<%BANNER%>

Household Participation, Resource Dependency, and Diversification

Permanent Link: http://ufdc.ufl.edu/UFE0041208/00001

Material Information

Title: Household Participation, Resource Dependency, and Diversification A Livelihoods Assessment in the Kanchenjunga Conservation Area, Nepal
Physical Description: 1 online resource (157 p.)
Language: english
Creator: Parker, Pete
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: areas, community, decentralization, dependency, livelihoods, management, nepal, participation, protected, sustainable
Tourism, Recreation, and Sport Management -- Dissertations, Academic -- UF
Genre: Health and Human Performance thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: HOUSEHOLD PARTICIPATION, RESOURCE DEPENDENCY, AND DIVERSIFICATION: A LIVELIHOODS ASSESSMENT IN THE KANCHENJUNGA CONSERVATION AREA, NEPAL The impact of decentralized conservation initiatives among households was assessed on a socioeconomic level at the Kanchenjunga Conservation Area Project (KCAP), in Nepal. Participation roles in sponsored programs, associated distribution of benefits, dependency on natural resources, and the diversification of livelihood strategies were assessed on a household level. The primary methods of data collection were participant observation, Participatory Rural Appraisal wealth ranking, and a survey administered to 215 household throughout the region. An ordered probit regression model was conducted to assess the existing household participation patterns in sponsored programs. A linear regression was utilized to determine the distribution of benefits based on the predicted role of participation in sponsored programs. A logistic regression model was conducted to identify household factors that affect natural resources dependency (a ratio of total household income compared to natural resources based income). A multi-step cluster analysis was utilized to identify discernable barriers to existing livelihood strategies. A multinomial logistic regression model was then applied to examine the existing correlate of access or barriers to graduate to more effective livelihood strategies. Factors that were significantly related to household participation roles in sponsored programs included the education and age of household head, education of household adults, size of household, household distance to KCAP office, household visitation to KCAP offices, and household visitation by KCAP staff. Greater roles of participation in sponsored programs were significantly related to increased access to benefits including financial credit, lower interest rates, and training programs. Natural resources dependent activities made up 64% of the average household?s total income. Higher elevation households were more dependent than lower elevation households. Significant predictors of dependency on natural resources included remittances, household size, and the number of dependents. Ten of the most prevalent livelihood activities found throughout the region were identified and analyzed. Five discernable livelihood strategies were identified with varying levels of remuneration. Significant predictors of specific livelihood strategies included landholdings, dependents, livestock, remittance, ethnicity, and participation roles in sponsored programs. The results of these analyses were used to suggest community management strategies to encourage and promote maximum participation by the communities to enhance their sustainable livelihood strategies.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Pete Parker.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: Thapa, Brijesh.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0041208:00001

Permanent Link: http://ufdc.ufl.edu/UFE0041208/00001

Material Information

Title: Household Participation, Resource Dependency, and Diversification A Livelihoods Assessment in the Kanchenjunga Conservation Area, Nepal
Physical Description: 1 online resource (157 p.)
Language: english
Creator: Parker, Pete
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: areas, community, decentralization, dependency, livelihoods, management, nepal, participation, protected, sustainable
Tourism, Recreation, and Sport Management -- Dissertations, Academic -- UF
Genre: Health and Human Performance thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: HOUSEHOLD PARTICIPATION, RESOURCE DEPENDENCY, AND DIVERSIFICATION: A LIVELIHOODS ASSESSMENT IN THE KANCHENJUNGA CONSERVATION AREA, NEPAL The impact of decentralized conservation initiatives among households was assessed on a socioeconomic level at the Kanchenjunga Conservation Area Project (KCAP), in Nepal. Participation roles in sponsored programs, associated distribution of benefits, dependency on natural resources, and the diversification of livelihood strategies were assessed on a household level. The primary methods of data collection were participant observation, Participatory Rural Appraisal wealth ranking, and a survey administered to 215 household throughout the region. An ordered probit regression model was conducted to assess the existing household participation patterns in sponsored programs. A linear regression was utilized to determine the distribution of benefits based on the predicted role of participation in sponsored programs. A logistic regression model was conducted to identify household factors that affect natural resources dependency (a ratio of total household income compared to natural resources based income). A multi-step cluster analysis was utilized to identify discernable barriers to existing livelihood strategies. A multinomial logistic regression model was then applied to examine the existing correlate of access or barriers to graduate to more effective livelihood strategies. Factors that were significantly related to household participation roles in sponsored programs included the education and age of household head, education of household adults, size of household, household distance to KCAP office, household visitation to KCAP offices, and household visitation by KCAP staff. Greater roles of participation in sponsored programs were significantly related to increased access to benefits including financial credit, lower interest rates, and training programs. Natural resources dependent activities made up 64% of the average household?s total income. Higher elevation households were more dependent than lower elevation households. Significant predictors of dependency on natural resources included remittances, household size, and the number of dependents. Ten of the most prevalent livelihood activities found throughout the region were identified and analyzed. Five discernable livelihood strategies were identified with varying levels of remuneration. Significant predictors of specific livelihood strategies included landholdings, dependents, livestock, remittance, ethnicity, and participation roles in sponsored programs. The results of these analyses were used to suggest community management strategies to encourage and promote maximum participation by the communities to enhance their sustainable livelihood strategies.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Pete Parker.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: Thapa, Brijesh.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0041208:00001


This item has the following downloads:


Full Text

PAGE 1

HOUSEHOLD PARTICIPATION, R ESOURCE DEPENDENCY, AND DIVERSIFICATION: A LIVELIHOODS ASSESSMENT IN THE KANCHENJUNGA CONSERVATION AREA, NEPAL By PETE PARKER A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2009 1

PAGE 2

2009 Pete Parker 2

PAGE 3

To my father for provid ing the needed inspiration 3

PAGE 4

ACKNOWLEDGMENTS This dissertation would not have been possible without the assistance and encouragement of several peopl e. I thank my parents for their encouragement. I gratefully acknowledge my advisor, Brijes h Thapa, for being both a friend and a mentor (and for knowing which role was needed at the appropriate time throughout the process). The other members of my committee also have my sincere thanks for all their input and assistance: Dr. Stephen Holland, Dr. Taylor V. Stein, and Dr. Madan Oli. I would also like to thank the departm ent of Tourism, Recreation & Sport Management the for the graduate assistantship that allowed me to complete my program. Within my departm ent, I would also like to thank Nancy Gullic for her friendship and administrative support. My time spent in Nepal would not have been as fruitful nor as enjoyable without the assistance of several individuals. S pecifically, I extend my gratitude to Megh Bahadur Pandey for his assistance at the D epartment of National Parks and Wildlife Conservation, Praveen Bisht for his logistic al support, and Dan Putnam for providing a home away from home. My friends also deserve my sincerest appreciation for their encouragement. In particular, I would like to thank Caroline, Ly le, Amy, John, Dave, and Phil. I would also like to thank the Top for providing the v enue for several organizational meetings, both scheduled and impromptu. 4

PAGE 5

TABLE OF CONTENTS page ACKNOWLEDG MENTS..................................................................................................4 LIST OF TABLES............................................................................................................8 LIST OF FI GURES..........................................................................................................9 LIST OF ABBR EVIATIONS...........................................................................................10 ABSTRACT ...................................................................................................................11 CHAPTER 1 INTRODUC TION....................................................................................................13 Defining the Problem ..............................................................................................13 Sustainable Livelihoo ds Framew ork.......................................................................18 Statement of Probl em.............................................................................................23 Purpose of Study....................................................................................................26 Project Signi ficance................................................................................................27 Dissertation Format................................................................................................29 2 PARTICIPATION ROLES AND BEN EFITS IN SPONSORED PROGRAMS AMONG HOUSEHOLDS WITHIN KANCH ENJUNGA CONSERVATION AREA, NEPAL ....................................................................................................................39 Introducti on.............................................................................................................39 Literature Review....................................................................................................41 Decentralized Conservation Partici pation.........................................................41 Decentralized Conservation Costs and Benefits..............................................43 Kanchenjunga Conservation Ar ea Project, Nepal...................................................45 Programs and M anagement .............................................................................46 Conservation Issues.........................................................................................48 Study Si te.........................................................................................................49 Methods ..................................................................................................................50 Data Coll ection.................................................................................................51 Selection of Pa rticipant s...................................................................................52 Instrument ation.................................................................................................53 Operationalization and Data A nalysis...............................................................54 Research question 1: What factors affect household participation roles in sponsored program s?.........................................................................54 Research question 2: Do participa tion roles in sponsored programs impact the distribution of benefits among hou seholds?..........................56 Result s....................................................................................................................56 Profile of Househo ld Responde nts...................................................................56 5

PAGE 6

Research Question 1: Household Pa rticipation Roles in Sponsored Program s.......................................................................................................60 Research Question 2: Household Part icipation Roles a nd Distribution of Benefit s.........................................................................................................62 Discussio n..............................................................................................................64 3 USE PATTERNS AND FACTORS AFFECTING HOUSEHOLD NATURAL RESOURCE DEPENDENCY WITHIN KANCHENJUNGA CONSERVATION AREA, NE PAL........................................................................................................ 69 Introducti on.............................................................................................................69 Literature Review....................................................................................................71 Kanchenjunga Conservation Area Proj ect..............................................................75 Methods ..................................................................................................................80 Data Coll ection.................................................................................................80 Selection of Pa rticipant s...................................................................................81 Instrument ation.................................................................................................82 Operationalization and Data A nalysis...............................................................82 Research question 1: W hat are the existing levels of natural resource dependencies among household s?........................................................83 Research question 2: What fa ctors affect natural resource dependencies among household s?........................................................83 Result s....................................................................................................................84 Profile of Househo ld Responde nts...................................................................84 Research Question 1: Levels of Natural Resour ces Dependenc y...................85 Research Question 2: Factors Associated With Natural Resources Dependency..................................................................................................86 Discussio n..............................................................................................................87 Conclusi on..............................................................................................................90 4 LIVELIHOOD STRATEGIES AND DI VERSIFICATION FACTORS AMONG HOUSEHOLDS WITHIN KANCHENJUNG A CONSERVATION AREA, NEPAL....92 Introducti on.............................................................................................................92 Literature Review....................................................................................................95 Sustainable Livelihoo ds Framew ork.................................................................95 Factors that Affect Li velihood Diversificati on....................................................99 Kanchenjunga Conservation Area Proj ect............................................................102 Enhancing Local Livelihoods ..........................................................................104 Livelihood Chal lenges ....................................................................................105 Methods ................................................................................................................107 Data Collec tion...............................................................................................107 Selection of Pa rticipant s.................................................................................108 Instrumentat ion...............................................................................................109 Operationalization and Data Anal ysis.............................................................109 Research question 1: What is th e household level of livelihood strategy diversificat ion?......................................................................................109 6

PAGE 7

Research question 2: What are the factors associ ated with dominant livelihood strat egy adopti on?................................................................111 Results ..................................................................................................................111 Profile of Re spondents ...................................................................................111 Research Question 1: Diversifica tion of Livelihood Strategies......................113 Research Question 2: Factors A ssociated With Dominant Livelihood Strategy A doption ........................................................................................117 Discussio n............................................................................................................120 5 CONCLUS ION......................................................................................................125 Participation Roles................................................................................................125 Natural Resource s Dependency ...........................................................................127 Livelihood Diversi fication......................................................................................128 Recommendations to E nhance KCAP Mana gement............................................ 130 Research Limi tations............................................................................................131 Future Res earch...................................................................................................132 APPENDIX A HOUSEHOLD SURVEY QUESTIONAI RRE.........................................................134 B HOUSEHOLD SURVEY QUESTIONAIRRE (Translated into Nepali)...................141 C INSTITUTIONAL REVIEW BOARD APP ROVAL..................................................148 LIST OF REFE RENCES.............................................................................................149 BIOGRAPHICAL SKETCH ..........................................................................................157 7

PAGE 8

LIST OF TABLES Table page 2-1 Operationalizat ion of va riables ...........................................................................55 2-2 Profile of res pondents .........................................................................................57 2-3 Cross-tabulat ion of age by years of educat ion....................................................58 2-4 Descriptive statis tics for participation role analysis va riables.............................59 2-5 Households participation roles in sponsor ed program s......................................61 2-6 Chi-squared percentage of parti cipation roles and bene fits received.................62 2-7 Perceived household co mmunity benefits from KCAP.......................................63 2-8 Parameter estimates for benefits received based on part icipation role..............64 3-1 Household c haracterist ics..................................................................................84 3-2 Sources of household income and dependency on natural resources access...86 3-3 Logistic regression of househo ld natural res ources dep endency.......................86 4-1 Livelihood activiti es by elev ation.......................................................................113 4-2 Cluster analysis of activiti es within each live lihood strat egy.............................115 4-3 Factors of liveli hood strategy adoption .............................................................119 8

PAGE 9

LIST OF FIGURES Figure page 1-1 Sustainable livel ihoods fram ework .....................................................................21 1-2 Concept ual model...............................................................................................27 1-3 Village of Taplec hok...........................................................................................31 1-4 Village of Olanchan Gola....................................................................................31 1-5 Concept ual model...............................................................................................32 1-6 Concept ual model...............................................................................................32 1-7 Prayer wheels in Olanc han Gola ........................................................................33 1-8 Yak train in Olanchan Gola.................................................................................33 1-9 Lelep Village.......................................................................................................34 1-10 Typical hous e in Lelep Village ............................................................................34 1-11 KCAP headquarters o ffice in Lelep Vill age.........................................................35 1-12 Village of Phere..................................................................................................35 1-13 Village of Ghunsa...............................................................................................36 1-14 Tent camp in Ghunsa Village ..............................................................................36 1-15 Typical Ghuns a Village kitchen ...........................................................................37 1-16 Close-up of cooking area....................................................................................37 1-17 Local resident..................................................................................................... 38 2-1 Map of KC AP......................................................................................................50 3-1 Map of KC AP......................................................................................................77 4-1 Sustainable livel ihoods fram ework .....................................................................97 4-2 Map of KCAP....................................................................................................103 9

PAGE 10

LIST OF ABBREVIATIONS ACAP Annapurna Conservation Area Project CAMC Conservation Area Management Committees CAMR Conservation Area Management Regulation CBC Community-Based Conservation CBD Convention on Biological Diversity CBNRM Community Based Natural Resources Management DFID United Kingdom Department for International Development DNPWC Department of National Parks and Wildlife Conservation FAO Food and Agriculture Organization HDI Human Development Index ICDP Integrated Conservati on and Development Project ICIMOD Integrated Centre for Integrated Mountain Development KCAP Kanchenjunga Conservation Area Project MBCAP Makulu-Barun Conservation Area Project NTFP Non-timber forest products PRA Participatory rural appraisal UNDP United Nations De velopment Program USAID United States Agency fo r International Development VDC Village Development Committee WHO World Health Organization WWF World Wildlife Fund for Nature 10

PAGE 11

Abstract of Dissertation Pr esented to the Graduate School of the University of Florida in Partial Fulf illment of the Requirements for t he Degree of Doctor of Philosophy HOUSEHOLD PARTICIPATION, R ESOURCE DEPENDENCY, AND DIVERSIFICATION: A LIVELIHOODS ASSESSMENT IN THE KANCHENJUNGA CONSERVATION AREA, NEPAL By Pete Parker December 2009 Chair: Brijesh Thapa Major: Health and Human Performance The impact of decentralized conserva tion initiatives among households was assessed on a socioeconomic level at t he Kanchenjunga Conserva tion Area Project (KCAP), in Nepal. Participation roles in sponsored programs, associated distribution of benefits, dependency on natural resources, and the divers ification of livelihood strategies were assessed on a household level. The primary methods of data collection were participant observation, Participatory Rural Appraisal wealth ranking, and a su rvey administered to 215 household throughout the region. An ordered probit regression model was conducted to assess the existing household participation patterns in sponsor ed programs. A linear regression was utilized to determine the distribution of benefits based on the predicted role of participation in sponsored progr ams. A logistic regression model was conducted to identify household factors that affect natural resources dependency (a ratio of total household income compared to natural resour ces based income). A multi-step cluster analysis was utilized to identify discernable ba rriers to existing livelihood strategies. A 11

PAGE 12

12 multinomial logistic regression model was then applied to examine the existing correlate of access or barriers to graduate to mo re effective livelihood strategies. Factors that were significantly related to household participation roles in sponsored programs included the education and age of househ old head, education of household adults, size of household, househol d distance to KCAP office, household visitation to KCAP offices, and household visi tation by KCAP staff. Greater roles of participation in sponsored programs were signi ficantly related to increased access to benefits including financial credit, lower inte rest rates, and training programs. Natural resources dependent activities made up 64% of the average households total income. Higher elevation households were more dependent than lower elevation households. Significant predictors of dependency on natural resources included remittances, household size, and the number of dependents. Ten of the most prevalent livelihood activities found throughout the region were identified and analyzed. Five discernable li velihood strategies were identified with varying levels of remuneration. Significant predictors of specific livelihood strategies included landholdings, dependents, livestock, re mittance, ethnicity, and participation roles in sponsored programs. The results of these analyses were used to suggest community management strategies to encourage and promote maximu m participation by the communities to enhance their sustainable livelihood strategies.

PAGE 13

CHAPTER 1 INTRODUCTION Defining the Problem Local people, especially in developing countries rely on resources and products from parks and protect ed areas. Their demands for the use of natural resources are either in conflict with, compatible, or comp lementary to the conser vation objectives of local and national government agencies as well as with local and international conservation organizations (Salafsky & Woll enberg 2000). However, means of how to avoid conflict and ensure complementarit y has been the focus of much discussion during the last three dec ades (McNeeley 1988). The initial proponents of parks and pr otected areas advocated fortress conservation, fines and fences, or coercive approaches, which were based on preservation of natural re sources with strong support from western international conservation organizations via financial and technical assistance (McNeely 1988; Wells et al. 1992). The concept of parks and prot ected areas and the aforementioned management approaches first became exp licit in the creation and custodial management of Yellowstone National Park, USA in 1872. In the 1970s and 80s, scientists examined biological preservation not as merely ac reage preserved, but rather the quality of biological diversity that exis ted within parks and prot ected areas (Brechin & West 1991). Given that an abundance of ri ch biological diversity existed in developing countries (Western & Wright 1994), the 1982 World Parks Congress in Bali recommended that all nations should strive to allocate 10% of their lands under protection (McNeely 1988). However, withi n the fortress conservation approach, especially in developing countries, local people who resided within proposed boundaries 13

PAGE 14

were relocated with little or no compensation (Western & Wright 1994). Those people who lived outside protected areas were denied access to those natural resources located within the park boundaries. In most cases, their livel ihoods depended on the natural resources (Manning 1994), and local residents participated in behaviors that undermined the whole conservation init iative (Brechin & West 1991). In the 1990s, it was acknowledged that successful parks and protected area management depended on integration and partic ipatory conservation by the local population (Brandon & Wells 1992). In 1992, delegates from 179 countries signed the landmark United Nations Convention on Biologi cal Diversity (CBD) which pledged to create a system of protected areas based on three primary objectives: 1) conservation of biological diversity, 2) sustainable use of the components of bi ological diversity, and 3) fair and equitable sharing of the benefits arising out of the utilization of genetic resources (CBD 1992). Therefore, parks and protected areas are now charged with missions beyond biodiversity conservation; they must also improve social welfare, guard local security, and provide economic benefit s across multiple sectors, objectives traditionally relegated to the dev elopment sector (Naughton-Treves et al. 2005). Based on the aforementioned objective s, the Integrated Conservation and Development Project (ICDP) framework has been promoted in developing countries as a means of maximizing benefits for local pe ople while minimizing the costs endured due to biodiversity conservation measures (Brandon & Wells 1992, p. 557). Since its advocacy, ICDPs have proliferated throughout developing countries and have garnered major support for conservation (McShane & Wells 2004). ICDPs are based upon the idea that if programs improve the livelihoods of local people, then they would consume 14

PAGE 15

fewer natural resources, and thus minimize pressures within protected areas (Brandon & Wells 1992; Brandon et al. 1998). Additionally, ICDPs are based on the paradigm that projects need to be cent ered on providing for the loca ls and positively respond to changes in their livelihoods (Berkes et al. 2007). The inclusion of local participation in ICDP management and decision making helps to ensure an equitable distribution of benefits that compensate for costs borne thr ough environmental conservation practices (Brandon 1997). By devolving management to the local people, it saves time and money from the central governing body and provides immediate benefits and empowerment to the local peop le while utilizing existing traditional knowledge (Ribot 1995). ICDP terminology has been used interchang eably in the literature along with Community Based Natural Resources Management (CBNRM), and Community-Based Conservation (CBC). However the focus of ICDPs is primarily relegated to parks and protected areas (Naughton-Treves et al. 2005), while CBC and CBNRM operate within a community level at a smaller scale (Horwich & Lyon 2007). Although, all three concepts share the same major components of conservation through maximized local participation, and management vi a decentralization of institutional power (Murphee 1994; Agrawal & Gibson 1999; Sala fsky & Wollenberg 2000; Stem et al. 2003; Child 2004; Berkes 2007). In order to protect the environment and provide for an improved and sustained quality of life for local people, programs need to be based on local initiatives, with training assistance provided by non-government organizations (NGOs), in accordance with local public institutions (Ribot 1998). 15

PAGE 16

However, results from existing ICDPs have shown varied results and have led to a debate arguing both for and against them (Wells et al. 1992; Brandon 1997; Oates 1999; McShane & Wells 2004). Naughton-Treves et al. (2005, p. 240) note that most ICDP failures reported in the literature are primar ily caused by the erroneous labeling of any conservation project that incorporates local people as an ICDP, the terminology has become all things to all people. A fundamental flaw with ICDPs relates to the creation and formation of the project plan (Berkes et al. 2007). Most ICDPs are created by western-based international organizations with large amounts of funding and little site-specific knowledge of culture and institutions (Oates 1999; McShane & We lls 2004). Neuman (1997) states that ICDPs do not invite true participation, but rath er serve as coercive forms of conservation that resemble ill fated colonial efforts to convert shifting cultivators into progressive farmers (p. 564). Al though ICDPs are designed to ensure an equitable distribution of benefits from conservation measures, evidenc e suggests that this rarely occurs, especially among the most dis advantaged of households (Berkes et al. 2007). It is not enough for project designers to simply assume that community leaders will assure that benefits accrue equally to all individuals. Mc Shane & Wells (2004) state that one of the main shortcomings of ICDPs is the lack of trade-offs between the interests and claims of multiple stakeholders. Accountability is improved if whole communities, including disadvantaged members such as women and mi norities are involved in decision-making processes. Success of ICDPs is heavily co rrelated to existing levels and roles of participation (Strusaker et al. 2005), and there needs to be a complete devolution of 16

PAGE 17

authority to local communities over biologic al resources so that they have a vested interest (Neuman 1997; Ribot 2003). Critics of ICDPs argue agains t the existence of an economic threshold that if met will lead residents to cease use of natural resources extraction (Stem et al. 2003). For example, if $1000 dollars is deemed minimal to survival without natural resources extraction, proponents of ICDP s denote that residents will stop extracting and/or committing illegal activities, once they rece ive $1000 dollars worth of alternatively generated funds. Conversely, those in opposition indicate that residents will still continue to consume and generally want to maximize consumption, at least until complete absorption of labor into new activities (Brandon 1997). Another argument against ICDPs occurs in the debate between social scientists and natural scientists with respect to the primary objectives of an ICDP (Brechin et al. 2002). Natural scientists emphasize a conservation dominated focus while social scientists advocate for community development related initiatives. Conservation and development are ultimately incongr uent goals (Oates 1999), as ICDPs have demonstrated to be problematic when equal emphases are implemented in practice (Wells et al. 2004; Berkes et al. 2007). However, local perspectives need to be recognized, as user groups within ICDPs ar e not necessarily homogenous with respect to conservation or development. For example, Baral et al. (2008) found that the longer established user groups (>9 years) wit hin an ICDP (Annapurna Conservation Area Project) were focused on conservation; m edium length established user groups (5-9 years) emphasized institutional capacity building while more recently established usergroups (<5 years) were focused on economic development. 17

