Demand and Supply of Forest Products in Bandevi Buffer Zone Community Forest, Chitwan National Park, Nepal

Buffer Zone Community Forestry (BZCF) in the Chitwan National Park (CNP) started with an objective to engage locals and fulfill their resource needs without jeopardizing conservation. This study estimates the forest product demand and supply of fuelwood, fodder, and timber in Bandevi BZCF user group. Data was collected using stratified random sampling and forest inventory using the quadrate method. Results indicated Bandevi BZCF conditions improved since its handover to the community. However, the study estimated a deficit of 26173 cubic meters per year of timber, 3.21 million tonnes per year (Mt/yr) of fodder, and 0.12 Mt/yr of fuelwood. Deficits were fulfilled from agricultural lands outside BZCF and illegal collection from the Bharandavar corridor forest and CNP. BZCF program is a success in improving forest conditions and needs continuation. Policy and programs must focus primarily on livelihood improvement and income generations to reduce the dependency of local people on the forest.

species-level conservation was of primary concern with top-down conservation approach. The model with the exclusion of people in the early seventies was heavily criticized for restricting local level usury rights and debarring local people (Heinen & Shrestha, 2006). Nepal and Weber (1995) have identified major issues of park-people conflicts prevailing in the park including illegal transactions of forest products, livestock grazing, illegal hunting and fishing, crop damage, and threats to human from wild animals. Likewise, Shrestha (1994), Nepal and Weber (1995), Heinen and Mehta (2000), and Budhathoki (2003) studied about the resource conflict between park conservation and adjoining settlements and found a serious threat to the survival of endangered animals and plants because of poaching and illegal use of park resources. To achieve socially favorable and ecological sound conservation, it is imperative to address local livelihood needs (Gurung, 2005). Fines and fences measures faded with increased park-people conflict and established the notion that local peoples' needs, and aspirations hold priority for the better conservation of biodiversity. Natural resources professionals had recognized the need to work beyond PAs if they are to sustain viable populations of wildlife species and large-scale ecological process (Heinen & Mehta, 2000). This irrefutable conservation thought led to amendment in National Parks and Wildlife Conservation Act of Nepal in 1993 that had facilitated the legal foundation for biodiversity conservation to establish and manage the buffer zone areas outside the PAs (GoN, 1973;Heinen & Kattel, 1992). Since then, the buffer zone has been institutionalized as an operational approach and large-scale ecological process to ensure the ecological integrity of PAs and enabling of local communities to sustain their livelihood through active management of natural resources outside the park (Heinen & Mehta 2000;Wells & McShane 2004;Straede & Treue, 2006).
The buffer zone development program is designed to address the local communities' needs through active participation in Buffer Zone Community Forest (BZCF) and other community development activities. In 1996, 750 square kilometers (Km 2 ) of adjacent areas in Chitwan National Park (CNP) was declared as a buffer zone which encompasses 35 Village Development Committees (VDCs) and two municipalities that have 510 settlements (DNPWC, 2000). The buffer zone area comprises a mosaic of forests, agricultural lands, settlements, cultural heritage areas, village open spaces and many other types of land use (Budhathoki, 2005). CNP had released the budget of approximately Nepali Rupees 2007). Straede and Helles (2000) have also raised a question over the capability of the BZCF to supply resources. However, the strategy of a buffer zone in the PA is ambitious, and many anticipated it to resolve the much-contested linkages diminishing societal support for PAs (Sanderson & Redford, 2003).
