may be more than one monitoring action for each feature
and attribute being monitored. For example, two methods to monitor wood collection for
fuel are specified in Table 4.6. The work plan lists for each action its conservation or
management objective, where the monitoring is to be undertaken, when, its costs, who is
to carry it out, and what protocol is to be used (see Box 4.3). A section of a hypothetical
workplan for ACAP is illustrated in Table 4.6.

c - a a er —————— 7
Protected Area Monitoring Guidelines l

SOJJO PAI 103995
dVOV-ONLNY

SOJO PJalf 10}99g

dVOV-ONLINY

SA1JJO PJOL4 10399
dVOV-ONLINY

S2O1JJO PJaty 10199
dVOV-OINLNY

SIO JO
UOTJPRAIOSUOD
aouredy / 10}9e..U0D

S2YFO PJAl4 10}99S
dVOV-ON.LIO

dVOV-ONLNY
“UUed} STD

Ayijigqisuodsay

SOCA JO a[duies

SOCA IV

S}SO10J
asn dATSUOIUT

ae.

SBOIB JSOIOF Koy 9

sevaie yso10y Kay 9

VOV IIV

a

sieak dai} AIOAg

jenuuy

sea
¢ AIDAd UDUI “OQOT

owiueirs01d
Surp[OI 9K F “HOOT

saeak ¢ AIOAq

sivok
¢ AlaAa Udy} ‘SOOT

sieak
¢ Alana uay) S007

Buloyiuow
jo Aouanbaly
pue Buiwiy

saaoalgo afdiyjnuu 07 aye[a1 pynom sway! SulsoyuOW Aueu asiyoeid uy ,

(6) SMOIAIO}UT
(QIN) Sese][IA SuLIo}Ua

spro] JO syunoos ajdures

(SIA) SALOUDAUT 4Sa104

(ZIW)
sjoasue.y JUDURULIOg

(Cd)
syoo]q jusuvULIOd

UI S}UNOD sa1dads SUT |,
(ZdIW) S}0]d JuoueuLIeg

(Td) sisAyeue
SID / Rep [98S

jo20}01d
pue poujsi

uejd y10M Bullojiuow e jo ped ajdwexea uy “9p ajqeL

(79°¢

u01}99S

das)

9S9Al}Ia[qo
pej}ejay

[ery
JO} UOTIA][OI Poop

jong
JO} UOTIATJOI POOA\

auoz asN dATSUOzUT
UI SSBUIOIG 1S910.]

asurl
piedoay] Mous
ssouyoll saroads

P1Iq 10JeOIpUT }sd104

ssouyou
satoads oar],

}SO10} Povo]
-peoig [je JO wWd}XY

s]uswainseaij|

Protected Area Monitoring Guidelines

N
~

4.14 Carry out necessary training

The establishment and maintenance of a biodiversity monitoring programme requires a
range of skills and so training is likely to be necessary. All training activities should be
based on an assessment of individual needs and build on the existing experience of staff
and collaborators.

The first step in the identification of training needs is to determine the staff, local community
groups and contractors involved in assessment and monitoring, from the collection and
processing of data through to its interpretation and use in management decisions (see Figures
3.1 and 4.1). Much of this information will have been defined in the process of defining the
monitoring programme and the preparation of a work plan and the monitoring protocols.
Figure 4.6 (The co-ordination and data flow system for ACAP biodiversity monitoring)
can be used as a guide for this step. For each actor involved in biodiversity monitoring, a
short description of their responsibilities and the skills required should be prepared. This
description of the ideal situation then provides a basis for an assessment of the current
skills that exist and the needs for training.

Training will require a combination of teaching basic concepts and practical exercises to
consolidate the learning of the participants. It is important to explain how field work and
data analysis will be used in decision-making. Once the monitoring programme has been
established many of the training requirements and procedures will be standardised
procedures, as defined by the monitoring protocols. Each training session should include
an anonymous feedback form, so as to continue improving the effectiveness of training.

For a protected area where management 1s in conjunction with local communities, such as
the ACA, training needs will include appropriate data collection and analysis with designated
community members. This training should be very practical, with practice of all the stages
from identifying monitoring sites and completing forms to the interpretation of the results.
The value of standardised procedures to allow the detection of change should be emphasised.
On-going support should be provided to participatory monitoring work, to overcome any
obstacles and maintain enthusiasm and standardised approaches.

4.15 Analyse data

4.15.1 Key steps in statistical analysis
Before starting an analysis, the steps are to:

1. Remind oneself of the question that the monitoring has been set up to answer (which
should be linked to establishing whether or not a specific objective is being met).

2. Assess the quality and completeness of the available data, and fill data gaps where
possible.

3. Choose an appropriate statistical method and package to conduct the analysis.

This section offers a quick guide in Table 4.7 to selecting appropriate significance tests, as
this is often the most challenging decision in data analysis.

Protected Area Monitoring Guidelines —&K$__$_$= $= <$ <$ $ $ $$ $ $< $< <$< $< $<$<$< <<< —— 2)

Table 4.7. Choice of tests for different types of hypothesis

Test for

Data normality for
single continuous
variables

Data of equal
variance

Linear correlation
between variables

Difference between
two unrelated
groups

Difference between
two related groups

Difference between
three or more
unrelated categories =
relationship between
continuous dependent
variable and one or
more categorical
independent variables

Time series

Date
description

Variable

Continuous
variable with
categories to
compare

Two
continuous
variables

Two
continuous
variables

Two
continuous
variables for
same cases

Binary
variable for
same cases

Continuous
variable with
nominal
categories

Continuous
variable with
ordered
categories

Continuous
variable with
samples
through time

Parametric | Non-parametric

alternative

Pearson’ s
correlation
coefficient

Unpaired t-
test

Paired t-test

t-test ANOVA
(one-way)

t-test ANOVA
(one-way )

alternative

2-tailed
Kolmogoroy-
Smirmov test

Levene statistics

Spearman’ s rho
coefficient (+1 to
=j1))

Mann-Whitney
U-test

Wilcoxon paired
signed-ranks test

McNemar’s test

Kruskal-Wallis
H-test ANOVA

Jonckheere-
Terpstra
ANOVA

Time series analysis (e.g. via

regression) — or see “Difference

between two related groups’ to test for
difference between two time periods)

Example use

Find out if a single
variable has a normal
distribution

Find out if a single
variable has equal variance

Find out whether spatial
variation in a species’
abundance is correlated
with an environmental
variable such as pH or
slope; or temporal
variation is correlated with
a change in a factor such
as grazing intensity

Compare species richness
between two sets of
samples collected at
different times

Compare relative
abundance in two sets of
permanently located
samples from two points
in time

Test significance of an
individual species being
present or not present in
two sets of permanently
located samples from two
points in time

Compare abundance at
three or more sites

Compare mortality or food
intake (continuous) with
age or size classes
(ordered)

Simulate future pattern
based on past history.

Protected Area Monitoring Guidelines

YUuUOI
(Hoa) UY OTR MOS

/oyeurp10/Auerj}oq/10s
ye/npaayeysyo MMM //:dy1Y “7 SOANLULID}L
juifti~ | SNOLIRA :saQ4 ‘7

few ayAinjuso‘awoy//:dyyy *] ON ‘I

dse*xaput/AjISIaAIporq/sao
INOSOI/YSY/Sas/3.10°91yUIdA
SO1OfOISepLIOAV MMM //:dyVY

[UHY UOoeUTXepUI/WOS" UO
BAJasUOS-sadsId- MM M//:dY

/Alasuogjyeu/qeiyeop/atdoad (posmbes [aoxq

/npaoson-ABojoIq’Mmm//:dyYy | yyosors1\) SOX

WY AODATP
/Aysjopod~/jouyye awio0y//:dyy

é aiqeyeae

d}S JOUII}U| UOISJ9A 9914

*So1jsiqeR}S A}ISIOAIC] “SOAINO
voie saisadg “(yOd) sasAjeur Jojsnjo “(NVdSNIM.L
‘VOd ‘VOO) sasdjeue dsdouapuodsaii0s ‘sul[eos
[BPUOISUSWIPH[NW d1IAW-UOU :suOT}eUIPIO snoLIe A

“OIBMIJOS UTeUOp d1]qnd Jo uoNeyIdiu09
[njasn AOA YW “UONLUIPIO “SuLIO}sNjO ‘stsAjeuR a9dUejSIP
‘SQOIPUL AJISIOAIP PUL SSOUYOII “UOISSAISA1 IOJ S}OOL,

ole)
‘suONNgLYSIp souepuNqe “uoNRUasoides Byep jeorydess
‘spoyjoul payejor AjISIOAIp ejOq pue eydye yo asuvi opim
B SIQJJO “SONSHRIS pue soInsvow AjIsIOAIP SuNe[noeD

(SOdINOSOI IDY}O, SV PAphyoUut osye qs]
QOUdIOJOI JUSTJoOXY ‘sjordun jroulaseuew Surpuvjssopun
pue sIsA[eue 0} YSnNoIYy) UsISOp WoIZ SulIO}IUOW
JO ssooo0id ot} UI JSISSe 0} popuazUT ‘[IOx| 10J s[OOT,

“‘surjdures
Joy 10J svore APVe.ys day 0} asp], ‘syods yoy AjIsIOATporg
Jo sdeut jno-jsuly sonpoid 0} asp—, “AVISIOAIPOIQ 105
OJLFOLINS IANIAFJO-1SO9 ‘[nyasn AISA & ST 19}[1f PEPISIOATG
OY} ‘AdUd}]] “APISIOAIPOIQ YSTY YM dJe[OI1I09 Uso svore
asay], “Ajauasoiaj9y YSIYy JO Svose aJeoIpul 0} se APA
B YONS Ul HI SJo}[y pure Aroseu yea suvos pepIsioAiq

SOATBUIO}L
pue GuO-Od

aseyord
sisAyeuy AjIsIOAIpolg

IIT ssouyory
pure AjIsIoAIq

satoads

Req UONeAIOSUOD
JO sisAjeuy pue
UBISAC] JO} S[OO],
SUIapOW /[RousHeIS

Pepmsioalc]

SoNSHeIS [PoIsO;ONq

uonrordde
pue usIsop SuLIO} UO

sisAjeue dseul]

Joo} yo adAy

Protected Area Monitoring Gutdeli1€0S —$<$<=<—<=_<_—<—$— — << —$— — _<—<——————————————————

eyep Hurojyiuow Huljasdiajzul 10 BuisAjeue pue ‘Hulsoyuow Huruueyd 10} jnjasn sjoo} aiemyos “g*p ajqeL

“UOTRIO}SA1

JO} SO}IS JO UOTVN]LAD pue “Sul[japoul souepunge soroods

unyspuradcyar ‘Suruuvyd y1OMjou dAr1AsaI apnyout soydiuexg ‘UoNeUoOyUr
jspwa/npa‘ysio"|sj-Mman//:d vy SuISSIUU JO sjoeduut ay} UO sIseYyduua ue sadv]q ‘soto (joo, woddng
‘unyxapur |HO'SI9eq WINLIAILID *g yuowaseueul JO siskjeue ‘ayv}s JO JUAUISsassy “oddns UOIsIoaq JUOWIASeUR
sspua/npa'ysio'|sp mavan//:dyyy_ | SIDAV sounbat) sax] UOrIstoap jusAseURU IOJ Wd}sAs paseq aspo;Mouy [eo1sojooq) SCWA

