Applied Geography 69 (2016) 25e34

Contents lists available at ScienceDirect

Applied Geography
journal homepage: www.elsevier.com/locate/apgeog

Assessments of ecosystem service indicators and stakeholder's

willingness to pay for selected ecosystem services in the Chure region
of Nepal
Pratima Bhandari a, Mohan KC b, Sujata Shrestha c, Achyut Aryal d,
Uttam Babu Shrestha e, *
a
Institute of Forestry, Pokhara Campus, Tribhuvan University, Nepal
b
REDD Implementation Center, Ministry of Forest and Soil Conservation, Nepal
c
Department of Biology, University of Massachusetts Boston, USA
d
Institute of Natural and Mathematical Sciences, Massey University, Auckland, New Zealand
e
Institute for Agriculture and the Environment, University of Southern Queensland, Toowoomba, Australia

a r t i c l e i n f o a b s t r a c t

Article history: The Chure region of Nepaldthe area sandwiched between the hills in the north and the plains in the
Received 9 August 2015 southdis considered an ecologically fragile, structurally weak, and highly erosion-prone region. The
Received in revised form forest in the Chure region provides several ecosystem services to people living in the downstream areas.
3 February 2016
However, assessment and quantification of ecosystem services in this region are very limited. This study,
Accepted 6 February 2016
conducted in a watershed of the Chure region of western Nepal, combined local users' perspectives with
Available online 1 March 2016
experts' opinions to identify and rank ecosystem services based on land use types, to investigate the
downstream users' willingness to pay for ecosystem services, and to explore the socio-economic factors
Keywords:
Ecosystem services
affecting their willingness to pay. The study found that forests offered the highest number of ecosystem
Nepal goods and services in this area. Local people were familiar with 10 different ecosystem services provided
Himalaya by the watershed and ranked drinking water service at the top. The downstream beneficiaries would be
Payment for ecosystem service willing to pay a higher amount for drinking water service than they were currently paying if the quality
Watershed of the service and its sustainability were assured. The amount they were willing to pay for ecosystem
services increased significantly with monthly income. The results of this study are useful for other areas
in which an upstreamedownstream linkage exists and the upstream communities play a crucial role in
maintaining ecosystem functions and the resulting supply of ecosystem services to the downstream
communities.
© 2016 Elsevier Ltd. All rights reserved.

1. Introduction (Nelson et al., 2009; Turner & Daily, 2008). This concept has been
used as a policy instrument in biodiversity conservation and nat-
Ecosystem services, defined as the benefits that humans obtain ural resource management (Burkhard, Kroll, Müller, & Windhorst,
from natural ecosystems (Daily, 1997), have received significant 2009; Fisher & Turner, 2008; Koschke, Fuerst, Frank, &
attention in global environmental policies in recent years (CBD, Makeschin, 2012; Seppelt, Dormann, Eppink, Lautenbach, &
2010; MEA, 2005; Perrings, Duraiappah, Larigauderie, & Mooney, Schmidt, 2011). Therefore, the identification and valuation of
2011; TEEB, 2010). The concept of an ecosystem service highlights ecosystem services supports decision making and policy aimed at
the role of natural ecosystems in providing goods and services for biodiversity conservation and sustainable development (Burkhard
human well-being, economic development, and poverty alleviation et al., 2009; Fisher & Turner, 2008; Koschke et al., 2012; Seppelt
et al., 2011). However, the assessment of ecosystem services in
developing countries, where people are dependent on natural
* Corresponding author. ecosystems such as forests and rangelands for their livelihoods
E-mail addresses: [email protected] (P. Bhandari), mohankc. (Shrestha & Bawa, 2014) is limited due to the lack of appropriate
[email protected] (M. KC), [email protected] (S. Shrestha), savefauna@ data, methods, tools, and management framework (Paudyal, Baral,
gmail.com (A. Aryal), [email protected] (U.B. Shrestha).

http://dx.doi.org/10.1016/j.apgeog.2016.02.003
0143-6228/© 2016 Elsevier Ltd. All rights reserved.
26 P. Bhandari et al. / Applied Geography 69 (2016) 25e34