Several studies have been conducted on BZCF in Nepal with a discussion on forest management and resource allocation. On a brighter side, BZCF has helped improve livelihood through plantations and agroforestry that has created additional natural resource base for firewood and fodder and generated household income (Maskey & Bajimaya, 2005). Such community-based forest management has a potential of income-equalizing among the rural households (Mamo et al., 2007). Strade and Trene (2006) demonstrated the economic importance of forest product of CNP to the livelihood of people in Bacchauli VDC and revealed the pressure correlated with the economic value of the product. Poudyal (2007) detailed the socio-economic characteristics of CNP buffer zone residents and concluded that local residents heavily depend on national park and surrounding forests to meet their forest product demand. Mulepati (2009) concluded that about 62% claimed decreasing availability of forest products and 68% complained that there Baghauda BZCF is not fulfilling their demand. Dangol (2009) reported that only about 17% of the fuelwood and 8.4% of the fodder could be sustainably supplied from Bacchauli BZCF. Also, in Piple and Bandevi BZCF, as estimates of annual forest yield and household demand for forest products do not fulfill the needs, deficits are met through park resources and other nearby community forests outside the buffer zone (Poudyal, 2007;Pokharel, 2009). Straede et al. (2002) concluded that non-timber forest products, fuelwood, and timbers are not always compatible with what the locals want, and the pressure of resource theft is still on the PAs. People near to national park sneak inside to meet their requirements illegally thereby compromising conservation. Nyaupane and Poudel (2011) also reported that people of the BZCF user group exerted pressure on resources in the national forest and national parks. Although BZCF has been a better conservation model than restricting local level usury rights and debarring local people, the reality had been festering with little success for biodiversity conservation as well as adequate forest resources supply (Nepal & Spiteri, 2011). Poudyal (2007) concluded that all buffer zone households, irrespective of their land holding size, need forest products for fodder and fuelwood, and have not yet been able to sustain demand despite the establishment of BZCF and restoration of adjoining areas of CNP to present forest size. Although everyone wants fuelwood and fodder, the quaintly demanded mostly depends on household economic conditions. In general, socioeconomic factors such as land and livestock holdings, caste, education, household income, societal status exerts a strong influence on the benefits from common resources (Adhikari et al., 2004;Pokharel et al., 2017Pokharel et al., , 2018. Large agricultural households demand larger quantities of grasses and leaf litter for subsistence uses since they hold more livestock. However, due to the lack of land and/or livestock, the poor use lesser quantities of these products (Adhikari et al., 2004). Therefore, there is need of investigating the success of BZCF in Nepal not only in terms of improved forest conditions, but also in terms of socioeconomic condition, the livelihood of the surrounding  (Siwakoti & Karki, 2010). The reservoirs provide suitable habitat for migratory and local birds, and therefore, has high avian diversity. More than 250 different bird species have been reported in these forests of the buffer zone (Baral & Upadhyay, 1998;BES, 2007

Local Demand Estimation
The in-person household survey was conducted in September 2007 to collect information on demand of three primary forest resources-fodder, fuelwood, and timber and socioeconomic attributes. Stratified random sampling was applied to the survey by the settlement size with two parameters: a) land holding size, and b) settlement size by population. Data on land holding and settlement size of User Group was obtained from the office of Bandevi BZCF user group (BMC, 2006). The survey sample size (n) was determined using Equation 1 (Arkin & Colton, 1963;Poudyal, 2007).
n=[N Z 2 P (1-P)]/[N e 2 +Z 2 P (1-P)] where N is a total number of households, Z is the standard critical value of normal distribution at 95% confidence level (Z=1.96), and P is the estimated population proportion (0.05, this maximizes the sample size), and e is the error limit (±0.05). This calculated sample size was then stratified by settlement size and land holding by drawing random lottery without replacement. A random sampling of 71 households stratified across seven settlement sizes (villages) and four land holding sizes were conducted as specified in Table 1. Before conducting the formal questionnaire survey, focus group discussion and pre-testing of the questionnaire was conducted in some households. Modifications were made to make the questionnaire easier to understand for the respondent and surveyor and eliminate misunderstanding and misinformation. Although the data is a decade old, we believe that it provides relevant information on demand and supply gap of the BZCF.  The demand for green fodder (D fd ) and fuelwood (D fw ) per household was calculated as specified in Equations 2 and 3. The survey questionnaire asked respondents to report their demand for fodder and fuelwood in addition to their farm supply in local units per month. The demand for the timber (D t ) was collected from the Bandevi BZCF management office based on the application for timber supply (BMC 2006).
where Q fw and Q fd are the quantities of fuelwood and fodder demanded in local units per month, τ fw and τ fd are the conversion factors to convert local units to International System of Units (abbreviated as SI, from the French Système International) ( Table 2). The total fodder and fuelwood demand for Bandevi BZCF were obtained by multiplying fodder and fuelwood demand per household by total household in the User Group, respectively.