‘TSTUC] Ul gynpout ‘uiayed yeoriduua ayy

Protected Area Monitoring Guidelines

B se papnjout ‘ON “Z| Jo sIsAyeue [eosRIs UO paseg asuLY asn pur] oNjny

B10'sqepyejo- mavan//:duy °Z (adeyord] Jo uoNRoo] ay} saye[NuMIs pue ‘ommyny oy) OVW ayeI RY}

[WY powoas/sjapourorumeudp |NVYLUOA ® se wanbai] soyepodv.yxa “sorOJ-UOU 0} Paysoroy WOAZ UOISIOAUOD asn
qddgsi8mporapeA-yoreasary/:dyy | UO ageiear) (S94 ]} PUR] Jo ayes ayy souTULIa}ap a8vyoed poseq-uOIssossaI Sty | GOWOdD

[UY XoLIOA ‘dnoin jsije1oads suipacig uoleArasuod
/B10°6X91IOA'MMM//:d11Y ISS/NONI 4q pas “Surjepour Aqyiqera uoneijndog XALYOA

(ISTUCI 19 “SSVuD SID WWM
“MOIA OLY ‘OJu] sayeAdazuy “oJa ‘satoads Fo snjv}s Jeary] SuIssasse ‘AjI[IQeIA
suioo'seuer MMM //: dy -[ory saumnbar) ‘ON} uonendod surjjapour uo Sutsnsoj sjonpo.d jo zaquinu y SVNVU SUOISIOAp JUdLUASeUR |
‘A[UO UOISIOA UOTE ‘oRjiojut A[pudLy asueyo
/uoo'qeyturUr MM A\//:d))Y -suowlap Aep (¢}] -1asp) ‘s}stsojo1q Aq pasn Ajapim “jonpold yerosoWUWOD AVLININ | asn pur, suljjopow
“Ayyenuue
juioo'ssds' mm ar//:dyay [eMouar asuaol] soumnbay ‘aseyoed yeonsyeis pasn Aap, SSdS AqyiqeraA uonepndog

“MATA OLY YA o[qeviodo19jur ore

YOIYM SayNpoUur aos sey sn{q-S “UleWOp s1;qnd ay} UT sty
/310°}99 f01d-1 MMM //:dY *Z (QD) SAA °Z] IS[IYA Jonpo.d [erossuU109 v sI sn g-S ‘asensur] jeonuapl uone1ojdxe vjep
‘Il

woo |NAYsIsur MMM //:d AY (snjd-s) ON ‘[}] jsowye ue asn 0} padojaAap uaaq savy Y pue sn{g-S yY pure snfg-S | pue sonsieys [vorsse[D

\o
(u09) eyep Buroyiuow Bunaidsajzul so HuisAjeue pue ‘Hulsoytuow Huluueld 104 jnjasn sjoo} aieMYOS ‘g"p 9/GeL =

Classical statistics use parametric, non-parametric and exact tests to identify the probability
that a null hypothesis is correct (Table 4.7). This formal analysis of a question attaches a
level of significance (p-value) to the result. For example, a paired t-test may be used to test
the null hypothesis that there is no difference between samples taken from the same
permanent plot in different years. If there is a 95% probability that the null hypothesis is
rejected, we say that it is rejected at the 95% significance level and accept the alternative
hypothesis, that there has been a change through time.

Parametric tests require normality in the datasets tested - that is, when value on the x axis
is plotted against the number of data points on the y axis, the results should be a symmetrical
bell-shaped curve about the mean value. A two-tailed Kolmogorov-Smirnov test can be
used to assess the null hypothesis that the observed distribution function cannot be
distinguished from normal. The t-test ANOVA also assumes that each category comes
from an underlying symmetric distribution within groups of equal variance (a measure of
spread), which can be tested using Levene statistics.

Non-parametric alternatives with exact or Monte Carlo significance can be substituted if
these assumptions are not met. These tests make no assumptions about the population
distribution. An exact test takes the groups being compared and repeatedly shuffles the
values to calculate the test statistic for every possible combination of the observed numbers.
The test statistic for the dataset, a, is calculated as usual. The exact p-value is then the
frequency at which a occurs within the collection of statistics. That is, if it occurs less than
5% of the time, the null hypothesis is rejected at the 95% level. Due to computational
limitations the suggested maximum number of cases for exact tests is 30. When there are
more cases than this a Monte Carlo estimate of the exact significance level can be used.
This method uses repeated random generation of values within the observed margins to
calculate probabilities, rather than investigating every combination of the observed values
like the exact test.

Many of these tests can be carried out using software that is available via the internet.
Table 4.8 therefore lists some statistical resources that exist in the public domain, and
offers pointers to resources for other types of analysis, which may be helpful in planning
other monitoring or interpreting results. These range from simple tools for single analyses
through to management decision-making aids to help you make wider use of monitoring
results. They vary considerably in the level of expertise required for their use.

Further guidance on statistical analysis can be found in the references listed in Box 4.4. In
addition the ‘Statistical/Modeling Tools for Design and Analysis of Conservation Data’
and the ‘Biodiversity Analysis Package’ listed in Table 4.8 are worth a look for their clear
explanations of techniques and issues as well as their tools. If an internet site is not accessible,
the Wayback Engine at http://www.archive.org can be used to access archived versions of
most of these webpages.

yay

Protected Area Monitoring Guidelines

Box 4.4. Recommended sources of further information on statistical
analysis.

Ecological statistics:
Krebs, C.J. (1998). Ecological Methodology. 2" edition. Longman. Great help from
experimental design through to analysis.

Legendre, P., Fortin, M-J. (1989). Spatial pattern and ecological analysis. Vegetatio 80: 107-
138. Classic on pattern exploration and spatial autocorrelation.

Digby, P.G.N. & Kempton, R.A. (1987). Multivariate analysis of ecological communities.
Chapman & Hall, London, UK. Helpful chapter on data exploration.

Wilkinson, L. (1992). Graphical displays. Statistical Methods in Medical Research 1: 3-25.
On presenting information.

Hurlbert, S.H. (1984). Pseudoreplication and the design of ecological field experiments.
Ecological Monographs 54(2): 187-211. Argues for replication of plots to avoid this issue.

Statsoft (2005). Electronic Statistics Textbook. http://www.statsoft.com/textbook/stathome.html.
Includes guide to time series analysis.

Statistical packages:

PsychNet-UK (n.d.). Software Packages — Statistics. http://www.psychnet-uk.com/
experimental design/software_packages.htm. Accessed 14/4/04. - An excellent list of public
domain software, put together for behavioural scientists but as useful for ecologists.

Statlib. http://lib.stat.cmu.edu/. [Not available 14/4/04] - A more technical resource

Ordination and regression:

Palmer, M. (n.d.) Ordination Methods for Ecologists. http://www.okstate.edu/artsci/botany/
ordinate/. Accessed 13/4/04.

Ecological software:

Illinois Natural History Society (2004). Clearing house for Ecology Software. http://
nhsbig.inhs.uiuc.edu/. Accessed 14/4/04. - A few programs for wildlife ecology & statistics

Species distributions:
Anon (2004). Habitat modelling algorithms. http://www.conserveonline.org/2004/03/a/en/
habitat_modeling_algorithms.doc Accessed 14/4/04.

Buckland, S.T. & Elston, D.A. (1993). Empirical models for the spatial distribution of wildlife.
Journal of Applied Ecology 30: 478-495.

Fielding, J.H., Bell, A.F. (1997). A review of methods for the assessment of prediction errors
in conservation presence/absence models. Environmental Conservation 24(1): 38-49.

Pearce, J. & Ferrier, S. (2000). Evaluating the predictive performance of habitat models
developed using logistic regression. Ecological Modelling 133(3): 225 - 245. — using the
ROC curve for model thresholds.

78 ee Protected Area Monitoring Guidelines

4.15.2 The use of Geographical Information Systems (GIS) in analysis of
monitoring data

The first role of GIS is in data visualisation and exploration. The ability to overlay datasets
gives GIS a unique role in map design and exploratory analysis. Field data can be brought
together with remote sensing data or existing digitised maps. In a monitoring situation,
maps of species or habitat distributions may be compared to assess change through time,
and the area and locations of change calculated.

More advanced analytical functions are also available to an expert user, or GIS datasets
may be exported for analysis in statistical software. Most GIS packages will allow summary
Statistics to be calculated, will perform correlations between maps of different variables,
and may be programmed to carry out a range of modelling activities limited only by available
data, computational capacity and technical skill.

Species distributions

Species surveys based on a stratified or random sample do not by themselves create a
comprehensive distribution map. However, such field data may be interpolated using GIS
to fill in the gaps. A map of the probable species distribution can be created, by analysing
species observations together with maps of the key factors known to influence their
distribution. These may include climate, snow cover, altitude, soil type and vegetation
cover. A number of different techniques of varying complexity are employed by the tools
listed in Table 4.7. Some are fully integrated with a GIS, whilst others require the relevant
data to be extracted prior to analysis. One commonly-used technique is logistic regression,
which requires records of species presence and absence.

A set of such species maps can be compiled to allow the identification of areas likely to
have a high species richness. This exercise can be carried out for endemic or threatened
species to highlight areas of particular conservation interest. These estimates can be “ground-
truthed’ by field surveys.

Ecological indicators

As well as helping to interpret field data, GIS procedures can be used to derive new variables
from the vegetation datasets. Landscape ecology metrics can themselves serve as indicators
for protected area management objectives. For example, a conservation objective may be
to limit forest fragmentation, which can then be monitored for the reserve as a whole using
landscape metrics. Measures of different aspects of fragmentation, such as local forest
density, patch area and distance to core areas of forest can be monitored alone or combined
(Kapos ef al. 2000).

Scenarios and management planning

When a species distribution has been successfully modelled and mapped onto the current
environment of the reserve, it is possible to use the same model to evaluate the potential
long-term impacts of environmental change. For example, if a management plan involves
altering an area’s habitat type, the model can indicate which species one would expect to
lose from the area, and which might be expected to eventually colonise the new habitat
patch. These techniques have been frequently used to simulate the impacts of climate and
land use change.

ieee ————————— eee 7C
Protected Area Monitoring Guidelines )

For a more sophisticated analysis, population viability models can be applied within the
estimated range of the species, encompassing issues such as minimum habitat requirements
and population dynamics (Lindenmayer ef al. 1995). This requires much more information
about the species in question, but in return provides an estimate of whether the species has
a long term future in the areas currently occupied.

4.16 Report results to stakeholders

The results of monitoring need to be communicated to several different categories of
stakeholders, each of which will have different interests and abilities to interpret and use
the results. The manager of the biodiversity monitoring programme needs to identify the
stakeholders or users of the monitoring results and assess both their decision-making needs
and the most appropriate forms for communication of the information. For example, the
manager of the protected area will require written reports and maps, with an analysis of the
extent of progress in reaching the conservation objectives. The report should have a summary
and may include recommendations for management actions, based on the interpretation of
the results. Another key stakeholder group is local communities and their leaders within or
near to the protected area. The most appropriate means to present the monitoring results to
these groups may be in a much more graphical format, in the local language, and without
technical terms. Other important stakeholder groups include government agencies and
statutory bodies concerned with the protected area, donors and supporter groups, tourists,
and the news media. Whilst monitoring is a repeated activity the timing of the reporting of
results should be designed to meet the decision-making needs of the stakeholders.