Burkhard, Bhandari, & Keenan, 2015). monitor, has straightforward financial benefits to payers, and is
Critical ecosystem services in developing countries are gener- valued greatly by local people (Van Hecken et al., 2012).
ated from forests, agroforests, farmlands, and rangelands that are The effective design of a PES mechanism requires not only
managed by low-income people (Milder, Scherr, & Bracer, 2010; knowledge about the relationships between ecological function
Molnar, Scherr, & Khare, 2004). About 400e500 million people in and the resulting ecosystem services but also a clear understanding
developing countries are dependent on forests for their livelihoods of the role of local communities and their insights about the ben-
(Charnley & Poe, 2007). Large areas of forest in developing coun- efits and tradeoffs of PES schemes (Kinzig et al., 2011; Simelton &
tries are managed or owned by local communities in various forms Dam, 2014). The involvement of local stakeholders in ecosystem
of community-based management practices such as community services assessment is crucial for selecting appropriate ecosystem
forestry (White & Martin, 2002). Since the implementation of the service indicators, evaluating possible management options
community forestry program in Nepal three decades ago (Shrestha, through ranking of different ecosystem services, and the validation
Shrestha, & Shrestha, 2010), community forestry has become a of different management options (Ananda & Herath, 2009; Seppelt
dominant forest management practice. Local communities et al., 2011). Further, better understanding of the behavioral and
currently manage about 1.7 million hectares of forest and about 35% governance dimensions of local communities is required before
of the population of Nepal is involved in the community forestry initiating a PES scheme as a policy option (Muradin et al., 2010).
management program (DoF, 2015). This is relevant in the context of Nepal, where PES schemes for the
Community-managed forests in Nepal have been instrumental hydrological services of watersheds and the carbon sequestration
in fulfilling the local communities' subsistence needs for the service of forests have been receiving attention in the policy sector
various ecosystem goods (timber, fodder, and fuel) and in providing as a means of promoting the sustainable management of water-
services (watershed protection, erosion protection, ground water sheds and forests (Khatri, Paudel, Bista, & Bhandari, 2013; Regmi
recharge, and water purification) to local as well as distant com- et al., 2009).
munities (Birch et al., 2014; Oort et al., 2014; Shrestha et al., 2010). This current study was conducted in the Northern watershed
Likewise, Nepal's community forests have significantly contributed (NW) of Chure region of Nepal, where the potential of PES schemes
to climate change mitigation through the sequestration of carbon for hydrological and other ecosystem services is large. The study
(Pandey, Cockfield, & Maraseni, 2016). Community forests also play aimed to combine local users' perspectives with experts' opinions
an important role in preventing forest degradation and restoring to identify and rank various ecosystem service provisions at
degraded forests in many watersheds of Nepal (Birch et al., 2014; watershed level, based on the land use types. It also investigated
Paudyal et al., 2015). Therefore, to incorporate all of the potential the willingness to pay (WTP) of the downstream beneficiaries for
benefits provided by community forests to local, regional, and ecosystem services such as drinking water provided by the water-
global communities, an ecosystem services framework might be shed as a result of protection by upstream conservators. Finally, the
useful because the concept of ecosystem services emphasizes the socio-economic factors affecting the WTP for ecosystem services
multiple benefits of ecosystems to human well-being (MEA, 2005). were explored.
Further, quantifying ecosystem services also helps to explore the
prospect of ‘payment for ecosystem service’ (PES) in community- 2. Materials and methods
managed forests.
The PES scheme is a market-based mechanism in which the 2.1. Study area
beneficiaries (buyers) of ecosystem services pay the communities
or land owners (sellers) for maintaining the ecosystem that pro- The study was conducted in the semi-urban areas of Birendra-
vides goods and services to the beneficiaries (Ferraro & Kiss, 2002; nagar municipality and the adjoining Gadi and Jarbuta village
Wunder, 2005). Incentives provided under the PES scheme for ‘best development committees (VDCs) of Surkhet district, mid-western
practices’ is thought to be more motivational and effective than Nepal, which covers a total area of 9427 ha (see Fig. 1). The re-
imposing ineffective sanctions for ‘bad practices’ (Van Hecken, gion comprises 12,029 households in Birendranagar municipality,
Bastiaensen, & Va
squez, 2012). However, the notion of PES is also 1,837 households in Jarbuta VDC, and 685 households in Gadi VDC,
subject to considerable criticism (Redford & Adams, 2009). For with a population of 47,914, 8,580, and 3,050 respectively (CBS,
example, in some cases the commodification of ecosystem func- 2014). Birendranagar municipality is the major trading center of
tions as ecosystem services to humanity may prove harmful to the mid-western region of Nepal and is undergoing rapid urbani-
biodiversity conservation (Peterson, Hall, Feldpausch-Parker, & zation. The Northern watershed, situated in the northern part of the
Peterson, 2010). Nevertheless, PES schemes are gaining mo- Birendranagar municipality, provides various ecosystem services
mentum, with several active programs of PES globally, although the (including drinking water) to about 60,000 residents living down-
outcomes of such projects are mixed (Kinzig et al., 2011; stream in the Birendranagar municipality and in Jarbuta, Latikoili,
Pattanayak, Wunder, & Ferraro, 2010; Schomers & Matzdorf, and Uttarganga VDCs. Due to the significant contribution of this
2013). PES schemes have supported water quality improvement watershed in delivering various ecosystem services, it was declared
in Ecuador (Quintero, Wunder, & Estrada, 2009) and forest cover a protected watershed in 1988 by the late King Birendra during his
enhancement in Costa Rica (Arriagada, Ferraro, Sills, Pattanayak, & visit in that region (the municipality is named after him). It has
Cordero-Sancho, 2012). On the other hand, a PES scheme in Mexico been protected through the active involvement of local govern-
promoted a short-term utilitarian view of conservation at the cost mental and non-governmental agencies, as well as community
of decreased intrinsic motivation (Rico García-Amado et al., 2013). forest user groups (CFUGs), which are local, community-based
PES schemes for carbon (Corbera, Soberanis, & Brown, 2009) and natural resource management groups formed under the commu-
biodiversity (Sanchez-Azofeifa, Pfaff, Robalino, & Boomhower, nity forestry program of Nepal and responsible for the protection,
2006) can be large in size and cover a huge geographical area, management, and distribution of forest resources. Community
whereas PES schemes for water and related hydrological services forestry, which was initiated about 30 years ago, is now the major
are generally smaller in size and focus on local watersheds (Lopa forest management program in Nepal. Various forest-related pol-
et al., 2012; Mun ~ oz-Pin
~ a, Guevara, Torres, & Bran ~ a, 2008; icies, such as the Master Plan for the Forestry Sector (1989), Forest
Simelton & Dam, 2014; Turpie, Marais, & Blignaut, 2008). Further, Act (1993), Forest Regulations (1995), and Forestry Sector Policy
a PES scheme for drinking water or hydrological services is easy to (2000), were prepared and implemented to support the
 P. Bhandari et al. / Applied Geography 69 (2016) 25e34 27