Annual Yield and Sustainable Supply of Forest Products
The sustainable yield and supply of fodder, fuelwood, and timber from the Bandevi BZCF were estimated using forest inventory method developed by the Forest Survey and Statistics Division, Ministry of Forest and Soil Conservation, Nepal (Sharma & Pukkla, 1990), now restructured as Forest Research and Training Center, Ministry of Forest and Environment, Nepal. The forest inventory was carried out between September and November of 2007. The Bandevi BZCF map was divided into 30 minutes by 30 minutes (longitude and latitude) grids, and 25 random locations were selected. At each location, a quadrate plot of 20 m by 20 m (400 m 2 ) was laid to measure trees. Within the tree plot, in two opposite corners (southeast and northwest) nested sub-plots of 5m by 5m (25 m 2 ) and 1m by 1m (1 m 2 ) were laid to study shrub and herb species, respectively. Diameter at breast height (dbh) was measured with the help of D-tape for different stand sizes as specified in Table 3. Height (h) of the tree was calculated using simple trigonometry as specified in Equation 4 using the angle (θ) measured with clinometers 6 m away from the base of the tree. Thus, the calculated h in meters and dbh measured in centimeters were used to calculate the volume and biomass of the tree. The standing volume of the tree (V), was calculated using Equation 5 (Sharma & Pukkla, 1990). The volume measured using Equation 5 is corrected by dividing V by 1000 such that the reported volumes are in cubic meters. The total dry weight of the stem is estimated using Equation 6, and the total weight in kilograms of the tree is estimated using equation 7.
h=6 * tan(θ)+1.6 (4) W=W s +x * W s +y * W s where ln(.) refers to the logarithm, α, β, and γ are the volume parameters, ρ is the wood density that is constant for each species but differs between species (Table 4), x and y are the branches to stem and foliage ratios for tree species. x=0.300 and y=0.0620 were used to calculate branch and foliage biomass since almost all trees had dbh <28 cm (Sharma & Pukkla, 1990). We used the same ratio for all tree species due to lack of data and comparable stem and foliage compositions. The volume parameters and wood density were obtained from the Forest Survey and Statistical Division, now known as Forest Research and Training Center (Sharma & Pukkla, 1990) are reported in Table 4. Source: FSSD (1989), Sharma and Pukkla (1990).
The annual yield of the Terai mixed hardwood forest was used for the annual yield of tree species. The annual forest product yield of annual stem biomass (Y s ), annual branch biomass (Y b ), and annual leaf biomass (Y l ) were estimated using ratios of branch and foliage with total biomass (Table 5) as specified in Equations 8, and 9, and 10 respectively.
where a, b, and c are the annual yield parameters (Table 5) (FSSD, 1989). The TND yields for hardwood forest with grazing is 34% of the annual foliage biomass. Sustainable timber supply (S s ), Sustainable fuelwood supply (S fw ), and sustainable fodder supply (S fd ) per hectare of forest were calculated as specified in Equations 11, 12, and 13. The total annual yield of the forest was calculated by multiplying annual yields per hectare with the total area of the BZCF area.
Also, ecological indices were calculated for all tree, shrub, and herb species. These indices provide insight into forest types, habitat types, and the value of forest to the local community (Odum et al., 1971).