When reporting monitoring results it is important to ensure credibility in the information.
This may be obtained by requesting a review of the results by other experienced and
technically-competent colleagues. Similarly, it is important to have documentation of the
methodologies and field results in an accessible form, such as the monitoring protocols
and database and GIS information management documentation. Communication of the
monitoring results should also be considered a two-way process, with the managers of the
protected area and its monitoring programme listening to feedback on both the interpretation
of the results and the ways in which they are presented.

4.17 Review the monitoring programme

It will normally take a couple of years to establish a monitoring programme. And even
once it is established, it might be necessary to further develop the programme, as original
ideas may not work out as expected or new questions may arise.

During all stages, it is important to continuously review the development of the monitoring
programme, by comparing it with its original or adapted objectives and plans. Such a
review is important to be able to take action when things are not developing as planned.
The review is also important in relation to the quality assurance of the scheme and its
outcomes. The review will show whether a scheme has been implemented according to
the agreed work plan and protocols, and hence whether scientifically valid results can be
derived from it.

80 —$—$—$—$—$—$—_—————— __ ————___ Protected Area Monitoring Guidelines

Issues that should be considered within the review include:

m was the intended number of sample plots achieved or is the number increasing
according to schedule?

was the intended frequency of visits to the plots achieved?
was the intended sampling strategy well applied?

did the planned field methods work?

were personnel sufficiently trained?

have methods been applied consistently?

were the data and analysis adequate to reliably establish if each objective was
achieved?

m were there unforeseen events that affected the monitoring or the achievement of the
objectives?

were the monitoring resources adequate, and are more required?

If the review shows that there are some problems in the development of the programme,
then constraints should be identified and solutions sought. However, in some cases the
original plans may turn out to be impractical and may therefore have to be modified.

But one should be aware of the impacts of changing the monitoring programme, as it
might severely influence the possibilities for analysing long-term trends. Every change of
the program e.g. in methods or sampling strategy, should be clearly registered and it’s
consequences should be analysed beforehand.

Protected Area Monitoring Guidelines: —<$<$<=<$<— << — ————— 81

References

Abbot, J. and Guit, I. 1998. Changing views on change: participatory approaches to
monitoring the environment. International Institute for Environment and Development,
London, UK.

Bibby, C.J., Burgess, N.D. & Hill, D.A. 1992. Bird Census Techniques. Academic Press,
London.

Bliss J., Aplet G., Hartzell C., Harwood P., Jahnige P., Kittredge D., Lewandowski S.,
Soscia M. L., Gray G. J. e., Enzer M. J. e. and Kusel J. 2001. Community-based ecosystem
monitoring. Journal of Sustainable Forestry 12: 143-167.

Bliss J., Aplet G., Hartzell C., Harwood P., Jahnige P., Kittredge D., Lewandowski S.,
Soscia M. L., Gray G. J. e., Enzer M. J. e. and Kusel J. 2001. Community-based ecosystem
monitoring. Journal of Sustainable-Forestry 12: 143-167.

Bonham, C.D. 1989. Measurements for Terrestrial Vegetation. John Wiley, New York.
Borrini-Feyerabend, G., M. T. Taghi Farvar, J. C. Nguinguiri, and V. A. Ndangang. 2000.
Co-management of natural resources: organising, negotiating and learning-by-doing. GTZ

and IUCN, Kasperak Verlag, Heidelberg, Germany.

Burrough, P. A. and McDonnell, R. (1998) Principles of Geographical Information Systems.
Oxford University Press.

Buckland, S.T., Anderson, D.R., Burnham, K.P., Laake, J.L ., Borchers, D.L. & Thomas L.
2001. Introduction to distance sampling: estimating abundance of biological populations.

Oxford University Press, Oxford.

Callicott J. B., Crowder L. B. and Mumford K. 1999. Current normative concepts in
conservation. Conservation Biology 13: 22-35.

CCW. 1996. A guide to the production of management plans for nature reserves and
protected areas. Countryside Council for Wales, Bangor.

Cochran, W.G. 1977. Sampling Techniques (3rd edn). John Wiley & Sons, Chichester.

Daily, C. G., editor. 1997. Nature’s services: societal dependence on natural ecosystems.
Island Press, Washington, D.C.

82 Protected Area Monitoring Guidelines

Danielsen, F., D. S. Balete, M. K. Poulsen, M. Enghoff, C. Nozawa, and A. E. Jensen.
2000. A simple system for monitoring biodiversity in developing countries. Biodiversity
and Conservation 9:1671-1705.

Danielsen, F., M. M. Mendoza, P. Alviola, D. S. Balete, M. Enghoff, M. K. Poulsen, and
A. E. Jensen. 2003. Biodiversity monitoring in developing countries: what are we trying to
acheive? Oryx 37: 407-409.

Elzinga, C. L., D. W. Salzer, J. W. Willoughby, and J. P. Gibbs. 2001. Monitoring plant and
animal populations. Blackwell Scientific Publications, Abingdon, UK.

Fabbro, L. 2000. Amazonia Biodiversity Estimation using Remote Sensing and Indigenous
Taxonomy. Associagdo Amazonia. Project presented at the European Space Agency
Symposium 2000, Gotheborg, Sweden, 16 - 21 October 2000. http://www.amazonia.org/
SustainableDevelopment/Jauaperi/Biodiversity/ALMA/ABDE/index.htm. Accessed
14/4/04.

Feinsinger, P. 2001. Designing field studies for biodiversity conservation. 212 pp. Island
Press, Washington D.C.

Fuller J. 1998. Participatory Monitoring of forest resources: Current methodologies being
developed in Thailand. Rural Development Forestry Network Paper 23e. ODI, London.

Goldsmith, B., editor. 1991. Monitoring for conservation and ecology. Chapman and Hall,
London.

Hill, D.A., Fasham, M., Tucker, G., Shewry, M. & Shaw, P. (Eds.) in press). The handbook
of biodiversity methods: Survey, evaluation and monitoring. Cambridge University Press.

Hirons, G., B. Goldsmith, and G. Thomas. 1995. Site management planning. in W. J.
Sutherland and D. A. Hill, editors. Managing habitats for conservation. Cambridge
University Press, Cambridge.

Hockings, M., S. Stolton, N. Dudley, and J. Parrish. 2001. The enhancing our heritage
toolkit - Book 2. The workbook on how to build assessment, monitoring and reporting

systems on the management effectiveness of World Heritage Sites. UNESCO / IUCN.

Inskipp, C. 1989. A popular guide to the birds and mammals of the Annapurna Conservation
Area. ACAP, Kathmandu. 54pp.

IUCN 2004. 2004 IUCN Red List of Threatened Species. <www.redlist.org>.

Jackson, R. 1990. Threatened wildlife, crop and livestock depredation and grazing in the
Makalu, working paper publication series. Report 12 . 75pp

83

Protected Area Monitoring Guidelines

Jackson, R. M., G. Ahlborn. M. Gurung, and S. Ale. 1996. Reducing livestock depredation
losses in the Nepalese Himalaya. R. M. Timm and A. C. Crabb (eds.). Proceedings of the
17th Vertebrate Pest Conference. University of California, Davis, pp 241-47.

Jackson, R. Richard, C. and Gurung, P. 2001. Enhancing participatory assessment capacity
in range and management and livestock wildlife interaction

Jackson, R. Ahlborn, G. Ale, S. Gurung, D. Gurung, M, and Yadav, U.R. 1994. Reducing
livestock depredation losses in the Nepalese Himalaya: A case study ACA B

Kapos, V., Lysenko, I. & Lesslie, R.G. 2000. Assessing forest integrity and naturalness in
relation to biodiversity. FAO/UNEP-WCMC, Cambridge, UK.

KMTNC 1994. Biodiversity Conservation Data Project, Final Report. King Mahendra
Trust for Nature Conservation, Kathmandu, Nepal.

KMTNC 1997. Annapurna Conservation Area Management Plan. King Mahendra Trust
for Nature Conservation, Kathmandu, Nepal.

Krebs, C. J. 1999. Ecological methodology 2nd ed. 620 pp. Benjamin/Cummings, Menlo
Park, California.

Kremen C., Merenlender A. M. and Murphy D. D. 1994. Ecological monitoring: a vital
need for integrated conservation and development programs in the tropics. Conservation
Biology 8: 388-397.

Lawrence A. 2002. Participatory assessment, monitoring and evaluation of biodiversity:
Summary of the ETFRN internet discussion 7-25 January 2002. Website: http://
www.etfm.org/etfrn/workshop/biodiversity/index.html.

Lawrence A., Malla Y. B., Paudel K. C. and Barnes R. 2003. Why biodiversity? A holistic
approach to participatory monitoring of community forestry in Nepal. In Proceedings of
International conference on Himalayan Biodiversity. Kathmandu, Nepal.

Linden, H., Helle, E., Helle, P. and Wikman, M. 1996. Wildlife triangle scheme in Finland:
methods and aims for monitoring wildlife populations. Finnish Game Research 49 4-11.

Lindenmayer, D.B., Burgman, M.A., Akeakaya, H.R., Lacy, R.C. & Possingham, H.P. 1995.
A review of the generic computer-programs ALEX, RAMAS/space and VORTEX for
modelling the viability of wildlife metapopulations. Ecological Modelling 82: 161-174.

Margules, C., and M. B. Usher. 1981. Criteria used in assessing wildlife conservation
potential: a review. Biological Conservation 21:79-109.

NASA 2003. Remote Sensing Tutorial. Vegetation Applications: Agriculture, Forestry, And
Ecology. Available from: http://rst.gsfc.nasa.gov/Sect3/Sect3_1.html. Accessed 7-Oct-2003.

84

Protected Area Monitoring Guidelines

Nature Conservancy Council. 1989. Guidelines for selection of biological SSSIs, 1998
edition. Nature Conservancy Council, Peterborough.

Oaks, J.L., Gilbert, M., Virani, M.Z., Watson, R.T., Meteyer, C.U., Rideout, B.A.,
Shivaprasad, H.L., Ahmed, S., Chaudhry, M.J.I., Arshad, M., Mahmood, S., Ali, A. and
Khan, A. A. 2004. Diclofenac residues as the cause of vulture population decline in Pakistan.
Nature 427: 630 — 633.

Longley, P. A., Goodcjild, M. F., Maguire, D. J. and Rhind, D. W. (2001) Geographical
Information Systems and Science, John Wiley and Sons

Obura D. O. 2001. Participatory monitoring of shallow tropical marine fisheries by artisanal
fishers in Diani, Kenya. Bulletin of Marine Science 69: 777-791.

Oli, M.K, 1991. The Ecology of Conservation of the Snow Leopard ( Panthera uncia) in
the Annapurna Conservation Area, Nepal

Pimbert M. P. and Pretty J. N. 1995. Parks, people and professionals: putting ‘participation’
into protected area management. Discussion Paper 57 UNRISD / IED / WWE.

Posey, D. A., editor. 2000. Cultural and spiritual values of biodiversity. A complementary
contribution to the global biodiversity assessment. UNEP.

Pretty J. N. 1994. Alternative systems of inquiry for sustainable agriculture. JDS Bulletin
25: 37-48.

Rabinowitz, D. 1981. Seven forms of rarity. Pages 205-217 in H. Synge, editor. The
biological aspects of rare plant conservation. Wiley, Chichester.

