Evaluation of community-based conservation approaches

Material Information

Evaluation of community-based conservation approaches management of protected areas in Uganda
Mugisha, Arthur Rwabitetera ( Dissertant )
Jacobson, Susan ( Thesis advisor )
Alavalapati, Janaki ( Reviewer )
Hildebrand, Peter ( Reviewer )
Hostetler, Mark ( Reviewer )
Chapman, Colin ( Reviewer )
Place of Publication:
Gainesville, Fla.
University of Florida
Publication Date:
Copyright Date:


Subjects / Keywords:
Agricultural land ( jstor )
Animal poaching ( jstor )
Communities ( jstor )
Crops ( jstor )
Environmental conservation ( jstor )
National parks ( jstor )
Protected areas ( jstor )
Wildlife ( jstor )
Wildlife conservation ( jstor )
Wildlife management ( jstor )
City of Gainesville ( local )
bibliography ( marcgt )
theses ( marcgt )
government publication (state, provincial, terriorial, dependent) ( marcgt )
non-fiction ( marcgt )


Unsustainable resource use and increasing human population threaten protected areas (PAs) in developing countries. Uganda is endowed with diverse natural resources that are becoming isolated in PAs. Loss of wildlife and its habitats, and conflicting relations between PA management and neighboring communities continue to challenge conservation efforts and threaten the survival of Uganda's PAs. To secure support from neighboring communities and ensure long-term sustainability for Uganda's PAs, a policy of community-based conservation (CBC) was adopted in 1988 on a pilot basis, as a management approach in some PAs. This research evaluates the effectiveness of CBC approach in comparison to the traditional PAs' management in Uganda, using the threat reduction assessment indices, group discussions and document review at seven CBC and nine non-CBC PAs. The average threat reduction assessment indices at CBC PAs (x bar = 49.0 plus or minus 12) were not different from those at non-CBC PAs (x bar = 37.96 plus or minus 21.6) (chi square = 16.0; p > 0.05). However, general trends indicate that CBC PAs tend to have higher indices than non-CBC PAs. Management approaches at both CBC PAs and non-CBC PAs mitigate fewer than 50% of identified PAs threats. Three hundred personal surveys were administered at communities surrounding three CBC and two non-CBC PAs to measure and compare residents' attitudes, and knowledge about the PAs. Results were not significantly different. Surveys with leaders of districts with CBC and non-CBC PAs showed no difference in their understanding and support of PAs. A comparison of perceived CBC benefits and costs with benefits and costs of non-CBC PAs indicated that communities from non-CBC PAs receive 30 times more benefits than communities from CBC PAs. Findings indicated that local community members are interested in utilitarian values of natural resources, not their preservation. They are interested in security of their crops, not of wildlife, and in access to resources rather than their protection. Incentives that link people's livelihood and ecosystem health are proposed for successful CBC intervention. Approaches that go beyond PA boundaries, address PA threats, involve different government departments, and strengthen PA institutions are recommended to address human welfare concerns and conserve PAs into the future. ( , )
Thesis (Ph. D.)--University of Florida, 2002.
Includes bibliographical references.
General Note:
Title from title page of source document.
General Note:
Document formatted into pages; contains x, 247 p.; also contains graphics.
General Note:
Includes vita.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
81204352 ( OCLC )

Full Text













Copyright 2002


Arthur R. Mugisha

I would like to dedicate this work to my wife Martha for her inspiration and support, and
to my children Jean and Dean for all their love, patience and encouragement.


My study program was made successful by the generous support of many

individuals and organizations, which supported and financed this research. I am

particularly indebted to the McArthur Foundation, the Wildlife Conservation Society, the

Compton Fellowship Foundation, the Beineckle Brothers Foundation and the Ford

Foundation. I cannot forget to thank the Program of Studies for Tropical Conservation,

which played a key role in developing and implementing this research project.

I wish to thank Dr. Lauren Chapman and Dr. Colin Chapman, who gave me the

inspiration, encouragement and assistance to join the University and embark on this

degree. I am indebted to my committee chair, Dr. Susan Jacobson, for her support and

encouragement during my entire study program. Special thanks go to my wife Martha,

and my children, Jean and Dean, for their patience support, and understanding throughout

my studies. Members of my committee: Dr. Janaki Alavalapati, Dr. Peter Hildebrand Dr.

Mark Hostetler and Dr. Colin Chapman also helped me greatly and I thank them

sincerely. My friends gave me confidence to finish this undertaking. I am especially

indebted to Matthew Udziela, Andrew Lepp, Karl Pomeroy, Elly Amani-Gamukama and

many others who helped me in numerous ways. My research assistants in Uganda: Mr.

Jimmy Mugisha, Mr. Kenneth Sabila, and Mr. Martin Asiimwe, did superb work

collecting the field data. The staffs of Uganda Wildlife Authority also assisted me during

the data collection. I am indebted to Executive Director Dr. Robbie. Robinson, Research

Director Ms. Apophia. Atukunda, and the field wardens, who gave me their cooperation

and support. I also acknowledge the support of my sister Joan Nyakato, my parents, and

all my friends during the fieldwork in Uganda.



ACKNOW LEDGM ENTS .............................................................................................. iv

ABSTRACT....................................................................................................................... ix


1 INTRODUCTION ...........................................................................................................

Protected Areas in Africa............................................................................................. 1
Establishment of Protected Areas ............................................................................ 2
Introduction of Community-Based Conservation.................................................... 5
Literature Review...................................................................................................... 6
Community-Based Conservation (CBC) ................................................................. 6
Factors for Success ................................................................................................ 13
Research Hypotheses and Objectives ........................................................................ 15
Hypotheses .......................................................................................................... 15
Objectives ........................................................................................................... 16

PROTECTED AREAS IN UGANDA ....................................................................... 25

Introduction................................................................................................................... 25
M eth o d s......................................................................................................................... 2 7
Threat Reduction Assessment (TRA) Technique .................................................. 27
Document Review.................................................................................................. 32
Interviews of the District Leaders; Local Council Five (LCV)............................. 32
Survey of Protected Area M managers ...................................................................... 33
R e su lts........................................................................................................................... 3 3
Threat Reduction Assessment Indices and Threats Percentage Occurrences........... 33
Documented Illegal Activities ............................................................................... 34
Responses from the District Leaders and Protected Areas Wardens..................... 35
Discussion..................................................................................................................... 36
Effectiveness of Management in Mitigating Threats to Protected Areas .............. 36
Conservation Status of the Protected Areas........................................................... 39
Threat Reduction Assessment and Illegal Activities ............................................. 39

Linkages between People's Aspirations, Threat Reduction and Survival of
Protected A reas ......................................................................................... 42
Linkages between Political Leadership and Protected Area Management ............ 48
Lim stations of the Study......................................................................................... 50
Conclusions and Recommendations.......................................................................... 52
C onclusions......................................................................................................... 52
R ecom m endations.................................................................................................. 53

NEIGHBORING PROTECTED AREAS IN UGANDA ..........................................95

Introduction ................................................................................................................... 9 5
Hypotheses and Objectives of the Study ................................................................... 99
H ypotheses .......................................................................................................... 99
O objectives ........................................................................................................... 99
M ethod s....................................................................................................................... 100
Study Sites and Respondents Selection ............................................................... 100
Survey D esign ...................................................................................................... 10 1
Administration of the Survey............................................................................... 103
R esu lts ......................................................................................................................... 10 4
Respondents' Demographic Characteristics ........................................................ 104
Attitudes toward Protected Areas and Conservation.............................. .... 106
Explanatory Responses to the Open-ended Questions......................................... 108
Responses to Knowledge Questions .................................................................... 109
Responses to Behavior Questions........................................................................... 111
Responses to Costs and Benefits Questions......................................................... 112
Comparison among Protected Areas.................................................................... 114
D iscu ssion ................................................................................................................... 1 16
Comparison of Attitudes, Knowledge and Behavior of Community Members
between CBC and Non-CBC Protected Areas........................................ 116
Conflicts between Conservation and Peoples' Needs.......................................... 116
Linkage between Community Needs and Protected Areas Values...................... 119
Association between Socio-demographic Variables with Attitudes and
K now ledge .............................................................................................. 119
A access to R esources............................................................................................. 120
Importance of Communication and Dialogue in Influencing Attitudes............... 120
Problem A nim als ................................................................................................. 121
Comparison of Attitudes, Knowledge and Behavior among All Protected Areas.. 122
Influence of Community-Based Conservation, on Attitudes and Behavior ........ 122
Benefits and Costs Analysis between CBC and Non-CBC Communities........... 123
Guiding Principles for Effective Protected Areas Management in Uganda ............ 126
Lim stations of the Study....................................................................................... 131


A W ARDEN S' SURVEY............................................................................................. 169


Introduction........................................................................................................... 171
Bokora W wildlife Reserve (Non-CBC)..................................................................... 172
Bugungu W wildlife Reserve....................................................................................... 173
Bwindi Impenetrable National Park......................................................................... 176
Karum a W wildlife (Non-CBC) .................................................................................. 178
Katonga W wildlife Reserve (Non-CBC) ................................................................... 180
Kibale National Park (CBC) .................................................................................... 182
Kidepo Valley National Park (Non-CBC) .............................................................. 184
Kigezi W wildlife Reserve (Non-CBC)...................................................................... 187
Lake M buro National Park (CBC)........................................................................... 189
M atheniko W wildlife Reserve (Non-CBC) ................................................................ 192
M gahinga gorilla National Park (CBC)................................................................... 194
M t. Elgon National Park-North (Non-CBC)............................................................ 195
M urchison Falls National Park ................................................................................ 198
Pian-Upe W wildlife Reserve (Non-CBC) .................................................................. 199
Queen Elizabeth National Park................................................................................ 200
Semuliki W wildlife Reserve (Non-CBC)................................................................... 203
Conclusion .................................................................................................................. 205

C QUESTIONNAIRE SURVEY.................................................................................. 206

D SUMMARY OF DOCUMENTED ILLEGAL ACTIVITIES..................................215

E M AP I .........................................................................................................................231

F INFORM ATION FROM DISTRICT LEADERS ....................................................... 232

LIST OF REFERENCES............................................................................................... 235

BIOGRAPHICAL SKETCH ......................................................................................... 246


Protected Areas In Africa

The importance of biological diversity as natural resource capital for economic

development and sustaining human welfare has been documented (Costanza et al. 1997).

However, the rate at which natural resources continue to be degraded and the continued

deterioration of human welfare in developing countries have been a concern at local,

national, and international levels. One of the approaches to conserve the dwindling

biodiversity under the constraints of limited funding has been through designation of

protected areas (PAs) in identified biodiversity "hotspots" (Myres et al. 2000; Bruner et

al. 2001).

Africa is endowed with rich biodiversity, ranging from rainforests to savannahs,

wetlands, and deserts. It is also a continent where its people heavily rely on natural

resources, for their livelihood, which, in combination with other factors such as

demographic, social and economic, threatens PAs (World Bank 1996). The searches for

interventions that would achieve conservation and human development goals have

therefore been of much relevance at all levels in the continent.

Uganda, in East Africa, is endowed with diverse natural resources that are

increasingly becoming isolated in PAs. Loss of wildlife and its habitats, and conflicting

relations between PAs and local people continue to challenge conservation efforts. This

research evaluates two approaches of PA management: (1) the traditional PAs exclusion

model, and (2) the community-based conservation model that involves local people.

These have been used to varying degrees to manage Uganda's PAs. Findings will

contribute to the understanding of the PAs' management challenges and

recommendations for future management.

Protected areas are defined as areas of land and/or sea set aside and dedicated to

protection and maintenance of biodiversity for cultural, aesthetic, educational, scientific,

and tourism values (International Union for Conservation of Nature [IUCN] 1994). One

of the characteristics of protected areas is that they cover a large area, often thousands of

square kilometers (although small protected areas also exist, to protect special

biodiversity). The need for a large area is to allow for migration of wildlife, and to

support diverse ecological processes. Historically, large areas would also ensure the

notion of wilderness or pristine environment as viewed by the elite from the European

countries in the late 18th century (Grove 1987; MacKenzie 1988). Another characteristic

of protected areas is the principle of human exclusion on which they were established.

Human settlement within protected areas was outlawed, causing serious conflicts with

local people in the process of establishing and managing PAs. Also, there were concerted

efforts to prevent human consumptive use of resources to preserve the wilderness status

of these areas.

Establishment of Protected Areas

Establishment of PAs in Africa has its roots in the hunting ethos and natural

history studies that were popular at the end of the 18th and the beginning of the 19th

centuries in the western world (Beinart 1994). The dire need for a pristine natural place

where manhood could be proven through hunting by colonial masters led to enactment of

detailed legislation that restricted game to the few elite, and separated human settlement

from the land that was deemed suitable for game (Makombe 1993; Adams and Infield

2001). Also, by the end of the 19th century, natural historians and hunter elites started to

sound alarms about the rate at which African game was being hunted and also decimated

by rendeerpest.

As a result of these concerns, pressure groups mostly comprised of colonial

governors, aristocrats, sport hunters, and leading landlords in the colonies began to

advocate for game preservation (MacKenzie 1988). These pressure groups wanted to

have wildlife protected in Africa for a number of reasons. First, Africa had a rich

endowment of the mega-fauna, such as elephants, rhinos, and lions that were of special

interest to hunters. These hunters had turned around to be promoters of conservation.

Second, Africa had a high diversity of game, which was also at a high density in

comparison to other colonies at that time. Third, African tribes had a high population

growth rate, and their hunter-gatherer lifestyles, plus shifting cultivation practices, were

feared to increase agricultural expansion, which was viewed as a threat to the survival of

wildlife. The colonial masters and the conservationist pressure groups felt that they

needed to do everything possible to stop the disappearance of African game. Colonial

officers, game wardens and rangers were appointed and given wide-ranging

administrative and judicial powers over both the PAs and local people that lived within or

around these designated areas.

The interests and concerns of the local African people were not considered in the

establishment of these PAs. As MacKenzie (1988) rightly argues, foreign interests and

not the interests of the African peoples influenced the legislation for wildlife management

and PAs in particular. In many incidents, creation of these PAs deprived local people of a

resource that they had been accessing for a long time, for both their cultural and

economic values (Barrow and Murphree 2001). By 1980 the United Nations listed 29

conservation provinces or biomes that were under a protection status in Africa. These

provinces had a total of 324 protected areas covering a total area of 1,046,291.71 square

kilometers out of a total of 30,244,050 square kilometers of Africa's land surface, an

equivalent of 3.5% (IUCN 1981; Almanac 2001). This approach of establishing and

managing protected areas could partly explain the unsympathetic behaviors toward

wildlife management in Africa by local people. The increasing human population and the

resultant increased pressure on land resources increase the conflicts between PAs

managers and neighboring communities.