Density of a species (Di) (N i /A) * 1000 (14) Relative Density (%) (RD i )=(D i /∑D i ) * 100 (15) Frequency (f i )=(q i /Q) * 100 (16) Relative Frequency (%) (RF i )=(f i /∑f i ) * 100 (17) Basal Area (BA i )=(π * dbh 2 )/4 (18) where N i is the total number of i species recorded in the study area, A is the total area of the study plots, D i is the density of i th species in number per hectare, ∑D i is the sum of all density of all species, f i is the frequency of i th species, q i is the number of quadrate with species i, Q is the total number of quadrats, ∑f i is the sum of frequencies of all species, dbh is the diameter at breast height, π is a constant, BA i is the basal area of species i, ∑BA i is the total basal area of all species in the study area, and IVI i is the important value index of species i.

Socioeconomic Status of the User Group
The socioeconomic status of the User Group surveyed is presented in Table 6. The sex composition of the survey was 38% female and 62% male. Approximately, 79% of the population was working population (Figure 2). This comprises of 52% male and 27% female working population. About 21% of the population was dependent on others. The female dependent population was slightly higher than the male.   handful of livestock (4 or less per household) and very little access to energy and forest resources.
Janajati also shares a very small portion of land (7%; 3% of them were landless), energy and forest products. For BZCF management, Brahmins/Chettris were more active than the other ethnic groups.
The representation of the Dalit was very poor. Regarding education, 38% of the respondents were illiterate, 31% can read and write but lack formal education, 2% had completed primary school, 16% had completed secondary level, and 14% had a college degree. Highest literacy was observed in Brahmin-Chettri. Tharus, Dalits, and Janajati rarely had a college degree. About 37% of the respondents worked in agriculture, 14% in service and 24% in the business sector. Approximately, 7% of the respondents worked as wage laborers. Rest of the respondents were involved in politics, housework, and/or other occupations (

Demand and Supply of the Forest Products
The Largest Livestock Unit (LSU) per household was observed among medium landholdings followed by landless. One LSU is equal to two goats or one and a half cows or one buffalo (Poudyal, 2000;Poudyal, 2007;Mulepati, 2009). Households with large land holdings had lesser livestock units.
Therefore, the demand for resources was higher in a household with medium land holdings ( Table 7).
The demand for timber, fodder, and fuelwood in one BZCF User Group household was 14 m 3 /yr, 1725 t/ha/yr, 64 t/ha/yr, respectively (Table 7). The total demand was 3.23 Mt/yr for fodder, 0.12 Mt/yr for fuelwood and 26,208 m 3 /yr for timber in Bandevi BZCF user group (Table 9). indicating that it is the most valuable and dominant species (Table 10). Adina cordifolia (Karma) and Dalbergia sisso (Sisso) were reported with about 2% relative frequency in the Bandevi BZCF. Both were planted species with few numbers found at the edge of the forest. There was a deficit of 26173 m 3 /year of timber, 3212105 t/year of fodder, and 119697 t/year of fuelwood in the Bandevi BZCF User Group (Table 9). About 23% of the respondents reported that they did not have any deficiency in forest products. About 48% of the respondents said they buy their deficit from other sources like sawmills, local markets and even from the landless peoples who illegally get the resources from the Bharandavar Corridor forest, BZCF, and/or the national park. Approximately, 27% of the respondents said that their deficit is fulfilled from their farms (Figure 4). About 1% of the people fulfill their deficit from their farm as well as buy from the local market (others farm), and 1% of the respondents do not use any forest products. Thus, the deficiency and its management are very important in the study area as 48% of people buy or borrow it, which may encourage sellers to get it from protected forests illegally. The usual sellers are landless who get the resources illegally from corridor forest or the BZCF.