Ramsar Bureau. 2002. New guidelines for management pianning for Ramsar sites and
other wetlands. Ramsar Bureau, Gland, Switzerland.

Ratcliffe, D. A. 1977. A nature conservation review. The selection of biological sites of
national importance to nature conservation in Britain. Cambridge University Press,
Cambridge.

Salafsky, N., and R. Margoluis. 1999. Threat reduction assessment: a practical and cost-
effective approach to evaluating conservation and development projects. Conservation
Biology 13:830-841.

Salim A., Brocklesby M. A., Tiani A. M., Tchikangwa B., Sardjono M. A., Porro R., Woefel
J. and Pierce Colfer C. J. 2001. In search of a conservation ethic. In: Pierce Colfer C.J. and
Byron Y. (Ed) People managing forests: The Jinks between human well-being and
sustainability. pp. 155-170. Resources for the Future, Washington.

PAV AS AUN ATI OTS Dg CTA OS I ET ST eT 85

Sheil, D., Puri, R.K., Basuki, I., van Heist, M., Syaefuddin, Rukmiyati, Sardjono, M.A.,
Samsoedin, I., Sidiyasa, K., Chrisandini, Permana, E., Angi, E.M., Gatzweiler, F., Johnson,
B. and Wijaya, A. 2002. Exploring biological diversity, environment and local people's
perspectives in forest landscapes. Centre for International Forestry Research, Bogor, Indonesia.

Shultz, S., Baral, H.S., Charman, S., Cunningham, A.A., Das, D., Ghalsasi, G.R., Goudar,
M.S., Green, R.E., Jones, A., Nighot, P., Pain, D.J. and Prakash, V. 2004. Diclofenac
poisoning is widespread in declining vulture populations across the Indian subcontinent.
Proceedings of the Royal Society London B (supplement). DOI 10.1098/rsbl.2004.0223

http://www.vulturedeclines.org/gypdiclol .pdf

Smith, P. G. R., and J. B. Theberge. 1986. A review of criteria for evaluating natural areas.
Environmental Management 10:715-734.

Southwood, T. R. E. 1978. Ecological methods, 2nd edition. Chapman & Hall, London.

Spellerberg, I. F. 1991. Monitoring ecological change. Cambridge University Press,
Cambridge.

Spellerberg, I. F. 1992. Evaluation and assessment for conservation. Chapman & Hall,
London.

Steinmetz R. 2000. Ecological surveys, monitoring and the involvement of local people in
protected areas of Lao P.D.R. Evaluating Eden Series, Discussion Paper 13. WWF Thailand
Programme Office, Bangkok, Thailand.

Stockdale M. C. and Ambrose B. 1996. Mapping and NTFP inventory: participatory
assessment methods for forest-dwelling communities in East Kalimantan, Indonesia. In:
Carter J. (Ed) Recent approaches to participatory forest resource assessment, pp. 170-
211. Overseas Development Institute, London.

Sutherland, W. J., editor. 1996. Ecological census techniques. Cambridge University Press,
Cambridge.

Thapa, K. 2000. Livestock depredation in Nar and Phu valley, Manang region of Annapurna
Conservation Area, Nepal, King Mahendra Trust for Nature Conservation/Annapurna
Conservation Area Project

Thomas, L. & Krebs, J.K. 1997. A review of statistical power analysis software. Bulletin
of the Ecological Society of America 78, 128-139.

Thomas, L., and J. Middleton. 2003. Guidelines for management of protected areas. UCN,
Gland, Switzerland and Cambridge, UK.

Usher, M. B., editor. 1986. Wildlife conservation evaluation. Chapman & Hall, London.

Van Rijsoort J. and Zhang J. F. 2002. Developing of participatory resources monitoring in
two Nature Reserves in Yunnan, P.R. China. Website: http://www.etfrn.org/etfrn/workshop/
biodiversity/index.html.

86

Protected Area Monitoring Guidelines

Acronyms

ACA
ACAP
CAMC
CE

GIS
GPS
KMTNC
VDC

Protected Area Monitoring Guidelines

Annapurna Conservation Area

Annapurna Conservation Area Project
Conservation Area Management Committee
Conservation Education

Geographical information system

Global positioning system

King Mahendra Trust for Nature Conservation

Village Development Committee

_———_—_——_——

Glossary

Accuracy The closeness of an estimated value to the true value.

Attributes Characteristics, qualities or properties of a feature that are inherent
to, and inseparable from, the feature (CCW, 1996). For species
these may include population size, structure, habitat requirements,
distribution and other parameters. Attributes of habitats may include
key species, composition, structure, supporting processes and other

parameters.

Feature A habitat, habitat matrix, species or a species assemblage occurring
on a site.

Key area An area that is subjectively selected for sampling within because it

is assumed to be important or representative of a larger area.
Statistical inferences can only be made for the key area.

Macroplot Relatively large regular shaped areas laid over the bulk of the target
area, within which sampling units such as quadrats, lines or points
are located. They facilitate the positioning of sampling units.

Monitoring "The collection and analysis of repeated observations or
measurements to evaluate changes in condition and progress toward
meeting a management objective" (Elzinga et al. 2001).

Observation A record (e.g. measurement of height, count of numbers) taken
from a sample unit.

Population Any collection of individual items or units which are the subject
of investigation. The population is the total number of units, from
which we usually take subsets or samples.

Precision The closeness of the sample measurements to each other. An
estimate is more precise if it has a smaller standard deviation.

Primary plot and Plots in two-stage (or multi-stage sampling), where further
secondary plots sampling (i.e. with secondary plots) is carried out within the primary
plots.

88 Protected Area Monitoring Guidelines

Sample A subset of the units in a population which represents the population
as a whole. Ifa sample is to be truly representative, the sample must
be drawn randomly (i.e. free from bias) from the population.

Sample population —_ The population or area over which samples may be drawn from.

Sample unit, A sample unit is an individual population unit from a sample. A
Sampling unit sampling unit is a collection of observations with specified
dimensions (e.g. a quadrat). A set of these comprises a sample.

Statistical population The entire set of observations across all the samples, from which
statistical inferences are made.

Surveillance An extended programme of surveys systematically undertaken to
provide a series of observations to ascertain the variability that might
be encountered over time (but without preconceptions of what these
might be).

Target population The population that we are interested in (e.g. the population of a
species, or an area of habitat, that we are managing and have set a
conservation objective for).

te)
Protected Area Monitoring Guidelines 8°

Example protocols from the ACAP
biodiversity programme

8.1 Introduction

These monittoring protocols and data collection sheets have been developed by KMTNC
with the assistance of UNEP-WCMC over the period 2003 to 2005. They are presented as
examples of the types of conservation objectives for which protocols can be developed and
the information and procedures required for monitoring

8.2 ACAP Monitoring protocol for Snow Leopard (Uncia uncia)

Monitoring objectives

Reasons for monitoring:

= Snow Leopard is an Endangered Species and is legally protected by the National
Parks and Wildlife Conservation (NPWC) Act 2029 of HMG/N, listed in Appendix I
of Convention on International Trade in Endangered Species of Wild Fauna and Flora
(CITES) and as Endangered in IUCN Red Data Book.

m tis an indicator species and one of the top level predator of alpine ecosystem within ACA

® Population in decline in last few years due to unknown reasons (possibly persecution
and poaching)

m= Flagship species (considerable international interest)

Conservation objectives for the resource:

To maintain the frequency of occurrence and distributional range of snow leopard within
ACA as indicated by baseline presence level (to be determined in 2004-2006).

Monitoring population | area and sub-units:

The target monitoring area is all suitable habitat in the ACA. But, due to the large size
of the ACA (7,629 km*), the steep terrain and numerous cliffs it is not considered feasible
to undertake monitoring of all suitable habitats. Monitoring fieldwork will therefore focus
on the following key populations in Sangta, Bhena (Mustang), the Nar and Pho valleys,
and Khangsar (Manang). [Sites to be confirmed and key areas mapped]

Sample areas within each key area will be land that is below 5,500 m, and excluding
areas of permanent snow or ice and ground that is too steep to safely access.

Presence in other areas to be monitored by reporting of incidental sightings to wildlife
recording scheme.

90 SSS SSSsSsSsSsSsSsSssSsSsSsSaSSSSSSSSSSsSs BG ER ERM ETT BEI HE TTS

Frequency:

There has been some evidence of declines in this globally threatened species. Therefore
establishment of a monitoring programme and assessment of population trends is a
high priority for ACAP. However, the potential range of the species is very large within
the ACA and survey work is difficult and time-consuming. It is therefore suggested that
monitoring is carried out on a five-year cycle.

All incidental sightings to be reported to the wildlife reporting scheme.

Monitoring Methods

These methods are largely based on the recommended Snow Leopard Information
Management System (SLIMS) survey methods described in the Snow Leopard Manual
(Bajimaya 2001). These have been adapted to increase their statisticai applicability and
to take into account the difficult terrain and resource limitations on ACAP staff.

Observation types:

Direct observation of animals (although very unlikely), calls; and signs (scrapes, scent
spraying, faeces, tracks (pugmarks, spoors), prey kills and snow leopard remains.

Pugmark size (to distinguish individuals).

Data type:

Presence / absence in sample squares and mapped positions of sightings / signs. Counts
of signs per transect (see attached recording form Part B).

Possible minimum number of individuals present if pugmarks are detected and measured.

Complete census or sample survey:

Sample survey.

Sample method:

Snow leopards are difficult to detect and wide ranging species, with home ranges of 12-
39 sq km in prime habitat (Bajimaya 2001). Multi-stage sampling will therefore be
used to ensure high interspersion of primary samples and more intensive sampling in
secondary samples.

Primary sample units will be 3km x 3km sample survey squares. A relatively large
primary sample area is used to allow for the likelihood that some of the area will be
inaccessible. Secondary samples are 5 m wide transects within each primary sample.

Sample area / time period:

The time taken to carry out surveys will vary according to terrain etc. However, the
actual time taken searching transects on the Ist survey must be recorded and the same
time taken on all repeat surveys.

, ¢
Protected Area Monitoring Guidelines: —<=<=_=_=_— @—@—@A@ i $@ —$ )1

Timing of observations:

Surveys will be conducted in February, March or April, and will start at sunrise. Return
surveys should be conducted within the same month of the year as the previous survey.

Potential causes of bias and rules for standardization:

Snow Leopards are mostly active during dawn and dusk or at night in disturbed areas.
There may therefore be a bias against sightings in areas near human habitation or if
daily survey periods vary. However, Snow Leopards are extremely difficult to observe
directly, and therefore most data are expected to consist of indirect signs of presence.
These potential biases are therefore unlikely to be significant in practice. Counts should
however start early in the day to maximize potential chances of sightings and because
Blue Sheep counts will be conducted at the same time, and these are best carried out
near dusk or dawn.

Underestimates may occur where livestock densities are high as trampling tends to
destroy Snow Leopard tracks and scrapes etc. Transects should avoid tracks used
intensively by livestock.

The effort used to detect signs must be consistent between years. All surveyors must be
trained in the recognition of Snow Leopard tracks and other signs. Observers must
therefore be very diligent when surveying each transect within the sample squares.
Surveys should be carried out by 2 surveyors. If more are present, no more than 2
should be actively searching at any one time.

Care should be taken in extrapolating results where a high percentage of the survey area
or primary sample was inaccessible.

Sampling methods

Temporary or permanent sample location:

Permanent primary and secondary samples.