PAGE 18

Conservation programs are only valid an d sustainable when they have the dual objective of protecting and improving loca l livelihoods and ecological conditions (Ghimire & Pimbert 1997, p. 8). Creation of valid and sustainable conservation programs depends on decentra lized management and the in corporation of local participation. This will ensure an equitable di stribution of benefits while simultaneously diversifying income generating activi ties thus, reducing dependency on natural resources (Chapin 2004; McShane & Wells 2004). Within an assessment of ICDPs worldwide, the UK Department for International Developm ent (2007) identified three major shortcomings: 1) local development is usually assessed in terms of generation of cash, increased production or jobs, while wide r social issues and livelihoods concerns are often ignored, 2) project assessments usually focus on the numeration of outputs achieved and often create unintended cons equences for peoples livelihoods, and 3) assessments usually focus on generation of local incomes instead of the promotion of commercial viability. Given these issues, the use of the sust ainable livelihoods framework for project assessment s in order to overcome these major shortcomings has been advocated (DFID 2007). Sustainable Livelihoods Framework The primary objective of ICDP s is to provide for diversified sustainable livelihood strategies while emphasizing sustainable development that ensures biological conservation (Scoones 1998). In 1987, the World Commission on Environment and Development first advanced the term sustainable livelihoods in relation to alleviation of poverty and protection of the environment through the promotion of participation and sustainable development (WCED 1987). Subsequently, the 1992 UN Conference on Environment and Development promoted sust ainable livelihoods as a means of linking 18

PAGE 19

socioeconomic and environmental concerns. These two international forums were critical to the formation of the sustainable livelihoods framework (Scoones 1998). The sustainable livelihoods framework has been championed in the community development and environmental management disciplines, including ICDPs, with the shift in paradigm toward a more people-c entered approach within a community (Ashley & Carney 1999). It emphasizes that communiti es have existing assets that should be promoted rather than the focus on comm unity needs. The sustainable livelihoods framework has been heavily uti lized to assess regions and projects with the emphasis on people living within a given place or si tuation (Chambers & Conway 1992; Scoones 1998; Ashley & Carney 1999; Carney et al. 1999; Ellis 2000; DFID 2007). The sustainable livelihoods framew ork is currently used by international development agencies such as, the United Nations Deve lopment Program (UNDP), the Food and Agriculture Organization (FAO), the United St ates Agency for Intern ational Development (USAID), CARE-International, Oxfam, and the United Kingdom s Department for International Development (DFID) (Frankenberger et al. 2000). Chambers & Conway (1992: p. 7) first defined a livelihood as comprised of the capabilities, assets (stores, resources, cl aims and access) and activities required for a means of living. A livelih ood is sustainable when it c an cope with and recover from stresses and shocks and maintain or enhance its capabilities and assets both now and in the future, while not undermining the nat ural resource base. Ellis (2000) adapted this definition with more emphasis on access in re lation to social institutions that might impact an individual from capacity to fulfill its consumption r equirements based on social inequalities such as, caste, race, gender, ethnicity, beliefs, et c. His definition is a 19

PAGE 20

livelihood comprises assets (natural, physica l, human, financial and social capital), activities, and access to assets (mediated by institutions and social relations) that collectively determine the living gained by an individual or househol d (p. 10). The sustainable livelihoods framework deviates from purely economic measurements such as primary income gross domestic product, and consumption metrics in poor regions, and levies emphas is on breaking accessibility barriers to improve the quality of life vi a institutions, policies, and processes (Scoones 1998). However, Ellis (2000) delineates the impor tance of a lack of emphasis on temporal changes in access to assets. Individuals are continually changing their livelihood strategies concurrently with the social institutions that govern access, especially in developing countries (Ellis 2000). The sustainable livelihoods framework e ssentially has four main components (Figure 1-1). First a populati on lives within a vulnerability context in which they are exposed to risk in the forms of sudden shoc k, trends over time, and seasonal change. Examples of risk include drought, tourism seas onality, or labor migration. Second, people utilize assets to meet t heir livelihoods. These assets include social (networks, norms, trust, institutions), human (skills, knowledge, and labor), natural (natural resources), physical (infrastructure includ ing transportation, water, and energy), and financial capital (savings, income, and credit). Third, these assets are utilized by residents through livelihood strategies in or der to create their livelihood outcomes. Lastly, transforming structures and pr ocesses shape and influence the ability of individuals to access livelihood assets as a means to overcome risks within the vulnerability context. It is within these tr ansforming structures that programs can impact 20

PAGE 21

the linkage between policies, inst itutions, and livelihood activi ties (DFID 2007). Figure 1-1 depicts the accessibility of resources, controlled by institutions and norms that are used within livelihood strategies to create an outcome that enhances pre-existing assets to overcome vulnerabili ty to risks and shocks. Figure 1-1. Sustainable liveliho ods framework (Source: DFID 2007) The emphasis on diversification of livel ihood strategies within the sustainable livelihood framework has been the catalyst for numerous international development agencies to utilize the framework (Brocklesby & Fisher 2003). It prom otes the creation of livelihood enhancement beyond just employ ment. Projects that have used the framework have realized that one source of income is not sufficient to truly aid in poverty alleviation and envir onmental management in developing regions (Ashley & Carney 1999). For example, tourism promot ion as a pro-poor strategy in developing regions has been unsuccessful due to the primary emphasis levied on one single strategy (Tao & Wall 2009). Tour ism seldom occurs in isolat ion as it competes for the 21

PAGE 22

use of scarce resources; therefore, tourism development projects need to be initiated as a complement to existing activities (Ash ley 2000). A comprehensive understanding of the link between tourism and other sectors requires an analysis of the whole context in which tourism exists (Tao & Wall 2009). Ba sed on the first study of the relationship between tourism and the sustainable liveliho ods framework, Tao & Wall (2009) state that a sustainable liveliho ods approach is more practical, especially in common situations in which communities and individuals sustain themselves by multiple activities rather than discrete jobs. T ourism is not a panacea for all problems faced in developing countries, and is not a reliable source of in come in many marginal economies, but it may supplement incomes derived in other ways and help disperse vulnerabilities to risk (Tao & Wall 2009). Therefore, advocates of sustainable development should explore how tourism might fit into a su ite of livelihood strategies, c ontributing to the achievement of sustainable livelihood outcome s (Tao & Wall 2009, p. 8). From a sustainable community devel opment perspective, the sustainable livelihoods approach analyzes a communitys assets, beyond needs and determines ways to promote the means to meet thos e needs while simultaneously increasing their capacity to combat shocks and threats (Carney et al. 1998; Scoones 1998; Ellis 2000). Projects that utilize the sustainable livelihoods framework also emphasize resident empowerment and capacity building to e ffectively enhance access to available resources that allows them to choose livelihood strategies to negotiate vulnerabilities, and incrementally improve their livelihood assets (Ellis 2000). If an individual retains a useful skill, it should lead to empowerment and thus may lead to further levels and roles of participation (Agrawal & Gupta 2004). 22

PAGE 23

According to local, regional, and national le vel institutions and norms, access to necessary livelihood assets is often avail able disproportionately (Ellis 2000; Agrawal 2001; Blaikie, & Springate-Bagi nski, 2003). Associated access is often correlated to the role of participation in project planning, management, and capacity building programs which is based on preexisting power struct ures that is linked to socio-economic characteristics (Rew & Rew 2003; Agrawa l & Gupta 2004). P oorer households are usually more dependent on natural resources for their livelihoods but have less access to them (Adhikari et al. 2004). Without access to res ources, poorer households are more likely to have less diversified liveliho od strategies, and thus are more dependent on natural resources in protected areas (Barret et al. 2001; Bhandari & Grant 2007). Households are better suited to cope with vulnerability to shocks such as drought and seasonality in markets through a more dive rse livelihood strat egy (Ellis 2000). However, disadvantaged households more often bear an unequal s hare of the costs associated with conservation. Furthermore high levels of resource dependency and the lack of participation role s in community management limit s them from opportunities to either enhance their livelihood strategies or to graduate to higher order strategies (Scoones 1999; Ellis 2000). Given that disadvantaged households are held to mere subservient roles in decision making and only allowed minimal access, if at all, to regional social, physical, financial, and envir onmental assets; the ability to achieve a diverse livelihood strategy is severely impacted (Agrawal 2001; Agrawa l & Gupta 2005). Statement of Problem Nepal is considered a l eader among least developed nations with respect to environmental policy and conservation programs Nepal has designated over 18% of the country as protected (26,695 km2) in 17 different protected areas. More importantly, 23

PAGE 24

Nepal has been a pioneer in decentralized conservation programs in parks and protected area management. Such programs highlight the link between conservation and socio-economic development among residents within protected areas. In 1986, the Annapurna Conservation Area Pr oject (ACAP) became the first Integrated Conservation and Development Project (ICDP) implemented by a loca l non-government organization in collaboration with the World Wildlife Fund fo r Nature (WWF). The concept behind the project was economically self-sustainab le conservation with development that emphasized popular local participation vi a a decentralized form of conservation management (Sherpa et al.1986). ACAP has been utilized as a successful case study within the discourse of ICDPs because it has provided increased development, improved distribution of benef its, empowered local resi dents, and enhanced biological conservation (Nyaupane et al. 2006). However, equal distribution of benefits and influence has been illusive in ACAP (Nyaupane & Thapa 2004; Bajracharya et al. 2005; 2006). Disadvantaged community members, including the poor, lower caste, and females are excluded from decision-making roles within conservation management and community development issues (Baral et al. 2008). Participatory exclusion has led to an inequitable distribution of benefits, greater burden of conser vation related costs, and higher levels of dependency on natural resources (Bajracharya et al. 2006). Furthermore, the regulatory and financial stru cture is still largely top-down as the ultimate authority still remains with the Department of National Parks and Wildlife Conservation (DNPWC) (Heinen & Mehta 1999). The most recently created conserva tion area in Nepal is the Kanchenjunga Conservation Area Project (KCAP) that wa s established in 1997. The government 24

PAGE 25

officially handed the management authority to the local populati on on September 22, 2006, although WWF still provides technical and financial assistance. It is the first protected area solely managed via local aut hority in Nepal, and possibly Asia (WWF 2006). Nepals DNPWC and WWF jointly managed KCAP for the previous ten years of existence. During the past 10 years, enorm ous positive strides have been achieved with respect to the management of KCAP. KCAP functions as an ICDP with major international donors including the Worl d Wide Fund for Nature, the MacArthur Foundation, the Kadoorie Agricultural Aid Associ ation, and the Darwin Initiative. KCAP, named after the third highest mountain in the world (Mt. Kanchenjunga, 28,169 ft), is a sparsely populated region comprised primaril y of poor ethnic minorities who are heavily dependent on the natural resources for their subsistence. Pr imary livelihood activities include agriculture, livestock, non-timber fo rest products, and tourism (Muller-Boker & Kollmair 2000). WWF has utilized the sustai nable livelihoods framework in their management and evaluation of project objectives (WWF 2006). Successful management of KCAP is not only a priority fo r Nepal, but also fo r the international conservation community. KCAP is a vita l component for the eventual creation of an international Kanchenjunga mountain comple x conservation area that includes the Khangchendzonga Biospere Reserve in Sikkim, India and the Qomolangma Nature Preserve in Tibet, China. In KCAP, a hierarchy of management has been instituted which advocates for equitable access to decision-making roles an d benefits distribution for all residents. There has been several capacity building in itiatives formulated while numerous community development projects have been successfully implem ented (WWF 2006). 25

PAGE 26

However, major conflicting issues especially among local residents are still evident. Residents have complained about spatial dispar ities, in that higher elevation villages receive less access to community dev elopment programs and capacity building initiatives despite residing in less productive lands (Muller-Boker & Kollmair 2000; Ikeda 2005). Without alternatives, these villa ges continue to practice slash and burn agriculture, overgraze pastures, and over harvest non-timber forest products (WWF 2006). Equitable distribution of benefits has also been identified as being disparate along socio-demographic lines, and possibly highl y correlated with levels and roles of participation in community development and decision-making in management (MullerBoker & Kollmair 2000; WWF 2006). The unequal distribution of benefits has been especially taxing for poorer households who are now inter-dependent on access to community natural resources (WWF 2006). Therefore, programs need to be able to identify and target those households most reliant on resources access and extraction activities, and subsequently, provide a m eans for creating sustainable livelihood strategies that will allow them the opportunity to overcome and persist while pressures on the natural resources are alleviated (Ellis 2000). Purpose of Study The purpose of this study was to examine the role of participation in sponsored programs1, levels of natural resources dependency, and the existing levels of livelihood strategy diversification among households in the Kanchenjunga Conservation Area Project. Specifically, the following thr ee major research ques tions were examined (Figure 1-2): 1 Sponsored programs include community based user groups that provide households with financial or technical assistance in decision-making, both individual and communal. 26

PAGE 27

1. What factors affect participation roles in sponsored programs by households within KCAP? Do participation roles in spons ored programs impact the distribution of benefits among households? 2. What are the levels of natural re sources dependency for households within KCAP? What factors affect a households le vel of natural resources dependency? 3. What is the household level of livelihood st rategy diversification within KCAP? What are the factors associated with domi nant livelihood strategy adoption? Figure 1-2. Conceptual model Project Significance This study examined the factors that affect household ro les of participation in sponsored programs, the associ ated distribution of benefits based on participation roles, the factors that af fect household levels of natur al resources dependency, and the potential factors or barriers associated with household adoption of more diversified and P P a a r r t t i i c c i i p p a a t t i i o o n n R R o o l l e e s s D D i i v v e e r r s s i i t t y y o o f f L L i i v v e e l l i i h h o o o o d d S S t t r r a a t t e e g g i i e e s s N N a a t t u u r r a a l l R R e e s s o o u u r r c c e e D D e e p p e e n n d d e e n n c c y y I I n n c c o o m m e e E E d d u u c c a a t t i i o o n n C C a a s s t t e e / / E E t t h h n n i i c c i i t t y y H H o o u u s s e e h h o o l l d d S S i i z z e e D D i i s s t t a a n n c c e e t t o o P P r r o o g g r r a a m m O O f f f f i i c c e e s s L L a a n n d d h h o o l l d d i i n n g g s s a a n n d d L L i i v v e e s s t t o o c c k k Research Question 2 Research Question 3 Research Question 1 27

PAGE 28

remunerative livelihood strategi es in the Kanchenjunga Conser vation Area Project. This research was conducted in a mountainous protected area within a completely decentralized community based management structure in Nepal, if not Asia (WWF 2006). This research is important for pr ogram designers and evaluators in both KCAP and beyond as it contributes to the discourse for a need of a robust quantitative method to operationalize diversified livelihood strategies. Results of this study demonstrate the importance of tar geting and empowering disadvantaged households within decentrali zed conservation programs. Education programs that emphasize community partici pation are better able to reach these disadvantaged households and provide for a mo re equitable distribution of benefits. Findings also aid conservation management by providing a methodology that enhances project abilities to identify households who lack access to the assets necessary to graduate to an improved livelihood strategy. Since these households are most heavily dependent on natural resources, and therefore are most susceptible to management actions or policy changes. Findings also aid policy makers by enhanci ng their ability to identify environmental pressures that are most contingent on further policy applications, while simultaneously illuminating those strate gies that are most critical to the communities livelihood. This research in corporated previous findings of WWF and intends to further enhance WWFs and other (I)NGOs ability to balance the needs of both nature preservation and community devel opment. The framework utilized in this research has been used by USAID, Oxfam, DFID, UNEP, and Care International. This research also further enhances the field implementation of conservation and development programs in the Kanchenjunga C onservation Area. Re search results will 28

PAGE 29

be shared with local community based dec ision makers, WWF staff, and the Department of National Parks and Wildlife Co nservation. Recommendations based on this research will augment their abilities to design management strategies that alleviate detrimental environmental pressures while c oncurrently promoting more sustainable community development via c apacity building programs. Fi gures 1-3 through 1-17 are provided for visual c ontextualization. Dissertation Format The dissertation is segmented into five chapters via three thematic areas based on each research question. The three re search questions build upon each other with assessments of household participation ro les in sponsored activities, household resource dependency, and household levels of livelihood diversification. Chapter 2 addressed household role of participation in s ponsored activities within KCAP. More specifically, socio-cultural factors were identified that influenced decentralized conservation participation and det ermined if participation roles affected the distribution of benef its. The factors that affect participation roles were analyzed using an ordered probit to assess the influence of these factors on participation roles from members, discussion le aders, suggestion makers, and decision-makers. A two stage linear model was utilized to determine benefits received from participation based on their role of participation. Results determined if those who were more heavily involved in decision-making processes withi n the decentralized conservation framework accrued more benefits than costs. Chapter 3 addressed natural resource use patterns of the local communities within KCAP. Natural resources dependency was estimated and converted to a binary scale which divided the population into high and low dependency households. A logistic 29

PAGE 30

regression assessed the affect varying economic, social, and access variables had on high or low dependency by the hous eholds in the community. Chapter 4 utilized cluster analysis to identify discernable differences between existing livelihood strategies, and employed a mu ltinomial logistic regression to identify existing correlates of access to graduate to more effective livelihood strategies. Results indicated how a household distributes its access to available capital to activities chosen livelihood strategies. Chapter 5 provided an overall summary of how the research questions utilized within chapters 2, 3, and 4 build upon each ot her and resultant utilit y provided. This chapter presented summaries of how parti cipation in decentralized conservation strategies in a mountainous protected area in Nepal have impacted household dependencies on natural resources and the associ ated effect on the household level of livelihood strategy diversification. 30

PAGE 31

Figure 1-3. Village of Taplechok (Source: Author 2008) Figure 1-4. Village of Olanc han Gola (Source: Author 2008) 31

PAGE 32

Figure 1-5. Mountainsi de (Source: Author 2008) Figure 1-6. Typical valley floor (Source: Author 2008) 32

PAGE 33

Figure 1-7. Prayer wheels in Ol anchan Gola (Source: Author 2008) Figure 1-8. Yak train in Olanc han Gola (Source: Author 2008) 33

PAGE 34

Figure 1-9. Lelep villa ge (Source: Author 2008) Figure 1-10. Typical house in Lelep village (Source: Author 2008) 34

PAGE 35

Figure 1-11. KCAP headquarters office in Lelep village (Source: Author 2008) Figure 1-12. Village of P hele (Source: Author 2008) 35

PAGE 36

Figure 1-13. Village of G hunsa (Source: Author 2008) Figure 1-14. Tent camp in Ghunsa village (Source: Author 2008) 36

PAGE 37

Figure 1-15. Typical Ghunsa vill age kitchen (Source: Author 2008) Figure 1-16. Close-up of cook ing area (Source: Author 2008) 37

PAGE 38

Figure 1-17. Local resi dent (Source: Author 2008) 38

PAGE 39

CHAPTER 2 PARTICIPATION ROLES AND BENEFIT S IN SPONSORED PROGRAMS AMONG HOUSEHOLDS WITHIN KANCHENJUNG A CONSERVATION AREA, NEPAL Introduction Nepal formally embarked on decentralized pa rticipatory conservation programs in 1990. However, the policies that endorse decent ralization and participation to achieve conservation goals are a debatable and cont entious issue, especially in parks and protected areas. Some authors have vehement ly advocated for additional regulations and management that pot entially excludes local participation and cohesion (Robinson 1993; Brandon et al. 1998; Terborgh et al. 2002), while others s upport further devolution of power within parks and protected areas in Nepal (Brandon & Wells 1992; Gibson & Marks 1995; Ribot 2003; Baral & Heinen 2007). Nevertheless, the need for effective integration of conservation and development in protected areas should not be viewed as islands of conservation divorced from the social and economic context within which they are located (IUCN 2003). Participatory m anagement strategies that emphasize the need for local communities to share in protected area benefit s and decision-making should be promoted and achieved. Additi onally, decentralization and resident participation in conservation programs s hould provide a balance between social and conservation goals (IUCN 2003). Advocation for decentralized conservation management is based on the assumption that lower-level decision make rs, who are otherwise generally excluded are now more likely to have enhanced access to technical and financial assistance, minimized organizational costs, and wider pa rticipation within an in creasingly creative and enriched economic sector (Ribot 2002; Agrawal & Gupta 2005). Protected area managers need to identify the potential positiv es and negatives that should be targeted 39

PAGE 40

and simultaneously avoided while conserva tion and livelihood issues are addressed (Baral & Heinen 2007). In Nepal, participation in decentralized conservation management determines the quantity and types of benefits and costs received (Maskey et al. 2006), which are based on socio-economic variables, including hous ehold income, education, and race/caste/ ethnicity (Agrawal 2000; Allendorf 2007; Baral & Heinen 2007), and higher levels of forest dependency (Jumbe & Angelsen 2007). The most significant benefits are consumptive (e.g., firewood, timber, food, herbs, medicines, construction materials, etc.), and the ability to earn income from recreation and touris m (Wells 1992). However, associated costs include loss of crops and livestock due to wildlife, injury and death, loss of access rights, and time and resources spent guarding against wildlife encroachment (Baral & Heinen 2007). Overall, there is a paucity of empirical research that examines the distribut ion of costs and benefits of de centralized participatory conservation at the household level (Baral & Heinen 2007). Kanchenjunga Conservation Area Project (KCAP), with its complete devolution of management authority to the local population pr ovides an exceptional opportunity to assess integrated decentralized participatory conservation pr ograms in protected area management in Nepal. This research seeks to assess if socio-cultural factors influence decentralized conservation parti cipation, and to determine if participation roles in sponsored programs affect the distribution of benefit s at the household level. Results will determine if those househo lds who are more heavily involved in decision-making processes within the decentralized conser vation framework accrue more benefits than costs. Specifically, this study will examine two questions: 1) What factors affect 40

PAGE 41

participation roles in sponsored program s by households in KCAP; and 2) Do participation roles in sponsored programs2 impact the distribut ion of benefits among households? Results of this study will demon strate the importanc e of targeting and empowering disadvantaged households, while educ ating them about the significance of participation in order to have a more equitable distribution of benefits within decentralized conservation programs to enh ance their sustainable livelihoods. Literature Review Decentralized Conservation Participation Participation in decentralized conserva tion management programs in Nepal is highly correlated with demogr aphic variables such as income, education, spatial location, caste/race/ethnicity and gender (Me tha & Kellert 1998; Sharma 2000; Blaikie & Springate-Baginski 2003; Dev et al. 2003; Neupane 2003; Smith et al. 2003; Adhikari et al. 2004; Maskey et al. 2006; Allendorf 2007). Income is a statistically significant predictor of participation in sponsored and decentralized programs. Households with gr eater landholdings have a higher degree of participation (Maskey et al. 2006), while poorer households participate less than wealthier and politically-connected household s (Agrawal 2000; Baral & Heinen 2007; Dev et al. 2003; Smith et al. 2003; Adhikari et al. 2004; Maskey et al. 2006; Allendorf 2007). Additionally, poorer hous eholds are not necessarily interested in participation as they have higher associated costs since it takes time away from more productive employment opportunities (Agraw al 2000; Timinsa 2003). 2 Sponsored programs include community based user groups that provide households with financial or technical assistance in decision-making, both individual and communal. 41

PAGE 42

Education can have both positive (L ise 2000) and negative (Agrawal & Gupta 2005) affects on participation. Residents with hi gher levels of education are more likely to understand the importance of forest cons ervation, and are thus more likely to participate (Jumbe & Angelsen 2 007). However, higher education levels might also lead to a greater involvement in alternative income generating activities, and subsequently create less interest in decent ralized conservation participation (Jumbe & Angelsen 2007). Older individuals from wealthier and higher caste/race/ethnicity households are more likely to participate and to be involv ed in decision-making from a leadership position especially within comm unity forest groups (Maskey et al. 2006). Mehta & Kellert (1998) found that wo men and low-caste people were highly underrepresented in executive committees responsib le for the management of loca l resources. Given the social system in Nepal, community based natural resources programs need to ensure equal participation of all associated interest groups to reduce to the risk of exclusion (Baral & Heinen 2007). Agrawal & Gupta (2005) assessed the major factors that influenc ed participation in the United Nations Developm ent Program sponsored Parks-People-Program in four protected areas buffer zones in Nepal. Re sults identified that supplementary income, landholdings, caste, firewood consumpt ion, household size, and the frequency of household visits to program o ffices were statistically signi ficant positive predictors of participation. However, education level was a statistically signif icant negative predictor of participation. Higher levels of educat ion led to more lucrative wages and less dependency on natural resources, and therefore decreased interest in sponsored 42

PAGE 43

management programs. Overall, frequency of visits to progr am offices and benefits received from participation had the greates t impact on participation. The authors argued for strategic management in favor of marginalized and poorer households in order to improve accessibility to government offices and respec tive officials (Agrawal & Gupta 2005). Additionally, Lachapelle et al. (2004) identified three broad hindrances to popular participation in decentralized conservation m anagement as: 1) vulnerability (lacking private resources), 2) inferiority (bas ed on discrimination due to caste, gender, and education), and 3) the pot ential for corruption, especially related to lack of financial transparency. Decentralized Conservation Costs and Benefits In Nepal, poorer households are more re liant on natural resources, but the absence of participation among disadvantaged households in sponsored programs has excluded them from decision-making processes in product distribution, and hence they receive fewer benefits (Agrawal 2 000; Baral & Heinen 2007; Dev et al. 2003; Smith et al. 2003; Adhikari et al. 2004; Maskey et al. 2006; Allendorf 2007). Within decentralized conservation programs, it is often the wealthier households that dominate decisionmaking, and push for higher income timber production which takes a longer time for maturity, thus creating greater co sts for poorer households (Adhikari et al. 2004). Primarily, people who are wealthy and politica lly-connected receive the most benefits derived from participation and from communi ty development projects (Adhikari et al. 2004; Agrawal & Gupta 2005). Domination of decision-making by the wealthy elite can also influence community development proj ects that exclude disadvantaged households (Timinsa 2003). Thompson-Slayter & Bhatt (1994) found the richest seventeen families 43