Post-colonial African governments also continued to implement conservation

policies that excluded local communities as an approach to managing PAs in Africa

(Gibson 1999). Local communities which used to have access to wildlife resources were

excluded from the established protected area management. This exclusion was effected

through deployment of para-military trained rangers whose job was to enforce wildlife

laws by apprehending lawbreakers and either levying a fine on them, meting out

punishment or having them prosecuted in the courts of law. Local community members,

in efforts to secure their means of survival, were the majority of culprits of this wildlife

management set up, and it caused much tension and conflicts between PA managers and

the local people. Serious conflicts would arise where some of the communities claimed

traditional access rights to resources in protected areas, such as the Balabaig and the

Maasai in Tanzania and Kenya (Lane 1996) and the pastoral Bahiima in Uganda (Barrow

and Murphree 2001).

Introduction of Community-Based Conservation

In more recent years, however, the goals of protected areas have shifted from

strict preservation of big game to conservation of biodiversity, and from the closed access

for the rich few to be open to all, especially for tourism. The increased poaching of game,

famine, human tragedies and the resulting environmental degradation of the 1970s led to

the questioning of the traditional ideologies and conservation approaches of "setting

aside" PAs for preservation of wildlife. For example, Hulme and Murphree (200 la) argue

that the colonial designs were mostly based on scientific considerations and lacked a

human dimension that would integrate conservation with human development needs. In

the 1980s, conservationists, international conservation organizations and African wildlife

departments conceded that the exclusion approach of managing PAs was increasingly

becoming ineffective for a number of reasons (Jones 2001). First, the approach was

believed to be too expensive to be sustained over a long period, as it would require many

rangers to patrol vast areas of the PAs. Second, it was realized that local people are the

main offenders of wildlife laws, so if they could become the guardians of wildlife, then

Africa wildlife would have a secure future. Third, it was pointed out that local people

bear the biggest costs from wildlife by way of damaged property such as crops, loss of

human lives and lost opportunities to use PA land, yet, they benefited least from wildlife

conservation programs (Gibson 1999). This revelation, which coincided with the more

general global trend in development studies, led to the initiatives to include local

communities in wildlife management in Africa's PAs (Gibson and Marks 1995).

In essence, the approach to involve local people, referred to as community-based

conservation (CBC), aspires to turn the would-be wildlife conservation "enemies" to

wildlife guardians. The underlying assumptions of this approach, which are examined in

this study, are that if communities benefit from wildlife, are knowledgeable and

understand the importance of conservation, only then would they change their behavior to

support conservation initiatives (McNeely. 1989; Sibanda and Omwega 1996). This

community-based approach has been applauded both at local and international levels, and

supported by international non-governmental organizations, as well as donor agencies.

Literature Review

Community-Based Conservation (CBC)

The concept of CBC originated from the international concerns for conservation

and ecological balance at international levels by environmental organizations such as the

UNESCO's Man and Biosphere Program -1971 and the World Conservation Strategy-

1980 (Redclift 1984). For example, the 1982 third World Congress on National Parks in

Bali called for expansion and consolidation of PAs. A decade later, the preservation tone

had changed, and the fourth Congress in Caracas in 1992 advocated involvement of

partnerships with wide ranging interests to benefit people, PAs, and biodiversity: under

the theme, "Protected Areas for life." Also, the United Nations Earth Summit of 1992 in

Rio de Janeiro reaffirmed that PAs can play a part in sustainable development process

(McNeely 1993). In the context of this study PAs will refer to national parks and wildlife


Although interpreted differently by different people and in different contexts,

community-based conservation philosophy is based on the premise that communities

neighboring PAs are forced to exploit resources in the PAs because they lack any

alternatives for their survival. Second, because they do not legally benefit from the

existence of the PAs, they do not have enough incentives to conserve it. Third, they are

not given an opportunity to participate in the management processes of the PA, as they

are alienated from the resources, and perceived as an obstacle to conservation initiatives

(Gibson 1999).

The CBC approach strives to reverse this situation by the following: (a) changing

the roles of such communities into partners in the PA management process and (b)

demonstrating that PAs are beneficial to local communities (Gezon 1997). There are

many descriptive combinations of words that are commonly used to refer to the concepts

of community-based conservation. Such terms include integrated conservation and

development, community conservation, collaborative management or co-management.

For the purposes of this study, the term community-based conservation (CBC) will imply

programs conducted in areas surrounding PAs, with the main purpose of soliciting

support of neighboring communities for PAs, through such activities that raise

conservation awareness and seek to benefit these communities.

A number of researchers have noted that for conservation measures to succeed,

there should be a sharing of both responsibility and benefits of managing PAs between

government agencies and neighboring communities, (Adams and McShane1992;

McNeely et al. 1992; Pimbert and Pretty 1997).

Schmink (1999) points out that community-based conservation should seek to

achieve social equity as its strategy, through community participation. She distinguishes

community-based conservation from other forms of development projects by an emphasis

on resource use and regard for both the community and the environment.

Further, CBC is a term that links development of livelihood systems and

conservation of the resources in question. Murphree (1996) stresses the point of defining

community conservation from people's perspective. He defines people/stakeholders in

Africa as small-scale farmers with usufruct rights to the natural resources. These farmers,

he argues, are the de facto managers of the resources, by virtue of their remote location

where the state's management arm cannot effectively and easily access.

Other definitions of CBC programs are based on the interrelationships between

non-human and human systems. For example, Barrett and Arcase (1995) define CBC,

which they describe as integrated conservation and development, as a link between rural

development and species conservation. They use the notion of exchange of access to

natural resources by local communities for material gains, as a distinguishing character of

this approach.

In the recent past, another term relating to natural resource management known as

collaborative management (CM) has come up in the conservation literature. This is a

form of community-based conservation, which seeks to create agreements between

resource users and conservation authorities in respect of accessing and managing natural

resources in the PAs. It aims at giving more autonomy to stakeholders in terms of

decision-making (Barrow and Murphree 2001). Understanding the degree to which

communities participate is important in understanding and defining community-based

conservation. Pretty (1994), quoted in Pimbert and Pretty (1994), gives seven levels of

participation ranging from passive participation to self-mobilization.

The agenda of getting communities involved in conservation is to make nature

and natural resource conservation beneficial and meaningful to rural communities. It is

argued that, only when conservation directly benefits those who incur costs of

conservation, will rural communities take on resources management responsibility (Bell

1987). A number of conservationists believe that communities in control of natural

resource management are better managers than state institutions, and through

conservation practices, they will be able to improve their economic well being (Western

1994). The urge to get people involved in conservation initiatives came about after

realization that success of the top-down approaches to conservation over the past several

decades in the developing countries, fell short of its expectations. This limitation was

attributed to the following: (a) limited central governments' capacities to coerce their

citizens to comply with unpopular programs, and (b) unpopular government policies that

were not understood by local communities (Agrawal and Sivaramakrishnan 2000).

However, recently, the CBC approach has come under heavy critique. Brandon

(1998) raises some concerns that involving community interests in conservation has

become a slogan laden with assumptions, which has constrained creative thinking to

address park protection and biodiversity conservation problems. Also, Hackel (1999)

raises concerns about the ability of the CBC approach to change people's behavior

through economic incentives, and argues that CBC cannot generate enough benefits to

offset costs communities bear from wildlife. He further asserts that CBC does not

engender the communities to comply with conservation laws, and by itself, CBC does not

address the development needs of the people, hence it is most unlikely that local people

will support it. He concludes that CBC can only act as a means by which old-style

protectionist policies and approaches can be modified to suit socio-economic realities in

Africa. More recent research findings indicate that CBC does not meet conservation

goals. For example, Lewis and Phiri (1998) found that in Lupande, Zambia, the Luangwa

Integrated Resources Development Project (LIRDP) had failed to reduce the incidences

of killing animals using wire snares. Wainwright and Wehrmeyer (1998) also found that

the LIRDP project had failed to benefit the rural communities. In a similar tone, Wells

(1998) points out that few community-based conservation programs have enhanced

conservation of biodiversity in Indonesia. Songorwa (1999) argues that the interests of

communities are largely their own survival rather than conservation. It is therefore

difficult to meaningfully engage them in conservation. Other literature indicates that

community institutions have been eroded and compromised, to the extent that they cannot

manage to take on conservation responsibilities (Barrett et al. 2001). Other weaknesses of

involving local communities include failure of national governments to give the

communities full responsibility to manage, as well as lack of capacity on the part of the

communities (Songorwa et al. 2000).

The challenges of involving neighboring communities have been well examined

(Kiss 1990; Child 1995). One of the concerns noted is the degree to which communities

are or should be involved in the management processes. Schmink (1999) emphasizes the

importance of the degree of participation in the success of natural resource management.

She cautions about the social and political dynamism, which outsiders may encounter in

trying to implement community-based projects.

Further complications in community conservation practices revolve around the

definition of community, and the goals of conservation. By advocating for PAs that are

designed in such a way that they benefit the people living around them, McNeely (1989)

implies that community should be defined by the proximity to the PA. He emphasizes the

resumption of the custodianship of natural resources by local institutions, where people

directly depend on the resources, and where there is evidence of traditional management.

But even when the traditional management is in place, it cannot act in isolation. There

could be other actors outside such a community with divergent interests that will have a

big influence on the effectiveness of involving communities in PA management.

Naughton-Treves (1999) cautions about the challenges of re-assigning property

rights of wildlife, since it is not a landed property. Because the ownership of wildlife is

only decided when it has been killed, due to its transient and transitory status, assigning

property rights for effective management may threaten the already threatened wildlife

populations, particularly where human population densities are high. She advocates

partnerships between people's democratically elected committees and government

agencies as the best institutions to involve in wildlife management, as well as the

communities that are in proximity to a PA.

Ostrom (1997) has developed a set of tools to predict when community based

conservation will be successful in forestry resource management. These conceptual tools

are based on the attributes of the forest resources to be managed and the communities

involved in management of these resources. Following on Ostrom's work, Gibson et al.

(2000) and Baland and Platteau (1996) further considered the attributes of common pool

resources and of users that will enhance formation of self-governance. Attributes related

to CBC programs include the following: (a) attributes of the resources, such as the

resource is in good condition, management information about resource is easily, cheaply

available, and that the resource is small in size for easy management; and (b) attributes of

the users, such as dependency on the resource for a major portion of their livelihood, a

common understanding of the resource, trust, and beneficial relationships among the

users. In addition, there must be a degree of autonomy, so that the users can determine

the rules of harvesting and access, and the users should have undergone some form of

training about community organization.

Pye-Smith and Borrini-Feyerabend (1994) give a detailed analysis of success

stories of community-based conservation in Africa. They indicate a number of factors

that lead to successful implementation. In Zimbabwe, in the CAMPFIRE program, they

cite dedication of the community members to conservation. This dedication stems from

tangible benefits accruing from use of wildlife. They attribute the success to fully

empowering communities to manage, and the communities' understanding of their rights

and agreement to take on responsibilities. Some critiques, though, have noted that the

local communities do not fairly benefit from the program, as the decision-making powers

are still central at the district council level (Patel 1998). In Mauritania they point to the

user group institutions and self-governance, which promote the community's sense of

ownership of the project. Training of the local people in livestock management is also

cited as one of the successes of the livestock project, funded by the World Bank (Pye-

Smith and Borrini-Feyerabend 1994). In Uganda, with the Pallisa Community

Development Trust, they point out a clear mission statement, sense of control and

responsibility, a participatory process of problem identification, and a sense of ownership

and control by community leaders themselves. These attributes help in understanding the

conditions under which community-based conservation is effective, and point out

indicators for success in a community-based conservation programs. Policy, institutional

reforms, and capacity in policy research are also mentioned as important factors in

conducting community-based conservation (Mugabe and Clark 1998).

Factors for Success

Table I lists potential factors leading to successful community-based conservation

programs. These data are generated from researchers' work evaluating programs at single

sites and multiple sites, as well as scholarly reviews of programs in conceptual terms. For

purposes of this research, these factors are broadly categorized as follows:

Institutional factors, such as (a) authority and responsibility, (b) social factors, (c)

leadership within a community, (d) division of labor and gender, (e) membership

composition and definition of a community.

Factors that act as incentives for conservation in a community, such as (a)

financial, (b) economic, (c) cultural values, (d) drastic changes in policy or (e)

natural calamities such as hunger that threatens people's livelihood security.

Resources-related factors, including (a) knowledge of the resource, (b) type and

definition of the resource (e.g., tenure rights, size, location, and any other

information relating to the resource).

Supporting factors, including (a) policy and legislative factors, (b) political

support, (c) governmental and NGO support (d) social support, (e) technical

support e.g., collaboration with technical and professional people such as the

researchers, (f) Social support, (g) financial support.

Management factors, such as conflict resolution and negotiation skills.

In spite of this burgeoning literature on CBC evaluation it would be unwise to

make generalized policy prescriptions about CBC approaches in PAs' management.

Particular case studies put emphasis on different factors with no mention of other factors

that would enable comparison of cases or test alternative hypothesis. Besides,

methodologies of carrying out these case study reviews are often hurried, and project

implementers are the ones who provide most of the information. As such, many of the

findings could be differently interpreted. This kind of evaluation approach can be biased

to the expert opinion of each study's author and fail to take into consideration real

management challenges on the ground. In addition, each country has its own unique

conditions that may influence implementation of conservation programs differently. Even

within a country, there are variations from PA to PA. This research study compares two

approaches to PA management, and hence, provides empirical evidence of how different

approaches work under different conditions in Uganda.

In Uganda, wildlife is legally owned by the state, for the benefit of its citizens

(Uganda Wildlife Statute No. 14 1996 sect. 4 (1)). Like in any other African country,

serious legislation for protected areas establishment started in the early 1930s in the form

of forest and game reserves. Establishment and management of PAs was on the principle

of "setting aside" and "exclusion" of the people. This modus operandi of establishing and

managing PAs continued up to the late 1980s with the declaration of more PAs.