Method for sample location:

Samples should be located randomly within each key monitoring area, but due to the
wide ranging nature of the species, good interspersion and independence of samples is
required. Primary samples will therefore be located by restricted random sampling.

At least 50% of each randomly located survey square must fall within the sample area
and accessible areas. Primary squares must also be > 1km apart.

Transects within survey squares will be placed by judgement along routes that are
considered likely to be used by Snow Leopards, e.g. along ridgelines, cliff bases and
river bluffs, according to guidance given in the Snow Leopard Manual. A sufficient
number of transects should be selected to take up 6 - 8 hours of searching (not including
travel time between). Transects should be subdivided according to appropriate land
features (e.g. extent of cliff edge) and the amount of time take to survey each one recorded
to aid consistency between subsequent resurveys. Each survey square should contain at

92

Protected Area Monitoring Guidelines

least | km of defined transects, and transects should aim to be 100 - 500 m long, but may
be longer if they match obvious continuous topographical features.

Transects should not cross the 5,500 m contour, permanent snow or ice or dangerously
steep ground.

The location of transects must be accurately mapped and frequent waymarker positions
noted accurately by GPS and photographed.

Note: As transects are located deliberately (by judgement) in areas of good habitat and
where signs are likely to be found they produce biased estimates of Snow Leopard
abundance. These results can therefore only be used as a rough guide to abundance and
should not be compared statistically. However, subsequent changes in relative abundance
on transects and the presence of snow leopards in each sample square may be tested if
the same transects and methods are used to search these in each subsequent year (see
analysis section below).

Number of samples:

Four squares in Khansgar key area, six in Nar and Pho valleys. [Numbers to be decided
for other key areas]

Monitoring Requirements

Personnel responsible and time required:

Monitoring to be coordinated by ACAP Monitoring Manager, with sample locations
identified by GIS team. Surveys must be carried out by specifically trained ACAP Field
Staff.

Experience training necessary:

Training must be given to all staff undertaking the surveys.

Licence and access permission requirements:

Not applicable

Equipment required:
GPS, map of each primary sample square (preferably at 1:10,000 or 1:20,000 scale) with
marked transects, photographs of key landmarks to aid relocation. Sections on Snow Leopard
signs from Snow Leopard Manual (Bajimaya 2001). Standard recording forms must be
used, with copies of codes sheets from Snow Leopard Manual. Binoculars, camera, small
tape measure (for measuring tracks and scrapes etc). Field safety kit.

Data storage

Original Data Recording Forms and maps should be safely stored within each Field
Office and the data entered onto the standard Excel summary form and sent to the
Monitoring Manager.

, ¢
Protected Area Monitoring Guidelines|$_ —_—___A__ CO )3

Field monitoring reports must be submitted with the data. These should document the
detailed methods used, including any deviations from this protocol. The reports should
include: large scale maps indicating boundaries of the primary sample square (with

‘coordinates for each corner). Transect locations with start and end coordinates:

photographs of transect start-points and other important features on the transect (e.g.
boulders used as scent marking sites); notes describing each transect (e.g. ridge,
streamside), the order in which each was examined and the time taken to examine each
transect. The reports should include the original data forms (or cross-checked typed
copies) and the SLIMS Codes.

Data analysis

The main analysis will be of the proportion of sample squares that have signs of Snow
Leopard. As sample locations are permanent, changes in the proportion of squares with
Snow Leopard may be tested using Mc-Nemar's test (See Elzinga et al. 2001), but
sample sizes are unlikely to be sufficient to allow formal statistical significance testing.
Any decline in presence should be considered to be of concern and should trigger further
investigations (e.g. further more intensive surveys) and appropriate management
measures.

A general indication of the relative abundance of Snow Leopards may also be gained
from the number of signs recorded per km of transect (see Snow Leopard Manual for

details). Changes in this relative abundance measure may be tested by a paired t-test.

Longer term trends may be examined by regression analysis.

Reporting procedures:

Every five-years on completion of survey cycle.

Reference

94

Bajimaya, S. 2001. Snow Leopard manual. Field techniques for the kingdom of Nepal.
WWE Nepal, Kathmandu.

Elzinga, C. L., D. W. Salzer, J. W. Willoughby, and J. P. Gibbs 2001. Monitoring plant
and animal populations. Blackwell Scientific Publications, Abingdon, UK.

EEE Protected Area Monitoring Guidehnes

8.2.1 ACAP Snow Leopard data recording form

Part a: count details

Date (day/month/year):

Counter details

Name of lead counter:

Address:

Other observers / trainees present:

Primary sample square

ACAP Unit Conservation Area: Altitude - min (m):
Nearest town / village: Altitude - max:
Survey square number: Aspect:

Coordinates (plot center):
Visit details
Start time: End time:

Weather conditions:

Temperature (circle): Cold ( (<5 °C)) / Cool (5-10 °C) / Mild (10-15 °C) / Warin
(15-20 °C) / Hot (>20 °C)

Cloud cover (to nearest 10% or give range):

Wind direction: Wind speed: still / light / breeze / strong breeze / near gale / gale

’ . Cc <4
ES OTCOLE ARE Ca IMOTISSONI TIE GTA LE )5 11S )5

Part b: Observations (use 1 form for each transect) Page ..... of ......

Comments
Lead observer Survey Nearset village Transect no.
square

Start coordinates End coordinates Length (m)

See notes below and use Codes for SLIMMS Form 2

ange use

Landform| Topo Marked
feature | feature

~“

Sign age /|Substrate

vis type

Notes* 1: | Number each site where sign observed and number map. 2: Number each observation at each
site. 4: Record all large predators (excluding domestic dog).

Comments: (with respect to site & obs no.)

96

Protected Area Monitoring Guidelines

8.3 ACAP Monitoring protocol for Himalayan Griffon
(Gyps himalayensis) and other vultures

Monitoring objectives

Reasons for monitoring:
m Population in decline in last few years due to unknown reasons

w Flagship species

Conservation objectives for the resource:

To determine baseline relative population abundance levels and to maintain these over
the next ten years (to 2015) within the Annapurna Conservation Area (ACA).

Monitoring population | area and sub-units:

The target population is the breeding population with the ACA, covering all Unit
Conservation Areas (Lomanthang, Jomson, Ghandruk, Lwang, Sikles, Bhujung,
Manang). The sample population will be birds visible from selected lookout points
within the target population area that are within accessible areas and which are not
above 5000 m, or are not rock or permanently covered in ice or snow.

Frequency:

There have been major declines in many vulture species and populations with the Indian
subcontinent in recent years, including Himalayan populations of White-rumped Vulture
(Gyps bengalensis), Cliff Vulture (Gyps indicus) and Slender Billed Vulture (Gyps
tenuirostris). There has also been some indications of a decline in Himalayan Vulture
populations within the ACA (Baral et al. 2002). Establishment of a monitoring
programme and assessment of population trends for Himalayan Vultures and other vulture
species is therefore a high priority for ACAP.

Annually for first 5 years and then every 2 years afterwards. Kill counts will be made
whenever encountered by ACAP staff.

Monitoring methods for soaring birds

Due to the large size of the ACA (7,629 km’), the steep terrain and numerous cliffs it is
not considered feasible to undertake monitoring of birds at nesting colonies. Most would
be inaccessible and monitoring of only larger key colonies could produce substantially
biased results.

Observation types:

Direct observation of soaring vultures (all species) and other large soaring raptors and ravens.

Data type:

Counts of total bird sightings (i.e. maximum possible number of individuals), estimated
number of individuals and minimum number of individuals during a timed count. See
attached Soaring Vulture recording Form.

Protected Area Monitoring Guidelines] _—<—LRA 97

Complete census or sample survey:

Sample survey.

Sample method:

Sample units will be point counts from selected vantage points.

Sample area | time period:

All visible birds will be counted (i.e. no fixed area) during a standard 4 hour count
period

Timing of observations:

Counts will be made between 10.00 and 14.00, between Ist June and 30th June (i.e.
when breeding birds are present and feeding young). [Seasonal timing to be confirmed]

Potential causes of bias and rules for standardization:

Soaring is dependent on the presence of thermals or updrafts and is therefore greatly
affected by weather conditions. Counts should, therefore, only be made during suitable
soaring conditions, 1.e. in the absence of rain, fog, low cloud or complete cloud cover.
Counts should also only be carried out during the middle of the day when solar energy
and hence thermal activity is at its greatest, i.e. 10.00 - 14.00 hrs.

It is difficult to avoid double counting of vultures as they often back track during their
foraging flights. Care must therefore be taken in estimating the number of individual
birds seen. Details of each bird's age and plumage should therefore be noted during
sightings and consistent decision rules used to estimate the number of individuals seen.
A minimum number of individuals seen will also be calculated on the basis of
simultaneous sightings plus any subsequent birds of clearly different age or plumage.

The effort used to detect birds must be consistent between years. Only one counter
should detect birds (i.e. if other observers / trainees are present, then they should not
indicate birds to the lead counter or include birds seen by them and not the lead counter
on the recording forms). Observers should remain vigilant throughout the four hour
period and should make regular scans of the sky with their binoculars (e.g. a 3600 scan
every 5 minutes). Telescopes should not be used to detect birds that are not normally
visible with binoculars.

Sampling methods for soaring birds

Temporary or permanent sample location:

Permanent points

Method for sample location:

Vantage points will be selected by judgment from within randomly placed 5 x 5 km
squares. Each randomly placed square must, however, be more than 10 km apart from
any previously located squares. If this is not the case then the selected square must be
discarded and another square randomly placed.

98

Protected Area Monitoring Guidelines

Randomly placed squares that overlap with the ACA boundary should be retained if a

suitable vantage point is present and counts made of all birds seen whether or not they are
over the ACA.

Vantage points should be selected so that they give as wide a view of the surrounding landscape
as possible. Within forested areas they should be at or above the treeline or as high as possible.
Suitable vantage points will include peaks and ridges. North facing slopes should be avoided
if possible, unless they provide good vantage points overlooking nearby south facing slopes.
Considerable care should be used in selecting vantage points as these will be permanently
used for all subsequent monitoring counts. It is therefore recommended that vantage points
are selected by careful reference to maps, supplemented with reconnaissance visits to several
potential sites before final selections of vantage points are made.

The location of vantage points should be accurately recorded by GPS, photographed
and mapped.
Number of samples:

One vantage point will be used in each 5 x 5 km square and twenty random sample
squares will be selected and counted in the Ist year. The data obtained from the Ist year
will then be used to calculate a suitable sample size for subsequent monitoring. If less
than twenty squares are acceptable, then excess sample squares will be randomly discarded.

Two counts should be made at each vantage point each year during the survey period.

Use of multi-stage sampling:

As vultures are very wide raging species, high interspersion of samples is required and
therefore multi-stage sampling is not applicable.

Monitoring methods for birds at carcasses
This method is based on the standard method used by the Vulture Decline Project.

Observation types:

Direct counts of birds at carcasses. If possible several photographs of birds present at
the carcass (and soaring above it) should be taken for verification by an ornithologist of
the species present and their age classes.

Data type:
Counts of vultures according to species and age class. Counts of birds showing signs of
neck drooping (i.e. disease). Counts of other scavengers.

Complete census or sample survey:

Sample survey

Sample method:

Birds present at carcasses, forms to be competed for all carcasses found, irrespective of
whether vultures are present.