Vegetation Analysis
The total number, species, density, relative density, the frequency of occurrence, and relative frequency of all reported species are also reported in Table 10. This study identified 65 plant species in the Bandevi BZCF. Imperata cylindrica was the densest (21,620/ha) and most frequent (30.8/ha) herb species. In Shrubs, Eupatorium adenophorum had the highest density (1,481/ha) and frequency of occurrence (19.46/ha). Shorea robusta (Sal, RD=41%, RF=62%) and Terminalia alata (Saj, RD=51, RF=32%) were the two main tree species found in Bandevi BZCF. Bandevi BZCF is a tropical forest with the very high percentage of saplings (83%) in the sampled plot indicating regenerating forest ( Figure 3). Only 6.15% of the stand was large saw timber.   (Dhakal & Yadava, 2011;Aryal et al., 2012). Bandevi BZCF also borders Bharatpur municipality, one of the fastest growing urban centers of Nepal (Dhakal & Yadava, 2011). Therefore, Bandevi BZCF User Group is a mixed community depending every day on resources of the BZCF and people who earn their living through other activities in the city and do not depend on the forest for livelihood. There is a growing pressure of urbanization and land use change on the forest. With development and urbanization, the majority of the population becomes indifferent towards resources of adjoining BZCF, but the marginal people who cannot meet the challenge of urban sprawl are becoming more and more dependent on the BZCF for their livelihood (Nepal & Spiteri, 2011). Bandevi BZCF served as a buffer between the wildlife corridor forest and the surging urban community delivering powers and authority to the locals in the User Group to make decisions and protect the forest against encroachment and illegal activity. On the other hand, the marginally poor people out of the user group are deprived of resource allocation and opportunity and has been reported to be involved in illegal activities in the BZCF and corridor forest. This has also given rise to conflicts in the community. This challenge has made Bandevi BZCF a unique and interesting natural resource conservation practice with an urban-wildland interface. The CNP, District Forest Office and BZCF User Group have to work together to maintain park-people relation and allocation of resources while keeping off illegal activities such as poaching, theft, and land encroachment.

Figure 3. Stand Size of Trees in Bandevi BZCF of Nepal
Bandevi BZCF User Group comprises populations from various ethnic groups and social status. The Brahmin/Chettri's were a dominant ethnic group with larger land holdings and higher levels of education in BZCF User Group followed by Dalits, Tharus, and Janajati. Tharus, Dalits, and Janajati depend largely on community forest for their fodder and fuelwood demand. About 38% of the population were illiterate, mostly from marginalized groups. Such disparity indicated there were still a lot of people who depended on manual labor for a livelihood. They were more likely to depend on forest resources for livelihood and least aware of sustainability and conservation issues (Nepal & Spiteri, 2011). A major source of income for the User Group was agriculture, wage labor, and small businesses. Livelihood improvement and active participation of backward ethnic groups must be the focal point of people park relations and effective buffer zone management strategy. We don't recommend sidelining the major ethnic group. However, we strongly suggest higher participation of the marginal ethnic groups and landless people for the better success of buffer zone since they are the most dependent people on BZCF. The park system should focus its extension programs on conservation issues and conduct training(s) on income generating programs to reduce the pressure on community forest. These programs should also include people outside the user group to reduce the illegal poaching, encroachment, and forest product theft.
The results based on the household landholding, livestock unit per household and green fodder and fuelwood supply options indicated the demand for green fodder and fuelwood had not been supplied sufficiently from Bandevi BZCF. The highest demand was observed among small and medium landholders since they have limited land to supply fodder and fuelwood from their farms. A household http://www.scholink.org/ojs/index.php/se Sustainability in Environment Vol. 3, No. 4, 2018 with a medium farm required more green fodder as they had large number of livestock compared to others. However, the fuelwood demand was the highest amongst the households with small landholdings. They were more dependent on fuelwood from the BZCF as they have less land to substitute resources from their farms. Large landowners were less dependent on community forest for their fodder and fuelwood demand. Households with larger land holding were limited and had the least number of livestock, therefore, supplied their fodder demand from their lands and were less dependent on BZCF. Households with larger landholdings were richer families who could afford other forms of energy and opt out of the tedious process to collect fuelwood from the forest. Households with medium landholding had more livestock and were more likely to get biogas, which is incentivized through the government. Without the land and few LSU, the landless and households with small land holdings are the most dependent on BZCF for resources.