Presently, there are 20 PAs (excluding forest reserves) that cover about 14% of the total

land surface area of the country (Mugabe and Clark 1998). In the late 1980's community-

based conservation programs were initiated on a pilot basis, as an alternative

management concept to the traditional top-down approach in order to improve PAs'


The ultimate goal of CBC programs is to achieve behavioral changes that are pro-

conservation. In the CBC approach it is assumed that if local people are aware and

benefit from the PAs, then their behavior will change to support PAs' management

programs (Mordi 1991). Community-based conservation programs in Uganda can be

classified into three main components. The first is aimed at directly benefiting local

communities around PAs and is directly linked with wildlife conservation. It entails

sharing of monetary revenues generated from tourism in a PA regulated access to natural

resources (mostly plant materials) from the PAs, and providing employment

opportunities to neighboring local residents. The second component also aims at

benefiting local communities by contributing toward social infrastructure development. It

is indirectly linked to conservation, and entails provision of public goods to communities

neighboring PAs, such as schools, dispensaries, and health clinics. The third component

aims at generating and promoting environmental awareness as well as creating capacity at

local level for responsible behavior toward PAs. It entails education and extension

programs, institutionalization of environmental stewardship in community government

set up and decentralization of natural resources management to grassroots levels.

Research Hypotheses and Objectives


This research was designed and carried out to test three main hypotheses that

form the underlying assumptions of the CBC approaches.

Hypothesis I: Protected areas where CBC programs are being implemented will

have reduced threats to their natural resources, compared to PAs without CBC.

Hypothesis II: Protected areas where CBC programs are being implemented will

have more support from their neighboring communities than those protected areas

without CBC. The term support implies that members of the community will demonstrate

greater (a) knowledge and understanding about natural resources, (b) positive attitudes

toward the protected area, and (c) pro-environmental behavior.

Hypothesis III: Where CBC programs are implemented, the benefit/cost ratio of

the protected areas to the neighboring communities will be higher than those protected

areas without CBC programs.

To determine the effectiveness of CBC approaches in achieving conservation

goals in Uganda's PAs, the above hypotheses were tested by comparing CBC with non-

CBC PAs. Group discussions with PAs managers and community leaders were conducted

at 16 PAs to identify main threats and assess the effectiveness of management

approaches, using a threats reduction assessment (TRA) technique (Salafsky and

Margoluis 1998). Seven of the PAs practiced CBC, and nine practiced the traditional top-

down type of management. A structured questionnaire survey to measure knowledge,

attitudes, and behavior toward PAs was also administered to 60 respondents at each of

five PAs, three of which practiced CBC and two that did not. Another semi-structured

questionnaire was administered to district leaders in whose districts the PAs were located.

Responses of the district leaders where CBC was being undertaken were compared with

those responses of district leaders where CBC was not being practiced. Also, secondary

data were collected by reviewing documents, such as patrol, monthly and annual reports

compiled by wardens, from both sets of PAs for comparison.


The effectiveness and efficiency of community-based management as a means of

integrating divergent needs of wildlife conservation and human development is explored.

Specifically, this study set out to accomplish the following objectives:

* Analyze the effectiveness of the CBC approach in achieving conservation

objectives in comparison to the traditional approaches of PA management through

the threat reduction assessment techniques.

* Analyze if community conservation approaches, compared to traditional

management approaches, increase benefits and reduce costs of a PA to the

neighboring communities.

* Evaluate the effectiveness of the current CBC initiatives in imparting knowledge,

influencing attitudes and behavior, though investigating differences in attitudes

and behaviors of neighboring communities toward PAs, between communities

subjected to CBC programs and those with the traditional PA management


* Investigate socio-economic and institutional conditions under which CBC

programs succeed or fail.

* Generate guiding principles for effective protected areas management and

biodiversity conservation in Uganda.

Table 1-1: Factors leading to successful community-based conservation programs Based on reports evaluating 35 sites
Category Factors leading to success 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Institutional Authority and responsibility. x x
Spirit of cooperation in community. x x x x
Leadership in a community. x
Division of labor and gender. x x
Composition of a community. x x
Respected social / traditional x x x
institutions. x
Cohesive institutional set up. x x x x x x
Recognized and legitimate x x x
institutions. x x x x x
Ambitious community members. x x x x
Democratic processes.

Table 1-1 Continued
Category Factors leading to success
Incentives Financial incentives
Economic incentives
Cultural values
Social prestige
Religious values
Calamities (natural or man-made)
Perceived or real livelihood risks
Resources related Information about a resource
Secure resource / land tenure
Size and location of the resource
Dependency on the resource
Resource use not just conservation
Resource not irreparably damaged
Policy/legislation Community friendly
legal framework

Technical support NGO support
Government support
Social support
Research support
Adequate Financial support

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

x x
x x x
x x

x x
x x x x

x x

x x

x x

x x

x x x

x x

x x

Table 1-1 Continued
Category Factors leading to success 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Management Freedom of community to develop their own
technology and institutions. x x x x
Low human populations density. x x x x x x
When participating is a learning process. x
Need long time frame. x x x
Low cost conflict resolution mechanism x
Monitors are users or accountable to users x
Autonomy to make rules and regulations of x x x
Co-management arrangements arrived at x x x
through a consensus manner.
Agreement on conservation objectives by all x x
Training of community members x x x
Integration of biological and socio-economic x x
needs of the community.
Linkages between extension, government x x x
regulations and community.
Opportunity for experimental management. x
Awareness rising among the community x
Key to authors: 1=Veit et al. (1995); 2 = Pimbert (1994); 3 = Pye-Smith and Borrini-Feyerabend (1994); 5 = Murphree (1996); 6 = Ostrom (1997);
7 = Bodmer (1994); 9 = Hill and Press (1994); 10 = Donovan (1994); 11= Otto and Kent (1994); 12 = Wells (1994); 13 = Tengue and Associates
(1994); 14 = Nkana (1994); 16 =Little (1994); 17 = Lynch and Alcorn (1994).

Summary of reviewed evaluation cases
Field of Focus of evaluation


Power sector


Self-help organization for
rural development in
Rushinga district

Why are power projects
in Sub- Saharan Africa
(SSA) less successful
than in other regions.
Impacts of community
conservation on local

Singini Agriculture How the Farmer Support
and in the home Program (FSP) has
Rooyen lands of increased effectiveness in
(1995) South Africa agricultural development.
Provide guidelines for the
future course and
development of
agriculture. FSP as an
integrated approach to
rural development.
Ortmann Economic Agricultural support and
et al. evaluation structural change policies
in agriculture in Kwa-
Zulu Natal

Table 1-2:

Role of
Responding to

Provide access
to archives, and

Jassat, et
al. (1990)



Recommendations/ Findings.

Participants in
74 members of the
self help group
and 27 non-
members in the
district and Project
26 SSA countries/

Park staffs, parish
Mgahinga Gorilla
National Park.
Farmers in the
black homelands

Households from
the wards in Kwa-



Too much pressure on the project
Coordinator from the donor and the
community leadership. The poorest of the
poor people in Rushinga district did not fully
participate in the project.

The World Bank should play a more
supervisory role in management of the power

The program is not addressing the real issues.
The benefits from the park are not enough to
compensate opportunity costs of the park.

Land tenure policies are needed for further
development. FSP strategy has a far wider
development impact at substantially lower
cost than large-scale centrally managed
settlement projects and state farming

The impact of FSP appears to have been
small. The costs of implementation were high

Table 1-2 Continued
Author Field of
Scott et Levels of
al. (1973) living in

Patel H.

program in

Focus of evaluation

Standard of living in two
communities of Anogi and
Ballos on the island of

Improved welfare of local
communities from
utilization of wildlife.

Taylor Wildlife Income accruing from
(1993) Utilization wildlife utilization in
Nyaminyami district,

Participants in
The inhabitants of
the two villages.

Local chiefs,
officials, and
Only the researcher

Recommendations/ Findings.

Some of the international indicators were
not relevant to the two villages. Others
were obsolete, as they had been overtaken
by events.

Role of
Provision of
Using the
inhabitants to test
the indicators of
development as
stated by the
United Nations
Institute for

Only to be
informed of the

That the popularized campfire program
was forcibly evicting people to create
room for wildlife was not benefiting the
people whose land was taken by the
program and was not participatory.

Wildlife utilization is a feasible
investment. The dividends from wildlife
do not accrue to the landowners that bear
the cost s of conservation. There is
mismanagement of the program at the
local level.

Author Field of Focus of evaluation Participants in Role of Recommendations/ Findings.
evaluation Evaluation participants

Metcalf Development.
(1996) through

Namara Community
(1998) Perspectives
on Natural
resource mgt.
In Uganda

Promoting sustainable
development so that
pressure is taken off
protected areas.

Existing community
initiatives in natural
resources management.

Evaluation team,
project staff, Govt.
and SAID
officials and some
community leaders

The consultant and
the leaders of the
projects visited.

information and
ideas about the

information and
participate in

Develop a bottom-up institutional
development strategy. Support the
implementation of the parks mgt. Plans.
Promote and monitor multiple use
programs. Promote farmer participatory
research. Continue funding to the 3rd
It is important to keep dialogue between
communities and govt. institution. There is
a need for the govt. to remunerate
participating members of the community.
People need permanent access to
resources, and not to be limited by access
agreements. Economic incentives to
participating members are important for
appreciation of natural resources
conservation. Destruction of property by
wild animals from a PA is contrasted to
the denial of access to resources in the PA.
People see law enforcement as unfair and
a mockery of the realities on the ground.
Lack of tenure security is a hindrance to
community-based conservation.

Table 1-2.


Table 1-2 Continued
Author Field of Focus of evaluation Participants in Role of Recommendations/ Findings.
evaluation Evaluation participants

Namara Interactions Post community 380 randomly Providing Land use had changed from dominantly
and between L. conservation project selected families information, pastoral to mixed farming, cattle with
Infield Mburo evaluation of Socio- around L. Mburo cultivation. Land-less households were
(1998) National Park economic baseline, national park, and still at large. Perceptions about the park
and the the investigator. had not changed but park/people relations
neighboring had improved. Law enforcement rangers
communities were corrupt. Poaching levels had
reduced. Benefits were perceived to have
been got in those areas where the program
was active. Need to influence land use
outside the park to ensure dispersal areas
for wildlife. Education programs should be
intensified and expanded. Formulate
policies to manage problem animals
outside the national park.
Kazoora Economics of Assess the cost Investigators and Reviewing Develop a policy framework for
and Ray community effectiveness of the project staff. literature and community conservation and a strategic
(1997) conservation community conservation carrying out action plan.
in L Mburo program in Uganda interviews. Explore more cost effectiveness of
National Park. carrying out community conservation
programs. Identify how best to achieve
objectives at the least costs. Develop
indicators against which to measure the
effectiveness of CC. Budget for CC as a
basic park function. Integrate CC into the
overall planning process of UWA.

Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy



Arthur Rwabitetera Mugisha

May 2002

Chairperson: Dr. Susan K. Jacobson
Major Department: Wildlife Ecology and Conservation

Unsustainable resource use and increasing human population threaten protected

areas (PAs) in developing countries. Uganda is endowed with diverse natural resources

that are becoming isolated in PAs. Loss of wildlife and its habitats, and conflicting

relations between PA management and neighboring communities continue to challenge

conservation efforts and threaten the survival of Uganda's PAs.

To secure support from neighboring communities and ensure long-term

sustainability for Uganda's PAs, a policy of community-based conservation (CBC) was

adopted in 1988 on a pilot basis, as a management approach in some PAs. This research

evaluates the effectiveness of CBC approach in comparison to the traditional PAs'

management in Uganda, using the threat reduction assessment indices, group discussions

and document review at seven CBC and nine non-CBC PAs. The average threat reduction

assessment indices at CBC PAs (x = 49.0 12) were not different from those at non-

CBC PAs (x = 37.96 21.6) (X2 = 16.0; p > 0.05). However, general trends indicate

that CBC PAs tend to have higher indices than non-CBC PAs. Management approaches

at both CBC PAs and non-CBC PAs mitigate fewer than 50% of identified PAs threats.

Three hundred personal surveys were administered at communities surrounding

three CBC and two non-CBC PAs to measure and compare residents' attitudes, and

knowledge about the PAs. Results were not significantly different. Surveys with leaders

of districts with CBC and non-CBC PAs showed no difference in their understanding and

support of PAs.

A comparison of perceived CBC benefits and costs with benefits and costs of

non-CBC PAs indicated that communities from non-CBC PAs receive 30 times more

benefits than communities from CBC PAs. Findings indicated that local community

members are interested in utilitarian values of natural resources, not their preservation.

They are interested in security of their crops, not of wildlife, and in access to resources

rather than their protection.

Incentives that link people's livelihood and ecosystem health are proposed for

successful CBC intervention. Approaches that go beyond PA boundaries, address PA

threats, involve different government departments, and strengthen PA institutions are

recommended to address human welfare concerns and conserve PAs into the future.



Community-based conservation (CBC) approach has been lauded as a better

approach to protected areas (PAs) management in Africa (Pimbert and Pretty 1994;

Western 1994; Barrow 1997; Schwartzman et al 2000) than the traditional top-down

management approach. The later approach has been blamed for failing to provide

protection to PAs (Hulme and Murphree 2001 b).

Recently however, the community-based conservation approach itself has been

heavily criticized as having failed to achieve much needed conservation results (Hackel

1998; Songorwa 1999). As the debate continues about whether CBC is effective,

empirical data are needed to understand if indeed CBC works and under what conditions,

in order to propose appropriate policies for PA management and biodiversity


In Uganda, both the traditional top-down management and the community-based

conservation approach have been operating concurrently for a decade in different PAs.

This research evaluates and compares the ability of protected area management to

mitigate threats, at CBC and non-CBC PAs, as a proxy measure of conservation success.

Sixteen PAs in Uganda were examined to determine whether CBC-PAs have better

addressed threats than non-CBC PAs as far as their conservation status is concerned.

Seven of the PAs have been practicing CBC for more than ten years, while nine have not

been exposed to CBC approaches (Map 1). This study tests the hypotheses that: PAs with

CBC programs have reduced threats in comparison to non-CBC PAs. Specific objectives

were as follows:

Learn from primary stakeholders (neighboring communities and PA

managers) what they perceive as threats to the survival of respective PAs.

Evaluate and compare the effectiveness of the PA management, using the

threat reduction assessment technique, in mitigating the identified threats to

respective PAs.