9c
EEG LCCIELATCR MOMSIOTING GING CS )9

Sample area | time period:

No restriction, all birds visible at the carcass when discovered.

Timing of observations:

No restriction

Potential causes of bias and rules for standardization:

Vulture numbers at carcasses may vary depending on the number of carcasses available
(i.e. 1f carcasses are plentiful then birds may be widely dispersed with low numbers at
each carcass). Thus apparent changes in numbers may result from variation in carcass
availability. Trends should therefore be compared with the number of carcasses found.

The number of carcasses found will, however, vary according to survey effort including
the willingness to look for and record carcasses and the amount of time spent in the
field. The importance of looking for and recording all carcasses encountered should
therefore be emphasized to all field staff. Results should also be related to the amount
of time spent in the field by staff (which is to be recorded as part of the ACAP wildlife
recording scheme, see separate protocol).

Sampling methods for birds at carcasses

Temporary or permanent sample location:

Temporary

Method for sample location:

Chance encounters by ACAP staff with carcasses during other field activities.

Number of samples:

Variable depending on carcass numbers and field effort.

Use of multi-stage sampling:

Not applicable

Monitoring requirements

Personnel responsible and time required:

Monitoring to be coordinated by ACAP Monitoring Manager [to be identified], with
sample locations identified by GIS team. Counts of soaring birds to be carried out by
specifically trained ACAP Field Staff. Counts of birds at carcasses to be carried out by
all ACAP staff who have received basic training in the method and vulture identification.

Experience training necessary:

100

Training must be given to all staff undertaking the surveys. However, expertise in
identification of all birds is not necessary, and the surveys can be carried out by ACAP
personnel or other members of the local communities who are competent in vulture
identification.

Protected Area Monitoring Guidelines

Licence and access permission requirements:

Not applicable

Equipment required:

Binoculars (consistent magnification between surveys, and preferably 10 x
magnification), GPS, map of site location of vantage point and photograph of vantage
point to aid relocation. Bird identification guide and tables summarizing vulture
identification criteria (see Bombay Natural History Society 2001). Standard recording
forms must be used. Telescopes may also be used for verifying identification, age and
plumage features of birds detected through binoculars. Field safety kit.

Data storage

Original Data Recording Forms should be safely stored within each Field Office and
the data entered onto the standard Excel summary form and sent to the Monitoring
Manager [ACAP to devise based on the recording forms]. Carcass recording forms
should also be copied and then sent to the appropriate contact person of the Vultures
Decline Project.

Data analysis
For counts of soaring birds; the number of individuals (maximum number, estimated
number of individuals and minimum number of individuals) of each species shall firstly
be averaged over the two counts at each vantage point. Year to year changes in relative
abundance should then be examined by calculating a mean percentage difference with
confidence limits across the set of vantage points.

Longer term trends may be examined by regression analysis.

It is anticipated that the analysis of carcass counts will be made by the Vulture Decline Project

Reporting procedures:
Annually

References
Bombay Natural History Society (2001). Proceedings of a Gyps spp. Vulture monitoring
workshop. Available at http://www.vulturedeclines.org/home.asp
Baral, H.S., Giri, J.B., Choudhary, H. Basnet, S., Watson, R. and Virani, M. (2002).

Surveys of Himalayan Griffon Gyps himalayensis in the Nepalese Himalayas. Final
report 2002 to The Peregrine Fund, USA.

Protected Area Monitoring Guidelines 101

8.3.1 ACAP soaring vulture data recording form

Part a: count details

Date (day/month/year): Count ......... of 2

Counts must be undertaken between 10.00 - 14.00. Please note any discrepancies or
interruptions in comments section.

Counter details
Name of lead counter:
Address:

Other observers / trainees present:

Count location

ACAP Unit Conservation Area: Nearest town / village:

Sample point code: Coordinates:

Altitude: Aspect:

Visit details

Weather conditions at start of count:

Temperature (circle one): Cold (<5 °C) Cool (5-10 °C) Mild (10-15 °C)
Hot (>20 °C) Warm (15-20 °C)

Cloud cover (estimate to nearest 10%):
Weather conditions at end of count:

Temperature (circle one): Cold (<5 °C) Cool (5-10 °C) Mild (10-15 °C)
Hot (>20 °C) Warm (15-20 °C)

Cloud cover (estimate to nearest 10%):

Optical equipment used:

Broad habitat type (circle one main type present within 5 x 5 km square: a) Cultivation
Area b) Deciduous / mixed forest c) Coniferous forest d) Montane deciduous forest /
shrubland e) Grassland f) Barren soil / rock g) Permanent snow and ice

Comments

102 oo l——— Protected Area Monitoring Guidelines

Part b: Observations

Date:

Lead counter:

of .....

Page .....

Notes
(see
below)

Age
Unknown

Number of Vultures

Juveniles

a

* Record all sightings of vultures, Golden Eagle, other eagles, buzzards and Ravens.

Notes: (see code number in Table)

103

SRE LE BIE A RE LEIM OTA EOL 01 G00 (C1 1

Part c: Summary of observations

Lead counter: Date:

Number of Vultures
S| = ee
dace

SSeS
eS a
SS
Poe en) |
oc eT a
= Sa
a
mre | —
eT ala a
a ae
Eo ae a
SS ae ee a
SSeS —
SS ee
ee ee
ace a
[moxie |
aes
Tee
Pwoxrey |
eatin |_|
Pi [
Bea
a
4
mas
Laces

kent ener |
PaaS lier leo

Comments:

104 ee ——— Protected Area Monitoring Guidelines

Counter details Recording form for scavengers on carcasses

Date (day / month / year) / /

Town and District a> c d
e f g h

Time of start of count (HH: MM) : am/pm (circle one)

Time of end of count(HH: MM) : am /pm (circle one)

Temperature (circle only one) cold cool mild warm hot Broad Habitat Type

6-10 1-25 >25

) 1 2-5
Counts of vultures (include sick birds in the counts of perched birds)

Total number of Gyps vultures (all species) perched at the carcass

Number of fresh livestock carcasses (circle only one

Total number of Gyps vultures (all species) soaring above the carcass

Please fill out the table below for the first 50 that you see when scanning the dump,
or all birds if there are less than 50. If you do not have time to age the birds, put total
numbers of each species in ‘Age Unknown’

Total perched birds Rerchedibiids
Species Age Juveniles | Sub-Adults Adults Cllgees
Unknown combined)

White-backed vulture

Slender-billed vulture [Peirce here yf
Indian vulture (long-billed) Veeaeadhe altri

Himalayan griffon
Egyptian vulture
i Total counts - no age
King vulture ;
separation needed

Other scavengers Notes and comments
(number) (eg any dead birds or unusual incidents)

Protected Area Monitoring Guidelines. —___HX_A_ 105

8.3.2 Vulture Declines Project Instructions for counting
scavengers at carcasses

This form can be used for carcass dumps or individual fresh carcasses observed at the
side of the road or elsewhere. Simply record the number of fresh livestock carcasses.
Please note, it is equally important to record carcasses with and without vultures. Please
record any other scavengers present.

Repeat counts at carcass dumps are very useful. Please count once a month if possible. Less-
frequent counts are also very useful. It does not matter that you cannot do repeat counts of
scavengers on individual livestock carcasses seen. This information is still very useful.

Please record broad habitat type in the following categories:
(a) cultivation, (b) thorn forest, (c) dry deciduous forest, (d) moist deciduous forest

(e) coastal area, (f) municipal park, (g) roadside plantation (h) other

Count the total number of all Gyps species (all species combined) perched at the carcass
dump, and soaring above the carcass dump.

If there are less than 50 birds, please record species and age classes for all birds. If there
are more than 50 birds, scan the birds and record species and ages and species of the
first 50 birds (Gyps only) that you see when scanning the dump. If you do not have time
to age birds please just separate by species and record numbers in the 'Age unknown!
category. For non-Gyps species (i.e. Egyptian, King and Cinereous vulture) simply
record total numbers in the 'age unknown' category.

Record numbers of birds with obvious neck droop in the neck droop column. 'Neck droop'
indicates that the bird's head is hanging vertically or almost vertically towards the ground.

All ages can be combined. Again, if there are less than 50 birds record whether or not
each bird has neck droop. If there are more than 50 birds, simply record whether or not
each of the first 50 birds that you see when scanning the dump have neck droop.

Record any dead or very sick birds under 'Notes and comments’. Please also record rat
droppings under 'Notes and comments'. Rat droppings can be recorded as none, scarce
or common.

For further information please contact Dr Vibhu Prakash, Principal Scientist, Bombay Natural
History Society, Hornbill House, Shaheed Bhagat Singh Road, Mumbai - 400 0023

Forms can be downloaded from the project website: http://www. vulturedeclines.org

106 Protected Area Monitoring Guidelines

8.4 ACAP Monitoring protocol for broad-leaved forest habitat
quality

Monitoring objectives

Reasons for monitoring:

To assure sustainable utilization of the forest resources and maintain the diversity of the
broad-leaf forest species in a given area.

Conservation objectives for the key feature:

To increase or maintain the current level (2005) of broad-leaved forest species diversity
and growing stock in a given forest area over the next ten years

Monitoring population / area and sub-units:

Mixed broadleaved forest in Annapurna Conservation Area- Southern Sector - Ghandruk,
Lwang, Sikles, Bhujung and Southern region of Northern Sector- Lower Manang and
Lower Mustang.

Frequency of surveillance:

Every 5 years
Users of Results
Primarily Project Manager and committee members- will be used in Conservation Area

Management Operational Plan (CAMOP)

Monitoring methods

Observation | data types:

Forest inventory data are primarily collected as per the Inventory Guideline of Community
Forestry (revised), 2004 developed by HMG/Nepal, Ministry of Forest and Soii
Conservation.

Data type:

Measurement- diameter at breast height (dbh), height and canopy cover for Tree and
Pole category (life form); Canopy cover will be measured by Dendrometer

Count (sapling and seedling, later for regeneration purpose)
Count: shrubs

Deadwood abundance:

Diameter at breast height (DBH) and height for dead, dying and diseased trees (3Ds) if
standing,

Count (numbers) and measurement (circumference of stump) for stumps

RO ICIC PALE IN CTIET ORO GLA CU ae 10

Complete census or sample survey:

Sample survey- firstly stratification of the area based on timber stocking (which is influenced
by attributes like age class, forest type, physical terrain etc.) and secondly laying out the
sample plots systematically, sampling intensity at least 0.1 (of targeted population)

General rule: Even if you don't have a prior knowledge on forest, stratification will do
good, since forest as a biological entity is essentially complex.

Sample area | method:

Determining the sample area is crucial. It depends on many factors for e.g. level of
precision expected, resource availability, management objective etc.

Sample area is calculated by multiplying the size of the sampling units and its number.
Size of the sampling unit for different categories (life forms) as prescribed by the
Inventory Guidelines, 2004, is as follows;

25x20 m? (0.05 ha)- tree category (<30 cm dbh, overbark)

10x10 m? (0.01ha) - pole category (10-29.9 cm dbh, overbark)

5x5 m? (0.0025 ha)- sapling (4-9.9 cm dbh, overbark) and regeneration (4 cm<)

Number of sample plot can be calculated, as the sampling intensity is known (prescribed
by the guidelines).