PAGE 44

within one village in Nepal were the recipi ents of project aid targeted toward the disenfranchised. However, food-insecu re households (poor households) have also been found to participate more often to ens ure access to resources (Jumbe & Angelsen 2007). Spatial location of households and agency o ffices shares a significant affect on participation roles and accrual of benefits. Meht a & Kellert (1998) found that residents living in remote areas of the Annapurna Co nservation Area Projec t (ACAP) and the Makalu-Barun Conservation Area Project received substantially less community development related benefits, including dr inking water systems, bridges, and health clinics, than their counterparts who lived in closer proximity to project offices. Maskey et al. (2006) found that socio-economic factors affected the level of participation in decentralized conservation, and the role of participation was based on the benefits received in the middle hills of Nepal. Lack of participation excluded disadvantaged households from decision-making in product distribution; therefore they received fewer benefits from the community forest. Community forest members who held a decision-making role within user gr oups received a higher quantity of forest benefits that included fuel w ood, fodder, and timber compared to those members who merely attended meetings. The authors call for empowerment of people with lower socio-economic standing to understand the im portance of participat ion in order to enhance an equal distribution of accrued benefits (Maskey et al. 2006). Baral & Heinen (2007) assessed residents who had participated in communitybased conservation programs and the associated di stribution of benefits accrued in two western protected areas in the terai lowlands of Nepal: S huklaphanta Wildlife Reserve 44

PAGE 45

and Bardia National Park. Gender, educati on, household affluence, and conservation attitudes were significant predictors of parti cipation, while family size, ethnicity, and resource dependency lacked predictive validity. Respondents noted that more benefits than costs were associated with participation in conservation program s. Overall, the sustainability of conservation programs depended on the continued devolution of power to local communities to sustain achievem ents, and to provide a means for a more equitable distribution of benefits (Baral & Heinen 2007). Kanchenjunga Conservation Area Project, Nepal KCAP is one of the most remote and sparsely populated protected areas in Nepal and is named after the third hi ghest mountain in the worl d (Mt. Kanchenjunga, 8586 m). In April 1997, the governm ent declared the region with its unique biological characteristics, high density of glaciers high biodiversity indices, and endangered wildlife as a Gift to the Ea rth in support of the World Wide Fund for Natures (WWF) Living Planet Campaign (Hei nen & Mehta 1999). The conservation area was officially created in July 1997 and contains 2,035 km2 of area primarily comprised of high mountain landscapes of glacie rs and rocks, with only a small amount of arable land (2% of the total). KCAP boasts 16% of Nepals total floral species and seven mammals with conservation significance including the red panda, snow leopard, grey wolf, Himalayan black bear, Himalayan tahr, blue s heep, and musk deer (WWF 2006). The aim of KCAP is to safeguard the bi odiversity of the ar ea, and improve the living conditions of the local residents by strengthening the c apacity of the local institutions responsible for decision making, which will effect the long-term biodiversity conservation and economic development of t he area (WWF 2006, p 2). The projects 45

PAGE 46

primary focus is to strengthen the capacity of local communities to manage their natural resources while concurrently improv ing their livelihood opportunities. KCAP functions as an Integrated Conservation and Development Project (ICDP) with major donor agencies that include WWF, the MacArthur Foundati on, the Kadoorie Agricultural Aid Association, and the Darwin Initiative. The project was established with financial and technical assistance provided by WWF. WWF has invested over US $1.5 million and has pledged continued support fo r another five years through the Sacred Himalayan Landscape project via a partnership with the Integrated C entre for Integrated Mountain Development (I CIMOD) (WWF 2007). Programs and Management WWF and the government have created a dem ocratically elected hierarchy of management at KCAP based on local involvem ent and empowerment to achieve selfsustainability. The project has been implementing programs in partnership with local community-based organizations (collectivel y referred to as sponsored programs), namely the Kanchenjunga Conservation Ar ea Management Council, seven User Committees, forty-four User Grou ps, and thirty-two Mother Groups3. The government recently transferred managem ent of the area to the Kanchenjunga Conservation Area Management Committee on September 22, 2006 wh ich established KCAP as the first community-managed conservation area in Nepal if not in Asia (WWF 2006). Several achievements have been garnered by KCAP since its existence in regards to accomplishment of stated objective s (WWF 2006). KCAP further reassigned management responsibilities of 47,772 ha to sixteen Community Fo rest User Groups 3 Mother Groups are sponsored user groups ex clusively comprised of female members. 46

PAGE 47

that represents over 891 hous eholds and includes one Mother s Group. In addition to providing a system for int egrated local management, 32 womens savings and credit groups, 25 revolving funds, a NRs 1,000,0004 fund for Non-Timber Forest Products (NTFPs) enterprises, and a NRs 100,000 livesto ck insurance program were created. Community development projects have included the delivery of meta l prayer flagpoles to replace the traditional custom of annual harvesting of trees, im proved cooking stoves to decrease pressures on forest use, and a micro-hydro scheme in Ghunsa and Gola. Additional projects include a child day care center, drinking water systems, 1,200 meters of new trails, suspension bridges, to ilets, a health post, an irrigation canal, and green houses. Although various WWF projects have been in operation since its establishment, only five percent of the residents knew that WWF was the main operational organization. Also, only a quarter of respondent s were aware of conservation related objectives, while over half knew of econom ic development related objectives (MullerBoker & Kollmair 2000). WWF (2007) also found that very few residents knew about the objectives of KCAP or the main organi zation that managed it. Local resident expectations were grossly unrealistic as the most often cited perceived objectives were large infrastructure construction. Local resi dents did maintain positive attitudes towards capacity building programs and perceived the project had in creased wildlife populations (Muller-Boker & Kollmair 2000). However, an increase in wildlife has led to increased frequencies of predation and crop damage. Residents also expressed negative attitudes towards social stratification of the area and the inequitable distribution of 4 At the time of research (October 1st, 2008) 1 USD = 75 Nepali Rupees (NRs). 47

PAGE 48

benefits based on differences betw een ethnic minorities, proximit y to project offices, and the Maoist insurgency5 (Muller-Boker & Kollmair 2000). Conservation Issues There have been five major environmental pr essures exerted by local residents identified at KCAP (WWF 2006). Explanations and reasons for the pressures are a mix of unclear ownership or tenure rights, la ck of policy and enforcement, use of poorer households for labor, low agricultural productivity levels of existing lands, food security, low levels of land ownership, a lack of ca sh crop market linkages, and external market demands for traditional medicines. Access to natural resources is limited to local residents and regulated by local institutions however outsiders can buy access permits (Ikeda 2005). Traditional local institutions of communal natural resources management have existed without much government interv ention due to the remoteness of the area (Muller-Boker & Kollmair 2000). First, over 170 households practi ce slash and burn techniques for agricultural purposes at high elevation (above 3,00 0 m) forests in the northwestern part of the c onservation area. Second, alpine meadows are being destroyed due to overgrazing, fuel needs, and hous ing construction. Third, there is a large amount of timber being over-harvested for purposes rela ted to residential cooking, heating, and construction, especially by re sidents who are heavily dependent on those resources for their livelihoods. Higher elev ation villages are more heavily dependent on natural resources access for their livelihood s (Muller-Boker & Kollmair 2000). Fourth, an over-harvesting of NTFPs exists that includes sea buckthorn, cardamom, and 5 The Maoist insurgency was the decade long civil war between communist rebel group and the democratic national government. The insurgency ha s ended with the rebel group integrated into the mainstream politics. See Thapa (2004), Battarai et al. (2005), and Baral & Heinen (2006). 48

PAGE 49

medicinal plants (WWF 2006). Lastly, poaching and retaliatory killings of wildlife, that includes snow leopards6, are also evident (I keda 2005; WWF 2006). Study Site KCAP is located in the nor thern portion of the district of Taplejung (see Figure 21). According to the 2001 National Cens us, of the 134,698 people in the district, 42.35% of women and 62.5% of men were literate. The average annual population growth rate from 1991 to 2001 was 1.21%. The district of Taplejung has a Human Development Index score of 0.328, which is 6th from the bottom in Nepal (out of 75 districts). There are 33,675 people per doctor and 8,980 peopl e per hospital bed. Only 47.5% of the district population has access to toilet facilities, and 61% have access to improved water supplies (Central Bureau of St atistics 2001). There is only 37 km of roads that are also unpaved in t he entire mountainous district. Around 4,570 people in approximately 860 househ olds reside in the four Village Development Committees (VDCs) that co mprise KCAP. The four VDCs are Olangchung, Lelep, Yamfudin, and Tapethok (WWF 2006). Many ethnic minorities that include Sherpa (Bhote), Lim bu, Rai, Gurung, Magar, T hakali, Tamang, Tibetan, and Jirel reside in the area. Over 48% of the population is Limbu, while another 25% is Sherpa (WWF 2006). The main sources of in come include subsistence agriculture and animal husbandry. Additional sources of live lihood activities include cottage industries, tourism, employment in t he military, trade with Tibet, and seasonal labor out-migration (Muller-Boker & Kollmair 2000). The average food sufficiency in KCAP is 5.5 months 6 The Ghunsa Valley contains forty percent of the av ailable snow leopard habitat in KCAP and is also an important part of the regional tourist trekking route. 49

PAGE 50

and less than 10% of household s produce enough food for subsistence consumption (WWF 2006). Figure 2-1. Map of KC AP (Source: WWF 2006) The existing conditions in KCAP are quite different than thos e found at the wellknown and highly successful model of the Annapurna Conservation Area Project (ACAP) in central Nepal. While ACAP has received over 75,000 tourist arrivals annually, KCAP has only been open for trekki ng since 1988 and tourist arrivals peaked at 800 in 2001, and then decreased to 175 in 2006 (WWF 2007). Lack of transportation infrastructure, tourist facilities, and the Maoist domestic insur gency are the primary reasons for the decline in t ourist arrivals (WWF 2007). Methods This study utilized a mix of both qualit ative and quantitative approaches. The combination of qualitative and quantitative methodology has been gaining credibility and 50

PAGE 51

relevance in the development literature (Booth et al. 1998; Kanbur 2001; White 2002; Ellis & Freeman 2004). This study utilized a tr iangulation method to improve the validity of the data. Triangulation is described as collecting information from multiple sources and methods of inquiry. This process r educes the chance of researcher bias and improves research validity (Denzin & Lincoln 2000). Data Collection Data collection was conducted over thr ee months (October through December 2008) and employed four methods. First, projec t reports were reviewed from the World Wildlife Fund-Nepal office, the Department of National Par ks and Wildlife Conservation, and KCAP offices in the capital Kathmandu and the research site. Information on past and current projects, policies, programs, and other management issues were examined. Second, participant observation methods were conducted to develop a better visual context of the communitys perspective and behavior. Participant observation was focused on resident livelihood strategies, nonverbal cues during participation in sponsored programs, and sentiments toward the possible presence of equitable social inclusion within the decentralized conservation program. Third, a Participatory Rural Appraisal (PRA) wealth ranking (World Bank 2007) was conducted at each village. The PRA wealth ranking allowed each community to familiarize themselves with the presence of the researcher and voice thei r views before individual households were selected. PRA wealth ranking techniq ues are commonly used in conjunction with quantitative surveys as a means to categor ize populations for stratified sampling procedures or for categorical analysis pos t data collection (Grandin 1988; Ellis & Freeman 2004; World Bank 2007). Fourth, a detailed interview survey was conducted among a random selection of households throughout the sampling area. The 51

PAGE 52

combination of methods was adapted from the UK Department for International Development (DFID) sponsored LADDER7 project (Ellis & Freeman 2004). The primary researcher previously lived in Nepal for 27 months and is proficient in the national Nepali lan guage; however, a local person from the research area aided the interviewer in the field and provided assist ance for rapport building (Nyaupane & Thapa 2004). Hiring a local research assistant al so minimized cross-cultural bias and nonsampling error (Walpole & Good win 2002). The interviews lasted up to half an hour in duration and was dependent on t he respondent. Survey questions were spoken in Nepali or other local dialects when required by the assistant. In tr anslation, local words were used wherever possible and technical jargon avoided. A professor at Tribhuvan University, Kathmandu, Nepal, assessed and verified survey translation accuracy from English to Nepali, and t hen back to English. Selection of Participants Based on a recent census by WWF (2006) the sample population constituted the entire population that resided within KCAP borders and comprised of 4,570 people in 860 households. The population was spatially distributed over four Village Development Committees (VDCs) that incl uded Lelep (47%), Tapetok (31.3%), Yamphudin (15.1%), and Olachan Gola (6 .8%). A systematic sampling method was used to select respondents. Households were targeted, visited and the head of household was requested to participate. A comprehensive list of village households was created for each village dur ing the PRA wealth ranki ng process. From the comprehensive list, every fourth house was surveyed after a random start for a total 7 LADDER stands for Livelihoods and Diversific ation Directions Explored by Research. 52

PAGE 53

population of 215 (response rate = 100%), which was representative of 25% of the total households in KCAP8. The head of the household ( > 18 years of age) was personally interviewed at their respective residence. Attemp ts were made to contact the head of the household, but in cases of unavailability, the most knowledgeable adult member was requested to participate. If members of the household were unava ilable, the next house was selected. Since local residents are not familiar with survey procedures, and more importantly, due to the high rate of illiteracy, questions were verbally read to participants while the interviewer recorded the responses (see Nyaupane & Thapa 2004). Instrumentation The questionnaire consisted of nine sectio ns. The first section examined basic household data includi ng education, occupation, and race/caste/ethnicity. The second section assessed household levels of participation in sponsored programs including forest and conservation user groups. Se ction three measured income related to agriculture. Section four assessed income generated from livestock, privately owned natural resources and community based natural resources, that included NTFPs. Section five measured non-farm related so urces of income such as remittances9 and household responses to unforeseen crises. A pilot test was conducted among project and government agency staff in the distri ct headquarters of Taplejung, and questions were slightly altered or removed as deemed necessary. 8 A sample size of 205 within a population of 860 households is sufficient for categorical data analysis with the assumption of 0.05 alpha level and a 0.05 margin of error (Cochran 1977; Bartlett et al. 2001). 9 Remittance is considered payments to friends and family in a migrants home of residence, practiced by over 190 million migrants totaling US $433 billion in 2008, of which US $328 billion went to developing countries (World Bank 2009). 53

PAGE 54

Operationalization and Data Analysis This study assessed two research questions. Research question 1: What factors aff ect household participation roles in sponsored programs? An ordered probit regression model asse ssed the existing household participation patterns in sponsored programs (see Gr eene 2000; Agrawal & Gupta 2005; Maskey et al. 2006). Dependent variables within ordered probit models are represented by an arbitrary ranking of numbers wh ile normality is assumed (Maskey et al. 2006). Table 21 presents the twelve independent variables utiliz ed to assess participation levels within economic, social, and access variable t hemes. The dependent variable measured household roles of participation in sponsor ed programs and was op erationalized as: 1=attendance, 2=suggestion, 3=discussion, 4=decision making (see Maskey et al. 2006). However, due to a low distribution of frequencies among the varying categories, the participation variable was recoded as : 0=non-membership, 1=general member, and 2=leading member. The role of participation in the first model is a function of income, age, education, remittances, caste, hous ehold size, dependents, livestock, land holdings, distance to nearest program office, office visits, and staf f visits. The ordered probit model utilized for research question 1 is as follows: Pi = 1 Income + 2 Highest education + 3 Age of head of household + 4 Education of head of household + 5 Remittances + 6 Caste + 7 Household size + 8 Dependents + 9 Livestock + 10 Landholdings + 11 Distance to program office + 12 Office visits + 13 Staff visits + e 54

PAGE 55

Table 2-1. Operationa lization of variables Name of variable Operationalization Participationa Household participation role in sponsored groups, 1=attendance, 2=suggestion maker, 3=discussion leader, and 4=decisionmaking. Recoded due to low frequency as 0=non-member, 1=general member, and 2=group leader. Economic Variables Remittanceb A yes or no dichotomy based on the presence of remittance. Incomeb A yes or no dichotomy based on the presence of income in addition to agriculturally deriv ed income including government salaries, tourism, pensions, or other entrepreneurial activities. Landholdingsb,c Land under the control of the household in ropani (19.65 ropani = 1 hectare). Livestockb,d Collected in absolute value and converted to livestock units for analysis (1 Livestock unit = 1.2 yaks, 1.2 cows, 1.2 buffalo, 5 sheep, 4 goats, 2 calves, 5 pigs, 100 chickens). Social Variables Education of head of householdc Highest level of education attained by the head of household. Highest education Highest level of education attained by any adult individual (18 years of age or older) living within the household. Sizeb The number of people residing within the household. Dependentsa The number of dependents (<18 years old) residing within the household. Age of head of householda Age of the head of household. Caste/Ethncitya,b A categorical variable measured as 5=Brahmin, 4=Chettri, 3=Sherpa, 2=Limbu, and 1=Rai. This variable was recoded as 1=Sherpa and 0=other due to low frequency distributions in other categories. Access Variables Office visitsb A yes or no dichotomy based on the presence of household members visits to KCAP offices over the past year (1=yes, 0=no). Staff visitsb A yes or no dichotomy based on the presence of KCAP staff visiting the household over the past year (1=yes, 0=no). Distance of house to officec Duration of time to walk to a KCAP office from house. Benefits Loans borrowed Value of loans borrowed from sponsored programs by household. Communitydevelopment projects Number of community development projects perceived to be accrued by household. Trainings Number of trainings household members have participated in the past year. a Adapted from Maskey et al. (2006) b Adapted from Agrawal & Gupta (2005) c Adapted from Jumbe & Angelsen (2007) d Adapted from FAO (2003) 55

PAGE 56

Research question 2: Do participation roles in sponsored programs impact the distribution of benefits among households? A linear regression was utilized to determi ne the distribution of benefits (see Table 2-1) based on the predicted role of participation in sponsored programs identified within the ordered probit from research question 1 (Maskey et al. 2006). A two-stage linear model for benefits as a function of participat ion was constructed. Participation roles were then utilized as dummy variables with the value of 1 if the role was present and 0 if there was an absence of any participation among the three predicted roles. Model estimation used ordinary least squares r egression and dropped the intercept to avoid perfect collinearity. The benefit model utilized to answer research question 2 are as follows: Loans borrowed = f(no participation, general member, leadership position). Community development projects = f(no partici pation, general member, leadership position). Trainings = f(no participation, general member, leadership position). Results Profile of Househ old Respondents A total of 215 surveys were completed and 205 were usable. Males constituted the majority (86%) of household heads10. Sherpa and Limbu were the dominant ethnicities within the region. The aver age household was headed by an individual who was 51 years old with no education, compris ed of 5.9 members, of which 2.2 were under 18 years of age (see Table 2-2). The lower elevation ( 1,800 m) VDC of Lelep (59%) was the most surveyed of the four, followed by Tapetok (17%), Yamphudin (17%), and the higher elevation ( 3,500 m) VDC of Olachan Gola (7%). 10 Households in Nepal are generally male dominated. In addition, sons generally reside at their parents home with their wives after marriage. Generally, nuclear families are commonly in existence. 56

PAGE 57

Table 2-2. Profile of respondents Socio-demographic Characteristics Frequency Percentage (%) Gender Male 177 86 Female 28 14 Age 18 29 8 4 30 39 20 10 40 49 58 28 50 59 44 22 60 or older 75 36 Caste / Race / Ethnicity Sherpa 109 53 Limbu 84 41 Rai 8 4 Gurung 4 2 Highest Education Level of Household Head No education 180 88 1 5 years 13 6 6 10 years 12 6 More than 10 years 0 0 Highest Level of Education No education 68 33 1 5 years 21 10 6 10 years 100 49 More than 10 years 16 8 The education system within the KCAP r egion has been established relatively recently. The average head of household only had 0.8 years of education; however, the highest education of adults (18 or more ye ars old) within the hou seholds ranged from 0 to 16 with a mean of 5.8 years. Over half (50.1%) of the residents within the sampled households (n=1,180) and all re sidents over the age of 60 had never attended school. Only 16% of household members who were over 30 years of age had attended school. A cross-tabulation of age versus education level among all household members is illustrated in Table 2-3. Results indicate t hat 56.7% of all reside nts who have attended school were under 18 years of age; while 87.3% were under the age of 30. Additionally, 57

PAGE 58

only 4.6% of household members over 40 y ears of age had completed at least one year of education. Therefore, due to the lack of educati on among head of households, and considering the prevalence of nuclear families in Nepal, the variable of highest educated adult living within a household was utilized for subsequent analyses. Table 2-3. Cross-tabulation of age by years of education Years of Education Age 0 1 5 6 10 11 and up 0-17 115 (26%) 193 (43%) 141 (31%) 0 (0%) 18-29 69 (28%) 33 (13%) 133 (53%) 14 (6%) 30 39 95 (66%) 14 (10%) 30 (21%) 4 (3%) 40 and up 312 (92%) 15 (4%) 11 (3%) 1 (0%) Total 591 (50%) 255 (22%) 315 (27%) 19 (2%) Note: Not all rows will add up to 100% due to rounding error. Chisquared (9) = 547.8; Probability 0.001. Over 94% of respondents owned their homes. Houses were built between 1933 and 2008 with a mean age of 24 years. Alt hough 91.3% of respondents indicated their homes had running water, this result is mi sleading in that local running water systems consisted of gravity fed, above ground poly vinyl chloride pipes (PVC). Only 1.5% of households had indoor toilets. Although 63.1% of respondents stated their homes had electricity, the village of Ghunsa (6.5% of the sample) was the only place with a fixed micro hydro system. The ot her houses that reported electricity were based on the presence of small photo-voltaic panels. H ousehold walls were constructed of varying compositions of wood, brick, stones and m ud. Although, only 5.4% of household walls were built exclusively with wood; all houses utilized wood for construction. Bricks constituted 26.3% of house walls, and 68% utilized mud as a construction material. Roofs were primarily construc ted out of wood (62%), thatch (24.9%), and a mix of wood and thatch (13.1%). 58

PAGE 59

The main KCAP office is located in the village of Lelep, while three other branch offices are located in Gola, Ghunsa, and Yam phudin. The average time duration by foot to reach one of the four KCAP offices is one hour and twenty minutes; however, times ranged from a few minutes to five hours. Fo r the past year, only 14.6% of households stated a member had visited a KCAP office and only 22% stated a KCAP staff person had come to visit their household (Table 2-4). Table 2-4. Descriptive statistics fo r participation role analysis variables Variable Mean Standard deviation Minimum Maximum Observations Participation 0.89 0.58 0 2 205 Remittance 0.31 0.46 0 1 205 Income (non-ag) 0.53 0.50 0 1 205 Land (ropani) 22.83 23.62 0 150 205 Livestock (units) 4.52 5.21 0 49 205 Education (years/head) 0.80 2.32 0 10 205 Highest education 5.81 4.63 0 16 205 Size (number of persons) 5.87 1.50 1 10 205 Dependents 2.22 1.37 0 6 205 Age of household head 51.56 13.17 25 94 205 Caste/ethnicity 0.53 0.50 0 1 205 Officer visit (n/y) 0.21 0.41 0 1 205 Member visit (n/y) 0.15 0.36 0 1 205 Household distance to office (minutes) 84 61 1 300 205 The financial wealth of households varied. Remittances11 were received by 29.3% of households from both the district c apital of Taplejung, and abroad that included Qatar, Malaysia, Japan, and the United Kingdom. Annual va lues of remittances (in Nepali Rupees (NRs ); on October 1st 2008, 1 USD = 75 NRs) averaged NRs 96,320 and ranged from NRs 20,000 to NRs 400,000. Household land ownership averaged 23.5 ropani (1 ha = 19.65 ropani) and ranged fr om 0 to 150 ropani. Only 38% of 11 Nepal received US $1.2 billion in remittances which re flect 14.9% of total Gross Domestic Product in 2006 (World Bank 2008). 59