Document concerns and aspirations of neighboring communities and how

these relate to threat reduction and the future survival of the PA.

The study used a threat reduction assessment (TRA) technique (Salafsky and

Margoluis 1998) to collect data from the sixteen PAs, to assess the effectiveness of PA

management to mitigate threats to the PAs. In a participatory manner, impending threats

to PAs were identified and PAs' management evaluated based on how effective the

management approaches had been in addressing the identified threats.

The TRA technique is a cost-effective way of defining and measuring

conservation success for different interventions. It has practical advantages of using data

collected through simple techniques relevant to local primary stakeholders of resources

for which evaluation is being done (Salafsky and Margoluis 1998). It is relevant to local

people's perceptions and understanding, and therefore the evaluation objective becomes

clearer from a local standpoint. As a research method, this was the first time TRA

technique has been used to compare the effectiveness of management approaches in PAs

in Uganda, but it has been widely applied in the Asian/Pacific regions to evaluate

community-based conservation programs (Salafsky and Margoluis 1999).

Group discussions, were used to explore socio-economic and environmental

issues that could impact PA management. Three additional methods were used to provide

additional data and to triangulate the results of the TRA technique (Wholey et al. 1994).

A review was conducted, examining documents from the sixteen PAs, to catalogue types

of illegal activities and their occurrences. Also, a questionnaire was administered to the

highest political office in the district, Local Council Five (LCV) to measure the support

and appreciation of the PAs at the district level. Lastly, a questionnaire was administered

to the respective PA managers to understand their views about the conservation status of

their PAs.


Threat Reduction Assessment (TRA) Technique

Evaluation of the effectiveness of PA management in mitigating identified threats

to PAs was conducted through threat reduction assessment techniques (TRA) (Salafsky

and Margoluis 1999). This was carried out by organizing group discussions at each of the

16 PAs. The discussion groups were composed of representatives from: communities at a

parish level, PA management staff representatives especially from the research, law

enforcement, and community conservation departments as applicable, conservation non-

governmental organizations (NGO) staff members working at the PA, and research

institutions where applicable.

To get representative views from neighboring communities to the PAs, caution

was exercised in the participant selection process. First, all the parishes around the PAs

were identified and communities grouped into socio-economic groups according to their

livelihoods, e.g. pastoralists, cultivators, and fisher people. Second, a reconnaissance

visit was made to local council two (LC II) chairpersons who are the political heads of

parishes, to discuss the purpose and objectives of the intended plans to hold group

discussions with representatives of communities neighboring respective PAs. To keep the

number of participants to manageable sizes especially for bigger PAs, some participants

represented more than one parish, but where the number of parishes was few, each parish

was represented. The final decision of who would represent community interests from a

community was left to LC chairpersons with their councils. The ability of an individual

to: (1) express community views in a group, (2) understand local peoples' problems, and

(3) have knowledge of aspirations of the communities, were emphasized as criteria for

participant selection. The main focus of participant selection was guided by the need to

have a diversified group whose members could easily communicate, to avoid redundancy

during discussions (Alreck and Settle 1995). The number of community representatives

ranged from 15 to 20 depending on the size and the number of parishes neighboring the


The assessment is based on three main parameters of the environment: habitat

integrity, quality, and ecosystem functioning. Local people are assisted to internalize their

thinking about these parameters, and think back, to make an evaluation and value

judgment. The key principle of TRA as an evaluation tool is that if threats to a PA are

mitigated, then the management will have succeeded. Conversely, if the threats are not

mitigated, the PA management approach will have failed. It is therefore imperative that

the assessment group is able to identify threats to the PA, and with facilitation, estimate

the degree to which these threats have been reduced as a measure of PA management

success. A threat reduction index (TRA-I) is then used to evaluate the effectiveness of a

management approach.

First, participants were led through a "brainstorming" exercise, whereby in a

relaxed atmosphere, they were encouraged to think out loud and write down what they

thought were threats to the existence of the PA being discussed. Everything that

participants cited as threats were written on a flip chart to help participants visualize and

reflect on the identified issues. Threats, for purposes of this research were defined as any

human related phenomena that could be avoided, either by the PA management or any

other management agency that negatively affect the existence of the PA in question and

are viewed as the inverse of opportunities (Salafsky and Margoluis 1999). Natural

phenomena such as natural fires or earthquakes were not considered to qualify as threats.

Second, to assist participants to focus their thinking about habitat integrity,

quality, and ecosystem functioning, these threats were ranked according to their relative

importance. This was achieved by considering the speed at which the threats could

destroy the PA, their intensity of destruction, and the area that they could affect. A

ranking scale of one to five was used throughout the exercise as it was found to be

convenient and acceptable to participants. Five represented the maximum score for each

area of ranking, and a score of one was the minimum. A total sum score was computed

after all the threats were scored.

Third, a consensus building exercise was used with the group to assess the extent

to which the PA management had mitigated each threat. All participants were given

approximately five minutes to think about each threat and evaluate independently, to

what extent the management approaches had addressed a specific threat. Scores were

assigned on a percentage basis. If a threat had not been addressed at all, management

would score zero. Where management had fully mitigated a threat, the score would be

100 percent.

During the initial stages of the exercise, it was observed that personal assessment

and scoring could be influenced by some of the most vocal participants. To overcome

this, a means of writing one's score and keeping it secret from other participants until all

had finished scoring was devised to guard against such influences. When there were large

disparities in the scores, a discussion would be conducted to ensure that an objective

consensus was reached. After the scoring and ranking exercise, total ranking scores were

multiplied by the percentage of the threat met to get a raw score for each threat. Dividing

the sum of the raw scores for each threat by the total possible rankings of all the threats

and multiplying by 100 computed the threat reduction index (TRA-I): (TRA-I = Y Raw

Scores / Y Possible Rankings X 100) (Salafsky and Margoluis 1999). This means that the

higher the index, the more successful management has been in mitigating the threats.

This procedure was carried out for all sixteen PAs, to have a meaningful comparison of

the indices. To find out if threat indices differ between CBC and non-CBC PAs, a t-test

was carried to compare the indices at CBC with those at non-CBC PAs.

Some PAs had unique threats due to their locations and other development plans.

For example, planned development projects such as hydroelectric power dams were a

threat to some PAs, but did not occur in others. Also, due to insecurity from guerillas and

banditry activities in some parts of the country, some PAs identified insecurity as a threat

and others did not. Such threats greatly affect the final TRA index as they would score

zero, since the PA management could not mitigate them. To streamline the comparisons

across all the PAs, two TRA indices were computed for PAs that had unique threats such

as those mentioned above, that a PA management had no legal mandate to mitigate. The

first index includes the unique threats, while the second one excludes them. The index

without the unique threats is used for comparison purposes with other PAs while the one

with all threats is used to indicate the relevance of the management approaches in light of

the identified threats.

The discussions focused on the PA management covered a number of issues

relating to communities, natural resources management, and general environmental issues

(Table 2-1). The researcher and his assistants moderated these discussions. Further

discussions concentrated on the lifestyles in the communities, and how people interacted

with their environment. Ethnic composition, socio-economic services (such as schools),

community management of their natural resources (such as water, trees, soils, pastures),

and economic activities were also discussed. Environmental and natural resources issues

were discussed, focusing on the environmental changes that had taken place over the last

five-years, concerns communities had about these changes, and what communities were

doing about these concerns as a coping strategy. Discussions also focused on wildlife

issues, both outside and inside the respective PA. Participants were encouraged to use

their intuition and local knowledge of the area, based on their experience, to discuss the

future of the PA and wildlife in light of the socio-economic and political changes in their

communities. Lastly, the groups specifically discussed the threats to the PA and assessed

how effective the management had been in addressing the identified threats.

Document Review

Document review and analysis was carried out to gain an understanding of the

past management challenges, successes, and failures at the 16 PAs. The Uganda Wildlife

Authority (UWA) documents from headquarter archives, as well as warden's field reports

from specific PAs, project evaluation reports, and routine reports were reviewed to get an

understanding of the challenges for the PA management. Specific documents reviewed

include all of the following sources available for each PA:

Warden's monthly reports from 1990s to 2001

Warden's quarterly reports from the 1990s to 2001

Project reports from organizations working in collaboration with UWA in PAs

UWA Annual reports from 1990 to 2000

Interviews of the District Leaders; Local Council Five (LCV)

In Uganda, the local government is organized along a five-tier system: village,

parish, sub-county, county, and district level. At each level, there is a nine- person local

committee headed by a chairperson. The CBC initiatives target all levels of the local

councils, to enlist their support in PA management in the country. To evaluate the

impacts of these initiatives, a semi-structured questionnaire was administered to district

leaders, (local council five) to determine their understanding about and support to PAs in

their respective districts (Appendix F). Responses were analyzed using a contingency chi-

square with the SPSS computer software, and results from the district leadership where

CBC is implemented are compared with the responses from districts where CBC is not


Survey of Protected Area Managers

A questionnaire was designed to get information about the conservation status of

the PAs as evaluated by the PA managers (Appendix A). The questions were formulated

based on the possible indicators for a successful PA management. Responses were

categorized into factors for successful management and percentages were used to

compare CBC and non-CBC PAs.


Threat Reduction Assessment Indices and Threats Percentage Occurrences

Tables 2-2 through 2-10 show threats to non-CBC PAs with respective TRA

indices and Tables 2-11 through 2-17 show threats to CBC PAs, and the TRA indices.

Tables 2-18 and 2-19 show percentage occurrence of identified threats at CBC and non-

CBC PAs respectively. The average threat reduction assessment indices at CBC PAs (x

= 49.0 12.0) were not significantly different from those at non-CBC PAs (x = 37.96 +

21.6) (5 = 16.0; p > 0.05) (P = 0.382). However, 71.0% of the CBC PAs had more

than 50% of their threats addressed, compared to only 33% of non-CBC PAs.

In total, 20 threats were identified at all the 16 PAs. Local poaching occurred

more than 80% at both CBC and non-CBC PAs. Poor community relations occurred more

frequently at non-CBC (66.7%), than at CBC PAs (57.1%). Problem animals occurred

more frequently at CBC PAs (85.7%), than at non-CBC PAs (33.3%), and unclear

boundaries were more frequent at non-CBC (55.6%) than at CBC PAs (28.6%). Table 2-

18 shows the comparison of CBC and non-CBC threat reduction indices, Table 2-21

shows percentage occurrences of the identified threats at both CBC and non-CBC PAs,

Tables 2-22-2-24 show Wardens' responses to the survey about the status and

management issues of their respective PAs, and Table 2-25 shows responses from the

districts leaders. Results from group discussions at all the 16 PAs are presented in

Appendix B. These data indicate that communities are aware of environmental

degradation going on in their areas. They are concerned about the increasing scarcity of

natural resources such as firewood, medicinal plants and building materials, plus a

general deterioration in their welfare. Coping mechanisms vary from place to place and

for different problems. A common trend regarding wildlife among all the 16 sites is that

big game wildlife species have limited opportunities to survive into the future outside

PAs, and even the possible opportunities will depend on the ability to address problem

caused by wild animal such as crop raiding. The survival of PAs themselves will depend

on a number of factors such as: alleviation of poverty in the rural areas, intensified

education programs, and family planning measures among others.

Documented Illegal Activities

Appendix D presents documented illegal activities, and how the PA management

is handling them. The common illegal activities include: local poaching, bush fires,

snares, illegal grazing and encroachment. The common manner of handling these

illegalities both at CBC and non-CBC is to arrest the suspect and take them to courts of

laws, and there are few incidents when leaders of local people are involved in resolving

such conflicts. These results support findings from the group discussions method that

identified threats to the PAs, although more threats were identified than were

documented. Figures 1 to 5 indicate the percentage occurrences of these threats at Queen

Elizabeth National Park, Murchison Falls National Park, Lake Mburo, Bwindi, and

Mgahinga National Parks over 10 years. It was not possible to get enough data from the

reports for the other eleven PAs. Either the respective PA management did not compile

them, or due to filing problems, they could not be traced.

Responses from the District Leaders and Protected Areas Wardens

A contingency chi-square analysis of the responses from the district leaders with

CBC did not show any significant statistical differences to districts without CBC, with

regard to political support to PAs, involvement of local administration in PA

management and beneficial aspects of the PAs to local people (Table 2-25). Responding

to queries about the benefits of PAs, 71% at CBC and 67% at non-CBC agreed that PAs

provide environmental benefits, while 29% at CBC and 33% at non-CBC PAs disagreed.

About employment benefits, 57% at CBC and 100% at non-CBC agree that PA provides

employment benefits and 43% at CBC disagree. Regarding what should be done to

improve the interaction between the PAs and the communities, 14% at CBC and 67% at

non-CBC, suggested improvement of communications for conservation awareness, and

14% at CBC and 33% at non-CBC suggested that the PAs should be more transparent.

Responses from the PA managers (Table 2-24) indicate that 86% of the CBC PAs

have management plans while 14% have already embarked on the process to making the

plans. In contrast, 33% of non-CBC PAs have management plans, 44% have no plans,

while 22% are in the process of making plans. Only 22% of the non-CBC PAs have well-

organized tourism program, 33% have poorly organized tourism program, and 44% have

no tourism program developed. In contrast 86% of CBC PAs have well-organized

tourism program and 14% have poorly developed programs.


Effectiveness of Management in Mitigating Threats to Protected Areas

Environmental degradation and the precarious state of PAs in Africa have

generated much debate among academicians, developers, and conservation practitioners.

While some scholars have argued that increasing human population is the major factor

that threatens survival of PAs (Struhsaker 1998), others counter argue that it is the

alienation of local rural people from nature (Pimbert and Pretty 1997; Schwartzman et al.

2000). Other threats to PAs have been documented as human activities such hunting and

setting wire snares to trap animals (Gibson and Marks 1995; Cuar6n 2000), charcoal

production and unsustainable land use practices (e.g. Seddon et al. 2001). Macro

economic policies and market failures, poverty, and unsustainable agriculture also have

been documented as the main causes of environmental degradation and subsequent threat

to PAs (Barbier and Burgess 2001).