Timing of observations:
March/April/May

Potential causes of bias and rules for standardization

Bias due to measurement
- due to sampling, unrepresentative sampling, non-response and volunteers tree
- due to Instrumental error

Observer ability | training

Not applicable

Seasonal / daily timing. Month/season should be kept constant. Return visits should be
undertaken within the same 2-week period as in previous years.

Sampling methods

Temporary or permanent sample location:

Permanent plots.

108

Protected Area Monitoring Guidelines

Method for sample location:
Sampling type - stratified systematic sampling in each strata

Rectangular plot (20x25 m* for tree life form), square plot (10x10 m? for pole and 5x5
m2 for sapling and for regeneration) - nested type

While establishing the sample units in the slope, for those arms going against the slope,
needs slope correction (horizontal distance = slope distance X cos @)

Number of samples:
Statistically adequate number of sample plots (thereby sample size) is required to meet

the desired precision level

It also depends on various factors for e.g. condition of forest (intrinsic homogeneity),
size of targeted population (ACA is quite big in area), level of precision needed (at what
confidence level), however we will base the numbers on 0.1 sampling intensity as
prescribed by Inventory Guidelines, 2004

Monitoring requirements

Personnel responsible and time required:
Monitoring to be coordinated by ACAP Monitoring Manager [Natural Resources
Conservation Officer] with sample locations identified by applying systematic sampling
techniques by field staffs (alternatively sample locations would be identified by GIS
team using sample design software).

Surveys must be carried out by specifically trained ACAP Field Staff.

Experience training necessary:

Training must be given to all staff undertaking the surveys. New training (for new field
staff) and refreshment (follow up) training for existing staffs at least once in five year
(before field works)

License and access permission requirements:

Not applicable

Equipment required:

GPS, topographic map of each forest with marked sample, photographs of key landmarks
to aid relocation. Standard recording forms must be used, with copies of codes sheets
(if needed). Other equipment needed are camera, pain and brush for marking, 50 meter
tape, 30 meter tape, diameter tape, Abney's level or Clinometer and altimeter for altitude
verification. Field safety kit.

Data storage

Monitoring manager should provide excel forms to input data.

Peotectem@ area MONisOrine GUI ACMNeS, em en 109

Original Data Recording Forms and maps should be safely stored within each Field
Office and the data entered onto the standard Excel form and sent to the Monitoring
Manager

Field monitoring reports must be submitted with the data. These should document the
detailed methods used, including any deviations from this protocol. The reports should
include: large scale maps indicating boundaries of the intensive use zone and strata (with
coordinates for each zonation and strata). Transect locations with start and end coordinates;
photographs of transect start-points and other important features on the transect (e.g.
boulders); notes describing each transect (e.g. ridge, streamside). The reports should
include the original data forms (or cross-checked typed copies) and the Codes.

Data analysis

Species and DBH class wise - density (applies to all), volume (Tree), biomass (Tree) -
(timber/branch/leaf)

Data are tabulated using Excel programme. Growing stock of broad leaved forest is
calculated by using the formula and models as per the Inventory Guidelines.

Long term trends will be examined by correlation and regression analysis. Correlation
gives the casual relation while regression provides the association.

Reporting procedures:

Every five-years on completion of survey cycle.

Health and safety

Forest work involves some inherent risks and hazards because of the places we go to
and the activities we undertake. Following safety precautions applying to all field work,
may minimize the risk substantially;

If at all possible, avoid going alone to the field/forest.

Wear clothing and footwear suitable for the weather, the activity and terrain
Never smoke in forests or grassland, and take care when lightening fires
Show extra care on cliffs and steep slopes

Don't incur additional risks by e.g. climbing cliffs, walking on slippery rocks, or
wading

alone rivers, unless these activities are an essential part of the work

Familiarize yourself with the direction and location of the nearby village/settlement
and available communication networks

Make sure you carry the First Aid and Emergency Kit

110 OO —————$—=———=—=—=—" Protected Area Monitoring Guidelines

References
Freese, F. (1984): Statistics for Land Managers, Wiley, New York

Goldsmith, F.B (1991): Monitoring for Conservation. Chapman & Hall, London

HMG/N (2004): Guideline for Inventory of Community Forests (revised), 2004, Ministry
of Forest and Soil Conservation, Kathmandu

Hurlbert, S.H (1984): Pseudoreplication and the design of ecological field experiments.
Ecological Monographs 54, 187-211.

Johnson, D.H. (1999): The insignificance of statistical significance testing. Journal of
Wildlife Management 63, 763-772.

IDLE CLEARED MONON GENEL CS 11]

8.4.1 ACAP Forest inventory data record sheet

VDC s) GAR AV rs Ce enact
Recorded By Se eerercc re error Sooner Name of Strata Hiab dice nitive eae
Date Jee ees coe eee Plot No Ssadedead aeeaeen, See eee es
INaimeio ip ROLESU ia ee .ctssccsccnsh see eer GPS poimts\(INIE) Meee ee ee
Altitude (m) DE Eee eer Slope (degree) it ARTS .25) Sh eo
Aspect oe crore sane aecseie co soo eee % Crown COVET: ........--.-++ SOlicOviereee eee
Sign of human impacts:

mie WoppingyVes/INOs Iii West (SPECIES eae-cesseecee testo eee set cece seers ov eteetaere seccecereemmeenrenmen )

m Logging Yes/No, If Yes (Species (no of cut StumpS) .............:::cccsceceseseesceseeeseeeenes )

w Non Timber Forest Products and/or Medicinal and Aromatic Plants (MAPs) collection

NeSHINO MI ES (SPECIES AMG CN(S bees vecee teat sa ace ser onceescucecsncutacsneseaneees )

mi Grass cuttins, Yes/INoy If Yes (Species rises esccstecesaceccedcsvescte. +s <ceennes-snass ovens eee )
Other impacts:

mie Grazinoayes/NO Ite yes) (WWilnicinranimiall (S) eee ceescccsezsssecevecesses-eoeeessseceseeeeneere sees )

w Forest fire : Yes/No

Overall condition of the forest: Regeneration / Pole / Tree

Diameter or circumference at breast height (over bark) in cm
Regeneration Sapling Pole & Tree dbh 10cm & above Remarks

(counting) | dbh(4to9.9 [dbh | Treetop/ | Distance
dbh <4cm cm) (cm) bottom of the
and height > | Crf (12.5 to 31 | or Crf angle tree from

ay cm) Obs.

=
N“

21

aaa
as)
eo
EI
ee
oes
eB S|
Been
eat
eave
ee
oa
Ea
a
ae
a
Lees
Bet
ae)
aa
Eas

Note: Crf = Circumference at breast height, Obs = Observer

112

Protected Area Monitoring Guidelines

Evidence of wildlife

List the name of NTFP and MAP found within the plot

Other observations/comments of the surveyor

DESC GLE CARE LAM OPES OTE ITO| GIS5(0 0101 SN 113

8.5 ACAP Monitoring protocol for broad-leaved forest birds

Monitoring objectives

Reasons for monitoring:

w Forest birds are of conservation importance, and include some threatened or regionally
important species.

gw Some forest birds are indicators of high forest quality or ecological value.

Conservation objectives for the key feature:

To maintain or increase the current levels of breeding bird species richness and population
size in selected forest indicator species over the next 10 years, to 2015.

Monitoring population | area and sub-units:

Mixed broadleaved forest in Annapurna Southern Sector - Ghandruk, Sikles, Lower
Mustang.

Mixed broadleaved forest in Manang.

Frequency of surveillance:

Annual

Monitoring methods

Observation | data types:

Direct visual observations and calls/songs of birds, mostly ascribed to species.
Bird species richness (species detected in each sample plot)

Relative bird abundance from Timed Species Count (TSC)

Frequency of species occurrences per 10-minute observation period
Abundance estimates from point counts

Complete census or sample survey:

Two-stage sample

Sample area | method:

TSC methods over three 60 minutes counts over fixed routes. Route to be deduced
beforehand from reconnaissance and previous habitat monitoring on Ist visit (see Forest
Habitat Quality Protocol).

Observers should walk slowly and quietly along the preset and mapped route, recording
all birds seen or heard within the survey square on the standard survey form (see below).

114

Protected Area Monitoring Guidelines

Observers should stop walking and listen quietly for about 3 minutes, approximately
midway through each 10 minute period. Birds seen outside the square may be recorded
in the appropriate column on the survey form.

Timing of observations:

March/April, from | hour after sunrise for 4 hours (c. 7am - | lam).

Potential causes of bias and rules for standardization:

Observer ability / training. Surveyor teams should include at least one fully experienced
and trained ornithologist (see below).

Seasonal / daily timing. This should be kept constant. Return visits should be undertaken
within the same 2-week period as in previous years.

Weather. Surveys should not be undertaken during heavy rain or snow, or in high winds
(c. > 15 mph).

Habitat change. This cannot be controlled for. Therefore care should be taken in
interpreting changes where habitat changes may have affected visibility.

Sampling methods

Temporary or permanent sample location:

Permanent primary and secondary routes.

Method for sample location:

Stratified random location of primary samples (2km x 2km), stratification by altitude /
habitat type [to be carried out by ACAP]. Exclude areas outside mixed-broadleaved
zone, and exclude steep ground to produce a sample population area.

Secondary routes to be spread across plot to encompass all variation in forest habitats
types within the survey plot, including different age classes, densities, habitat type and
degrees of degradation. Minor paths may be used if they do not interfere with tree
cover. But large tracks where cover is broken such that edge species dominate must
NOT be used. Treeless shrubland and grassland etc should not be included.

The route should be mapped as accurately as possible with distinctive features (e.g. an
obvious type of large tree) noted and coordinates recorded using a GPS (where tree
cover allows).

Number of samples:

To be allocated depending on resources; minimum of 10 per key monitoring area (or 5
per stratum).

PERO LEEIECUA Seca: VOTIETD 11.01 GL LE 115

Monitoring requirements

Personnel responsible and time required:

Teams of two ornithologists. One or two mornings fieldwork per primary plot, depending
on terrain (i.e. a total of about 4 hours). Remaining time each day to be used for traveling
to vicinity of the next sample.

Experience training necessary:

At least one of the two ornithologists to be fully experienced with forest bird identification
and trained in method. The other ornithologist may be a trainee.

Licence and access permission requirements:

Not applicable

Equipment required:

Binoculars (8x or 10x magnification), map with marked sample areas and coordinates,
compass, GPS, field notebook, standard recording forms (see below), watch with
stopwatch facility, and field first aid / safety kit.

Data storage
Original Data Recording Forms should be safely stored within each Field Office and the
data entered onto the standard Excel summary form and sent to the Monitoring Manager
[ACAP to devise based on the recording forms].

Data analysis

Reporting procedures:

Annually

116 Protected Area Monitoring Guidelines

8.5.1 ACAP forest bird survey recording form

Part a: count details

Date (day/month/year):

Counter details

Name of lead counter:
Address:
Other observers / trainees present:

Count location

ACAP Unit Conservation Area:

Nearest town / village:

Primary plot number:

Primary plot coordinates: SE corner: NW corner:

Altitude - min (m): Altitude - max: Aspect:

Broad habitat types: % cover within survey plot:

a) Cultivated: e) Montane deciduous forest / shrubland:
b) Broad-leaved forest: f) Grassland:

c) Mixed broad-leaved / coniferous forest: g) Barren soil / rock:

d) Coniferous forest: ) WOUMETE cccet rssh te

Visit details
Start time: End time:
Weather conditions:

Temperature (circle): Cold (<5 °C) / Cool (5-10 °C) / Mild (10-15 °C) / Warm (15-20 °C)
/ Hot (=20 °C)

Cloud cover (estimate to nearest 10%):

Wind direction: Wind speed: still / light / breeze / strong breeze / near gale / gale

Comments

Protected Area Monitoring Guidelines: ——_—_—_AAR-p @A a 117

)

of ..