PAGE 60

households owned more than one hectare. Hous ehold Tropical Livestock Unit (TLU) values averaged at 4.56 and r anged from 0 to 49. Over three-quarters of households partici pated in sponsored activities via user groups, mother groups, father groups, and conser vation groups. Moth er groups had the largest representation in parti cipation (71.4% of all house holds), while only 8.3% of households maintained memberships within ot her groups. Roles of participation within these groups varied: 23.9% of households did not maintain a membership within any group; 64.4% considered themselves to be g eneral members; and 11.7% of households maintained a leadership position within a group. Research Question 1: Household Particip ation Roles in Sponsored Programs Results of the ordered probit model utilized to assess factors that influence household participation roles are presented in Table 2-5. The overall model was statistically significant with seven independe nt variables that were statistically significant. None of the econom ic variables were statistica lly significant; however, four of the social variables, and all three of the ac cess variables were statistically significant predictors of participation in sponsored programs. Four of six social variables were statis tically significant. The level of education completed by the head of household had a negative value and was statistically significant. This indicates that higher levels of education acquired by household heads resulted in a lower likelihood of househo ld members participating in sponsored programs. However, the highest level of education among adult household members had a statistically positive association with participation. This positive association indicated that as the highest education leve l within a household incr eased, so did the likelihood of participation. The size of a household was a statistically significant 60

PAGE 61

predictor of participation. Households with more members were more likely to participate in sponsored program s. Age was the fourth stat istically significant social predictor of participation in sponsored programs, but had a negative relationship. Households with older household heads were less likely to participate in sponsored programs. Table 2-5. Household participatio n roles in sponsored programs Variable Name Coefficient Standard error Economic Variables Remittance -0.005 0.254 Income 0.059 0.229 Land 0.001 0.004 Livestock 0.017 0.017 Social Variables Education of household head -0.087** 0.040 Highest education of household 0.038* 0.023 Size 0.205** 0.087 Dependents -0.098 0.100 Age of household head -0.015* 0.009 Caste/Race/Ethnicity 0.079 0.192 Access Variables Staff visit to household 0.369* 0.223 Household member visit to office 0.671*** 0.251 Household distance to office -0.004*** 0.001 Ancillary Parameters Cut 1 1.477** 0.592 Cut 2 0.730 0.585 Chi-Square (13) = 38.96; Prob (Chi-squared) 0.001; Psuedo-R-Squared = 0.210. Note: Results are based on an ordered probit equat ion where the dependent variable is ordered 0, 1, and 2. Statistical significance of each coefficient is denoted by *, **, and ***, that represents 10%, 5%, and 1% levels respectively. The ancillary parameters are used to calculate predicted possibilities of each participation role on the standardized normal distribution. All three of the access variables were statistically significant predictors of household participation roles in sponsored programs at KCAP. A visit by a KCAP officer to a household within t he past year was a statistically significant predictor of increased likelihood of participation. Hous ehold members who visited a KCAP office within the past year also had significantly increased likelihoods of participation in 61

PAGE 62

programs. However, the fa rther the distance a house was located from a KCAP office the greater the likeliho od of significantly decreased partici pation in sponsore d programs. Research Question 2: Household Participat ion Roles and Distribut ion of Benefits A linear regression was utilized to dete rmine the distributi on of benefits based on the predicted participation role in sponsor ed programs identifi ed within the ordered probit. A two stage linear m odel for benefits as a function of participation was constructed. Benefits associated wit h participation in sp onsored programs are represented in the form of an ability to access savings accounts, financial loans, and capacity development trainings (see Table 2-6). Table 2-6. Chi-squared percentage of participation roles and benefits received Savings Loans Training Value of Loans No. of Benefits Total Role No Yes No Yes No Yes <2k 2-4k 4k+ 2-3 4-5 6-7 Nonea 100 0 100 0 85 15 0 0 0 49 38 13 47 Generalb 13 87 2 98 87 13 41 43 16 49 49 2 134 Leaderc 13 87 13 87 67 33 19 38 43 46 50 4 24 Total 33 64 25 75 84 16 38 42 20 49 47 4 205 Chi Square 125.6 180.7 6.6 7.2 11.1 Df 2 2 2 2 4 p-value 0.001 0.001 0.05 0.05 0.05 a none refers to a lack of participation role in sponsored programs b general refers to a general member participation role c leader refers to a leadership role The majority (97.4%) of households that participated in sponsored groups had borrowed loans from their respective gr oups, and 92% had access to savings programs within those groups. The amoun t of household loans borrowed varied from NRs 200 to NRs 200,000 and averaged NRs 7,330. Almost a quarter (23.7%) of loans borrowed exceeded NRs 6,000 in value, wh ile 44.7% were for less than NRs 4,000. Households with members that maintained community l eadership positions were more frequently able to access loans of greater value. Al though interest rates on the borrowed loans 62

PAGE 63

varied from 1.0% to 18%; most (83%) held in terest rates less than 2.5%; the rest had interest rates over 10% per annum. Respondent s participated in training that included animal husbandry, NTFP processi ng/harvesting, accounting, se wing, cooking, forestry, and agriculture. About 16% of all households participated in training activities, with a disproportionate percent of i ndividuals from households that held leadership positions. The number of different community developm ent projects perceived to be initiated within the communities averaged 3.71 (see Tabl e 2-7). The most commonly perceived projects included trail construction, bridge construction, and drinking water facilities development. However, households with no membership in organized groups perceived a higher number of accrued community development projects. Table 2-7. Perceived household community benefits from KCAP Community Benefits Yes No Trail Construction 100 0.0 Bridge Construction 95.5 4.5 Drinking Water 93.1 6.9 Micro-Hydro 34.7 65.3 Improved Stove 32.5 67.5 Toilets 5.0 95.0 Healthcare Post 4.5 95.5 Day Care Center 3.0 97.0 Irrigation 1.5 98.5 Greenhouse 0.0 100.0 Mean household perception of community development projects received = 3.71. The estimates for the second stage lin ear model for accrued benefits as a function of participatio n are presented in Table 2-8. The model used accrued benefits as a function of participation role, as the firs t model concluded that participation role is a function of socio-economic factors. Non-members tended to be excluded and leadership positions were rewarded, and t hus benefits received were a function of participation roles. The estimates for community development and amount of loans 63

PAGE 64

borrowed were also statistically significan t. Therefore, the va lue of loans borrowed increased with the higher participation role in sponsored programs. Table 2-8. Parameter estimates for benef its received based on participation role Variables Community Development Value of Loans1 No participation 3.872*** 0 Standard error 0.14 t -Value 28.49 General member 3.649*** 4969.4*** Standard error 0.08 1251.7 t -Value 45.34 3.97 Leadership position 3.625*** 5825.0** Standard error 0.19 2957.7 t -Value 19.06 1.97 R2 94.1% 8.9% F -Value (3, 205) 1076.81 6.55 **Statistically significant at 5% ***Statistically significant at 1% 1 in Nepali Rupees Discussion The findings highlight the variables that influence household participation roles in sponsored programs and, ultimately underline the impact of participation roles on the distribution of benefits der ived from involvement in decentralized conservation programs. Economic, social, and access variables were incl uded within the analyses. Overall, the findings sugges t that none of the economic variables measured were significantly related to household participatio n roles. Other related studies about decentralized conservation m anagement programs in Nepal found economic variables to be statistically significant predictors of program participation that included income, landholdings, livestock, and remittances (Met ha & Kellert 1998; Sharma 2000; Blaikie & Springate-Baginski 2003; Dev et al. 2003; Neupane 2003; Smith et al. 2003; Adhikari et al. 2004; Agrawal & Gupta 2005; Maskey et al. 2006; Allendorf 2007). In this study, 64

PAGE 65

alternative forms of income, quantity of land owned, and number of livestock were all non-significant variables. The lack of stat istical significance among these economic variables could be associated with the econom ic homogeneity of the population. There were significant positive correlations between the four economic variables, which indicated that if a household had one of the variables, it usually had all four variables. Likewise, if a household was economically disadvantaged within one economic variable, they were often disadvantaged in all four va riables. However, overall the KCAP region is rather homogeneous in economic terms; i. e., economically poor. Quantities of livestock are highly correlated with altitude and inversely related to land ownership. Households at higher elev ations have larger herd si zes and less land compared to lower elevation households. It is likely t hat the lack of variation within the economic variables led to statistical insigni ficance and being poor predictors. Social variables were si gnificantly related to household participation roles and included education levels of the head of household, highest education levels of household adults, head of household age, and si ze of households. Findings were similar to other research conducted in Nepa l. Education levels have been found to be both negatively (Agrawal & Gupta 2005) and positively related (Maskey et al. 2006) to participation levels. Such incongruence in findings could be associated with discrepancies in the variables operationa lization. Agrawal & Gupta (2005) operationalized education as the summed years of education of all household adults (15 years of age or older). In this study, education was measured by two variables which included the total years of education co mpleted by the head of household, and the highest level of education completed by any household adult (18 years of age or older). 65

PAGE 66

Findings identified the two education va riables to have conflicting impacts on participation roles. Head of household educ ation levels was negatively related to participation roles; while the highest educat ion level of household adults was positively related to participation roles. Although Maskey et al. (2006) found age to be positively related to participation roles; this current research found a statistically significant negative relationship. Caution needs to be exercised when making policy suggestions based solely on education or age variables d ue to the unresolved association with participation roles. Although the number of dependen ts living within a household was not a significant predictor of participation, total household size did play a role. Households within KCAP are rather large with many extended family adults residing in one household. Family size has been proven to have an impact on partic ipation roles, but has been previously measured as total number of individuals within the household (Agrawal & Gupta 2005), and as the number of dependents (Maskey et al. 2006). Results in this study found that overall household size was a better predict or of participation roles than number of dependents. To ensure a more representative level of involvement within conservation areas and between socio-econo mic categories, managers should seek participation from smaller households. Similar to Agrawal & Gupta (2005), all thr ee access variables in this study were significantly related to participation roles. An increase in interactions with KCAP offices and staff was positively related to participation roles. These results highlight the importance of program staff to conduct social mobilization activities that include social capacity trainings and community development projects. A simple act of allowing the 66

PAGE 67

community to interact with agency staff pr ovides a major benefit. However, although the access variables were significant predict ors of participation roles, relatively few households exhibited access variables. WW F and KCAP have increased effectiveness by maintaining four field offices in proximity to the various minority populations, within a one hour walking distance. The spatial di stance of KCAP o ffices to individual households did have a minor negative impact on par ticipation roles. This finding is similar to Mehta & Kellert (1998). The results of this study are important in the contex t of mountain versus terai (lowlands) research in the arena of commu nity based natural resources management in Nepal. The mountainous KCAP area lacks the presence of the socially dominant groups found in other studies which include the Brahmin and Chettri castes found in large numbers in the lowlands. This study site predominantly comprised of Sherpa and Limbu ethnicities that were rather heterogeneous following topogra phical lines. Most villages were dominated by one caste, and thus limited caste interactions. However, the distribution of benefits within KCAP was found to be unequal. Households with higher community participation roles benefited from enhanced access to financial credit in terms of higher pr inciples and lower interest rates. Higher participation roles also provided for an incr eased level of involvement in capacity development training offered via program offices. Within KC AP, if equal distribution is a goal, then increased promotion of program participation is important. Decentralized conservation programs have been heralded in the literature based on their theoretical ability to: 1) decrease organization costs, 2) provide financial and technical assistance opportunities to previ ously excluded stakehol ders, and 3) create a 67

PAGE 68

68 creatively enhanced and competit ive economic sector (Ribot 2002, Agrawal & Gupta 2005). However, given the discrepancies of l oan and interest rate distributions across households, and the unequal dist ribution of capacity development training, further research is needed to determine the level of financial and social inequalities and possible financial corruption activities. There are some limitations to this study. Although the primary researcher had previously lived in Nepal for over two y ears and spoke the nationa l language, cultural differences were still present. Local res earch assistants were hired to alleviate response or researcher bias; however, the availability of trained social researchers at the site was minimal. Althoug h the operationalization of variables within this study was adapted from previous research conducted in decentralized conservation programs, additional improvements could be made. Due to low participation role frequencies, this study recoded participation into only three categories. A descriptive narrative of an example of each participation role shoul d accompany the questionnaire to ensure complete comprehension of role categories. In addition, the economic variables of remittance and non-farm income should be meas ured as absolute values to further distinguish variation wi thin the population. Overall, this study has shown that although KCAP is the first completely decentralized conservation program in N epal, a disproporti onate amount of the available benefits provided by conservation are concentrat ed within a small proportion of decision-making households. Though the ICDP literature calls for complete devolution of decision-making responsibilities and equitable di stributions of benefits for sustainability, neither are cu rrently occurring at KCAP.

PAGE 69

CHAPTER 3 USE PATTERNS AND FACTORS AFFECTING HOUSEHOLD NATURAL RESOURCE DEPENDENCY WITHIN KANCHENJUNGA CONSERVATION AREA, NEPAL Introduction Rural communities, particularly in dev eloping countries often depend on natural resources extraction, and make extensive us e of environmental products and services as integral components of their livelihood strategies (Ellis 2000). Natural resource based activities tend to require little skill and monetary investment s (Berkes 2007), for products such as fuel wood, medicines, pl ants, food, grazing fodder, hunting animals, and building materials (DeBoer & Baquete 1998) Over 60 million people globally are heavily dependent on forest resources for th eir livelihood, which on average represents 22% of total household income in developing countries (Vedeld et al. 2004). However, it is necessary to maintain heterogeneity in resource use patterns among local people especially in regions with a diverse range and variety of available resources (Brandon 1997). These resources serve to mainta in livelihoods, provide an opportunity to graduate to a higher level of livelihood strategy, or provide the ability to overcome crises (Scoones 1998). The magnit ude and details of resource use and dependency likely has consequences for conservation and strategies to make resources sustainable (Vedeld et al. 2004). Nevertheless, an assessment to understand dependency and pressures on local natural resources can assist restruct uring of future Integrated Conservation and Development Projects (ICDP ) for effective management (H errold-Menzies 2006). ICDPs are based on the premise that dev elopment improves local livelihoods while subsequently diminishing consumptiv e pressures exerted and dependencies on natural resources. The ICDP framew ork has been championed since the 1992 United Nations Convention on Biological Diversity which pledged to cr eate a protected area 69

PAGE 70

system based on three primary objectives including the conservation of biological diversity, sustainable use of the biological diversity co mponents, and providing for the fair and equitable distribution of benefits (CBD 1992). Therefore, parks and protected areas are now responsible for not only main taining biological di versity, but also improving local social welfare, guarding loca l security, and providing economic benefits across multiple scales; objectives traditional ly relegated to the development sector. The Kanchenjunga Conservation Area Project (KCAP) is an ICDP established in Nepal in 1997. Within KCAP, a comm unity based hierarchy of management has been instituted which advocates for equitable ac cess to decision-making roles and benefits distribution for all residents, that includes access to natural resources for livelihood activities. Although several alternative livel ihood capacity building initiatives have been formulated to alleviate environmental pre ssures exerted via traditional extractive livelihood strategies, major challenges are still evident (WWF 2006). For example, higher elevation villages receive less acce ss to community development programs and capacity building initiatives despite residi ng in less productive lands (Muller-Boker & Kollmair 2000; Ikeda 2005). Given the lack of alternatives sources of livelihood activities, these villages continue to pr actice slash and burn agriculture, overgraze pastures, and over harvest non-timber forest products (WWF 2006). The unequal distribution of benefits has been especially taxing for poorer households who are now heavily dependent on access to community natural resources (WWF 2006). Therefore, conservation programs should identify and target those households most reliant on resource access and extraction activities, and subsequently provide a means to create 70

PAGE 71

sustainable livelihood strategies that will allow opportunities to overcome poverty while alleviating natural resource pressures (Ellis 2000). There is a need for conservation managers to more efficiently assess and identify the characteristics and social aspects that are associated with varying levels of community natural resource dependencies, espec ially at the household level, in order to connect management with the local context. By identifying resource use patterns, programs are able to direct strategies at those individuals (or groups) with a greater success rate in terms of environmental sustainab ility (Vedeld et al. 2004). The purposes of this research was twofold: 1) Assess levels of natural resource dependencies for households within KCAP, and 2) I dentify factors that affect levels of natural resource dependencies among households. Literature Review Variability in natural resource dependencies within a community arise due to fluctuations in socioeconomic conditions values, beliefs, goals, and preferences (Adhikari et al. 2004). This variability in depe ndency also changes between and among individuals over time (Jumbe & Angelsen 20 07), and is associated with affluence (Baral & Heinen 2007). The wealth of a household does not necessarily lead to less utilization of natural resources (Herrold-Menzies 2006), but will generally affect the level of dependency on resources (Adhikari et al. 2004). An improved income might enable households the opportunity to choose livelihood strategies t hat may be more dependent on natural resources or cause detriment al environmental problems (e.g. timber harvesting technology or the purchase of mo re livestock) (Herrold-Menzies 2006). Also, households with higher incomes are often too busy with micro-enterprises and lack time to participate in resource dependent activities (Herrold-Menzies 2006). 71

PAGE 72

Households with higher incomes often have greater landholdings, which provide a private source of natural resources to consume. Poorer households with less landholdings rely more heavily on natural resource extraction and often own less productive land (Vedeld et al. 2004). Although poorer households may be more dependent on local natural resource s (Reddy & Chakravarty 1999; Dev et al. 2003), wealthier households often consume more resources (Agrawal & Gupta 2005). For example, Adhikari et al. (2004) found that affluent fam ilies received three times more forest benefits than poorer households in the middle hills of Nepal. Besides income, other variables have been shown to have significant impacts on household resource dependency in Nepal and elsewhere, which includes education, age, gender, household size, occupation, landholdings, and spatial location of household (Gunatilake 1998; Reddy & Chak ravarty 1999; Ellis 2000; Adhikari et al. 2004; Vedeld et al. 2004; Maskey et al. 2006; Chandool 2007; Jumbe & Angelsen 2007). Education acts as a proxy for higher wages, and thus higher income allows for a greater choice of economic activities that facilitates a lower level of resources dependency (Gunatilake 1998; Ellis 2000). It has been demonstrated that education has a negative effect on resource dependency (Gunatilake 1998; Adhikari et al. 2004). Similarly, age has varying affects on dependency because older individuals might find it easier to survive by collecting resources fr om the forest, instead of more laborious agricultural work which might provide more income (Vedeld et al. 2004; Maskey et al. 2006). Also, the size of a household can a ffect resource dependency, both positively and negatively. A greater availability of l abor within a household can lead to a greater availability in alternative income generating activities that do not rely on natural 72

PAGE 73

resources. However, larger households al so have a higher requirement levels to survive and likely have higher resour ce dependencies (Gunatilake 1998; Maskey et al. 2006). Furthermore, household spatial characteristics also affect resource dependency as those households located closer to fo rests are more dependent (Gunatilake 1998; Jumbe & Angelsen 2007), while households located near a market are less dependent (Gunatilake 1998). Jumbe & Angelsen (200 7) assessed natural resources dependency levels among two different communities in Malawi. One of the villages was a heterogeneous composition of varying migr ant ethnicities located near a market providing forest products. The other v illage had a primarily homogeneous population located far from a forest products market. Members of the homogenous community were less likely to be dependent on natural resources access and more likely to participate in decentralized forest m anagement practices. The heterogeneous community was less willing to participate in community natural resources management programs. The presence of higher forest product prices and access to a market also increased local willingness to concentrate inco me strategies on forest related activities (Jumbe & Angelsen 2007). Adhikari et al. (2004) further analyzed the relationship between household characteristics and common property resource s used in order to assess whether poorer households are able to gain greater access to community forests as a result of institutional change. Findings revealed that land and livestock holdings, caste, education, and household income str ongly influenced benefits accrued from decentralized conservation initiatives within fore st user groups in the middle hills of 73

PAGE 74

Nepal. The authors conclud ed that poorer households face more access restrictions compared to wealthy households (Adhikari et al. 2004). Additionally, a participatory rural appraisal (PRA) wealth ranking techni que was applied to categorize residents into three wealth groups of poor, middle wealth, and rich. Results indicated that wealthier households harvested more community forest der ived fuel wood, fodder, and leaf litter. Also, households from a higher caste/race/ethni city harvested more of all three forest products, while households headed by a male consumed more resources. In addition, wealthier households were more likely to per ceive that they benefited from community development activities funded through community forest initiatives revenue, included drinking water, irrigation, schools, health posts, and temples (Adhikari et al. 2004). Reddy & Chakravarty (1999) and Bahaguna (2000) assessed household dependency on natural resources in the foothi lls of northern India. Both studies identified households with le ss landholdings as more dependent, as were poorer households. Dependency levels ranged from 37-76% and were measured as the proportion of total household income derived fr om natural resources access. Income activities dependent on natural resource a ccess utilized within the study included wage labor, fuel wood, fodder, NTFPs, and tim ber. Fuel wood and fodder collection constituted approximately a third of the nat ural resources dependent income (Bahuguna 2000). Baral and Heinen (2007) assessed how natural resources dependency affected conservation attitudes among households loca ted in two protected area buffer zones in the western lowlands of Nepal. Resource s dependency had a positive relationship with conservation attitudes. The authors found th at ethnic minorities and households with 74

PAGE 75

more livestock were more dependent on natur al resources (Baral and Heinen 2007). Households with greater landholdings and affluence were less dependent on natural resources. However, although the ethnic mi norities were less affluent, they exhibited higher natural resource dependencies, and were also less literate, but more aware of forestry and wildlife management issues. A dditionally, they held more positive conservation attitudes. Household o ccupations and sizes lacked significant relationships with natural resources dependency (Baral and Heinen 2007). In Trinidad & Tobago, Chandool (2007) assessed household resource use patterns and the factors that influenced levels of resource dependency. Eight variables were examined and included: age, gender, educ ation, landholdings, diversity of occupations, household size, private agricultural land holdings, and length of residency. However, only age and gender were signific ant predictors of re source dependency. Younger and female-headed households were more dependent on community resources. Chandool (2007) noted a lack of statistical relationships among other associated variables due to the generally lo w level of natural resource dependency of the communities at the research site. Kanchenjunga Conservation Area Project KCAP was created in April of 1997 as an Integrated Conservation Development Project (ICDP) to safeguard the biodiversity of the area, and improve the living conditions of the local residents by strengthe ning the capacity of the local institutions responsible for making decisions, which wi ll effect the long-term biodiversity conservation and economic development of the area (WWF 2006, p. 2). KCAP was originally created with financia l and technical assistance pr ovided by the World Wildlife Fund for Nature (WWF); additional assistance was provided by the Integrated Centre for 75

PAGE 76

Integrated Mountain Developm ent (ICIMOD), the MacArthur Foundation, the Kadoorie Agricultural Aid Association, and the Darwin Initiative. Management largely focuses on enhancing local capacity to improve their live lihoods while simultaneously managing the natural resources (WWF 2006). KCAP is named after the third highest mountain (Mt. Kanchenjunga) in the world (8,586 m), and also contains eleven additional Himalayan m ountains with peak elevations over 7,000 meters. The 2,035 km2 landscape is primarily comprised of high elevation rocks and glaciers with only minima l arable land (2%). The area (see Figure 3-1) boasts very high biological diversity indices, and is home to several mammals of conservation significance including the snow leopard and red panda (WWF 2006). Although created following the m odel utilized during the cr eation of the successful Annapurna Conservation Area Pr oject (ACAP) in central Nepal, KCAP operates under several extremely constricting conditions (Heinen & Mehta 1999). Tourism at ACAP has been very successful with annual arrivals reaching over 75,000. The popularity of ACAP as a tourism destination has been aid ed by the regions close proximity to primary population centers of Pokhara and Kathmandu, and the associated highly developed tourism and transportation infr astructures (Nyaupane & Thapa 2004). However, at KCAP annual tourist arrivals peaked at 800 and recently dipped to 175 due to the lack of transportation infrastructure, tourism facilities, and the domestic Maoist insurgency12 (WWF 2007). 12 The Maoist insurgency refers to the decade long civil war that recently ended with the rebel group integrated into mainstream politics. See Thapa (2004), Battarai et al. (2005), and Baral & Heinen (2006). 76

PAGE 77

Figure 3-1. Map of KC AP (Source: WWF 2006) many people that reside within the KCAP re gion are very poor with little access to health and economic facilities. The conserva tion area is located in the northeastern district of Taplejung. The district mainta ins one of Nepals lowe st Human Development scores13 (0.328). Less than half t he population has access to toilet facilities, and there is only one doctor for every 33,675 people (C entral Bureau of Statistics 2001). The entire mountainous region only has 37 km of unpaved roads. Primarily, livelihood strategies have been subsistence agricultu re and animal husbandry largely due to the lack of external market linkages and lack of transportation infrastructures. Additional livelihood activities include government em ployment, cottage industries, trade with 13 Human Development Index (HDI) scores are used to define well-being as a composite of education, life expectancy, and standard of living. In 2005, Ne pal had an HDI Score of 0.534, and ranked 142 out of 177 countries (UNDP 2008). 77