There was a very high demand for timber but very limited supply among all socioeconomic groups of Bandevi BZCF User Group. Sal (Shorea robusta) is one of the best timbers used for housing purposes in Nepal and are expensive to buy from the market. Straede and Treue (2006) demonstrated the economic importance of forest products of CNP to the livelihood of people and also revealed the pressure correlated with the economic value of the product. Due to high-value Sal timber, all the members must have asked for higher volumes of Sal timber (higher allowed) although they did not need it right away. This might have made deficiency seem enormous. Another reason could be that the available Sal timber is allocated based on first come first serve basis and put in a queue for next year if demand is not met in the current year. Therefore, people might have asked for timber even if they do not need it now but might need it in future. Also, the area is adjacent Bharatpur municipality and therefore, the demand is much higher compared to other BZCF User Group due to ongoing urbanization.
Although there is a deficit in available resource and the demand of the local community, the quality of Bandevi BZCF has improved since its handover to the local people in 2006. The very high number and density of saplings suggests regenerating forest. It might increase the available forest resources and lessen the deficit in recent future. However, the impact of deficit might have simply spilled over to the Bharandabhar Corridor Forest and the National Park. An earlier study reported that 37.1 % of fuelwood and 55.5 % of fodder were collected from National Park (DNPWC, 2000). This study reported that half of the deficit was being bought from others, especially landless and poor locals, who were believed to obtain it illegally out of BZCF and Corridor forest. The pressing issue was essential for livelihood in the poorer household who dwell near the BZCF. The culture dichotomy fueled by the local community's urgency to illegally use forests in BZCF, Corridor forest, and the national park for cattle grazing, thatch and fodder grass cutting, firewood collection, timber cutting, hunting and fishing are the frontline issues to challenge the protected area management through buffer zone management programs (Straede & Treue, 2006;Poudyal, 2007;Nyaupane & Poudel, 2011;Thapa, 2013;Stone & Nyaupane, 2016;Bhattarai et al., 2017). Many recommendations have been put forward to solve the forest product deficiency and control illegal activities to meet the societal demand. The extension of the Bandevi BZCF beyond 300m from the boundary into the corridor forest was the major priority felt by the locals. This might help the User Group to increase their available resources as well as protect the corridor forest. Paudyal (2008) argued that economic and political structures and social institutions set the context for individual and group behavior. Therefore, park and other conservation authorities must focus on behavior and livelihood uplifting program to reduce the pressure on natural resources. The buffer zone approach has been successful in improving the forest conditions with local involvement. Therefore, it must be continued with other programs to reduce the dependency of local people in forests. For example, promoting agroforestry, farm-based high-quality fodder production, and other animal feedstock to lessen the demand for the fodder can help reduce fodder demand on BZCF. Alternative energy promotions, incentives for biogas, and solar energy installment can reduce the demand for fuel woods. Similarly, the deficit for timber can be mitigated with alternative building materials such as concrete, iron, etc.
Therefore, newer policy and programs must focus primarily on livelihood improvement and income generations to reduce the dependency of local people on the forest.

Conclusions
All buffer zone households irrespective of their land holding size need forest produce for fodder and fuelwood. The Bandevi BZCF conditions have improved since its hand over to the local community.
However, the BZCF has not been able to sustainably supply the demanded amount of fodder, fuelwood, and timber to the local community. The deficit is met by the products of private farms and/or buying from others, sometimes illegally obtained from the BZCF, Bharandabhar Corridor Forest, and/or CNP.
This has jeopardized the conservation efforts of the protected areas. The extension of the forest was the major priority felt by the locals to increase their available resources as well as protect the corridor forest. However, improving the livelihood of local people through income-generating activities, promoting alternative energy, and conservation education might be a better strategy in conservation issues of the buffer zone.