Although the threats to PAs have been studied and documented, (Lacy 1987;

Orians 1993) measures taken to manage and conserve PAs and biodiversity pay very little

attention to such threats. A total of 20 threats were identified as affecting the existence of

the PAs in Uganda. Some of the threats were common ones, such as poaching of game,

pitsawying, and denied access to resources, however, the data shows additional external

threats such as highways through PAs, hydroelectric power stations, and kraal making

activities in Karamoja region. Comparison of the TRA indices indicates that there is no

statistically significant difference between CBC and non-CBC PAs. Both management

approaches do not fully mitigate PAs threats. However, general trends seem to indicate

that CBC PAs tend to have higher indices than non-CBC PAs. For example, as shown in

Table 2-18, 71.0% (five out of seven) CBC PAs, have indices above 50%, while only

33.0% (three out of nine) non-CBC PAs have indices greater than 50%. This suggests

that CBC PAs may be more successful in mitigating threats than non-CBC PAs.

Differences may not be statistically significant because of the high variances, e.g. ranging

from 28-58% in CBC PAs, and 2-71% in non-CBC PAs.

Results indicate that management approaches at both CBC PAs and non-CBC

PAs do not address all threats to the PAs. The average TRA-I for non-CBC is 30.03%

when all the threats are considered. However, when those threats that the management is

not able to address are excluded, the index average increases to 37.96%. Similarly,

average TRA index for CBC PAs is 40.05% when all the threats are considered, and

when the threats to which the management is not able to address are excluded the average

TRA index increases to 49.0%. It is important to note that both CBC and non-CBC PAs

have average scores of less than 50%, implying that PA management at both CBC and

non-CBC PAs only mitigates about a third of the PAs threats. Although both

management approaches fall short of addressing all PA threats, trends in the data

suggested that threats such as bush burning, pitsawying, encroachment and unclear

boundaries had been better mitigated at CBC PAs than non-CBC PAs.

Tables 2-19 and 2-20 show the occurrence of the identified threats at both CBC

and non-CBC PAs. With few exceptions, such as cattle kraal construction, highways, and

corruption, threats are common at both CBC and non CBC PAs. The most common threat

across all the PAs was local poaching of game, occurring in all the non-CBC PAs, and

86% of CBC PAs. Denied access to resources occurred 14.3% at CBC PAs, but was not

mentioned at all in non-CBC PAs. At CBC PAs there are arrangements with communities

to have controlled access to resources within the PAs. At non-CBC PAs, such

arrangements are lacking, but local people stealthily access resources, risking being

detected and punished. The fact that denied access to resources came up as a threat from

CBC PAs could signal the unhappiness about the regulatory mechanism to access

resources by the communities. On the other hand, at non-CBC PAs the illegal access

detection chances could be so low that local communities do not feel that they are being

denied access after all or it may not be an issue.

Corruption was mentioned at only 11.1% of non-CBC PAs, and 28.6% of CBC

PAs. The district leaders expressed similar sentiments, asserting that PA management

needs to be transparent. The threat of interference from politicians seems to be larger at

CBC PAs, than non-CBC, potentially indicating increased political interests in PAs that

generate revenue. This sentiment was corroborated by PA wardens' surveys. The threat

of encroachment was lower at CBC PAs (57%) than non-CBC (77.8%), but it was still


Poor relations between PAs and communities were cited as a threat at more than

half of the non-CBC PAs (66.7%) and CBC PAs (57%). Lack of sufficient funds to invest

in community development, and failure to provide compensation for damaged property

were highlighted as the main sources of poor relations between the communities and PAs.

At CBC PAs where there are community projects with support from NGOs, poor

relations were associated with conflicts with wildlife. Problem animal threats occur in

85.7% of CBC PAs, but only occur at 33.3% of non-CBC PAs. The CBC programs have

promoted environmental awareness, but the challenge of sustaining livelihoods still

dictate increased demands and expectations from communities. Communities still view

PAs as a source of food and land for cultivation, and the CBC programs have not

adequately addressed such expectations and demands from community members in

Uganda (Adams and Infield 2001).

Conservation Status of the Protected Areas

From the wardens' responses, only 44% (two of nine) non-CBC PAs reported that

they had a management plan, five out of nine (56%) reported that they had a management

objective. Non-CBC PAs that reported they had management mission, also had a higher

TRA index. More than 80% of CBC PAs had management plans and 14% were in the

process of making the plans, and all had management missions. Although boundaries of

some CBC PAs were contested, they were all known and marked. In non-CBC PAs five

out of nine (67%) reported that the boundaries were not known and were not demarcated.

The Status of natural resources in non-CBC PAs was reported to suffer from uncontrolled

use by local communities. Only 29% (two out of seven) CBC PAs reported the status of

natural resources to be fully protected.

Threat Reduction Assessment and Illegal Activities

Comparing TRA indices with actual documents of illegal activities substantiates

the range of threats to the PAs. The review of PA documents found that there were

recorded illegal activities such as local conversion of trees into timber and vegetation

destruction at PAs, which reported full protection. In Kibale, eight types of illegal

activities were recorded during the past 7 years, 1994-2000. The most common problem

in 1994, 1995, 1998 and 2000 were wire snares set for bush meat. This is in agreement

with what Koojo (2001) found while recording wire snare occurrences in Kibale National

Park. Cattle grazing was a major problem in 1996, and game poaching was recorded in

1996, 1997, 1999 and 2000.

In Queen Elizabeth National Park, most frequent problems were charcoal burning

and vegetation destruction in 1996, and game poaching, prevalent in 1998. In Mgahinga

National Park, the most common illegal activities were wire snares and vegetation

destruction, recorded in 1995, 1997 and 2000. In Murchison Falls National Park, game

poaching was recorded in almost all years, with pitsawying and encroachment recorded

in 1992, 1993 and 1997. In Lake Mburo, the most occurring illegal activities were

vegetation destruction and game poaching.

The illegal activities that were documented in respective PAs were all identified

as threats in the TRA process, except in Mgahinga where snaring was not mentioned.

However, group discussions identified even more threats, than recorded illegal activities.

The reason for this is that not all the illegal activities at a PA are detected. Even when

they are detected, they may not be recorded or followed up. This is confirmed by the

wardens' responses indicating that at times illegal activities are not recorded nor followed


At non-CBC PAs, records of illegal activities were not available, or non-existent.

This highlights a weakness in record generation and keeping on the part of the non-CBC

PAs. All five PAs whose files were found are CBC PAs, and the 11 whose file could not

be located are non-CBC. It should be noted that even in CBC PAs, not all the data for all

the respective years could be found, indicating lack of organization in records generation,

keeping and management. These findings further reinforce results from other methods

that CBC PAs still face similar institutional problems as non-CBC PAs.

Although the district leaders reported that all PAs have political support from the

district leadership, the Wardens feel that it is only central government that supports PA

management and little support is offered from local government or local people.

These data show that the current PA management approaches, be it CBC or non-

CBC, do not fully address threats to the survival of PAs in Uganda. Part of this failure

can be explained by examining institutional arrangements for managing PAs. Wildlife

management in general and PAs' management in particular have been perceived as an

exclusive biological profession, a policy approach that was not understood by the local

people (Agrawal and Sivaramakrishnan 2000). The institutional set up for managing PAs

was therefore designed to exclude people, who were viewed as "the problem" (West and

Brechin 1991). However, people-related issues affect PA management and wildlife

conservation in developing countries like Uganda, which have no institutional set up or

mechanisms of addressing these threats (Hyden 1998).

Current protected area management cannot address land degradation outside the

PAs, or human population increase, yet, such issues pose threats to the survival of PAs as

revealed by TRA technique. Increasing human population and lack of buffer zones have

over-stretched the law enforcement mechanism that has remained the same since the

western model of nature conservation was introduced in Africa. This has rendered follow

up of illegal activities by rangers ineffective. From all corners, PAs are being besieged,

by settlement attempts and changing land use patterns. Current PA management

approaches may not address a settlement issue next to a PA nor can it keep a 24-hour

surveillance over the activities that are being carried out in such a settlement.

Linkages among People's Aspirations, Threat Reduction and Survival of Protected Areas

During group discussions, participants were asked to reflect on environmental

changes that had taken place over the past five years. On this issue, most discussions at

both CBC and non-CBC PAs cited reduced or unpredictable rainfall patterns, lack of

clean water, and decreased soil fertility as the main concerns. As to how the communities

were responding to such concerns, the discussions revealed that both communities at non-

CBC as well as CBC PAs are expecting help from government and non-governmental

organizations. A few discussions revealed that the communities had already taken

initiatives such as de-silting water dams to address their concerns. In general however,

there seems to be lack of community initiatives to address the identified concerns as a

community. A number of factors could cause lack of such initiatives but ones that were

gleaned from this research point to lack of traditional community institutions and

decision-making authority due to multi-ethnic communities. Only seven PAs, out of 16:

Mt. Elgon, Pian Upe, Bokora, Matheniko, Kidepo, Bwindi and Mgahinga are surrounded

by one or two dominant ethnic groups. Other PAs irrespective of whether they are CBC

or non-CBC, are surrounded by more than three ethnic communities. This raises

institutional concerns in the management of natural resources.

Most communities around PAs lack well-organized community structures that

would enable the community to act in a cohesive manner. Government local councils are

the only formal institutions that are politically instituted. Office bearers of these

institutions are voted for, and to keep such offices, they have to ensure that they do not

antagonize future voters (Hulme and Murphree 2001 la). These strategic considerations in

leadership can compromise the rationale of the community leadership in meeting natural

resources management goals. Group discussants also indicated that community members

lacked power to implement the decisions they wanted to. In Bugungu, community

members wanted to stop immigrants from the Democratic Republic of Congo (DRC).

However, central government officials thwarted their decision and illegal immigrants

have continued to increase as illegal fishing increases. The ambivalent stand of different

communities toward the concerns about the environment therefore, could be partly

explained by such lack of cohesion, and leadership in the community as well as lack of

empowerment to make decisions. A community that is characterized by lack of

leadership and decision-making powers could take long to take on collective

responsibility to support PA management, (Ostrom 1997; Baland and Platteau 1996).

Results from the group discussions also reveal that community concerns about

environmental changes are more or less uniform at both CBC and non-CBC PAs, and are

primarily based on personal interests, yet PA management is in public interest. Results

indicate lack of strong community institutions that are able to articulate community

interests and to negotiate for their roles and responsibilities based on such interests and

needs. This lack of communication can lead to mistrust among different stakeholders. For

example where natural resources were mentioned as a concern, such as decreasing

wildlife populations, it was in relation to loss of game meat as a source of protein. In

Bwindi, where CBC programs were pioneered, people expressed a high degree of

skepticism about NGOs involvement in the PA management. The group indicated that

they fear a loss of community lands to the PA as wildlife population increases inside the

PA. The 'mzungus' (slang for 'white' people) have a plan to acquire more land for the

PA, through buying off poor local farmers and refusing to compensate for destroyed

property." There are concerns that the aid support being given for conservation from

international NGOs is a way of "bribing the government, so that the white people can

take over the land and set it aside for conservation." Whereas conservation values are

eco-centric, peoples' values for managing natural resources are anthropocentric

(Struhsaker 1998). This finding negates one of the basic assumptions of the CBC

approach, that if neighboring communities were granted access to resources, they would

develop pro- environmental behaviors that support PAs' management as well as improve

their welfare (Western 1994). These findings indicate that both communities from CBC

and non-CBC PAs were far more concerned about their welfare than the status of the PAs

and natural resources.

In the design of CBC approach, it is often assumed that local communities would

be able to buy into and accept the non-consumptive conservation values of natural

resources and PA management that hinge on aesthetic, scientific, moral, and recreational

values as promoted by western cultures. Findings from the group discussions indicate that

community members are interested in utilitarian values of wildlife. It is often observed

that communities at lower rungs of Maslow's hierarchy of needs (Maslow 1998), and are

still struggling to meet their basic needs may not easily change their values from

utilitarian to moral or recreational (Mordi 1991). The concerns expressed in the group

discussions, imply interests of the communities, which are in conflict with PAs and

conservation objectives. If CBC is embarked upon before harmonizing such conflicting

values of different stakeholders, conflicts could intensify and result in raised

expectations, mistrust and despair.

Discussions about the future of wildlife outside PAs revealed that big game

species outside PA boundaries had little chance of survival. The main reason for this

assertion is that wildlife is destructive to people's property and threatens human life. In

addition to this, the government does not provide any compensation for the damages due

to wildlife. People have taken it upon themselves to ensure that wildlife outside the PAs

is destroyed. This is being done through both destruction of wildlife habitats and active

hunting of any game that is found on private lands. Although participants in the group

discussions pointed out that problem animals such as bush pigs and baboons are still

abundant outside PAs, the general belief among wildlife managers is that wildlife

numbers have been drastically reduced. Records from aerial surveys of 1996 confirm this

(Lamprey and Michelmore 1996). In most discussions, participants kept reflecting on the

fact that wildlife is no longer available outside PAs, except for baboons, vervet monkeys

and hyenas, species that are commonly cited as problem animals. At Mt. Elgon, the group

indicated that the only wild animal species whose population had decreased both within

and outside the PA was the black and white colobus monkey. This species is highly

valued for its skin, which is used in cultural ceremonies. These comments indicate that

people harbor negative perceptions about wildlife, which they perceive as problem

animals, but they cherish those that they derive consumptive values such as meat or skins

for their utilization but not necessarily for their conservation.

Asked about the future of the PAs themselves, discussions again revealed that

there are some PAs that may not continue to function as conservation areas with viable

wildlife populations, such as Bokora and Matheniko. Other PAs were believed to have

only a conditional future. A salient issue in these conditions is that of problem animals.

Discussions came up with different suggestions ranging from fencing off the PA and

establishing a compensation scheme for those who lose property to wildlife, to reducing

numbers of "vermin" in the PAs. Some suggested that community sensitization programs

around PAs on the importance of wildlife and PAs be increased. Other suggestions

included increasing the amount of money for revenue sharing and giving such monies to

individual landowners on whose land wildlife resided, rather than giving it to the

community as a whole. Planting buffer crops such as tea that are unpalatable to wildlife,

next to the PAs also was suggested during the discussions. In Lake Mburo, participants

suggested that the PAs should be managed as a communal wildlife area entirely managed

by local communities.