PAGE 78

neighboring Tibet, remittances14, and a low level of tourism activity (Muller-Boker & Kollmair 2000). Local livelihood strategies have exert ed several pressures on the natural resources of the region. More specifically, higher elevation households practice slash and burn agricultural techniques due to unc lear land ownership rights, poor soil conditions, lack of arable land, and nonexistent market alternatives (WWF 2006). Also, in higher elevations, meadows are negatively impacted by overgrazing, fuel needs, and housing construction. Simila rly, over harvesting of Non-Timber Forest Products (NTFPs) is a prevalent environmental threat throughout the region. This has been attributed to extremely high levels of household dependencies on these products for their livelihoods, unclear landownership rights, and lack of external market linkages (WWF 2006). KCAP, aided by WWF, has employed several strategies to both alleviate environmental pressures and enhance indivi dual, community, and institutional capacities for development and conservation. As of September 22, 2006, KCAP is the first completely co mmunity-managed conservation area in Nepal, if not in Asia (WWF 2006). Management is achieved via a democra tically elected hierarchy of management comprised of user groups, mother groups15, and the leading authority of the Kanchenjunga Conservation Area Management Council16. KCAP management has 14 Remittance is payments paid to friends and/or family members in a migrants home place of residence. The value of global remittances in 2008 was US $433 billion; US $328 billion was sent to developing countries (World Bank 2009). 15 Mother groups are a form of sponsored user groups intended to represent the interests of female residents. 16 The Management Council is the highest level KCAP management, comprised of locally elected individuals. 78

PAGE 79

implemented several steps to aid local resident s. For example, Community Forest User Groups, representing 891 households, manage 47, 772 hectares of forest lands, and a rangeland management strategy has also been completed (WWF 2006). In addition, a sustainable NTFP enterprises fund has been created to prov ide training on enhanced collection and processing techniques. Seve ral community development projects have also been implemented and include savings and cr edit groups, toilet facilities, improved drinking water systems, a health post, and female education funds (WWF 2006). However, the effectiveness of these programs has been in question. Local residents are unsure about KCAP goals and objec tives. Muller-Boker & Kollmair (2000) found that only 5% of residents were aware of the primar y administrative organization. Residents also held unrealistic expectations towards KCAP, in that while few could identify conservation related objectives, mo st residents mentioned several objectives that included large infrastruc ture developments. Furthermo re, while most residents felt KCAP made positive strides towards wildlife conservation, there has been an increase in human-wildlife conflicts and associated retaliatory killings. Retaliatory killings of snow leopards are common, especially within the G hunsa Valley that contains over 40% of the snow leopards habitat and is also the main tourist trekking route (Ikeda 2005; WWF 2006). Overall, KCAP presented an excellent opportunity to assess the impact of decentralized conservation efforts on natur al resources dependency. The region lacks infrastructure development and links to external markets (including tourism), which thus restrains local economic activity to natural resources dependent sources. The purpose of this study was to assess household leve ls of natural res ource dependencies within 79

PAGE 80

KCAP and subsequently identify the factors that lead to the associated dependencies. The following research questions were form ulated and empirically tested in this research: 1) What are the levels of natural resource dependencies for households within KCAP? 2) What factors affect a households level of natural resources dependency? Methods A combination of qualitative and quantitative methods were utilized in this study. The combination of both methods has been gai ning credibility and relevance within the development literature (Booth et al. 1998; Kanbur 2001; White 2002; Ellis & Freman 2004). A triangulation method was also utiliz ed to both improve the data validity and reduce the occurrence of researcher bias (Denzin & Lincoln 2000). Data Collection Data collection employed four methods and lasted for three months (October through December 2008). Firs t, project reports and publicat ions were identified at agency and government offices in the capital city of Kathmandu and in Taplejung. Literature was targeted that reported on past and current programs that pertained to policies and management issues with regards to natural resources access, use, and dependencies. Second, participant observation techniques were employed to discern household behaviors and patterns in order to develop a better understanding of community context of how programs impact livelihood strategies on a daily basis. Observation focused on access to and use of natural resources. Th ird, a Participatory Rural Appraisal (PRA) wealth ranking (W orld Bank 2007) was conducted within each village. PRA wealth rankings are used to create a comprehensive list of households, and also to assimilate the researcher into t he communities to facilitate data collection. 80

PAGE 81

Fourth, a detailed household interview was conducted throughout the KCAP region. The combination of data collection techni ques were adapted from the United Kingdom Department for International Developm ent (DFID) sponsored LADDER project17 (Ellis & Freeman 2004). Two local research assistants were hired to help with the administration of the household interview. Althoug h the primary researcher previously lived in Nepal and worked with communities in protected areas as well as being proficient in the national Nepali language, local assist ants aided in rapport building and minimized both crosscultural bias and non-sampling error. Technical words were avoided, and a professor at Tribhuvan University, Kathmandu, Nepal asse ssed the accuracy of survey translation from English to Nepali and then back to English. Selection of Participants A recent WWF (2006) survey iden tified 4,570 people that resided in 860 households within the KCAP region. Of the four Village Development Committees (VDC) within KCAP, Lelep contained the largest percentage of population (47%), followed by Tapetok (31%), Yamphudin ( 15%), and Olachan Gola (7%). Respondent selection followed a systematic sampling method that targeted every fourth household within every village throughout the region from a random start. Among selected households, the head of household ( 18 years old) was asked to participate. If the household head was unavailable, the next most knowledgeable adult was requested to participate. Due to the regions high rate of illiteracy, and over all unfamiliarity with survey procedures, survey questions were verbally read to participants while research 17 LADDER stands for Livelihoods and Diversific ation Directions Explored by Research. 81

PAGE 82

assistants recorded all responses (see Nyaupane & Thapa 2004). All approached households agreed to participate (response rate = 100%), which provided a 25% population representation (n = 205) that was deemed sufficient for categorical analyses (Cochran 1977; Bartlett et al. 2001). Instrumentation The questionnaire was adapted from Ellis and Freeman (2004) and consisted of nine sections. The questionnaire was pilot tested in the district center of Taplejung among KCAP staff. The questions were alter ed as needed to improve clarity. The first section assessed basic household compositi ons that included number of members, gender, education levels, ethnicity/caste/race, occupations, housing construction materials, and available facilities. The second section assessed household membership characteristics in sponsored programs, level of interaction with agency staff, and level of perceived benefits received from sponsored programs18 at both the community and household levels. Section th ree assessed household agricultural assets and productivity levels that that included levels of land ownership and quantity of individual crop yields. The fourth section assessed the value of livestock and natural resources consumption for t he household, which included thos e natural resources found within community forests. The fifth secti on identified the sources of income that originated from the farm (e.g. wage labor pensions, salaries, remittances, self employment, and tourism activities). Operationalization and Data Analysis This study assessed two research questions. 18 Sponsored programs are locally comprised and driven by community groups that are aided with financial and or technical assistance provided by an outside organization. 82

PAGE 83

Research question 1: What are the existing levels of natural resource dependencies among households? Income activities considered to be dependent on natural resources access included those related to livestock, agricul ture, and community fore sts. Income not dependent on access to natural resources in cluded all off-farm sources such as remittances, wage labor, salaries, cottage i ndustries, and other activities including tourism. Household depende ncy on natural resources was determined by assessing the ratio of total household income compared to household income derived from natural resources (see Gunatilake 1998; Bahugana 2000; Masozera, & Alavalapati 2004; Chandool 2007; Jumbe & Angelsen 2007). An index of dependen cy was created by dividing income dependent on natural resources by total income. Research question 2: What factors a ffect natural resource dependencies among households? A logistic regression model was conducted to identify household factors that affect natural resources dependency (see Chandoo l 2007). The mean value of natural resources dependency identified in the first res earch question was used as a cutoff to create a binary value of either not dependent=0 or dependent=1. Ten predictor variables were utilized within the model repr esenting the three constructs of economic, social, and access variables. Economic va riables included household landholdings and remittances. Social variables included the highest education level achieved by an adult household member, household size, number of dependents, and caste/race/ethnicity. Access variables included if the household was visited by a KCAP offi cer or had visited a KCAP office in the past year, had rece ived training sponsored by KCAP, or had received a loan from a sponsored group. 83

PAGE 84

Results Profile of Househ old Respondents Of the 215 completed surveys 205 were us able. The majority of represented households were of Sherpa (53%) and Limbu (41%) ethnicities, and headed by males (86%) with less than a year of comp leted education. Most households19 were located in Lelep VDC (59%), followed by Tapetok VDC (17%), Yamphudin VDC (17%), and Olanchan Gola VDC (7%). The majority of households (76%) were located in the lower elevations (between 1,380 and 2,249 meters), while others (24%) were at higher elevations (between 2,250 and 3, 000 meters). Additional household characteristics are presented in Table 3-1. Table 3-1. Household characteristics Characteristic Mean Std. dev. Max Min. Economic Variables Land owned (ropani1) 22.83 23.62 150 0 Remittance 37,282 69,576 500,000 0 Social Variables Highest education level 5.81 4.63 16 0 Household size 5.87 1.50 10 1 Number of dependents 2.22 1.37 6 0 Caste/ race/ ethnicity 0.53 0.32 1 0 Access Variables Household visits by KCAP staff 0.21 0.47 1 0 Household member visits to a KCAP office 0.15 0.36 1 0 Training participation 0.16 0.36 1 0 Borrowed loans (NRs2) 3,930 14,579 200,000 0 Additional Variables Total Livestock Units 4.51 5.21 49 0 Elevation (meters) 2,205 554 3,595 1,380 1 1 USD = 75 NRs 2 1 hectare = 19.65 ropani 19 The spatial distribution of communities was similar to that identified by WWF (2006). 84

PAGE 85

The average household had six member s (of which 2 were dependents), was located within an 80-minute walk of a KCAP office and owned 23 ropani20 of land. Over three quarters of households (78%) had bor rowed a loan from a sponsored program, while only a few had been visited by a KCAP staff member (21%) or had visited a KCAP office (15%). Six categories of off-fa rm sources of income were identified. The most prevalent source of of f-farm income was remittances (32%), followed by tourism (14%), private industry (6%), government (3%), wage (3%), and casual labor (3%). Research Question 1: Levels of Natural Resources Dependency The composition of stated household incomes varied throughout the region. Table 3-2 presents the mean househol d values of income derived via different activities. On-farm income values were heavily skew ed towards the maximu m value, the bottom half of households earned less than NRs21 30,000 annually from on-farm activities. In addition, over half the population earned less than NRs 10,000 annually from livestock activities, and less than NRs 20,000 from off-farm activities. The average level of dependency on natural resources was 63.75% The most dependent quarter of the population was at least 90% dependent on natural resources access, while the least dependent quarter of the populat ion was less than 38% dependent on natural resources for household income. Elevation was si gnificantly associated with household dependency on natural resources access (F = 9.557, p value = 0.002). Households located at higher elevations had the hi ghest levels of dependency with a mean of 74.61%, while households at lower elevations were slightly less dependent at 60.33%. 20 1 hectare = 19.65 ropani. 21 At the time the field research was conducted (October 1st, 2008) 1 USD = 75 Nepali Rupees (NRs). 85

PAGE 86

Table 3-2. Sources of household inco me and dependency on natural resources access Income Sources Mean (NRs) Std. dev Min. Max On-farm 43,464 39,874 0 230,000 Livestock 15,056 15,873 0 110,000 Off-farm 55,992 74,875 0 550,000 Total natural resources income 58,524 45,489 0 25,000 Total income 114,517 96,630 0 630,000 Natural resources dependency 63.75% 0.2874 0% 100% Higher elevation households 74% 0.2572 0% 100% Lower elevation households 60.03% 0.2876 8.47% 100% Research Question 2: Fact ors Associated With Natu ral Resources Dependency Results of the logistic r egression model utilized to asse ss the factors that affect household dependency on natural re sources access are presented in Table 3-3. The overall model was statistically significant in cluding four individual predictor variables. Table 3-3. Logistic regression of household natural resources dependency Variable Name Coefficient Std Error Wald Sig. Exp (B) Economic Variables Land ownership 0.007 0.009 0.623 0.430 1.007 Remittances (n/y) -3.976 *** 0.579 47.102 0.000 53.292 Social Variables Highest education -0.064 0.048 1.806 0.179 0.938 Household size 0.302 0.166 3.311 0.069 1.353 Dependents -0.403 ** 0.187 4.668 0.031 0.668 Caste/ Race/Ethnicity 0.176 0.394 0.200 0.655 0.838 Access Variables Staff visit (n/y) 0.280 0.483 0.337 0.562 0.756 Member visit (n/y) -0.470 0.586 0.643 0.423 1.600 Training (n/y) -0.175 0.528 0.109 0.741 1.191 Loans (n/y) 1.146 *** 0.429 7.125 0.008 0.318 Chi-squared (10) = 94.575; Prob (Chi-squared) 0.001; R-Squared = 0.210, -2 log likelihood = 187.029, Cox & Snell = 0.370. Overall percent correct = 81.5%. The economic variable of remittances was negatively associ ated with natural resources dependency; however, the access vari able of households that received loans was positively associated. Therefore, hous eholds that were able to supplement their incomes with remittances were less likely to be dependent on natural resources for their income. However, those households with access to available credit from sponsored 86

PAGE 87

programs were most likely to actually hold loans. The social variables of household size and number of dependents were also both significant. Households with more dependents were less likely to be dependent on natur al resources. However, larger sized households were more dependent on natural resources. Discussion The ability to identify household natural resource use patterns has been shown to drastically enhance the sustainability of conservation projects (Vedeld et al. 2004). An improved identification met hodology increases the effect iveness of capacity building programs by efficiently targeting the most heavily dependent households (Vedeld et al. 2004). Similar to Chandool (2007) and Jum be & Angelsen (2007), this study assessed the level of household depend ency on natural resources within KCAP, and identified associated factors that affect dependency. Results of this study indicate t hat households within KCAP were heavily dependent on natural resources for their income. The average household dependency on natural resources for income in developing countries is 22% (Vedeld et al. 2004). However, this study found that 64% of households in KCAP were dependent, and a greater proportion of those in higher elevations were dependent (74.6%) than households in lower elev ations (60.3%). Variations in household dependency on natural resources access have been attributed to socio-economic characteri stics, values, beliefs, and preferences (Gunatilake 1998; Dev et al. 2003; Adhikari et al. 2004; Vedeld et al. 2004; HerrodMenzies 2006; Maskey et al. 2006; Allendorf 2007; Baral & Heinen 2007). Household affluence has been shown to decrease natural resources dependency by allowing an opportunity to choose less resource dependent livelihood strategies (Dev et al. 2003; 87

PAGE 88

Adhikari et al. 2004; Herrod-Menzies 2006; Baral & Heinen 2007). In addition, less affluent households are often more reliant on natural resources access due to inadequate land ownership (Dev et al. 2003; Vedeld et al. 2004). However, in this study land ownership was not a significant econom ic predictor of dependency. Remittances were found to be a significant indicator of natural resources dependency. Households that received remittances were able to supplement their income without reliance on natural resources based activities. This finding is common throughout Nepal. Remittances in Nepal were valued at over US $1.2 billion in 2006 and comprised more than 14.6% of Nepals national GDP in 2006 (World Bank 2008). Other social variables that have been show n to have significant effects on natural resources dependency in Nepal include educati on, household size, and spatial location. Gunatilake (1998) and Adhikari et al. (2004) found that education led to higher paying livelihood activities which resulted in decreased natural resources dependency. However, education was not a significant predictor of natural resources dependency in this study. The lack of significance may be a ttributed to the low over all rate of education within the KCAP region. Heads of househo lds averaged less than one year, and less than six percent of individuals at least thirty years of age had ever attended school. The size of a household has been s hown to have both positive and negative impacts on natural resources dependency (Gunatilake 1998; Maskey et al. 2006). Households with more member s require more income, but have more people to help in natural resources related activities. Howe ver, more members could also provide an opportunity to participate in alternative liveli hood activities that are not reliant on natural resources. This study found that households with more mem bers were more likely to be 88

PAGE 89

dependent on natural resources; however, households with fewer dependents were less dependent. Caste/race/ethnicity among households in Nepal have been significant predictors of natural resources dependency in the lit erature (see Gunatila ke 1998; Agrawal & Gupta 2005; Allendorf 2007). However, thes e studies were conducted in the lowland terai region of Nepal where the caste system is more prevalent. The population in this study primarily consisted of the two different ethnic minorit ies of Sherpa and Limbu. Therefore, caste/race/ethnici ty failed to have any predictive significance in regards to natural resources dependency. Household spatial characteristics have been shown to be significant predictors of household natural resources dependency (G unatilake 1998; Jumbe & Angelsen 2007). However the spatial distribution of households in this study was rather homogenous. All households were extremely disconnected from external commercial market linkages (at least by a three day walk), and the placem ent of four KCAP offices throughout the region provided easy accessibility to spons ored programs for most households. Variables associated with access have been frequently identified and researched the decentralized conservation literature. Common significa nt variables have included access to savings and credit programs, participation in sponsored programs, participation in capacity building training programs, and frequent communication with project staff via household visits and office visits (Dercon & Krishna 1996; Rew & Rew 2003; Agrawal & Gupta 2004; Acharya et al. 2005). Access to loans provided within the sponsored program had a positive significant affect on natural resources dependency. Participation in training progr ams, and interactions with KCAP staff failed to significantly 89

PAGE 90

affect resources dependency. The lack of signi ficant affect of KCAP staff interaction and training programs could be explained by the overall small frequency of households experiencing those situations. Less than 20% of households had direct communication with KCAP staff within the past year, and only 16% had participated in a training program. Social mobilization techniques and training programs are targeted toward achieving the overall aim of KCAP which is to protect the areas biol ogical diversity by alleviating environmental pressures and c oncurrently improving livelihoods by strengthening the capacity of the local institutions for making decisions (WWF 2006, p 1). The lack of social mobilization stra tegies and the absence of a region-wide system of KCAP sponsored capacity development proj ects has detrimentally limited the availability of opportunities for households to choose livelihood activities that are less dependent on natural resources. Conclusion Results of this study indicate that households within KCAP are heavily dependent on access to natural resources for their liv elihood strategies. The sustainability of KCAP as an ICDP is therefore contingent on project intervention via social mobilization, training programs, and participatory inclusion in decision making. Results of this study highlight the need for further research. An empirical study of consumption levels among households should be assessed in terms of f uel wood, timber for construction, and grazing (see Adhikari et al. 2004). Additionally, given t he prevalence and impact of remittances on natural resources dependency (31% of households received remittances that constituted about two-thir ds of their off-farm income), a study should be conducted in regards to the factors that affect household acquisition of remittan ces. In addition, 90

PAGE 91

providing externally funded training programs that target the adopti on of alternative livelihood activities are futile if there is disconnect with commercial markets. Further research needs to be conducted on the potential to develop market linkages within the KCAP region. 91

PAGE 92

CHAPTER 4 LIVELIHOOD STRATEGIES AND DIVER SIFICATION FACTORS AMONG HOUSEHOLDS WITHIN KANCHENJUNG A CONSERVATION AREA, NEPAL Introduction Nepal is considered a l eader among least developed nations with respect to environmental policy and conservation progra ms. The country has designated over 18% (26,695 km2) of its land as protected status in 17 different locations. More importantly, it has been a pioneer in establishing decentralized conservation management programs within parks and protected areas. Such programs highlight the link between conservation and socio-economi c development among residents within protected areas. In 1986, the Annapurna Conservation Ar ea Project (ACAP) became the first protected area In tegrated Conservation and De velopment Project (ICDP) implemented by a local non-government organi zation in collaboration with the World Wildlife Fund for Nature (WWF ). The concept behind the proj ect was economically selfsustainable conservation with a focus on development that emphasized popular local participation via a decentralized form of management (Sherpa et al. 1986). ACAP is regarded as a model of success within the di scourse of ICDPs because it has provided increased development, improved distributio n of benefits, empowered local residents, and enhanced biological conservation (Nyaupane et al. 2006). However, an equitable distribution of benefits, costs, influence, and a diversification of livelihood strategies has been elusive in ACAP, largely due to the participatory exclusio n of disadvantaged groups, varied levels of natural resource dependencies, and dispro portionate regulatory and financial structures (Heinen & Mehta 1999; Nyaupane & Thapa 2004; Bajracharya et al. 2005; 2006; Baral et al. 2008). 92

PAGE 93

The ICDP framework has been promoted in developing c ountries as a means to maximize benefits for local people and mi nimize costs endured due to biodiversity conservation measures (Brandon & Wells 1992). Since its advocacy, ICDPs have proliferated throughout developing countries and have garnered major support for conservation (McShane & Wells 2004). ICDPs are based on the idea that if programs improve the livelihoods of local people, then they would consume fewer natural resources, and thus minimize pressures wit hin protected areas (Brandon & Wells 1992; Brandon et al. 1998). Additionally, ICDPs are based on the paradigm that projects need to be focused on providing for the loca l population, and positively responding to changes in their livelihoods (Berkes 2007). However, results from existing ICDPs have shown varied results and have led to a debate arguing both for and against them (Wells et al. 1992; Brandon 1997; Oates 1999; McShane & Wells 2004; Berkes 2007). Neuman (1997) states that ICDPs lack true participation, and serve rather as a coerci ve form of conservation that resembles ill fated colonial efforts to conv ert shifting cultivators into progressive farmers (p. 564). Additionally, most ICDPs are created by western-based internationa l organizations with large amounts of funding, and little site-specific knowledge of cultural and institutional frameworks (Oates 1999; McShane & Wells 2004) Furthermore, doubts have arisen as well regarding the presence of an economic threshold that once achieved would alleviate the need for natural resources conservation (Stem et al. 2005) Conservation programs are only valid an d sustainable when they have the dual objective of protecting and improving loca l livelihoods and ecological conditions (Ghimire & Pimbert 1997, p. 8). Pr ogram success is inherently dependent upon 93

PAGE 94

ensuring an equitable distribution of benefits while simultaneously diversifying income generating activities thus, reducing depende ncy on natural resources (Chapin 2004; McShane & Wells 2004). Based on an asse ssment of ICDPs worldwide, the UK Department for International Development (DFID) (2007) identified three major shortcomings: 1) local development is usual ly assessed in terms of generation of cash, increased production or jobs, wh ile wider social issues and livelihoods concerns are often ignored, 2) project assessments usually focus on the numeration of outputs achieved and often create unintended c onsequences for livelihoods, and 3) assessments usually focus on generation of local incomes instead of the promotion of commercial viability. Given these limitatio ns, and in order to overcome these major shortcomings, the use of the sustainable livelihoods framework for project assessments has been advocated (DFID 2007). Within the theoretical lens of the sustainable livelihoods framework, the purpose of this study was to assess two fundamental elements identified as essential to sustainable integrated conservation and devel opment projects. Spec ifically, within the Kanchenjunga Conservation Area Proj ect (KCAP) ICDP in Nepal, this research will: 1) identify the current levels of diversificat ion among livelihood strategies utilized by households; and 2) identify potential factors associated wit h the adoption of existing dominant livelihood strategies. Findings of this research are beneficial along a range of scales from the local community level to the broader body of ac ademic knowledge. Results will enhance the management of community based natural res ources, and promote st rategies that will create more sustainable community capacit y building programs while concurrently 94

PAGE 95

alleviating pressures on the areas extravagant biological diversity. Results will aid program managers to identify household charac teristics strongly associated with those who lack access to livelihood assets (financial, social, physical, natural, and human) necessary to graduate to an improved livelih ood strategy. The potential for ICDP sustainability would be greatly enhanced by the ability to identify disadvantaged households and target project spending while organizational costs are minimized. Findings will also aid policy makers to st rengthen their ability to identify current environmental pressures that are most cont ingent for further policy applications, while simultaneously apply strategies that are most critical to a households livelihood. Literature Review Sustainable Livelihoods Framework The primary objective of an ICDP is to pr ovide for diversified sustainable livelihood strategies with an emphasis on sustainable development that ensures biological conservation (Scoones 1998). In 1987, the World Commission on Environment and Development first advanced the term sustainable livelihoods in relation to poverty alleviation and environmental protection through the promotion of participation and sustainable development (WCED 1987). Subsequently, the 1992 UN Conference on Environment and Development promoted sust ainable livelihoods as a means to link socioeconomic and environmental concerns. These two international forums were essential to the formation of the sustainable livelihoods framework (Scoones 1998). The sustainable livelihoods framewor k has been championed in both the community development and environmental m anagement disciplines, as it maintains a shift of paradigm toward s a more people-centered dev elopment approach within a community (Ashley & Carney 1999). The framework emphasizes addressing 95