(page ...

Part b: Observations

Transect number

Primary plot no.

Lead observer

Coordinates

m)
®
2
©
=
i}
=
°
fe)
~e)

Coordinates

Coordinates

Analysis

Time finished

Time period seen in

Time started

n
®
2
©
ic
Z
fe)
fe)
O

Coordinates

Coordinates

Coordinates

Protected Area Monitoring Guidelines

8.6 ACAP Monitoring protocol for remote sensing of
habitat extent and quality
This monitoring protocol is designed to assess, at the ACA level, the quality of habitats
using remote sensing data as indicated by spatio-temporal changes in habitat extent and

quality. It will also assess fragmentation characteristics of the habitat patches from a
landscape level perspective.

Monitoring objectives

Reasons for monitoring:

m A general idea of the trend of biodiversity in a habitat scale can be derived from
monitoring of habitat extent and location as it directly affects the distribution and
abundance of floral and faunal diversity

m Habitat extent and quality is a direct measure of biodiversity

m Spatio-temporal changes in habitat types and their extent is necessary to access the
effectiveness of management interventions

m Data on habitat fragmentation is useful to plan, monitor and evaluate habitat and
species conservation.

Conservation objectives for the key feature:

m To generate information on coverage, spatio-temporal changes and fragmentation
characteristics of habitats of ACA region for use by ACAP management team to
plan, monitor and evaluate the habitat and species conservation activities.

Users of monitoring results:

a ACAP management planning team (includes KMTNC program and monitoring unit,
ACAP team and Conservation Area Management Committees)

Monitoring population | area and sub-units:
mg Whole ACA region

Frequency of surveillance:

m Every 5 years

Monitoring methods

Observation | data types:

mw ASTER (Advanced Spaceborne Thermal Emission and Reflection) data set -
for 2000 onwards (This data set is available free of charge via http://
asterweb.jpl.nasa.gov/) and has spatial resolution of 15m x 15m).

a Landsat data for years before 2000 (http://landsat.gsfc.nasa.gov/)
@ Digital elevation model - it will be generated from the elevation contour data of

HMG/N (2002)

EXO LECLEIATCIMOPATOL IIE GENETICS) nr 119

w Ground truth data (collected using GPS set); complemented by:
e data generated from broad-leaved forest habitat quality monitoring and

@ reports of KMTNC/ACAP Natural Resources Conservation Program related
to forest inventory

Complete census or sample survey:

Sample ground truth survey (from representative habitat types all over ACA)

Sample area | method:

Unsupervised classification of satellite image will be done - resulting classes will be
the strata for ground truth data collection. To ensure better representation, unsupervised
classification will be done to get 3 times classes of the habitat types. Ground truth data
will be collected from all over the ACA using stratified random sampling.

Timing of observations:
Satellite image data of June-August has to be acquired because most of the vegetation
is detectable at this time.

Potential causes of bias and rules for standardization:

For time series analysis, satellite image data have to be of the same season as season has
impact on the reflectance value of vegetation. While using data from multiple sensors,
care should be taken of their spatial resolution. Different classification methods and
softwares might also affect results.

Sampling methods

Temporary or permanent sample location:

Temporary

Method for sample location:

Stratified (as per the result of unsupervised classification) random sampling. Good
interspersion and independence of samples is required.

Number of samples:

At least 10 from each strata of every Unit Conservation Offic

Monitoring requirements

Personnel responsible and time required:
= GIS Officer for co-ordination, data storage and analysis - 6 months
= Concerned UCO Monitoring Co-ordinator for field data collection - 2 weeks
= ACAP Monitoring Co-ordinator for overall co-ordination - 2 weeks

m= Natural Resource Conservation Assistants (NRCAs) of concerned UCOs -
approximately 15 man months for collecting ground truth data from all over the ACA

120 _——

Protected Area Monitoring Guidelines

Experience training necessary:

GIS Officer - Knowledge of satellite image interpretation in Erdas Imagine remote
sensing software, theoretical knowledge of satellite image interpretation in digital
environment, and Knowledge of Fragstats software and landscape indices

NRCAs - GPS operation and data recording, locating ground truth points on the
ground

License and access permission requirements:

None

Equipments required and cost:

Computer (preferably Pentium IV) with minimum of 512 MB RAM, 80 GB hard
drive with a mirror drive for backup - | set (Approx. NRs. 120,000)

Erdas Imagine remote sensing software (www.erdas.com) - | license (Approx. NRs.
800,000 on discounted price)

GPS set - 2 in each UCO (14 total) (Approx. NRs. 33,000/set)

Fragstats (version 3.0) software - 1 set (Free) (http://www.umass.edu/landeco/
research/fragstats/fragstats.htm1)

ArcView or ArcGIS software (for maps layout and production) (www.esri.com) - |
license (Approx. NRs. 500,000 for ArcGIS with Spatial Analyst)

Colour printer (for final maps production) preferably plotter of AO size (to print in
large size to show more details) - 1 set (A4 size printer - NRs. 5,000 to 300,000
depending on quality; Plotter - Approx. NRs. 800,000 to 2,000,000 depending on
size and quality)

Data storage

All the satellite data will be stored in the GIS lab computer at ACAP HQ under the
designated software domain. Ground truth and other inventory data will be primarily
stored in MS-Excel format. Proper back up of the data has to be done in digital and hard
copy (printed) format also.

Data analysis

Data will be analyzed to get:

ma Habitat classes (Refer to Annex I) using Supervised Classification method (Richards,
1999 and Lillesand et.al. 2004 will be used as reference for detailed procedures
and quality insurance). Digital elevation data might help improve the classification
result (Shrestha and Zinck, 2001). Comparison of time series habitat classes will
give spatio-temporal changes in the habitat types and quality.

ma Landscape indices (McGarigal and Marks, 1994; Rutledge, 2003) - for accounting
fragmentation (Southworth et. al., 2002).

Protected Area Monitoring Guidelines —_—_—_—_—__—___—$5—5——$— SS — $— $$ $$ $< ————————— ——————————————_ 121

Reporting procedures:

Every 5 years to the senior management (within 3 months of completion of ground truth
data collection). Report will include the result as maps, tables and graphs to show status
and changes in habitat extent, quality and fragmentation. Powerpoint presentations and
webpages will also be part of the report. Posters and booklet will be produced in Nepali
language to report the results to the local communities. Seminars will be held at local (at
least UCO level) and national level to disseminate the findings.

References

KMTNC/ACAP/BCDP (1994): Final Draft Report. King Mahendra Trust for Nature
Conservation, Annapurna Conservation Area Project, Biodiversity Conservation Data
Project. March 1994.

His Majesty's Government of Nepal (2002): Digital Topographic Base Maps in ArcInfo
format - based on 1992/1996 aerial photographs (field verification done in various years
- 2000 for ACA region) - 1:50,000 for mountains and 1:25,000 for Terai. HMG/N,
Topographic Survey Branch, Min Bhawan, Kathmandu.

Lillesand, T.M., Kiefer, R.W. and Chipman, J.W. (2004): Remote Sensing and Image
Interpretation. Fifth Edition. John Wiley & Sons, Inc. ISBN: 0-471-15227-7.

McGarigal, K. & Marks, B.J. (1994): Fragstats: spatial pattern analysis program for
quantifying landscape structure, v. 2.0. Corvallis, OR, Oregon Forest Science Lab,
Oregon State University. p. 134.

Richards, J.A. (1999): Remote Sensing Digital Image Analysis. Springer-Verlag, Berlin. p. 240.

Rutledge, D. (2003): Landscape indices as measures of the effects of fragmentation:
can pattern reflect process? DOC Science Internal Series 98. New Zealand Department
of Conservation. ISBN 0-478-22380-3.

Shrestha, D.P. and Zinck J.A. (2001): Landuse Classification in Mountainous Areas:
Integration of Image Processing, Digital Elevation Data and Field Knowledge
(application to Nepal). JAG, 3 (1), pp 78-85.

Southworth J. Nagendra H. & Tucker C. (2002): Fragmentation of a Landscape:
incorporating landscape metrics into satellite analyses of land-cover change. Landscape
Research, Vol. 27, No. 3, pp 253-269.

122 SEE. nen Gina wT olccread Area Monitoring Giiaennes:

8.6.1 Habitat Classes

SN

Habitat Type

A. Sub-tropical Zone

A.l

A.2
A.3

Broad leaved forests (Includes sub-types Hill Sal forests, sub-tropical
deciduous hill forests, semi-evergreen forests and Schima-Castanopsis
forests)

Shrublands

Grasslands

B. Temperate Zone

B.1

B.2
B.3
B.4

Broadleaved forests (Most of the patches dominated either by Quercus species
or by Rhododendron)

Conifer forests
Shrublands

Grasslands

C. Sub-alpine Zone

C.1

Cy
C.3
C.4

Broadleaved forests (Dominated in some places by Rhododendron and in
some places by Betula with fir, spruce and scrub)

Conifer forests (Pine trees associated with Taxus, Tsuga and Betula)
Shrublands

Grasslands (numerous herbs and forbs dominated by Carex spp)

D. Alpine Zone

D.1

Alpine scrub and grassland (It consists of moist and dry alpine scrub and
grassland)

E. Special Habitats

leh
eZ
BS
E.4
E.5

Agricultural fields and settlements
Rivers and Ponds

Abandoned agricultural land
Bareland

Permanent snow cover

Source: Modified from KMTNC/ACAP/BCDP (1994)

Protected Area Monitoring Guidelines. 123

~ nuzauhd maids POs

bora

muinwihete To) ‘ rie) Bove? Gece

PTR me ‘ See Goyal '

| | ya

. ie
Dien | <
'* Body eiotd es Vu (aetis A b at |
r oe ote Fad
‘8
yy =
)
° 1
; 4
“aT i i

iy SRR (iy ort Meg rie mals Beenie 2 >t Gower @ «i
: io ae a ate 09) ali shd egg ; Wi : ey
| 7 ay

Leiboat! Dirrobad | i ASS ialisope sh, Sa? gary ect ieee 4

: Awol arate - 2

a ‘hye gel) Wii Oe/lan tel Ps TN flee hd! no aes =,
> 7 ( M “~~

7 ‘ ~ ~

f -

tian Ory wi wile Al ae ae el Pe a Ste enw =

he

Piagtien tas kere
iz — a :

Photo: Siddhartha B. Bajracharya

King Mahendra Trust for

Nature Conservation
P O Box 3712 Kathmandu

Nepal
Tel: +977 1 5526571

Email: info@kmtnc.org.np
Website: www.kmtnc.org.np

Photo: Siddhartha B. Bajracharya

Biodiversity assessment and
monitoring is required to identify
the problems of parks &
evaluate priorities for

responding to them.

UNEP World Conservation! 2

Monitoring Centre

219 Huntingdon Road, Cambridge CB3 ODL
United Kingdom

Tel: +44 1223 277314

Email: info@unep-wemc.org

Website: www.unep-wemc.org

Photo: Siddhartha B. Bajracharya