These findings indicate that the survival of both CBC and non-CBC PAs is still

precarious, in light of increasing human populations, changing land use patterns and

decreasing human welfare. At the time of gazetting PAs between the early 1930s and

1960s in the colonial era, there were fewer people and less pressure on land. PAs were

planned to have buffer zones around them that mitigated negative human impacts on the

PAs and vice versa. Today, most of the PAs lack any buffer areas, and the "front line" of

human activities is right to the border posts of PAs. Hence, the conflicts between wildlife

and communities are becoming bigger issues today than they were before. Coupled with

this reality of increasing conflicts are negative perceptions resulting from denied access

to resources. As one participant observed in the CBC PA discussions, You stop our

animals from going to the park, but when your animals come to our land, we are punished

for "chasing" them away". This finding is in agreement with assertions made by Hulme

and Murphree (2001b) that although CBC approach is a step in the right direction, it has

made a very limited improvement. Solutions to such kind of issues lie beyond current

approaches to PA management, and call for formulation of realistic policies that

adequately address conservation challenges in a changing world.

Ambiguous resource tenure, which was expressed in terms of unclear boundaries,

can affect PA management (Lynch and Alcon 1994). Uncertainty of boundary location

creates a vacuum and high insecurity of resource tenure outside the PAs. Local people

and PA managers alike keep guessing about whether or not a piece of land is within or

outside the borders of a PA. On the part of community members, it takes one daring

person to take a risk and find out if he/she will be punished for settling on a piece of land

whose ownership is unclear. In the likely event that PA rangers do not apprehend

him/her, other community members will follow suit. On the other hand, if a PA

management decides that a piece of land is inside the borders, the bona fide occupants

risk being evicted not withstanding the number of years of residence on the land. For

such reasons, communities neighboring PAs lack the security to invest and properly

manage the resource for long-term benefits. In Bwindi, the discussions revealed the fear

of the people that the PA borders will be expanded in the future. Yet ownership is central

to the determination of issues of access, distribution, and sharing of benefits from

resource conservation.

Civil unrest also leads to internally displaced people who end up using resources

without any secure resource tenure. Such displaced people lack alternative income, and

they resort to activities such as charcoal burning, poaching and unsustainable agricultural

practices as a means of securing their livelihood. As they expect to go back to their

homes once peace is restored, they have no incentives to invest in the temporary land

holding. They have no tenure rights to the land, and their main interests are to satisfy

their short-term needs without respect for the future. Issues pertaining to resource tenure

have traditionally been perceived as non-related to wildlife and PA management. It is

therefore not surprising that neither the CBC, nor the traditional top-down approaches, in

the management of the PAs in Uganda, address such an issue, which continues to threaten

the survival of PAs.

Linkages between Political Leadership and Protected Area Management

Responses from the district leaders (Table 2-25) showed no significant differences

between districts with CBC and those without CBC PAs, and indicate that the status of

the resources and the PAs are generally in a good condition. This is in contrast to the

wardens' responses, which indicated that over 60% of the non-CBC PA boundaries were

not demarcated, and were feared to be encroached, while at CBC, 14% of the PAs had

unknown boundaries with conflicts. Response frequency distributions for all districts

showed a well-informed awareness and interest in PAs' management at district levels.

For example, only 28.6% of the district leaders indicated that they did not know how

illegal cases are handled in CBC PAs but at non-CBC 50% said that they did not know.

Over 80% at CBC and 67% at non-CBC PAs agreed that they get involved in resolving

conflicts, while all the respondents agreed that PAs enjoy central government political

support. When asked about their understanding of the wildlife legislation in the country,

at CBC PAs, 29.0% said that their understanding was good, 43.0% said it was neither

good nor bad, while only 29.0% said it was bad. At non-CBC PAs, in contrast, 83.3%

rated their understanding as good, and 17% rate the understanding as neither good nor

bad. Respondents from CBC PAs, 71.4% rated the PA-community interactions as good,

and only 14.3% rated it as bad. At non-CBC PAs, 83.3% rated the interaction as good.

Whether such support and understanding promotes conservation or not can be

gleaned from the results of the TRA technique that identified political interference as a

threat at a quarter of both CBC (28.6%) and non-CBC (22.2%) PAs. Much as the district

leadership may be knowledgeable about the PAs, if there is lack of effective

environmental governance, the use of such knowledge may end up just as political

rhetoric, without practical application (Okoth-Ogendo and Tumushabe 1999). Tumushabe

(1999) advocates for civil institutions to assist local people in taking on responsibilities of

managing their environment. The civil institutions would help in the mobilization of local

communities, protect them from political manipulation and promote good environmental


Regarding the involvement of local people in PA management, 43% at CBC and

83% at non-CBC of district leaders agreed that local people were involved, and 57.1% at

CBC and 17% at non-CBC disagreed. About who was actually involved, 86% at CBC

and 100% at non-CBC indicated that it was government officials who are involved. This

indicates that at CBC PAs, district leaders feel the level of local involvement is not

enough and they need to get more involved.

Of note, there is misunderstanding and lack of clear guidance as to what

"participation" means and how to "participate" in the PA management programs. During

the group discussions, some participants equated local participation and awareness with

access to community development funds from donor supported community programs. In

one of the interviews with the district leaders, a sense of mistrust was detected between

the PA management and the district administration. The districts leadership believes that

PAs have huge budgets, but they refuse to support district-based programs. This mistrust

is further revealed in the recommendations by the district administration on how to

improve the PAs-community relations. One of the recommendations is the need for PAs'

management to be more transparent, and to implement its commitments. This finding

agrees with Slocum and Thomas-Slayter (1995) who assert that what passes as

community participation, could be lip service intended to maintain the system. Such

relations create confusion in who should be responsible for what and at which level.

Songorwa et al. (2000) argue that for CBC to be effective, governments and wildlife

institutions should relinquish some or even most of its powers to the local people. This

would empower the local communities to make their decisions and enable governments

to play a facilitation, coordination and educational role.

Results from the district leaders' survey, on one hand, could signify how CBC has

become political rhetoric, without addressing local people's needs and concerns. On the

other hand, the findings highlight expectations of local politicians from PA management.

One district leader pointed out that as a district, they will not allow the PA management

to evict their people from the PA, and lamented that the district should be involved in the

planning and budgeting for the PA. Also the results highlight a latent conflict and power

struggle between the local political leaders and institutions managing PAs. This struggle

could be a result of the policy gap for institutionalization of CBC programs leading to

uncertainty of expectations from different stakeholders, and/or the unwillingness of the

wildlife institutions to relinquish power to communities that are not well organized.

Limitations of the Study

In general, TRA is a useful tool for evaluating conservation interventions. It is

easy to understand and use by those implementing management programs, and is

sensitive to changes in the entire project area. However, it has specific weakness as an

evaluation tool. It does not directly measure the biodiversity conserved, instead it

indirectly measures the threats met. There is lack of consistency and ambiguity, the

results could be subjective, and the scores for management performance are not directly

linked to a specific biodiversity (Margoluis and Salafsky 2001).

Specific limitations of this study included involvement of appropriate

stakeholders. Biases could have occurred in the process of selecting community

representatives to participate in this evaluation exercise. Such participants may

potentially be ignorant about local people's situations, although the community

leadership selected them. Even if knowledgeable, they may decide to give information

that they think will please the groups facilitator or other members of the group.

It was difficult to control for the flow of information between some group

discussions among sites. The communities from CBC PAs at times share information

with communities from non-CBC PAs. It is therefore possible that information about

what was discussed at CBC PAs could have been known at the next PA before the

research team arrived. This could lead participants to say what the previous group had

stated, and bias the results. The interviews with the district leaders could have suffered

from "expectation biases". Some district leaders may have felt that this information could

be used to bring more projects to their districts, so they would give answers that were

intended to show how supportive they were toward PA and wildlife management.

The sample size of CBC and non-CBC PAs was small and the variances in the

responses large. This could explain why the results could not show statistical


Conclusions and Recommendations


In conclusion, the research findings indicate that the current management

approaches fall short of addressing PA threats. Nevertheless, a trend in the data suggested

that threats such as bush burning, pitsawying, encroachment and unclear boundaries had

been better mitigated at CBC PAs than non-CBC PAs, indicating a potentially more

successful approach to conservation than the traditional top-down approach. It can

therefore be concluded that CBC has marginally performed better, as an approach to PA

management than the traditional top-down approach although it does not address all the

threats to PAs. Since the current traditional top-down management and the CBC

approaches, do not fully mitigate threats to PAs in Uganda, it is imperative that more

pragmatic approaches that go beyond the PA boundaries and which address PA threats be

pursued to address human welfare issues and conserve PAs into the future. The

recommendations do not only target PA management, but are made in a broad sense to

encompass other institutions, as the threats to PA have proven to be multi-dimensional in


The results indicate that the management approaches mitigate less than 50% of

the PA threats. There is a need to design CBC programs that target specific PA threats as

identified by stakeholders. For example, CBC programs need to work with communities

neighboring PAs, to foster land use practices that promote conservation of biodiversity

outside PAs. The integrity of PAs will depend on conservation of natural resources such

as soils, wetlands and vegetation, outside these areas. The current CBC design that puts

emphasis on community infrastructure development such as schools and health centers

may have little positive impact on mitigating PA threats. A community with schools and

other infrastructure built under the auspices of CBC, will continue to degrade PA

resources, if residents cannot meet their basic needs.


Unmarked PA boundaries were identified as a threat to PAs, because they abet

PAs encroachment by local people who either feign or are ignorant about boundaries

location. It is therefore important to ensure that all the boundaries around PAs are

properly surveyed and marked. Land that is not under the PA management but is adjacent

to the PAs should then be properly managed by ensuring that its ownership rights are

defined and its use properly planned. This will also go a long way to minimize conflicts

between PAs and communities, and avoid open access type of land tenure rights.

Formulate community policy guidelines that clearly define functions and roles of

local political leaders in PA management. One of the factors for successful CBC

implementation is the devolution of authority and responsibility (Veit et al. 1995). One of

the findings of this research is that there are suspicions between district leadership and

PAs' management. Such suspicions revolve around securing funds and unclear means

and ways of getting involved in PAs' management. A policy developed in a participatory

manner with local people and district leaders, will help to build trust and understanding

between the district leaders and the PAs managers. It will also provide guidelines and a

common understanding about "what" participation is and "how" it should be

implemented. This will also provide a basis for the government institutions responsible

for PA management such as Uganda Wildlife Authority (UWA) in deciding whether or

not to promote CBC approach in PA management. For CBC to succeed, local people

should be given full responsibility to manage, and there should be a full commitment on

the side of the government institutions to work toward such a status. Where such

commitment is not a possibility, efforts must be re-directed to strengthening law

enforcement capacity of the PAs, with an emphasis on public relations.

The traditional top-down approach was designed to protect mostly wild animals

from local hunters. The current threats to PAs have changed their dimensions, from being

simply subsistence poaching to human survival and improving welfare of rural local

people. These issues are exacerbated by global and macro-level development and

economic policies. There is a need to involve different government departments, such as

that of agriculture, and population. Government institutions need to be strengthened for

them to play their role, such as research, monitoring and policy formulation.

Lack of manpower to manage PAs was one of the threats identified during the

group discussions. The existing PAs' management is over-stretched in relations to the

implementation of PAs' management activities. There are fewer rangers for an increasing

human population, and this poses a threat as identified by these data. There is therefore a

need to develop strong government institutions to take on the challenges of PA

management. Capacity building for government institutions to manage natural resources

takes a long time and resources. It entails identification of individuals with personal

interest and commitment, short and long-term training, on-job training as well as

developing appropriate policy and legislative instruments. It also entails hiring of an

appropriate number of staffs and remunerating them appropriately and according to the

tasks they perform.

Another large threat identified through this research is that of armed poaching,

which often escalates into bloody conflicts. Protected areas in Uganda are negatively

affected by uncontrolled possession and use of firearms. There are different armed groups

such as the local defense units, the national army, vigilantes as well as armed rebels who

use PAs to launch their attack on government. This creates insecurity for the natural

resources within the PAs, the staffs, and visitors alike. Such a threat needs concerted

government efforts to ensure security for the survival of PAs. Also since local people do

not flee from areas of conflicts like expatriate conservationists, the payoffs of training

local people would be great, in times of escalating conflicts.

The district survey revealed that PAs are not adequately valued for environmental

non-monetary benefits, as a service to improve human welfare. Over 30% of all the

district leaders do not believe that PA provide environmental benefits, although over 80%

concur that PAs provide educational values. There is therefore a need to expand the

ongoing educational programs to target local politicians to promote intangible benefits of

PAs. It is also apparent that although PAs are well politically supported at the central

government level, these data reveal that the support is weaker at the local political level.

A well designed, directed educational program would contribute to increasing

appreciation of PAs by local politicians, and help to garner their political support.

Many conservation biologists have tended to view conservation and PA

management, as a biological professional discipline. However, as Oneka (1996) argues,

designing a PA management plan is like a jigsaw puzzle, whereby you have to consider

different parameters that are related in different perspectives. Protected areas issues are

multidisciplinary in nature, and they need pragmatic, integrated and coordinated

approaches. The use of TRA methods and other techniques for evaluating the

effectiveness of programs, will continue to be important in monitoring the progress and

test new conservation initiatives in the future.

Table 2-1: Discussion guide for group discussions

Discussion Issues/Topic
Community issues

Community economic
Resources issues

Future of the PA and

Discussion evaluations

Table 2-1: Discussion guide for group discussions

Self introductions led by facilitator to act as "ice-breaker."
Issues that could have an impact on the management of natural
resources in their areas as well as the neighboring PAs such as
community composition, population density and economic activities.
Community skills for natural resources management explored
Activities that could have impacts on natural resources and PA, such
as charcoal burning etc.
Environmental changes that took place over the past five years,
concerns about these changes and what the communities were doing
about these concerns
In light of the past experience and current social, economic and
political changes, reflections on the future of the PA and wildlife.
Identify threats to PA through a brain storming exercise, rank these
threats according to the area, intensity and urgency. Assess the
effectiveness of the PA management in mitigating the threats over
the past five years, by awarding scores out of 100. Later, compute
the TRA-Indices
At the end of the discussions, participants were asked to evaluate the
content and quality of the discussions for the benefit of the
facilitators in conducting other discussions at other PAs.