PAGE 96

community assets rather than primarily focusing on a communitys needs (Chambers & Conway 1992; Scoones 1998; Ashl ey & Carney 1999; Carney et al. 1999; Ellis 2000; DFID 2007). The sustainable livelihoods framework is currently used by international development agencies such as: United Nati ons Development Program (UNDP), Food and Agriculture Organization (FAO), Unit ed States Agency for International Development (USAID), CARE-Internati onal, Oxfam, and the United Kingdom Department for International De velopment (DFID) (Frankenberger et al. 2000). Chambers & Conway (1992) first defined a livelihood as comprised of the capabilities, assets (stores, resources, cl aims and access) and activities required for a means of living. A livelih ood is sustainable when it c an cope with and recover from stresses and shocks and maintain or enhance its capabilities and assets both now and in the future, while not undermining the natur al resource base (p. 7). Ellis (2000) adapted this definition with more emphasis on a ccess in relation to social institutions that might impact an individual from capacit y to fulfill its consumption requirements based on social inequalities such as, caste, race, gender, ethnicity, beliefs, etc. His definition is a livelihood comprises asse ts (natural, physical, human, financial and social capital), activities, and access to a ssets (mediated by institutions and social relations) that collectively determine the liv ing gained by an individual or household (p. 10). The sustainable livelihoods framework has four main components (DFID 2007). First, a population lives within a vulnerability cont ext in which it is exposed to risk in the forms of sudden shock, trends over time and seasonal change. Examples of risk include drought, tourism seasonality, or labor migrati on. Second, people utilize assets 96

PAGE 97

to meet their livelihoods. These assets include social (networks, norms, trust, institutions), human (skills, k nowledge, and labor), natural ( natural resources), physical (infrastructure including transportation, water, and energy), and financial capital (savings, income, and credit). Third, thes e assets are utilized by residents through livelihood strategies in order to create t heir livelihood outcomes. Lastly, transforming structures and processes shape and influenc e the ability of individuals to access livelihood assets as a means to overcome ri sks within the vulnerability context. It is within these transformative structures that programs can impact the linkage between policies, institutions, and livelihood activi ties (DFID 2007). Figure 4-1 depicts the accessibility of resources, controlled by in stitutions and norms t hat are used within livelihood strategies to create an outcome that enhances pre-existing assets to overcome vulnerability to risks and shocks. Figure 4-1. Sustainable liveliho ods framework (Source: DFID 2007) The sustainable livelihoods framework deviates from purely economic measurements such as primary income gross domestic product, and consumption 97

PAGE 98

metrics in poor regions, and levies emphas is on breaking accessibility barriers to improve the quality of life vi a institutions, policies, and processes (Scoones 1998). However, Ellis (2000) delineates the importa nce of a lack of priority on temporal changes in access to assets. Individuals are continually changing their livelihood strategies concurrently within the social in stitutions that govern access, especially in developing countries (Ellis 2000). Projects that utilize the sustainable livelihoods framework also emphasize re sident empowerment and capacity building to effectively enhance access to available resour ces that allows them to c hoose strategies in order to negotiate vulnerabilities, and incrementally impr ove their livelihood assets (Ellis 2000). The emphasis on diversification of livel ihood strategies within the sustainable livelihood framework has been the catalyst for numerous international development agencies to utilize the framework (Brocklesby & Fisher 2003). It prom otes the creation of livelihood enhancement beyond just employ ment. Projects that have used the framework have realized that a si ngle source of income is not sufficient to truly aid in poverty alleviation and envir onmental management in developing regions (Ashley & Carney 1999). For example, tourism promot ion as a pro-poor strategy in developing regions has thus far been unsuccessful due to a primar y emphasis levied on one single strategy (Tao & Wall 2009). Tourism is not a panacea for developing countries and is not a reliable source of income in many margi nal economies, but it may supplement other income sources and help disperse vulnerabili ties to risk (Tao & Wall 2009). Tourism seldom occurs in isolation as it competes for the use of scarce re sources; therefore, tourism development projects need to be initiat ed as a complement to existing activities 98

PAGE 99

(Ashley 2000). A comprehensive understanding of the link between tourism and other sectors requires an analysis of the whole co ntext in which tourism exists (Tao & Wall 2009). A sustainable livelihoods approach is more practical, especially within common situations during which communities and indi viduals sustain themselves by multiple activities rather than discrete jobs. Therefore, advocates of sustainable development should explore how tourism might fit into a su ite of livelihood strategies, contributing to the achievement of sustai nable livelihood outcomes (Tao & Wall 2009, p. 98). Factors that Affect Livelihood Diversification Rural livelihood diversification is def ined as the process by which rural households construct an increasingly diverse portfo lio of activities and assets in order to survive and to improve their standard of living (Ellis 2000, p. 15). Ellis delineates five main categories of livelihood activities: 1) non-farm (unskilled labor, wage employment, and non-farm business profits), 2) agriculture (net income from crop sales, ownconsumption, and off-farm labor), 3) transfe rs (national and international remittances22, pensions, and payments to the poor), 4) livesto ck (net income from sales of cattle and poultry, and own consumption), and 5) rent al income (from ownership of assets, including land, machinery and water). Varying levels of livelihood diversif ication are often associated with the disproportionately available access to nece ssary livelihood assets based on local, regional, and national institutions and norms (Ellis 2000; Agrawal 2001; Blaikie & Springate-Baginski, 2003). Associated access is also often correlated to the role of 22 Remittance is the money that someone sends to friends or family members that live somewhere else. Globally, remittances are a large source of financia l transactions with over 190 million migrants sending US $433 billion in 2008, of which US $328 billion went to developing countries (World Bank 2009). 99

PAGE 100

participation in project planning, management, and capacit y building programs which are based on pre-existing power structures linked to socio-economic characteristics (Rew & Rew 2003; Agrawal & Gupta 2004). Less affluent households are usually more dependent on natural resources for their liv elihoods but have less access to them (Adhikari et al. 2004). Without access to resource s, poorer households are more likely to have less diversified livelihood strategies, and thus are more dependent on local natural resources in protected areas (Barret et al. 2001; Bhandari & Grant 2007). Households are better suited to cope with vulnerability to shocks such as drought and market seasonality with a more diverse livelihood strategy (Ellis 2000). However, disadvantaged and poorer households more often bear an unequal sh are of the costs associated with conservation. High leve ls of resource dependency and the lack of participation roles in community managem ent prevent poorer households from the opportunity to either enhance their livelihood st rategies or to gradu ate to a higher order strategy (Scoones 1999; Ellis 2000). Giv en that disadvantaged households are held to subservient roles in decision-making and only allowed minimal access, if at all, to regional social, physical, financial, and envir onmental assets; the ability to achieve a diverse livelihood strategy is severely limited (Agrawal 2001; Rew & Rew 2003; Agrawal & Gupta 2005). Several income based strategies have been employed to characterize livelihood strategies. Barret et al. (2006) compared off-farm and on-farm income proportions to total incomes and found that as total income rose, the number of strategies followed suit. Dercon & Krishna (1996) identified that total income was related to number of income sources, household characteristics, and access to credit. Ellis & Mdoe (2003) 100

PAGE 101

indicated that poverty was strongly associat ed with lack of land and livestock, as well as the inability to secure off-fa rm employment alternatives. These methods of analysis are counterintuitive to the sust ainable livelihood framework t hat emphasizes an analysis of existing asset allocations beyond need discrepancies. In Nepal, Bhandari & Grant ( 2007) found that agricultural production alone was not a viable livelihood option, and livelihood security was bas ed on landholdings, manpower within the family, and access to agricultural technological inputs. The authors call for empowerment and policies to provide local residents with adaptable income generating activities, resilient resource management institutions and enhancement of knowledge, skills and social capital (Bhandari & Grant 2007). These methods are also based on externally driven assumptions that categoriz e activities into arbitrarily significant employment sources. Additionally, Acharya et al. (2005) found that while functional literacy programs are good initiatives in Nepal, participati on in group savings and credit programs provide the impetus to help illiterate rural women in itiate small scale economic activities that are effective in the abs orption of the female workforce. In Kenya, Brown et al. (2006) found several key charac teristics that statistically differentiate households that pursue low, medi um, and high return livelihood strategies. Brown et al. (2006) utilized cluster analysis and identif ied five livelihood strategies that include part-time subsistence, mixed smallh olders, staples producer s, off-farm skilled employment, and diversified commercial. The latter two strategies provided the highest rates of return. Furthermore, household lo cation, size, and farming experience were significant predictors of househo lds that pursued a livelihood st rategy with lower rates of return. Access to credit and re mittances were significant pr edictors to pursue a higher 101

PAGE 102

rate of return livelihood st rategy. The authors state that the resu lts should facilitate specific interventions designed to improve household livelihoods (Brown et al. 2006). Kanchenjunga Conservation Area Project Kanchenjunga Conservation Ar ea Project (KCAP) is located in the northeast portion of the district of Taplejung, and was o fficially created as an ICDP in 1997. The World Wide Fund for Nature (W WF), MacArthur Foundation, Kadoorie Agricultural Aid Association, and the Darwin Initiative have provided financial and technical assistance. KCAP received its namesake from the th ird highest mountain in the world (Mt. Kanchenjunga) and is also home to eleven other mountains over 7,000m. The area consists of 2,035 km2 of land composed primarily of gl aciers and rocks (65%), forests (24%), and a small amount of arable land (2%). The area has biological significance with several endangered animals including th e red panda, snow leo pard, grey wolf, Himalayan black bear, Himalayan Tahr, bl ue sheep, and musk deer (WWF 2006). KCAP has an international trans-boundary conservation significance by maintaining shared borders to the East with the Khangchendzonga Bios phere Reserve in Sikkim, India and to the North with the Qomolangma Nature Preserve in Tibet, China (See Figure 4-2). KCAP is a vital component for the eventual creation of an international Kanchenjunga mountain complex c onservation area (WWF 2006). 102

PAGE 103

Figure 4-2. Map of KC AP (Source: WWF 2006) Although KCAP creation para lleled the successful model utilized by the Annapurna Conservation Area Project (ACAP), several lim iting factors exist at KCAP. ACAP has an extensively developed tour ism infrastructure, is centrally located, and easily accessible with a transportation network that includes paved roads and an airport with several daily flights (Nyaupane & Thapa 2006). ACAP has also been highly successful in attracting tourists and trekkers with over 75,000 entrance permits sold annually (Nyaupane & Thapa 2004). Conversely, KCAP is one of the most sparsely populated and remotely located protected areas with only rudimentary tourism facilities that cater primarily to self-contained group tourists. Tourist arrivals peaked at 800 in 2001 and dropped to 175 in 2006 (WWF 2007). The histor ic paucity of tourist arrivals and more 103

PAGE 104

recent decrease has been attributed to the poor or non-existent transportation options, lack of tourism infrastructure, and the domestic Maoist insurgency23 (WWF 2007). Enhancing Local Livelihoods The KCAP region is sparsely popul ated and comprised of 4,570 people in approximately 860 households, of which 48% are of Limbu and 25% are of Sherpa ethnicity (WWF 2006). The main sources of income include subsistence agriculture and animal husbandry. Supplemental sources of income include cottage industries, trade with Tibet, remittances24, government, military, and a sma ll volume of tourism (MullerBoker & Kollmair 2000). The la ck of productive land, civil infrastructure, and access to markets and geographic isolation has contribut ed to the area being one of the poorest in Nepal. Taplejung district, where KCAP is located, has the sixth lowest Human Development Index25 score (0.328) in Nepal. Over one third of the residents live below the poverty level, and less than 10% have the capability to produce sufficient quantities of food for dietary subsistence (WWF 2006). The aim of KCAP is to safeguard the biodiversity of the ar ea, and improve the living conditions of the local residents by strengthening the c apacity of the local institutions responsible for making decisions, which will effect the long-term biodiversity conservation and economic development of t he area (WWF 2006, p. 1). Therefore, one of the projects primary obj ectives is to strengthen the capacity of local communities 23 The Maoist insurgency was a decade long civil wa r between communist rebel group and the democratic national government; however, the conflict officia lly ended and the rebel group has been integrated into the government. See Thapa (2004), Battarai et al. (2005), and Baral & Heinen (2006). 24 In Nepal alone, remittances accounted for over 14. 6% of the national gross domestic product in 2006, totaling over US $1.2 billion (World Bank 2009). 25 Human Development Index (HDI) scores are used to define well-being as a composite of education, life expectancy, and standard of living. In 2005, Ne pal had an HDI Score of 0.534, and ranked 142 out of 177 countries (UNDP 2008). 104

PAGE 105

to manage their natural resources while concu rrently promoting sustainable livelihoods and opportunities to reduce the pressure on natural resource, and thus enhance individual, community, and instit utional capacities for development and conservation. Several accomplishments were achieved at KCAP during the fi rst ten years of existence in regards to enhancement of loca l livelihoods. Multip le programs have been created to aid local financia l sustainability and livelihood enhancement that includes: several savings and credit groups, a non-timber forest product (NTFP) enterprises fund, and a livestock insurance program to compen sate instances of wildlife depredation. A few local residents were also provided the opportunity to attend a national level NTFP trade show. Additional community development programs initiated within KCAP include a child day care center, a micro hydro electr ic project, drinking water systems, over 1,200 meters of trails, suspension bridges, toilets, a health post, an irrigation canal, stone fences, and green houses (WWF 2006). Livelihood Challenges Although several positive strides appear to have been accomplished by WWF, there has been a dearth of independent resear ch conducted within KCAP. Muller-Boker & Kollmair (2000) conducted a household survey one year after project initiation and found that only a very small percent (5%) of individuals could identify WWF as the primary organizational agency. Residents also perceived that successful wildlife conservation activities had contributed to increased frequencies of livestock predation and crop damage (Muller-Boker & Kollmair 2000) Furthermore, local resident perceptions of project objectives were heavily skewed towards economic and infrastructure related initiatives, and only a few conservation related objectives. In addition to positive attitudes towards capacity building programs, there also existed 105

PAGE 106

perceptions of unequal access to such program s due to social stratification based on ethnicity, proximity to project offices, and the Maoist insurgency (Muller-Boker & Kollmair 2000). A survey conducted seven years later by WWF (2007) found that resident expectations of project objectiv es were still dominated by a hope for large infrastructure developments. Several challenges to obtaining a sustainabl e livelihood have been identified within KCAP. Reasons include diminished agricultura l capacities, unclear land tenure rights, minimal productive land available and a lack of land ownership, socially inequitable labor habits, food security issues, and the absence of established links to market functions (WWF 2006). Slash and burn tec hniques are practiced around villages in higher elevations (above 3,000 m) due to less available arable land, and are thus more dependent on natural resources for their livelihoods which include resources for cooking, heating, and economic purposes such as NTFPs (WWF 2006). Human-wildlife conflicts and internationa l demand for traditional medicine have contributed to an increased prevalence of poaching and retaliatory killing of large mammals, including snow leopards (Ikeda 20 05; WWF 2006). The degradation of snow leopard habitat and the corresponding decrease in natural prey has resulted in an increase in livestock depredati on and human-wildlife conflict (WWF 2006). Ikeda (2005) found an abundance of retaliatory killings of snow leopards especially in the Ghunsa Valley where the habitat coincided with yak pastoralism. The Ghunsa Valley contains 40 percent of the available snow leopard habitat in KCAP, and is also an integral portion of the tourist trekking route. The average yak herd size wa s 36, and small to medium herd sizes were most susceptible to killings, which jeopardizes herder livelihoods (Ikeda 106

PAGE 107

2005). The annual economic loss due to snow leopard predation ranged from $54 to $405, or from one to six animals and averaged $168 or 2.2 yaks. Average net income per head was $54. Herders with neutral a ttitudes toward KCAP had negative views towards snow leopard conservation policies. They also complained about increased restrictions on fuel wood collection, a ban on snow leopard killings without compensation, and insufficient project mana gement reliability and transparency (Ikeda 2005). Methods The principal objective of this study was to assess discernable differences between current livelihood strat egies, and the associated abili ty to evolve to more effective livelihood strategies within the Kanchenjunga Conservation Area. This study followed a series of both qualitative and quant itative methodologies similarly utilized by the DFID sponsored LADDER26 project (Ellis & Mdoe 2003; Ellis & Freeman 2004). A triangulation method was employ ed to improve the validity of the data collected from several sources, and also to reduce researcher bias (Denzin & Lincoln 2000). Data Collection Data collection occurred for three months (October through December 2008) and utilized four methods. First, a review of KCAP project documents available in the capital city of Kathmandu and the research site wa s completed. Information was examined regarding past and current projects, polic ies, programs, or any other management issues that might have been pertinent to the re search. Second, participant observation with respect to livelihood activities was c onducted among local residents, project staff, 26 LADDER stands for Livelihoods and Diversific ation Directions Explored by Research. 107

PAGE 108

and other associated individuals. Third, a participatory ru ral appraisal (PRA) wealth ranking (World Bank 2007) was performed which consisted of a selection of key informants and community households in as cending order based on affluence. The PRA wealth ranking exercise aids resear chers in assimilating into study site communities, and enhances the capability of a more robust data collection commonly used in conjunction with interview surveys (Grandin 1988; Ellis & Freeman 2004; World Bank 2007). The final step consisted of an on-site survey (adapted from Ellis & Freeman 2004), administered among a r andom sample of households. It is worth noting that t he primary researcher s peaks fluent Nepali and had previously lived in Nepal for over two y ears as a US Peace Corps volunteer in the Department of National Parks and Wildlife Conservation. In addition to the primary researcher, local research assistants were also hired and trained to administer the survey utilizing local dialect where needed, wh ich also helped to build local rapport in the sampled communities. Selection of Participants The survey was administered to adults following a systematic sampling method applied to the entire KCAP pop ulation among the four Village Development Committees (VDC) of Lelep, Tapetok, Yamphudin, and Olachan Gola. Participant selection utilized the comprehensive list created through the PRA wealth ranking, and then from a random start, whereby every fourth househo ld was selected. Due to the high prevalence of illiteracy, and lack of familiari ty with survey procedures, questions were verbally administered while the interv iewer recorded responses. The head of households ( 18 years of age) were targeted. In their absence the next most knowledgeable adult was interviewed at their place of residence (see Nyaupane & 108

PAGE 109

Thapa 2004). Every targeted house chose to participate (response rate = 100%), and 205 of the 215 surveys were usable in the analyses27. Instrumentation The questionnaire assessed household livelihood assets, levels of access, and coping abilities within nine det ailed sections. The first section assessed demographic characteristics in regards to education, o ccupation, and race/cas te/ethnicity. The second section measured earned agriculture-re lated household income. Section three pertained to income generated from livestock, privately owned natural resources and community based natural resources, including NTFPs. The fourth section identified the presence of off-farm income sources and hous ehold responses to unforeseen crises. The questionnaire was slightly altered as deemed necessary following a pilot test conducted among project and government agency staff in the district headquarters of Taplejung. A professor at Tribhuvan University in Kathmandu assessed survey translation accuracy from English to N epali and also back to English. Operationalization and Data Analysis This study assessed two research questions. Research question 1: What is the hou sehold level of livelihood strategy diversification? Households within the KCAP region parti cipate in a variety of both on and offfarm livelihood activities. Categories chosen within the analyses represented the most heavily adopted income generating activities. Agricultural crops were measured as reported yield in weight28. The most dominant crops we re corn, cardamom, millet, and 27 A sample size of 205 is deemed sufficient with a population of 860 for categorical data analysis with a 0.05 alpha level assumption and a 0.05 margin of error (Cochran 1977; Bartlett et al. 2001). 28 Crop yields were reported in maund (1 maund = 37.32 kg). 109

PAGE 110

potatoes. The less frequently ut ilized crops of barley, wheat, and rice were analyzed as cereals, while galaichi and charaito were assessed as medicinals. Livestock were analyzed via two categories of large ruminants (yaks, cows, buffalo, and ox) and small ruminants (pigs and goats). Within this r egion, remittances are a major source of household income; therefore, annual values of remittances received were also collected and utilized as a category. All off-farm income-generation activities (i.e. cottage industries, wage labor, salary, pension, etc.) were assessed as an annual household value. Therefore, a total of 10 distinct income-generation activities were utilized within the analyses: corn, cardamom, millet, potato, cereal, medici nal, large ruminant, small ruminant, remittance, and tota l off-farm income. Similar to Brown et al. (2006), a multi-step cluster analysis was utilized to identify discernable barriers to ex isting livelihood strategies. The authors deemed cluster analysis as a theoretically and statistically defensible means of aggregating the data into distinct livelihood categories (p. 26). This method allows the data to distinctly identify discernable livelihood strategies vi a an asset based approach by accounting for access to required financial, physical, so cial, human, and natural capital. The identification of a households distribution of access to available capital represented adopted livelihood strategies. The ten distinct income-generating activities were utilized within the cluster analysis. A z-score transformation was conducted among the independent variables (livelihood activities) prior to inclusi on within the cluster analysis (Hair et al. 2006). 110

PAGE 111

Research question 2: What are the f actors associated with dominant livelihood strategy adoption? The second research question applied a mult inomial logistic r egression to examine existing correlates of access or barriers to graduate to more effective livelihood strategies (Brown et al. 2006). The multinomial logistic regression analysis determined any discernable barriers to the adoption of additional or more dominant livelihood strategies. A total of twelve variables we re included within the model: the number of household members under 18 y ears of age, amount land ow ned, total livestock units, head of household education level, highest leve l of education completed by an adult household member, receivership of remittanc e, ethnicity, participation role within sponsored programs (non-member, general member, and leadership member), participation in capacity build ing training programs, and visi ts by household members to KCAP offices, or KCAP staff members to the households in the past year. Results Profile of Respondents The geographic distribution of respondents (n=205) within the four VDCs were similar to that identified by WWF29, and included: Lelep VDC (59%), Tapetok VDC (17%), Yamphudin VDC (17%), and Olachan Gola VDC (7%). Males constituted the majority of household heads (8 6%). Sherpa and Limbu ethnici ties were dominant within the region and represented 53% and 41% of t he sampled population respectively. The average head of household was 51 years ol d with no education, and lived with 4.9 additional members of which 2.2 were under 18 years of age. Over 94% owned their 29 Village Development Committee census results were Lelep (47%), Tapetok (31.3%), Yamphudin (15.1%), and Olachan Gola (6.8%) (WWF 2006). 111

PAGE 112

homes with the oldest house built in 1933 and averaged 24 years old. Drinking water systems were constructed of gravity fed above ground polyvinyl chloride (PVC) pipes. Only 1.5% of the households had indoor toilet s. Residents in t he village of Ghunsa (6.5% of the sample) had a fixed micro-hy dro system, while othe r village residents (63.1%) had small photovoltaic pa nels as their sole source of electricity. Houses were constructed with walls of varying compositions of wood, brick, stones, and mud. Houses at higher elevations primarily utilized wood for wall construction, while houses in lower elevation were built of mud. The roofs of houses at hi gher elevations were comprised of wood while lower elevation ho uses primarily had thatched roofs. Participant observation determined that t he most evident predictor of variance within the population was househ old elevation. Therefor e, additional analyses of variance were conducted to compare households located at higher and lower elevations30. All livelihood activities were f ound in both elevation levels; however, notable differences were apparent. Table 4-1 summarizes the frequencies and means for each type of livelihood activities within the lower (76%) and higher elevations (24%). Households at lower elevations ow ned more land per household (27.3 ropani31) compared to households at higher elevations (8 .5 ropani). Therefor e, lower elevation households tended to produce more corn, car damom, and millet, while those at higher elevations tended to produce more potatoes and medicinals. Households at lower elevations owned larger herd sizes of sma ll ruminants; however, households at higher elevations held a greater number of large ruminants. In ad dition, households located in 30 Lower elevation households were found between 1,380 and 2,249 meters, while higher elevation households were located from 2,250 up to 3,600 meters. 31 19.65 ropani equal 1 hectare. 112

PAGE 113

lower elevations on average received more than nine times the annual amount of remittances. Table 4-1. Livelihood activities by elevation Means Activities % of HHs in Lower1 elevations % of HHs in Higher2 elevations Chi2 value Lower (N=156) Higher (N=49) Fvalue Sample (N=205) Corn (weight) a 87 14 93.91 *** 6.00 1.35 37.63 *** 4.89 Cardamom (weight) 67 12 44.41 *** 3.53 1.04 11.71 *** 2.94 Millet (weight) 83 10 86.55 *** 5.01 0.20 58.54 *** 3.86 Potato (weight) 46 65 5.85 ** 2.42 5.49 4.05 ** 2.79 Cerealsb (weight) 13 14 0.02 0.72 0.63 0.05 0.70 Medicinals c (weight) 12 31 10.04 *** 0.50 1.80 11.01 *** 0.81 Large Ruminantsd 87 88 0.05 4.41 10.33 13.77 *** 5.51 Small Ruminantse 83 21 66.26 *** 3.49 1.45 13.03 *** 3.00 Remittancef (NRs) 38 8 15.41 *** 47,327 5,306 14.50 *** 37,283 Off-farm incomeg (NRs) 78 84 0.69 19,041 17,655 0.05 18,710 **=significant at 5%, ***=significant at 1%. a Weights are reported in maund (1 maund = 37.23 kg). b Households engaged in cereals activity produce at least some wheat, barley, rice, and/or buckwheat. c Medicinals include charaito and galaichi. d Large ruminants include cows, yak, buffalo, and ox (number of head). e Small ruminants include pigs and goats (number of head). f Remittance includes any household receiving consis tent money from family members not living in the house. (October 15th 2008, 75 NRs = 1 USD) g Off-farm income includes wage labor, salary, pensions and non-farm business profits, not remittances. 1 Lower elevation households are from 1,380 to 2,249 meters, and include Lelep, Hellok, and Yamphudin. 2 Higher elevation households are from 2,250 to 3,600 meters, and include Ghunsa, and Olachan Gola. Research Question 1: Diversifi cation of Livelihood Strategies Cluster analysis identified five distinct livelihood strategies (i.e. k=5). A hierarchical cluster analysis identified popu lation segments based on an examination of the agglomeration coefficient. K-means clus ter analysis then utilized the initial seeds 113

PAGE 114

derived from the hierarchical method to identi fy the final clusters. The largest observed increases in agglomeration coefficients between clusters were used. This nonhierarchical step confirmed the init ial cluster grouping seed points (Hair et al. 2006). Utilization of both non-hi erarchical and hierarchical steps confirms cluster seeds and increases the credibilit y of the resulting cluster groups (Andreu et al. 2005). Table 4-2 provides the mean values for each of the livelihood activities within the five identified livelihood strategies (cluster s). In addition, Table 4-2 pr esents the percentage of lower elevation households within eac h group, the total number of households within the cluster, and the overall percentage of households represented by the cluster. Twelve additional variables were also included in the table, of which seven were statistically significant and included: number of dependents, total land ow ned, total livestock owned, remittance, Sherpa ethnicity, and household s ponsored program participation roles general or non-existent. Over a quarter of the hous eholds (27%) adopted the leas t diversified livelihood strategy (Cluster 1: Low-income/lowest di versity, large livestock, medicinals, and offfarm). These households produce less than t he sample mean values for each livelihood activity besides large ruminants, medicina ls, and off-farm income. In addition to adopting the least diversified liv elihood strategy, these households also have the lowest annual per capita income of NRs 9,837. Most of the househol ds within this livelihood strategy are of Sherpa ethnicity (89%), and are located in higher elevations (70%) with an average elevation of 2,842 m. These households are the least productive for six of the livelihood activities. Due to the lowest level of land ownership (8 ropani), this population segment owns the least amount of small ru minants and produces the 114

PAGE 115

Table 4-2. Cluster analysis of acti vities within each livelihood strategy Clustering Variables Cluster 1 Cluster 2 Cluster 3 Cluster 4 Cluster 5 Sample means Corn (weight) 0.32 5.15 10.36 3.83 4.48 4.89 Cardamom (weight) 0.45 1.69 3.28 6.00 4.48 2.94 Millet (weight) 0.32 1.15 10.04 1.50 2.83 3.86 Potato (weight) 2.41 17.31 2.60 1.00 0.87 2.79 Cereals (weight) 0.13 0.69 0.36 12.50 0.43 0.70 Medicinals (weight) 1.75 2.15 0.00 0.17 0.51 0.81 Large Ruminants (head) 7.41 14.96 4.85 3.17 3.21 5.51 Small Ruminants (head) 0.92 4.08 3.49 2.83 4.01 3.00 Remittance (NRs) 9,285 63,076 84,000 25,000 22,090 37,283 Off-farm income (NRs) 28,559 31,923 8,477 21,175 16,167 18,710 Overall % of all households 27 6 26 3 38 100 % from lower elevations 30 54 100 83 96 76 No. (#) of households 56 13 53 6 77 205 Per capita income (NRs) 9,837 22,896 20,098 20,066 11,203 14,436 Additional Variables F-values No. of dependents 2.09 2.54 1.85 1.50 2.51 3.12** Total land owned 8.02 24.39 36.59 18.33 23.81 12.40*** Total livestock 5.69 12.62 3.98 3.98 2.80 13.66*** Education of household head 0.91 0.77 0.75 0.83 0.76 0.05 Highest household education level 5.71 6.38 6.42 4.83 5.63 0.41 Remittance (n/y) 0.11 0.38 0.64 0.17 0.22 12.86*** Sherpa (n/y) 0.89 0.92 0.43 0.00 0.29 21.25*** Participation-Leader (n/y) 0.13 0.00 0.09 0.33 0.14 1.214 Participation-General (n/y) 0.68 0.62 0.83 0.00 0.53 5.75*** Participation-None (n/y) 0.19 0.38 0.08 0.67 0.33 4.76*** Trainings (n/y) 0.21 0.08 0.08 0.17 0.18 1.27 Staff visits (n/y) 0.20 0.08 0.30 0.17 0.19 1.05 Office visits (n/y) 0.18 0.15 0.17 0.00 0.12 0.45 Strategy Name Cluster 1: Low income/lowest diversity large livestock, medicinals, and off-farm Cluster 2: High income/highest diversity on and off the farm Cluster 3: High income/high diversity farmers with remittance Cluster 4: Low income/low diversity cardamom and cereals farmers Cluster 5: Low income/low diversity cardamom and millet farmers, small ruminants Statistical significance is represented by *10%, **5%, and ***1%. 1 USD = 75 NRs. 1 hectare = 19.65 ropani. Participation refers to household role with spon sored programs that are community based groups which provide financial or technical assistance for both individual and communal decision-making (leadership, general, and non-member). 115

PAGE 116

smallest quantities of corn, cardamom, m illet, and cereal yields. This group also receives the lowest average annual value of remittances, and therefore supplements their income with the second highest val ue of off-farm work, and the biggest large ruminant herd sizes. The popu lation also has the largest percent of hous eholds without membership in sponsored programs. The second identified adopted livelihood strat egy (Cluster 2: High income/highest diversity on and off the farm) represents 6% of households, are primarily of Sherpa ethnicity (91%), and almost evenly distributed between the higher and lower elevations. This group of households outperforms the other groups in five livelihood activities that includes potatoes, medicinals, both large and small ruminants, and off-farm activities. In fact, households within this strategy produce over six times the average weight of potatoes. In addition, these households are t he second largest producers of corn and cereals. Households also s upplement their high ly productive on-farm activities with the highest average value of off-farm incomes, and the second highest average values of remittances received that provides them with the highest annual per capita income of NRs 22,896. The third livelihood strategy (Cluster 3: High income/high diversity farmers with remittance) was adopted by 26% of the households all located within lower elevations. Households with this livelihood strategy ow n the most land (36.6 ropani), produce the most corn and millet, and receive the larges t average value of annual remittances (NRs 84,000). The per capita annual income for this group is the second highest at NRs 20,098. However, although this group receives the highest values of remittances, it produces the lowest average values of off-farm income. 116

PAGE 117

The fourth livelihood strategy (Cluster 4: Low income/low diversity cardamom and cereals farmers) was adopted by the smalle st percentage of households (3%). The households within this group are primarily (83%) located in lower elevation villages, have the least average amount of dependents and are not of Sherpa ethnicity. Although households who have adopted this st rategy are the lar gest average producers of cardamom and cereals, they have the smallest average number of large ruminants and the second smallest average herd sizes of small ruminants. Average cereal production within this strategy is 12.5 maund, while the overall average was only 0.7. The fifth and most frequent livelihood strategy (Cluster 5: Low income/low diversity cardamom and millet farmers, small ruminants) was adopted by 38% of households. This group is predominantly fr om lower elevations (96%) with an average elevation of 1905 m. This group has the second lowest average annual per capita income of NRs 11,023, and second lowest average values of both remittances and offfarm income. However, these households do produce the second largest average yields of cardamom and millet. Research Question 2: Factors Associat ed With Dominant Livelihood Strategy Adoption The two most diversified livelihood stra tegies with higher levels of income previously identified were only chosen by 32% of the households. Possible barriers and factors could be related to access to necessary assets that include financial, technical, physical, social, human and natural. A mult inomial logistic r egression analysis was conducted to ascertain possible factors or barriers to the adoption of graduated livelihood strategies. Results of the mult inomial logistic regression utilized to assess factors or barriers associated with the adoption of the highest income livelihood strategy 117

PAGE 118

are presented in Table 4-3. However, prior to analysis, the fourth identified livelihood strategy (Low income/low diversity car damom and cereals farmers) (represented only 3% of households) was combined with the fifth livelihood strategy (Low income/low diversity cardamom and millet farmers, small ruminants) and referred to as low income/low diversity farmers. Four variables (number of dependents, tota l land owned, total livestock units, and Sherpa ethnicity) were significant predictor s of the likelihood that a household adopted the low income/lowest diversity large livestock livelihood strategy over the low income/low diversity farmers livelihood st rategy. Low income/off-farm workers, on average, were more often located in the higher elevations with less than a third of land under ownership, owned about twice the to tal livestock units, had less children, and were more often of Sherpa et hnicity. Living at higher elev ations limits the amount of available livelihood activities, especially farm activities. Theref ore, households within this strategy are forced to seek off-farm activities for employment (e.g. trade and tourism activities). Households that adopted this livelihood strategy were also more likely to not participate in sponsored programs. In the second comparison case, four predi ctor variables (total livestock owned, remittance, Sherpa ethnicity, and leader participat ion role) were statistically significant in regards to household adoption of the highest diversity and income producing livelihood strategy of high income/highest diversity on and off the farm. Households that adopted this diversified livelihood strategy on av erage owned five times more livestock units, received remittances more often, and were more likely to be of Sherpa ethnicity. 118

PAGE 119

Table 4-3. Factors of livelihood strategy adoption Variable Coefficient Std. error Sig. Low income/lowest diversity large livestock, medicinals, and off-farm vs. low income/low diversity farmers (Cluster 1 vs. Clusters 4 & 5) Dependents -0.364 0.194 0.060 Total land owned -0.112 *** 0.024 0.000 Total Livestock Units 0.374 *** 0.103 0.000 Education of head 0.060 0.113 0.595 Highest Education -0.042 0.060 0.482 Remittance (n/y) 0.116 0.718 0.871 Sherpa (n/y) 2.678 *** 0.593 0.000 Participation-Leader (n/y) 0.174 0.917 0.850 Participation-General (n/y) 0.896 0.605 0.151 Trainings (n/y) 0.504 0.740 0.496 Staff visits (n/y) 1.070 0.720 0.137 Office visits (n/y) 0.183 0.770 0.812 High income/highest diversity on and off the farm vs. low income/low diversity farmers (Cluster 2 vs. Clusters 4 & 5) Dependents -0.017 0.354 0.962 Total land owned -0.007 0.021 0.731 Total Livestock Units 0.528 *** 0.115 0.000 Education of head 0.113 0.166 0.495 Highest Education -0.061 0.099 0.536 Remittance (n/y) 1.809 ** 0.861 0.036 Sherpa (n/y) 2.418 ** 1.209 0.046 Participation-Leader (n/y) -18.210 *** 2.345 0.000 Participation-General (n/y) 0.417 0.854 0.625 Trainings (n/y) -0.173 1.511 0.909 Staff visits (n/y) -0.550 1.359 0.686 Office visits (n/y) 0.541 1.086 0.618 High income/high diversity farmers with remittance vs. low income/low diversity farmers (Cluster 3 vs. Clusters 4 & 5) Dependents -0.439 ** 0.184 0.017 Total land owned 0.013 0.010 0.186 Total Livestock Units 0.293 *** 0.106 0.006 Education of head 0.097 0.103 0.347 Highest Education -0.108 0.058 0.063 Remittance (n/y) 1.918 *** 0.487 0.000 Sherpa (n/y) 0.345 0.522 0.510 Participation-Leader (n/y) 1.204 0.965 0.212 Participation-General (n/y) 2.044 *** .0679 0.003 Trainings (n/y) -1.586 ** 0.798 0.047 Staff visits (n/y) 0.043 0.570 0.940 Office visits (n/y) 0.352 0.693 0.612 Number of observations 205 Percent correctly predicted 68.7 Log-likelihood value 298.561 Pseudo R-squared 0.703 Statistical significance is represented by *10%, **5%, and ***1%. 119

PAGE 120

However, none of the househo lds that adopted the most dom inant livelihood strategy held a leadership position wit hin sponsored programs. In the third comparison case, six vari ables (number of depend ents, total livestock units, highest education level, remittance, general participation role, and trainings) were significant predictors of the likelihood a household adopted the more diversified and financially lucrative livelihood strategy of high income/high diversity farmers with remittance. These households, on average, had fewer dependents, owned almost 50% more livestock units, had less educated adults within the family, and received remittances about three times as frequently. Households that adopted this livelihood strategy usually (83%) held general mem ber participation ro les within sponsored programs, and participated less often in training provided by KCAP, possibly due to the lack of necessity. Discussion The purpose of this research was to i dentify the current levels of livelihood diversification among households, and subsequent ly to identify potential factors or barriers to the successful adopt ion of potential dominant live lihood strategies at KCAP. Results are based on a cluster analysis of the ten most prevalent liv elihood activities, where households were segmented into five distinct livelihood strategies. The five livelihood strategies were statistically di fferent among seven household characteristic variables that included: number of dependents, total land ow ned, total livestock units, remittance, Sherpa ethnicity, and participation roles. Findings were similar to the literature with respect to di versified livelihood strategi es, ICDP, and participation in decentralized conservation (Barrett et al. 2001; Rew & Rew 2003; Ellis & Mdoe 2003; 120

PAGE 121

Adhikari et al. 2004; Agrawal & Gupta 2004; Bennett et al. 2006; Brown et et al. 2006; Bhandari & Grant 2007). Similarly, other results resonated with the literature indicating households have disproportionately adopted the more diversif ied and higher income generation livelihood strategies based on several socio-economic variables and geographic location. These households are characterized by access to livelihood assets that include human capital (having fewer dependents) (Brown et al. 2006), natural capital (owned more land) (Ellis & Mdoe 2003), financial capital (receive more remittances and own more livestock) (Barrett et al. 2001; Ellis & Mdoe 2003; Adhikari et al. 2004; Bhandari & Grant 2007), and social capital (participation in sponsor ed programs) (Rew & Rew 2003; Agrawal & Gupta 2004). Conversely, there were several predictor variables that failed to have statistical significance in this study. Education of both the head of household and fellow adult members had no impact on live lihood strategy adoption largely due to the overall lack of education within the region. The average level of educat ion among head of households was less than one year, and only six percent of all household members at least thirty years old had even attended school. In addition, access to saving and credit programs, social connections with agency staff, and capacity building training and programs were also all non-significant. Lack of statistical relationships could be attributed to the overall low rates (less than 20%) of visitations by hous ehold members to KCAP offices, visits to households by KCAP staff, and non-participation in sponsored training programs. In relation to the overall financial state of the region, caution should be utilized in the interpretation of the economic results. Although each of t he identified livelihood 121

PAGE 122

strategies were statistically discernable, ov er 75% of households failed to earn at least a dollar a day. The KCAP area is still considered very poor with lack of infrastructure, external markets, and comm unity development. In addition, remittances are a crucial determinant of livelihood strategy adoption in the region. Almost a third (31%) of the households received remittances with an annual mean value that varied from US $124 to $1,120, and a maximum of over US $6,600. This finding is important since that total annual off-farm household incomes (excluding remittances) only varied between US $113 to $426, with an annual maximum value of half th at of remittances or US $3,300. Overall, based on the findings of this research, several recommendations can be made to improve the management of KCAP. Livelihood dive rsification is dependent on access to participation in sponsored progr ams (Rew & Rew 2003; Agrawal & Gupta 2004). Therefore, household participation in capacity building programs needs to improve to reduce the impact of negative vulnerabilities (Ellis 2000), especially given the temporal viability among commercial markets and associated linkages (DFID 2007). KCAP staff needs to increase the frequency of social mobilization programs, and thus improve the rate of interaction with household members. Study re sults could enhance the ability of KCAP managers to target households that have the most need for community capacity building programs and strat egies. This would enhance the ability of KCAP to effectively manage the areas diverse biodiversity, while concurrently improve the livelihoods of its resident s. Successful management of KCAP is not only a priority for Nepal, but also for the international conservation community. KCAP is a vital component for the eventual creation of a trans-boundary Kanchenjunga mountain complex conservation area between N epal, India, and China. 122

PAGE 123

The limitations of the study are primarily related to t he nature of the data collected. Economically related household information is often reluctantly provided to researchers. Household values are based sole ly on what was provided by the respondent. However, several steps were taken to alleviate possibl e socially desirable res ponse effects. Even though the primary researcher previously live d in Nepal for over two years providing assistance to communities within protected areas, local research assistants were hired to help administer the surveys. An additional limitation of economically related research is due to temporal changes in market linkages, and the resultant affects on livelihood strategy adoption. Therefore, this study operationalized onfarm livelihood activities in yield weights and not economic value. Further research should assess the regions access to external markets, and the affect these linkages have on the associated value of livelihood activities. For example, prev iously cardamom was an extremely lucrative activity and was utilized as a livelihood strategy by most of the households. Recently, the price of cardamom dramatically decreased and households have attempted to switch livelihood activities. Findings of this research are relevant along a range of scales from the local community level to the broader body of academic knowledge. Results add to the literature on sustainable livelihoods, as it co uld possibly be the first study to utilize this research framework within a mountainous protected area that operates under a completely decentralized form of management. In addition, results add to the discourse on the need for a robust quantit ative method of operationa lization of diversified livelihood strategies. Th is research adapted Brown et al.s (2006) research whereby cluster analysis was utilized to identify discernable livelihood strategies, and subsequent 123

PAGE 124

techniques identified potential fa ctors or barriers to the adopt ion of more diversified or graduated livelihood strategies. Results wil l assist program managers in identifying household characteristics strongly associat ed with those who lack access to livelihood assets (financial, social, physical, natur al, and human) necessary to adopt improved livelihood strategies. The pot ential for ICDP sustainability would be greatly enhanced by the ability to identify disadvantaged hous eholds and target project spending while organizational costs are minimized. In addition, findings will aid policy makers abilities to identify current environmental pressures that are most important for futu re policy modifications, and to simultaneously apply strategies that are most critical to a households liveliho od. This ability will enhance the management of community based natural resources, and promote strategies that will create more sustainable community capacity building programs while concurrently alleviating pressures on the areas biological diversity. 124

PAGE 125

CHAPTER 5 CONCLUSION This study examined the role of partici pation in sponsored pr ograms, levels of natural resources dependency, and the existing levels of livelihood strategy diversification among households in t he Kanchenjunga Conserva tion Area Project, Nepal. Overall, this study has shown that although KCAP is the first completely decentralized conservation program in Nepal, a disproportionate amount of the benefits are being accrued by a small group of households that maintain decision-making roles. In terms of sustainability, the ICDP literature calls for a complete devolution of decisionmaking responsibilities and system that provides for an equita ble distribution of benefits; neither currently exist at KCAP. Participation Roles The findings of this study highlight the variables that influence household participation roles in sponsored programs, and ultimately underline the impact of participation roles on the distribution of benefits derived from involvement in decentralized conservation programs. Econ omic, social, and access variables were included in the analyses. Economic variables included remittances, existence of offfarm income, land, and livestock. Overall fi ndings suggest that none of the economic variables measured were significantly related to household participation roles. The lack of statistical significance among the economic variables is likely associated with the homogeneity of the population. There were significant positive correlations between the four economic variables which indicated if a household had one of the variables, it usually had all four variables. Likewise, if a household was economically disadvantaged on one economic variable, they were often di sadvantaged on all four variables. Overall 125

PAGE 126

the KCAP region is rather homogeneous in ec onomic terms; i.e., economically poor. Quantities of livestock are highly correlated wit h altitude and inversely related to land ownership. Households at higher elevations have lar ger herd sizes and less land compared to lower elevation households. It is likely that the lack of variation within the economic variables that led to statistica l insignificance and thus were ineffective predictors. Social variables were signi ficantly related to househol d participation roles and included education levels of the head of household, highest education levels of household adults, head of household age, and size of households. Although Maskey et al. (2006) found age to be positively related to participation roles; however, this current research found a statistically significant negat ive relationship. Results of this study found that overall hou sehold size was a better predictor of participation roles than number of dependents. Additiona lly, all three access variables in this study were significantly related to participation roles. An increase in interactions with KCAP offices and staff were positively related to participat ion roles. These results highlight the importance of program staff to conduct social mobilization activities that include social capacity trainings and community development projects. A simple act of allowing the community to interact with agency st aff provides a major benefit. The distribution of benefits within KCAP was found to be unequa l. Households with higher participation role s benefited from enhanced acce ss to financial credit in terms of higher principles and lower interest rates. Higher participation roles also provided for an increased level of involvemen t in capacity development training offered 126

PAGE 127

via program offices. With in KCAP, if equal distribution is a goal, then increased promotion of program participation is important and necessary. Natural Resources Dependency The sustainability of conservation projects has been shown to be enhanced by the ability to identify household resource use patterns, and thus better target and improve the capacity of those households (or communities) who are most heavily dependent on access to natural resources for their livelihood strategies (Vedeld et al. 2004). This study assessed the level of household dependency on access to natural resources within KCAP, and identified factors that affect the associated levels of natural resources dependency. Results of this stud y indicate that households within KCAP were heavily dependent on natural resources for their income. The average household dependence on natural resources access for in come in developing countries is 22% (Vedeld et al. 2004); however, this study found that 64% of households in KCAP were dependent. Households at higher elevations (74.6%) were more dependent than lower elevation (60.3%) households. Several economic, social, and access re lated variables were significant predictors of natural res ources dependency. Land ownership was not a significant predictor of dependency, but remittances were significan t. Households that received remittances were able to supplement their income without reliance on natural resources access. Education was not a significant predictor of natural resources dependency which may be due to the low overall levels of education within the KCAP region. The average education of a household head was less than one year, while less than six percent of individuals at least thirty y ears of age had even attended school. This study found that natural resour ces dependency increased among households with larger 127

PAGE 128

numbers of household members, while depe ndency decreased among households with more dependents. Caste/race/ethnicity failed to have any predictive significance. Spatial distribution was rather homogenous within the study site. All households were disconnected from external commercial ma rket linkages (at least three days walk), and the placement of four KCAP offices thro ughout the region facilitated accessibility to sponsored programs. T here was no predictive validity in regards to natural resources dependency on the varying levels of training par ticipation and the amount of interaction with KCAP staff (both within t he household and at the offi ces). However, access to loans within the sponsored program did have a positive significant affect on natural resources dependency. Less t han 20% of households have had direct communication with KCAP staff within the past year, and only 16% had participated in a training program. Social mobilization techniques and provided traini ng programs are targeted to achieve the overall aim of KCAP which is to protect biological diversity by alleviating environmental pressures while concurrently improving livelihoods via strengthening the capacity of the local institutions for making decisions (WWF 2006, p. 1). However, few households have benefited from project enhancement techniqu es, therefore they are limited to the utilization of loans for traditi onal livelihood techniques that are strongly reliant on natural resources. Livelihood Diversification This study identified the current leve ls of livelihood diversification among households, and subsequent factors or barriers to the successful adoption of potential dominant livelihood strategies. Based on a cluster analysis, the ten most prevalent livelihood activities among house holds were segmented into fi ve statistically distinct strategies. Significant pr edictor variables included: number of dependents, total land 128

PAGE 129

owned, total livestock units, remittance, S herpa ethnicity, and participation roles. Findings were similar the literature with re spect to diversified livelihood strategies, ICDP, and participation in decentralized conservation (Barrett et al. 2001; Rew & Rew 2003; Ellis & Mdoe 2003; Adhikari et al. 2004; Agrawal & Gupta 2004; Bennett et al. 2006; Brown et al. 2006; Bhandari & Grant 2007). Similarly, other results resonated with the literature indicating households that have disproportionately adopted more diversified and higher income generation livelihood strategies were associated wit h several socio-economic variables and geographic location. These households were characterized by access to livelihood assets that included human capital (having fewer dependents) (Brown et al. 2006), natural capital (more land under ownership) (Ellis & Mdoe 2003), financial capital (receive more remittances and ow n more livestock) (Barrett et al. 2001; Ellis & Mdoe 2003; Adhikari et al. 2004; Bhandari & Grant 2007), and social capita l (participation in sponsored programs) (R ew & Rew 2003; Agrawa l & Gupta 2004). Conversely, there were several predictor variables that failed to achieve statistical significance. Education of both the head of household and fe llow adult members had no impact on livelihood strategy adoption, la rgely due to the low levels of education within the region. In addition, access to saving and credit programs (Dercon & Krishna 1996; Acharya et al. 2005), social connections with agency staff, and capacity building trainings and programs were also all non-significant predictors. In relation to the overall financial state of the region, caution should be utilized in the interpretation of the economic results. Although each of t he identified livelihood strategies were statistically discernable, ov er 75% of households failed to earn at least a 129