Table 2-2: Threats to Bokora wildlife reserve
Threat Area Intensity Urgency Total Threat Raw TRA I
Ranking met Score
Insecurity* 5 5 5 15 0% 0
Poaching 5 3 4 12 0% 0
Demand for land for 5 3 4 12 0% 0
Bush fires 5 4 4 13 0% 0
Illiteracy* 5 5 4 14 0% 0
Cattle rustling* 2 3 5 10 0% 0
Shortage of rangers 5 3 5 13 5% 0.65
TOTAL 32 26 31 89 0.65 0.73%
TRA without threats 38 1.71%
* Indicates threats that the PA management has no mandate to handle.
The PA management had not done anything to the insecurity threat, and hence it scored
0%. Insecurity calls for immediate government intervention to provide security by
stopping cattle rustling. Illiteracy, cattle rustling and demand for land for cultivation are
also threats beyond the mandate of the PA management, although they were identified by
the participants as threats.

Table 2-3: Threats to Bugungu wildlife reserve
Threat Area Intensity Urgency Total Threat TRA- I
Poaching of game 4 1 3 8 80%
Bush Burning 5 4 4 13 20%
Deforestation 2 2 3 7 60%
De-gazzeting 1 5 5 11 0
Boundary Conflicts 1 1 1 3 70%
Increased human population 2 3 3 8 0%
Increased demand for resources from the 3 2 3 8 80%
Negative attitudes toward the reserve 1 2 2 5 60%
Bad relations between community and 2 2 2 6 50%
PA management
TOTAL 19 17 21 58 47.8%
TRA without threats 39 71.0%
De-gazzeting was not scored because the benefiting community could not be objective in
their ranking, and the PA management could not do anything about it, as it is the role of
the central government.
Poaching scored 80% because management had contained poaching, but according to
participants, they are using brutal methods such as killing people to stop poaching.
Increased demand on reserve resources scored 80%, for it was agreed that through
education programs and ranger patrols, illegal access on PA natural resources had been
greatly reduced.
Increased human population 0% -this is not a mandate of PA management, so nothing is
being done about it.

Table 2-4: Threats to Karuma wildlife reserve

War/ insecurity in the area*
Settlement and cultivation
Local conversion of trees into
PA Staff corruption
Unclear boundaries
Ignorance about values of the
Poor PA-community relations
TRA without the threats





Threat met TRA I



Table 2-5: Threats to Katonga wildlife reserve
Threats Area Intensity Urgency Total Threat Met TRA I
Misuse of political influence and abuse of 1 1 4 6 5%
Unclear boundaries 1 1 1 3 10%
Lack of awareness on the values of the 5 5 5 15 50%
Encroachment 5 5 5 15 65%
Tribal conflicts between Banyankore and 1 1 1 3 0%
Human population increase* 2 1 1 4 0%
Bush burning 3 3 5 11 10%
Poaching 3 5 5 13 80%
Little manpower in the reserve 3 3 5 11 30%
Totals 24 25 32 81 33.0%
TRA without the threats 74 36.0%
Lack of awareness scores 50 .The management has been serious on the encroachment
and has done some sensitization campaign in some areas. They still lack manpower.
They have established gates, which control people who come in the PA.
Encroachment scores 65. There are no new cases of encroachers, other than those
who were there by 1995. Those who are caught are taken to court for prosecution.
The management stays in the PA, but despite the gates, wrong people get in. The
encroachers have not yet been evicted.
Poaching scores 80%- Management has tried hard to trap and arrest hunters all over
the PA. Dogs and spears are not allowed in the PA.

Table 2-6: Threats to Kidepo valley national park
Threat Area Intensity Urgency Total Threat TRA I
Bush fire 5 4 5 14 50%
Local poaching 4 3 5 12 80%
Insecurity/armed poaching 5 3 5 13 10%
Encroachment 3 2 5 10 80%
Poor /inadequate management 3 3 5 11 40%
Poor relations between PA and 2 3 5 10 50%
TOTALS 22 18 30 70 50.4%
Bush fire 50%, management tried to control internal fires but has failed to control
external fires. The roads, which used to act as firebreaks, have overgrown. Also
internal roads have over-grown and park management sensitization of the
communities has not been so frequent.
Insecurity 10%, the management lacks adequate machinery/support weapons. They at
times leave cattle rustlers to pass through the park. There is lack of communication
between the communities and the park.
Poaching 80%, there are now serious patrols and have reduced poaching to low

Table 2-7: Threats to Kigezi wildlife reserve
Threats Area Intensity Urgency Total Threat TRA-I
Poaching 2 2 5 9 50%
Deforestation (charcoal and 3 3 3 9 40%
Insecurity armed poaching* 5 5 5 15 0%
Corruption among the PA 4 3 3 10 70%
Uncontrolled bush fires 5 5 5 15 30%
Poor PA / community relations 5 5 5 15 50%
TOTALS 24 23 26 73 37.1%
TRA without threats 58 46.7%
* Insecurity scores 0. The national army is handling, but not the PA management
* Corruption scores 70%. Corrupt staffs are often expelled. The bosses are not corrupt,
but the sub-ordinates are. However, in some cases, they release the culprits after
receiving bribes
* Bush burning scores 0. There are very few rangers to fight any fires, and even the few
who are they are not hard working. They do not have any facilities to fight fires.
Despite the sensitization carried out fires are still a big problem
* Poor communication scores 50. There are some workshops and meetings, which are
held regularly for the communities. They have few radios. But there are often delayed
responses to the issues discussed in the meetings. There is a need for more radios,
motorcycles and better roads

Table 2-8: Threats to Matheniko wildlife reserve
Threats Area Intensity Urgency Total Threat TRA-I
Demand for land for cultivation* 4 1 2 7 0
Armed poaching 5 5 5 15 50
Existence of traditional shrines 5 3 5 13 0
within PA
Non-involvement of locals in PA 5 3 5 13 0
Increasing human population* 5 3 4 12 0
Demand for grazing land 5 4 3 12 0
De-vegetation for kraal and fence 5 5 5 15 0
Bush burning 5 5 5 15 0
TOTALS 39 29 34 102 7.2%
TRA-I without threats 83 9.0%
* Armed poaching scores 50%. They were serious before, but they then instantly
stopped, giving reasons of lack of manpower. There used to be much trade in ostrich
egg, but it has now reduced due to PA management involvement in controlling the
trade. But reduction in ranger numbers has led to increased poaching.
* Burning scores 0- they have not tried to address the issues at all same as vegetation
destruction for kraal and fence construction. The PA management has no control over
the warriors and yet the nature of the insecurity warrants that one has to build a strong
fence for cattle protection against wild animal and cattle rustlers

Table 2-9: Threats to Pian Upe wildlife reserve
Threat Area Intensit Urgency Total Threat TRA-I
y met
Hunting/poaching 2 5 4 11 55%
Encroachment-PA boundaries 5 2 1 8 0
Bush fires 5 5 2 12 30%
Deforestation 4 4 4 12 50%
Lack of awareness about PA 5 5 5 15 60%
Insecurity* 5 5 5 15 0
Political changes* 3 4 5 12 0
TOTALS 24 28 25 77 26.8%
TRA-I without the threats 50 41.3%
* Bush fires 30%, PA management arrests suspects who set fires, but they are not
* Illiteracy lack of awareness of values of the PA 60%. Physical presence of the
managers sends messages about the importance of the PA. Managers also mobilize
people and carry out education programs. However, they have not yet reached all the
communities around.
* Insecurity, nothing is being done about this threat, for it is beyond the management of
the PA.
* Political changes, nothing is being done about this threat, because it is not the
responsibility of the PA
* Poaching 55 %, poaching has reduced a bit, but it is still taking place under cover
* Deforestation 50%, the management is educating people about the dangers of
deforestation, but they have not covered the entire community


Table 2-10: Threats to Semuliki wildlife reserve

Bush burning
Unclear boundaries
Petroleum prospecting*
Political influence
Land slides
Diseases (anthrax)
Migration of animals*
Human population
TRA index without *




23 23



50 105

Threat Met TRA-I


* Insecurity 50%, PA

staffs have trained rangers, to fight rebels, with the cooperation of

UPDF. Rangers track down thieves who steal people's cows. PA gives material
support in fighting rebels such as information and maps etc
* Bush Burning 65%, education and sensitization programs about fire have been carried
out. Collaboration with local leaders has been intensified. Operations have been
intensified in monitoring, prosecution of wrong doers, etc. The only problem is that
the staffs are still very few.
* Landslides 30%, mobilization for tree planting and environmental awareness has been
carried out.

Table 2-11: Threats to Bwindi impenetrable national park
Threats Area Intensity Urgency Total Threat TRA-I
Bush burning 2 3 5 10 70%
Poaching 1 1 2 4 40%
Local conversion of trees into timber 1 2 3 6 60%
Illegal use of resources (firewood, poles 1 1 1 3 50%
bean stakes)
Human population pressure 5 5 5 15 0%
Poor relations between PA (rangers) and 4 5 5 14 50%
Corruption (bribing PA management for 3 4 4 11 40%
Local conversion of trees into timber)
Totals 12 16 20 63 39.8%
TRA without threats 48 52.3%

* Poor relations score 50%-there are programs for creation of institutions such as
Community Protected Area Committees (CPAC). They have also allowed resources
use such as herbs, seedlings and bee keeping. The communities benefit from presence
of vehicles of the PA. However when communities are called upon to help, such as in
fire fighting they are not rewarded. Some of the PA staffs are harsh to the community
members. They delay in conflict solving between the PA and the communities. They
do not involve local leaders in deciding on the penalties to give to wrongdoers
* Corruption (Bribing) scores 40%-Corrupt staffs are only transferred. Some people are
also suspended. Management has conducted sensitization programs on roles and
responsibilities of community members in the management of the PA to discourage
community members to bribe PA staffs. However, some staffs are still corrupt and
even when they are reported they do not get punished. If the community reports a
person to the PA, staffs takes no action, or when he is arrested by the community
members and taken to the PA staffs, the culprit is released immediately.
* Bush fires scores 70% -they have allowed sharing of resources, such as bee keeping.
People know that they will lose their bees if they bum the forest. Projects like schools
which people appreciated have helped to reduce the incidences of fires. Sensitizations
of dangers of fires and their sources such as smoking have helped to reduce the
incidents of bush fires. However, more sensitization is still needed, especially the
fires that come from personal gardens. Also people should e rewarded every time they
help to put out a bush fire. The management should help the people around the park
to utilize their lands next to the park so that it can act as a firebreak.


Table 2-12: Threats to Kibale national park
Threat Area Intensity Urgency Threat TRA-I
Problem animals. 4 5 5 10%
Poor relationships between the park and 5 5 5 60%
neighboring communities.
Illegal access to resources (poaching, pitsawying). 5 1 5 50%
Failure to adopt extension techniques among 5 5 5 40%
Misuse of park resources by FACE employees. 1 1 5 10%
Bush burning. 1 1 5 30%
Insufficient monitoring due to little manpower of 5 5 5 30%
the park.
Lack of appreciation of park values by 1 5 5 40%
Human population increase*. 3 1 3 5%
Poor harvesting mechanism in softwood. 1 5 5 20%
TOTAL. 31 34 48 35.0%
TRA without the threats 38.0%

Table 2-13: Threats to Lake Mburo national park
Threat Area Intensity Urgency Total Threat TRA-I
Lack of permanent water sources 3 3 5 11 0%
outside the park*.
Less community support to the 2 3 2 7 50%
protected area (due to problem
animals, restricted access to water
and pasture).
Political influence/interference. 1 1 1 3 60%
Unclear boundaries. 1 2 1 4 90%
Poaching. 2 2 2 6 50%
Encroachment. 3 1 1 5 90%
Bush burning. 2 1 1 4 40%
Diseases. 1 1 1 3 5%
TOTALS. 15 14 14 43 42.2%
TRA-1 without threats 57.8%
Less community support to the park scores 50- management has sensitized people on
conservation issues. They have contributed toward community projects such as
schools, roads and dispensaries. The PA management is friendly to the communities.
But the projects are not continuous, and people still demand committees to bridge the
PA management and the communities
Poaching scores 50-hunting has reduced, but they lack manpower to carry out
constant monitoring. They lack a rear guard the communities who are not reporting
the hunting. They should be liaison offices around the PA where to report
Encroachment scores 90%-there are good bye-laws that promoted collaboration with
neighboring communities, when it comes to fining encroachers. The rangers have
been strategically deployed around the PA in parts that are liable to encroachment.
There is disciplined management otherwise they would have been bribed. However
there are some illegal settlers who have not yet been removed. There should be
continuous education programs about the dangers of encroachment.