PAGE 130

US $1 a day. The KCAP area is very economically disadvantaged, it lacks infrastructure, external mark ets, and community developmen t. In addition, remittances are a crucial determinant of liv elihood strategy adoption in t he region. Almost a third (31%) of the households received remittanc es with an annual mean value that varied from US $124 to $1,120, and a maximum of ov er US $6,600. This finding is important given the total annual off-farm household incomes (excluding remittances) only varied between US $113 to $426, with an annual maximum va lue of half that of remittances or US $3,300. Recommendations to Enhance KCAP Management Several recommendations can be made based on the findings of this research to improve the management of KC AP. Participation in sponsored programs was found to be related to the level of interaction with KC AP staff. Greater roles of participation in sponsored programs were also significantly related to increased access to associated benefits including financial credit, lower interest rates, and training programs. However, less than 20% of household had any interacti on with KCAP staff. The lack of social mobilization strategies and the absence of a region wide system of KCAP sponsored capacity development projects has detrimentally limited the availability of opportunities for households to choose livelihood activities that are less dependent on natural resources. KCAP needs to target thos e households that are not participating in sponsored programs, therefor e they received fewer benefits, and are more heavily dependent on natural resources for their livel ihoods. Household size was positively associated with participation roles. To ensure a more comprehensive level of involvement within conservation areas and between socio-economic categories, managers should encourage participation from smaller households. 130

PAGE 131

Results in this study indicate that households within KCAP ar e heavily dependent on access to natural resources for their liv elihood strategies. The sustainability of KCAP as an ICDP is therefore contingent on project intervention via social mobilization, training programs, and participatory inclusion in decision making. Livelihood diversification is dependent on access to participation in sponsored programs (Rew & Rew 2003; Ag rawal & Gupta 2004). Househ old participation rates in capacity building programs need to be improved which w ould reduce the impact of negative vulnerabilities (Ellis 2000), especia lly given the temporal viability among commercial markets and associated linkages (DFID 2007). KCAP staff needs to increase the frequency of social mobilizatio n programs and thus improve the rate of interaction with household member s. The results of this re search could be utilized to enhance the ability of KCAP managers to identify and target households that lack the necessary assets required for more viable livelihood strategy adoption. These households thus have the most need for c apacity building programs and strategies. This would enhance the ability of KCAP to effectively manage the areas diverse biodiversity, while concurrently improv ing the livelihoods of its residents. Research Limitations There are some limitations to this study. Although the primary researcher had previously lived in Nepal for over two y ears and spoke the national language, cultural differences still occurred. Local research a ssistants were hired to alleviate response or researcher bias; however, the availability of trained social researc hers at the site was non-exsistant. Economically re lated household information is often reluctantly provided to researchers, and household values were based solely on the information provided by 131

PAGE 132

respondents. However, several steps were tak en to alleviate possible socially desirable response effects. The operationalization of variables within this study was adapted from previous research conducted in decentralized conservation programs; however, additional improvements could be made. Due to low fr equencies of participation roles, this study recoded participation in only thr ee categories. A descriptive narrative of an example of each participation role could accompany the questionnaire to facilitate comprehension of role labels. Future Research Decentralized conservation programs have been heralded in the literature based on their theoretical ability to: 1) decreas e organization costs, 2) provide financial and technical assistance opportunities to previ ously excluded stakeholders, and 3) establish a creatively enhanced and competitive economic sector (Ribot 2002, Agrawal & Gupta 2005). However, considering the discrepancies of loan and interest rate distributions across households, and the unequal distribution of capacity development training present at KCAP, further research is neede d to determine the level of financial and social inequalities. Households within KCAP are heavily depe ndent on access to natural resources for their livelihood strategies. The sustaina bility of KCAP as an ICDP is therefore contingent on project intervention via soci al mobilization, tr aining programs, and participatory inclusion in decision maki ng. An empirical study of household consumption levels among households should be assessed in terms of fuel wood, timber for construction, and grazing (see Adhikari et al. 2004). Additionally, given the prevalence and impact of remittances on natural resources dependency (31% receive 132

PAGE 133

remittance that constitute about two-thirds of their off-farm income), a study should be conducted as to the factors that affect whether a household receives remittances. Providing externally funded training programs targeting the adoption of alternative livelihood activities are futile if there is a disconnect from a commercial market. Further research needs to be conducted pertaining to market linkages within the KCAP region. A limitation on economically re lated research is due to temporal changes in market linkages, and the resultant affect s on livelihood strategy adopti on. Therefore, this study operationalized on-farm livelihood activities in yield weights and not economic value. Further research should assess the regions a ccess to external mark ets, and the affect these linkages have on the associated value of livelihood activities. For example, previously cardamom was an extremely lucr ative activity and was utilized as a livelihood strategy by most of the hous eholds. Recently, the price of cardamom dramatically decreased and households have attempted to switch livelihood activities. 133

PAGE 134

APPENDIX A HOUSEHOLD SURVEY QUESTIONAIRRE Protocol Title: Participation, dependency, and dive rsification: A livelihoods assessment of Kanchenjunga Cons ervation Area Project, Nepal. Please read this consent document carefully before you decide to participate in this study. Purpose of the research study: The purpose of this study is to assess nongovernment organizational impacts on local livelihood strategies. What you will be asked to do in the study: To answer the interview questions. Time required: 15 30 minutes Risks and Benefits: There are no anticipated risk or other direct benefit s to you as a participant in this study. Compensation: There is no compensation for par ticipating in this research. Confidentiality: Your identity will be kept confidentia l to the extent provided by law. Your name will not be recorded, and so y our participation and responses will be completely anonymous. The findings will never discuss individual responses in any report. Voluntary participation: Your participation in this study is completely voluntary. There is no penalty for not participating. Right to withdraw from the study: You have the right to wit hdraw from the study at anytime without consequence. You do not have to answer any questions you do not want to answer. Whom to contact if you have questions about the study: Pete Parker, Graduate Student, Department of Tourism, Recreation & Sport Management, 325 Florida Gym, phone 208-731-0069. Email: idparker@ufl.edu Or Dr. Brijesh Thapa, Associat e Professor, Department of Tourism, Recreation & Sport Management, 325 Florida Gym, phone (352) 392 4042extension 1239. Email: bthapa@hhp.ufl.edu Whom to contact about your rights as a research participant in the study: IRB02 Office, Box 112250, University of Fl orida, Gainesville, FL 32611-2250; ph. 352392-0433. 134

PAGE 135

135

PAGE 136

136

PAGE 137

137

PAGE 138

138

PAGE 139

139

PAGE 140

140

PAGE 141

APPENDIX B HOUSEHOLD SURVEY QUESTIONAIRRE (TRANSLATED INTO NEPALI) 141

PAGE 142

142

PAGE 143

143

PAGE 144

144

PAGE 145

145

PAGE 146

146

PAGE 147

147

PAGE 148

APPENDIX C INSTITUTIONAL REVIEW BOARD APPROVAL 148

PAGE 149

LIST OF REFERENCES Aase, T. & Vetaas, O. (2007) Risk m anagement by communal decision in TransHimalayan farming: Manang Vall ey in Central Nepal. Human Ecology 35: 453460. Acharya, A., Yoshino, E., Jimba, M. & Wakai, S. (2005) Em powering rural women through a community developm ent approach in Nepal. Community Development Journal 42(1): 43-46. Adhikari, B., Falco, S. & Lovett, J. ( 2004) Household characteristics and forest dependency: Evidence from common proper ty forest management in Nepal. Ecological Economics 48(2004): 245-257. Agrawal, A. (2000) Small is beautiful, but is larger better? Forest-management institutions in the Kumaon Him alaya, India. In: Clark, G., Margaret, A. & Elinor, O. eds People and Forests, Communities Institutions, and Governance The MIT Press, Cambridge, pp.57-85. Agrawal, A. & Gibson, C. (1999) Enchantment and disenchantment: The role of community in natural resource conservation. World Development 27(4): 629-649. Agrawal, A. & Gupta. K. ( 2005) Decentralization and participation: The governance of common pool resources in Nepals Terai. World Development 33(7): 1101-1114. Agarwal, B. (2001) Participatory exclusions, communi ty forestry, and gender: An analysis for south Asia and a conceptual framework. World Development 29(10): 1623-1648 Allendorf, T. (2007) Residents attitudes toward three protected areas in southwest Nepal. Biodiversity and Conservation 16(7): 2087-2102. Andreu, L. Kozak, M., Avci, N. & Cifter, N. (2005) Market segmentation by motivations to travel: British tourists visiting Turkey. Journal of Travel and Tourism Marketing 19(1): 34-56. Ashley, C. (2000) The impacts of tourism on rural livelihoods: Namibias experience. Overseas Development Institute Working Paper 128. Ashley, C. & Carney, D. (1999) Sustainable Livelihoods: Lessons from early experience Department for International Development, London. Bajracharya, S., Furley, P. & Newton, A. (2005) Effectiveness of community involvement in delivering conservati on benefits to the Annapurna Conservation Area, Nepal. Environmental Conservation 32 (3): 239. 149

PAGE 150

Bajracharya, S., Furley, P. & Newton, A. (2006) Impacts of community-based conservation on local communities in the Annapurna Conservation Area, Nepal. Biodiversity Conservation 15: 2765-2786. Baral, N. & Heinen, J. (2005) The Maoi st peoples war and conservation in Nepal. Politics and Life Sciences 24(1-2): 2-11. _________________ (2007) Decentraliz ation and peoples participation in conservation: A comparative study from the western terai of Nepal. International Journal of Sustainable Development & World Ecology 14(2007): 520-531. Baral, N., Stern, M. & He inen, J. (2007) Integrated c onservation and development project life cycles in the Annapurna Conservation Area, Nepal: Is development overpowering conservation? Biodiversity Conservation 16(10): 2903-2917. Barret, C., Reardon, T. & W ebb, P. (2001) Nonfarm income diversification and household livelihood strategies in rural Africa: Concepts, dynamics, and policy implications. Food Policy 26(2001): 315-331. Barret, C., Marenya, P., McPeak J., Minten., B., Murithi, F. Oluoch-Kaoura, W., Place, F., Randrianarisoa, J., Rasambainarivo, J. & Wangila, J. (2006) Welfare dynamics in rural Kenya and Madagascar. Journal of Development Studies 42(2): 248-77. Battarai, K., Conway, D. & Shresta, N. (2005 ) Tourism, terrorism and turmoil in Nepal. Annals of Tourism Research 32(3): 669-688. Berkes, F. (2007) Community-based cons ervation in a globalized world. Proceedings of the National Academy of Sciences 104(39): 15188-15193. Bhandari, B. & Grant, M. (2007) Analysis of livelihood security: A case study in the KaliKola watershed of Nepal. Journal of Environmental Management 85(2007): 17-26. Blaikie, P. & Springate-Baginski, O. (2003) Is community forestry in contemporary Nepal pro-poor and sustainable? Improving policy-livelihood relationships in south Asia, Briefing Note 4. Booth, D., Holland, J., Hentschel, J. Lanjouw, P. & Herbert, A. (1998) Participation and combined methods in African Povert y Assessment: Renewing the Agenda Social Development Division, UK Department fo r International De velopment (DFID), Feb., mimeo. Brandon, K. (1997) Policy and practical consi derations in land-use strategies for biodiversity conservation. In Kramer C., van Schaik, C. & Johnson, J. eds. Last stand: Protected areas and the defense of tropical biodiversity. Oxford, UK: Oxford University Press. 150

PAGE 151

Brandon, K., Redford, K. & Sanderson, S. eds (1998) Parks in Peril: People, politics, and protected areas and the defense of tropical diversity New York: Oxford University Press. Brandon, K. & Wells, M. (1992) Planning for people and par ks: Design dilemmas. World Development 20(4): 557-570. Brechin, S. & West, P. (1991) Resident peoples and national parks: Social dilemmas and strategies in international conservation Tuscon, AZ: University of Arizona Press. Brechin, S., Wilshusen, P., Fortwangler, C. & West, P. (2002) Reinventing the square wheel: A critique of the new protectionist paradigm in international biodiversity conservation. Society and Natural Resources 15: 41-64. Brown, D., Stephens, E., Okuro Ouma, J., Muri thi, F. & Barrett, C. (2006) Livelihood strategies in the rural Kenyan highlands. African Journal of Agriculture and Resources Economics 1(1): 21-35. Budhathoki, P. (2004) Linking communities with conservation in developing countries: Buffer zone management initiatives in Nepal. Oryx 38(3): 334-341. Chambers, R. & Conway, G. (1992) Sustaina ble rural livelihoods: Practical concepts for the 21st century. IDS discussion paper 296. Brighton, UK: Institute of Development Studies, University of Sussex. Chandool, C. (2007) Households resource use, c onservation attitude and perceived socioeconomic impact of the protected areas policy in the Nariva Swamp, Trinidad and Tobago Dissertation, University of Florida, Gainesville, FL. Chapin, M. (2004) A chall enge to conservationists. Wolrdwatch 17(6): 17-32. Child, B. (2004) Parks in transition: Biodiversity, rural development and the bottom line United Kingdom: Earthscan. Cochran, W. (1977) Sampling techniques (3rd ed.). New York: John Wiley & Sons. Denzin, N. & Lincoln, Y., eds. (2000) Handbook of Qualitative Research Thousand Oaks, CA: Sage Publications, Inc. Dercon, S. & Krishna, P. ( 1996) Income portfolios in rural Ethiopia and Tanzania: Choices and constraints. Journal of Development Studies 32(6): 850-75. Dev, O., Yadav, N., Springate-Baginski, O. & Soussan, J. (2003) Impacts of community forestry on livelihoods in the middle hills of Nepal. Journal of Forest and Livelihood 3(1): 67-87. 151

PAGE 152

Ellis, F. (1998) Household strategies and rural livelihood diversification. The Journal of Development Studies 35(1): 1-38. Ellis, F. (2000) Rural livelihoods and divers ity in developing countries NY: Oxford University Press. Ellis, F. & Allison, E. (2004) Livelihood diversification and natural resource access. Overseas Development Group, Universi ty of Anglia, UK FAO, LSP WP 9. Ellis, F. & Freeman, A. (2004) Rural livelihoods and poverty reduction strategies in four African countries. The Journal of Development Studies 40(4): 1-30. Ellis, F. & Mdoe, N. (2003) Livelihoods and rural poverty reduction in Tanzania. World Development 31(8): 1367-1384. Frankenberger, T., Drinkwater, M. & Maxwell, D. (2000) Operationalizing household livelihood security: A holistic approac h for addressing poverty and vulnerability. Program Document, CARE USA, CARE, Atlanta, GA. Ghimire, K., & Pim bert, M., eds (1997) Social change and conservation London: Earthscan. Grandin, B. (1988) Wealth ranking in smallerholder communities: A field manual. London: Intermediate Technology. Greene, W. (2000) Econometric analysis fourth ed. Prentice Hall, Englewood, CA. Gunatilake, H. (1998) The role of rural development in protec ting tropical rainforests: Evidence from Sri Lanka. Journal of Environmental Management 53: 273. Hair, D., Babin, B., Anderson, R ., Tatham, R., & Black, W. 2005. Multivariate Data NY: Prentice Hall. Heinen, J. (1993) Park-people relationships in Kosi Tappu Wildlife Reserve, Nepal: A socio-economic analysis. Environmental Conservation 20(1): 25-34. Heinen, J. & Kattel, B. (1992a) A review of conservation l egislation in Nepal: Past progress and future needs. Environmental Management 16(6): 723-733. Heinen, J. & Kattel, B. (1992b) Parks, people, and cons ervation: A review of management issues in Nepals protected areas. Population and Environment 14(1): 49-84. Heinen, J. & Mehta, J. ( 1999) Conceptual and legal issues in the designation and management of conservation areas in Nepal. Environmental Conservation 26(1): 21-29. 152

PAGE 153

Heinen, J. & Mehta, J. (2000) Emerging issues in lega l and procedural aspects of buffer zone management with case studies from Nepal. Journal of Environment & Development 9(1): 45-67. Herold-Menzies, M. (2006) In tegrating conservation and dev elopment: What can we learn from Caoha i, China. The Journal of Environment & Development 15(4): 382406. Horwich, R. & Lyon, J. (2007) Community conser vation: Practitioners answer to critics. Oryx 41(3): 376-385. Ikeda, N. (2005) Economic impacts of livestock depredation by snow leopard Uncia uncia in the Kanchenjunga Conservation Area, Nepal Himalaya. Environmental Conservation 31(4): 322-330. IUCN. (1994) Guidelines for protected area management categories Cambridge, UK: IUCN. Jumkbe, C. & Angelsen, A. (2007) Forest dependence and participation in CPR management: Empirical evidence from forest co-management in Malawi. Ecological Economics 62(2007): 661-672. Kellert, S., Mehta, J., Ebbin, S. & Lichtenfeld, L. ( 2000) Community natural resource management: Promise, r hetoric, and reality. Society and Natural Resources 17: 705-715. Lachapelle, P., Smith, P. & McCool, S. (2004) Access to power or genuine empowerment? An analysis of three co mmunity forest groups in Nepal. Human Ecology Review 11(1): 1-12. LADDER Research Team. (2001) Me thods manual for fieldwork. LADDER Working Paper No. 2. Lise, W. (2000) Factors influenc ing peoples participation in fo rest management in India. Ecological Economics 34: 379-92. Maskey, V., Gebremedhin, T. & Dalton, T. (2006) Social and cultural determinants of collective management of community forest in Nepal. Journal of Forest Economics 11(2006): 261-274. McNeely, J. (1988) Economics and biological diversity: Developing and using economic incentives to conserve biological resources Gland, Switzerland: IUCN. McShane, T., & Wells, M., eds. (2004) Getting biodiversity projects work: towards more effective conservation and development Columbia University Press, New York. 153

PAGE 154

Mehta, J. & Kellert, S. (1998) Local atti tudes toward community-based conservation policy and programmes in Nepal: A case st udy in the Makalu-Barun Conservation Area. Environmental Conservation 25(4): 320-333. Mller-Bker, U. & Kollmair, M. (2000) Livelihood strategies and local perceptions of a new nature conservation project in nepal: The Kanchenjunga Conservation Area Project. Mountain Research and Development 20(4): 324-331. Murphee, M. (1994) The role of institutions in community-based conservation. In Western, D., Wright R. & Strum, S. eds Natural connections pp. 403-427. Columbia University Press, New York, USA. Naughton-Treves, L., Holland, M. & Brandon, K. (2005) The role of protected areas in conserving biodiversity and su staining local livelihoods. Annual Review of Environment and Resources 30(2005): 219-252. Neuman, R. (1997) Primitive i deas: Protected area buffer z ones and the politics of land in Africa. Developmental Change 28: 579-582. Neupane, H. (2003) Contested impact of community forestry on equity: Some evidences from Nepal. Journal of Forest and Livelihood 2(2): 43-51. Nyaupane, G. & Thapa, B. (2004) Evaluation of ecotourism: A comparative assessment in the Annapurna Conservation Area Project, Nepal. Journal of Ecotourism 3(1): 20-45. Nyaupane, G. & Thapa, B. (2006) Perceptions of environmental impacts of tourism: A case study at ACAP, Nepal. International Journal of Sustainable Development and World Ecology (13): 51-61. Nyaupane, G., Morais, D. & Dowler, L. (2006) The role of community involvement and number/type of visitors ecotourism im pacts: A controlled comparison of Annapurna, Nepal and Northw estern Yunnan, China. Tourism Management 27(6): 1373-1385. Oates, J. (1999) Myth and reality in rain forests: How conservation strategies are failing in West Africa Berkeley: University of California Press. Rew, A. & Rew, M. (2003) Development model s out of place: Social research on methods to improve livelih oods in eastern India. Community Development Journal 38(3): 213-224. Ribot. J. (1995) From Exclusion to Participation: Turning Senegals Forestry Policy Around? Harvard University, Cambridge, MA, USA. Ribot, J. (2002) Democratic decentralizati on of natural resources: Institutionalizing popular participation. Washington DC: World Resources Institute. 154

PAGE 155

Ribot, J. (2003) Democratic decentralization of natural resources: Institutional choice and discretionary power transfers in Sub-Saharan Africa. Public Administration and Development 23: 53-65. Robinson, J. (1993) The limits to caring: Sustainable living and the loss of biodiversity. Conservation Biology 7(1): 20-28. Salafsky, N. & Wollenberg, E. (2000) Linking liveli hoods and conservation: A conceptual framework and scale for asse ssing the integration of human needs and biodiversity. World Development 28(8): 1421-1438. Scoones, I. (1998) Sustainable rural livelihoods : A framework for analysis Brighton, UK: Institute of Development St udies, University of Sussex. Sherpa, M., Coburn, B. & Gurung, C. (1986) Annapurna Conservation Area, Nepal: Operation plan King Mahendra Trust for Nature Conservation, Kathmandu, Nepal. Smith, R., Mulr, R., Walpol e, M., Balmford, A. & Leader-W illiams, N. (2003) Governance and the loss of biodiversity, Nature 426: 67-70. Stem, C., Lassoie, J., Lee, J., Deshler, D. & Schelhas, J. (2003) Community participation in ecotouris m benefits: The link to conservation practices and perspectives. Society and Natural Resources 16: 387-413. Tao, T. & Wall G. (2009) Tourism as a sustainable livelihood strategy. Tourism Management 30(1): 90-98. Terborgh, J., van Shaik, C., Da venport, L. & Rao, M., eds (2002) Making parks work: Strategies for preserving tropical nature Washington DC: Island Press. Thomas-Slayter, B. & Bhatt, N. (1994) Land, livestock and livelihoods: Changing dynamics of gender, caste, and ethni city in a Nepalese village. Human Ecology 22: 467-494. UNDP 2008. Human Development Reports. Http://hdr.undp.org/en/repor ts/nationalreports/asiathepacific/nepal/name,3287,en.h tml. Vedeld, P., Angelsen, A., Sjaastad E. & Be rg G. (2004) Countin g on the environment forest incomes and the rural poor. Environmental Economics Series, World Bank Paper No. 98. WCED. (1987) Our Common Future: World Co mmission on Environment and Development Oxford University Press, Oxford. Wells, M., Brandon, K. & Hannah, L. (1992) People and parks: Li nking protected area management with local communities The World Bank, Washington, D.C. 155

PAGE 156

Wells, M., McShane, T., Dublin, H., OConnor, S. & Redford, K. (2004) The future of integrated conservation and development pr ojects: Building on what works. In McShane, T. & Wells, M., eds. Getting Biodiversity Projects to Work pp. 398-421. Columbia University Press, New York, USA. Wells, M. & Sharma, U. (1998) Socio-economi c and political aspects of biodiversity conservation in Nepal. International Journal of Social Economics 25(2-4): 226243. Western, D. & Wright, R. (1994) Natural connections: Perspectives in community based conservation Washington DC: Island Press. White, H. (2002) Combining quantitative and qualit ative approaches in poverty analysis. World Development 30(3): 311-322. World Bank (2008) Migration and Remittances Factbook. http://econ.worldbank.org/WBSITE/ EXTERNAL/EXTDEC/EXTDECPROSPECTS/0 ,,contentMDK:21352016~pagePK:64165401~piPK:64165026~theSit ePK:476883,0 0.html _________ (2009) Remittance market lookout. http://web.worldbank.org/WBSITE/EXT ERNAL/TOPICS/EXTFINANCIALSECTOR/ EXTPAYMENTREMMITTANCE/0,,cont entMDK:22121552~menuPK:2554287~pag ePK:210058~piPK:210062~theSit ePK:1943138~isCURL:Y,00.html WWF (2006) Understanding the c hanges in livelihoods assets with locals: A case study from Kanchenjunga Conservati on Area Project, Nepal. Sacred Himalayan Landscape in Nepal Series Number 3. ____ (2007) Biodiversity Governance. Sustainable Livelihoods Sacred Himalayan Landscape in Nepal Series No. 2. www.wwfint.org ____ (2007) Kanchenjunga Conservation Ar ea Project. 17-37. www.wwfint.org ____ (2007) Kanchenjunga Factsheet. www.wwfint.org ____ (2007) Nature, culture & conservation: Sacred Himalayan Landscape in Nepal. Thematic research working brief No. 1. www.wwfint.org 156

PAGE 157

157 BIOGRAPHICAL SKETCH Pete Parker is a sustainable development strategist with a background in community-based natural resources pr oject development and management. His emphasis has been working with protected areas in developing countries utilizing sustainable tourism as an economic alternative to natural resource consumption. He grew up in rural Idaho and received his bachelors degree in environmental science at the University of Idaho. He then joined the US Peace Corps in Nepal working as an environmental awareness specialist within the Department of National Parks and Wildlife Conservation. It was in Nepal, working alongside local communities, that he found his career calling. He has conducted re search within rural communities in the Himalayan regions of Nepal (Ph.D.) and Chi na (M.Sc. in environmental studies at San Jose State University). His research focuses on enhancing sustainable livelihoods through the equitable distribut ion of costs and benefits associated with natural resources conservation. He began his studies at the University of Florida in August 2006 and received his Ph.D. in the fall of 2009.