Table 2-14: Threats to Mgahinga gorilla national park
Threats Area Intensity Urgency Total Threat TRA-I
Poaching (killing of animals by 1 1 5 7 80%
the Batwa).
Bush burning (fires by the Batwa 3 5 5 13 75%
as they collect honey).
Human population pressure* 5 5 5 15 0%
Insecurity*. 1 1 5 7 0%
TOTALS 5 7 15 42 36.7%
TRA-I without the threat 27 56.9%
* Poaching scores 80%-poaching has decreased in the past three years. There is
effective enforcement and apprehending of culprits. However, a few Batwa still carry
out poaching and are not apprehended.
* Bush burning 75 %-there is sensitization of communities around the PA. When fire
occurs, they are often mobilized to fight it. There is constant monitoring by staff and
good relations

Table 2-15: Threats to Mt. Elgon national park
Threat Area Intensity Urgency Total Threat Met TRA-I
Increasing human 4 4 3 11 0%
Forest fires 2 2 3 7 60%
Pitsawying. 3 3 3 9 70%
Encroachment. 2 2 2 6 60%
Charcoal burning. 1 2 2 5 70%
Hunting. 5 2 2 9 90%
Insecurity (cattle rustling). 2 2 2 6 50%
Bad management practices. 1 2 2 5 40%
Bad politics. 2 4 2 8 60%
TOTALS. 22 23 21 75 47.3%
TRA-I without threat 64 55.4%
* Population pressure, nothing is being done about this threat for the PA management is
not the appropriate authority to handle such a threat.
* Local conversion of trees into timber 70%- there are still a few people carrying out
local conversion of trees into timber hideously
* Bad politics 60%-people have understood, but there are still problems in Kapchorwa
* Hunting 90%- hunting is only done for traditional reasons, there are no new cases of
hunting that are being reported.
* PA management and communities. However, PA management needs to be more
active in stopping Batwa who set fires as they collect honey. Some fires are

Table 2-16: Threats to Murchison falls national park
Threat Area Intensity Urgency Total Threat TRA-I
Hydro-electric power at 1 1 1 3 0
Lack of Awareness 5 5 5 15 35%
Less Park Manpower 3 3 2 8 50%
Insecurity 3 2 5 10 0
Poaching 5 5 5 15 50%
Problem Animals 2 2 2 6 30%
Bush Fires 5 5 5 15 10%
Diseases 2 1 2 5 0
Highway through the park* 1 1 1 3 0
Encroachment on park land 1 1 1 3 60%
Increase in human 2 2 1 3 0
TOTALS 28 28 30 86 24.8%
Totals without threats 77 27.7%

Notes: There was a big debate on the meaning and interpretation of "lack of
awareness" among the group. Apparently, there was a faction whose communities
had benefited from the buffer zone development project fund, a fund that is provided
by a German bank-KFW through a bilateral grant between the Uganda and German
governments. This faction was not supportive of the fact that lack of awareness was a
threat. On the other hand, the faction that represented the communities that had not
yet benefited from this fund strongly identified lack of awareness as a big threat. It
was apparent that "lack of awareness" was a need to implement development
projects in the communities and not education programs.
Lack of awareness 35%-because there has been some sensitization program in
some districts
Poaching 50%- there has been some reduction in poaching activities. It has
remained with the old people only
Bush fires 10%-there are no fire breaks, and both the rangers and poachers set
fires themselves

Table 2-17: Threats to Queen Elizabeth national park

Threat Area Intensity Urgency Total Threat Raw TRA-I
met Score
Poaching 1 2 5 9 50% 4.5
Bush fires 4 4 5 13 90% 11.7
Deforestation 1 1 5 7 40% 2.8
Encroachment 1 1 5 7 30% 2.1
Insecurity 1 2 5 8 40% 3.2
Corruption (park 1 3 5 9 50% 4.5
Poor park-community 5 4 5 14 55% 7.7
TOTALS 14 17 35 67 36.5 54.5%
Notes on the threat reduction by management approach
* Bush burning 90%-burning has generally reduced in most areas of PA. There have
been organized seminars on village levels to sensitize people against dangers of
burning. But there are some areas, where these seminars have not reached. Also, fire
set by poachers is still uncontrolled
* Corruption 50%- number of charcoal burners has decreased, corrupt staffs have been
dismissed, staff salaries have been increased, and there is strictness in prosecuting
corrupt people
* Poor relations 55%-management of park take law -breakers to LC and they decide the
punishment together.

Table 2-18: Summarized TRA indices to all protected areas
Name of PA TRA without TRA-I with all
threats threats

Pian- Upe
Mean TRA -I
Queen Elizabeth
Mt. Elgon
Murchison Falls
Lake Mburo
Mean TRA-I

37.96 21.6%

49.0 12.0

30.03 17.0

40.05 9.4

Table 2-19: Summary of percentage occurrence of threats to non-CBC protected areas
Threats Bokora Bugungu Kidepo Karuma Katonga Matheniko Pian- Kigezi Semuliki Occurrence

Local poaching
Armed poaching
Denied access to
PA relations

0 1

0 1 0

0 1

1 1

1 = Present, 0 = Absent







Table 2-19 Continu


Bokora Bugungu Kidepo Karuma Katonga Matheniko Pian- Kigezi Semuliki Occurrence

Deforestation 0 1 0 0 0 0 0 0 1 22.0%
Corruption 0 0 0 0 0 0 0 1 0 11.0%
Denied access 0 0 0 0 0 0 0 0 0 0
Lack of 0 0 0 1 1 0 0 0 22.0%
Problem animals 0 0 1 0 0 0 0 1 1 33.0%
Unclear 0 1 0 1 0 1 1 0 1 56.0%
Wildlife diseases 0 0 0 0 0 0 0 0 1 11.0%
Political 0 0 0 0 0 0 1 0 1 22.0%
1 = Present, 0 = Absent

Table 2-20: Summary of percentage occurrence of threats at CBC protected areas
Threats Queen Mgahinga Mt. Elgon Murchison Bwindi
Local poaching 1 0 1 1 1
Armed conflicts 1 1 1 1 0
Illiteracy 0 0 0 0 0
Pitsawying 0 0 1 0 1
Encroachment 1 0 1 1 0
Poor management 0 0 1 0 0
Land degradation 0 1 1 0 0
Charcoal burning 0 0 0 0 0
Human population 0 1 1 1 1
Bad community/PA relations 1 0 0 1 1
Deforestation 0 0 0 0 1
Corruption 1 0 0 0 1
Denied access 0 0 0 0 0
High-ways 0 0 0 1 0
Lack of manpower 0 0 0 1 0
Problem animals 1 1 0 1 1
Unclear boundaries 0 0 1 0 0
Wildlife diseases 0 0 0 1 1
Political interference 0 0 1 0 0

1 = Present, 0 = Absent

L. Mburo



Table 2-21: Summary of occurrence of threats at both CBC and non-CBC protected areas
Threats Occurrence at non-CBC PAs Occurrence at CBC PAs
Local poaching 100% 86.0%
Armed conflicts 89.0% 57.0%
Illiteracy 11.0% 0
Pitsawying 22.0% 43.0%
Denied access to shrine 22.0% 0
Encroachment 78.0% 57.0%
Poor management 11.0% 14.0%
Land degradation 56.0% 57.0%
Charcoal burning 33.0% 14.0%
Human population 67.0% 86.0%
Bad community/PA relations 78.0% 57.0%
Deforestation 22.0% 14.0%
Corruption 11.0% 29.0%
Denied access 0 14.0%
Highways 0 14.0%
Lack of manpower 22.0% 29.0%
Problem animals 33.0% 86.0%
Unclear boundaries 56.0% 29.0%
Wildlife diseases 11.0% 43.0%
Political interference 22.0% 29.0%

Table 2-22: PA Managers' responses about protected areas conservation status
Non- Mgt Mgt. Ecological information Boundary Status Resource Status Income Source
CBC PA object Plan

Bokora No No None



In Little info on bird and
process animal inventories

Kidepo Yes Yes Inventories of mammals,
birds, and wildlife
population trends.
Karuma Yes In Inventories of mammals
making and birds, vegt. Maps
and wildlife population
Kato No No Socio-economic data

Math No No No

Pian No Yes Mammal and bird
inventory, wildlife
population trends
Kigezi Yes Yes Inventories of birds and
mammals exist
Semuliki Yes No Inventories of mammal,
and socio-economic and
vegetation map

Not known, not demarcated, but
it is feared that there is
Fully and carefully demarcated
involving local communities
and are not contested
Known, not all are demarcated,
and there is no conflict.

Well demarcated but being

Known, not demarcated, there
is conflict, and it is feared that
there is encroachment
Not known, not demarcated, but
it is feared that there is
Not known, not demarcated, but
it is feared that there is
Known, not demarcated and no
Not known, not demarcated,
feared that there is

Suffers from uncontrollable
use by local communities

Suffers from commercial
poaching of wild animals and
Suffers from commercial
poaching of game

Suffers from uncontrolled use
by local communities

Suffers from uncontrolled use
by local communities

Suffers from uncontrolled use
by local communities

Suffers from uncontrolled use
by local communities

Suffers from uncontrolled use
by neighboring communities
Suffers from uncontrolled use
by local communities

Govt. subvention

Govt. subvention

Govt. subvention

Govt. subvention
and incomes
generated within.

Govt. subvention

subvention only

subvention only

subvention only

Table 2-22 Continued

Queen Yes

Mgahinga Yes

Mt. Elgon Yes

Murchison Yes

Bwindi Yes

L. Mburo Yes




Ecological information

Inventories of mammals,
birds, vegetation maps,
aerial photos and socio-
economic data.

Has Inventories of mammals,
birds, vegetation maps,
geological and soil maps,
wildlife population trends,
and socio-economic data
Draft Bird inventory and
form Vegetation maps.

In the

Inventories of mammals,
birds, vegetation maps and
socio-economic data

Has Inventories of mammals,
birds, vegetation maps and
socio-economic data
Has Inventories of mammals,
birds, vegetation maps and
socio-economic data
Has Inventories of mammals,
birds, vegetation maps and
socio-economic data

Boundary Status

Well demarcated but some are
being contested

Fully and carefully demarcated
involving local communities
and are not contested

Demarcated but contested

Known, not demarcated, but
there are some conflicts

Fully and carefully demarcated
involving local people and are
not contested
Well demarcated but are

Fully and well demarcated
involving local people and are
not contested

Resource Status

Suffers from
commercial poaching
of animals and trees.
Some are controllably
used by communities
Fully protected from
any exploitation

Controlled under CM
arrangements in some
areas. Others
uncontrollably used.
To an extent suffers
from commercial
poaching of wild
animals and trees
Fully protected from
any exploitation

To an extent suffers
from uncontrolled use
by local communities
To an extent suffers
from uncontrolled use
by local communities

Income Source

Govt. subvention

Govt. subvention
and NGO support

Govt. subvention,
donations and
generated income

Govt. subvention
and NGOs.

Govt. subvention,
NGOs and
generated revenue
Govt. subvention,
and generated
Govt. subvention,
NGOs and
generated revenue

Table 2-23: PA Managers' responses about protected areas management issues

Research Tourism Political Support Involve Locals Benefits Locals
/Management Organization.
Has no research Non existent Central govt. is Involves at least Through illegal

Non- Conflict
Bokora Reported,
and no
follow up
Bugungu Reported
Kidepo Reported
Karuma Reported,
Katonga Reported,
but not

Poorly developed
with very little

Well organized with
basic facilities in

Well organized with
basic facilities in

Poorly developed
with very low
visitation and
insufficient facilities

supportive. No support
from local government
and local communities

Local government
support but not from
local people and

Yes from central and
local govt. but partly
from local communities
and parliamentarians

Yes from central and
local govt. but partly
from local communities
and parliamentarians

Lacks central govt.
support. Has local govt.
support of home district

government officials in
decision making

Involves local people
and some government
officials in decision

Involves some govt.

Involves at least some
govt. officials

Does not involve local
people at all, only some
government officials

access to natural

Permit system to
access resources in

Some benefits -

opportunities, and
contributes to
Some benefits -

Has no research

Has basic and
applied research
that may support

Has basic and
applied research
that may support

Has no research

Table 2-23


Conflict Mgt.

Not reported,
not recorded
and not
followed up
recorded and
followed up

Kigezi Reported,
recorded and
followed up
Semuliki Reported, not
recorded but
followed up

Has no

Has integrated
research that
Has no

research that
does not
support mgt.




Poorly developed
with low visitation
and insufficient
tourists facilities

Political Support

There is no political

Local govt. and local
people with their

Central govt.

Support of central govt.
and local govt. but not
all parliamentarians

Involve Locals

Does not involve
people at all

Involves some govt.

Does not involve
people at all

Involves local people
at minimally

Benefits Locals

Illegal accessing the

benefits, and use of
resources from the
No benefits at all

opportunities and
provides market for
people's produce

Table 2-23 Continued
CBC PA Conflict Mgt.


recorded and
followed up.

Mgahinga Reported,
recorded and
followed up

Mt. Elgon Reported, not
recorded but
followed up

Murchison Reported,
recorded and
followed up

Bwindi Reported
recorded and
followed up

Only has basic
research that
does not support
Has a well
integrated basic
and applied
research that
Only academic
research does
not support

Only has basic
research that
does not support
Has a well
integrated basic
and applied
research that

Tourism Political Support Involve Locals Benefits Locals

Well developed
with basic
facilities in place

Well established
with high
visitation, but
there are no

Poorly developed
with low visitation
- insufficient
tourist facilities

Well developed
with basic
facilities in place

Well established
with high
visitation, but no
adequate facilities

Central government,
and local govt. but little
support from the local
people and their
Central govt. Local
govt. and local people
are all supportive.

High central govt.
support, low local govt.
and very low from local
people and their
Support of central and
local govt. but not from
local people and their

Central and local govt.
as well as local people
and their

Involves local people in
management activities,
and at least some govt.
officials in decision-
Involves local people in
management decision-
making, as well as some
government officials.

Involves local people in
management decision-
making, as well as some
government officials
plus NGOs.
Involves at least some
govt. officials in the
decision making

Involves local people in
their management
decision making

Provides natural
opportunities and
education programs.
Real benefits,
revenue sharing,
roads water
development and
Forest products and

opportunities and

Real benefits e.g.
revenue sharing, and

Table 2-23


CBC PA Conflict Mgt. Research Tourism Political Support Involve Locals Benefits Locals
/Management Organization.
L. Mburo Reported Has basic Well established Central and local govt. Involves local people in Real benefits e.g.
recorded and research that with high as well as local people their management revenue sharing,
followed up does not visitation, but no and their decision making and infrastructure
support adequate facilities parliamentarians development.
Kibale Reported Has a well Well established Central and local govt. Involves local people in Real benefits e.g.
recorded and integrated basic with high as well as local people their management revenue sharing,
followed up and applied visitation, but no and their decision making and infrastructure
research that adequate facilities parliamentarians development.

Table 2-24: Summary percentages of protected areas status as indicated by wardens
Factor CBC (n=7) Non-CBC (n=9)
Has management plan
* Yes 86% 33.0%
* No 44.0%
* In process 14% 22.0%
Has management objectives
* Yes 100% 56%
* No 44%

Boundary status
* Demarcated but contested 43% 11.0%
* Demarcated and not contested 43% 22.0%
* Not demarcated and encroached 14% 67%
Resource status
" Suffers from uncontrollable use by communities
" Suffers from commercial poaching 43% 89%
* Fully protected from exploitation 14% 11%
" Controllably used by communities 29% -
Conflict Management
Reported, not recorded not followed up 11%
Reported recorded followed up 86% 56%
Reported recorded not followed up 11%
Reported not recorded followed up 14% 11%
Not reported not recorded not followed up 11%

Table 2-24 Continued

Table 2-24 Continued

CBC (n=7) Non-CBC (n=9)

Research programs
* Has no research
* Has basic research that supports management
" Has academic research that does not support management
Tourism program
* Non-existent
" Poorly developed
* Well organized
Political support
* Central govt. support but not local govt.
* Local govt. support but not central govt.
* Local govt. support but not local people
* Central and local govt. support but not local people
" Total support
* No political support
Involves local people
* Involves at least govt. officials
* Involves local and govt. officials
* Involves govt. officials not local people
* Does not involve people at all
Benefits local people
* Through illegal access to resources
* Through legalized permit system
* Employment and infrastructure development.
* Revenue sharing
" No benefits at all