Fig. 1. Map of the study area.

community forestry program in Nepal. Eleven CFUGs, comprising a were conducted to collect data on the locals' perspectives on
total area of 2043 ha with 4606 households upstream, have been ecosystem services and to document the community's WTP for the
involved in protecting the NW (DoF, 2015). ecosystem services.
This study combined both quantitative and qualitative ap-
proaches and applied tools from social and spatial sciences for data
2.2. Land use/land cover (LULC) classification
collection. Satellite imagery was used to produce a land use/land
cover (LULC) map of the study area and expert knowledge was
A cloud-free satellite image of Landsat 8
utilized to identify the appropriate indicators of ecosystem services.
(Scene ¼ LC81430402014339LGN00), with spatial resolution of
Household surveys, focus group discussions, and field observations
30 m, was acquired on 5 December, 2014 from the United States
28 P. Bhandari et al. / Applied Geography 69 (2016) 25e34

Geological Survey's website (http://glovis.usgs.gov/). The satellite increasing confidence in the final results (Jacobs et al., 2015).
image was cropped for the study area and classified into six LULC Despite some limitations, the matrix model provides a foundation
classes (forestry, agriculture and agroforestry, water body, bush, for research in ecosystem service, especially in data-poor regions
grassland, and built-up areas), using the supervised classification such as Nepal, where ecosystem service is a new knowledge fron-
method in ArcGIS 10.2. Sixty-six polygons were created as training tier for the local communities and policy makers. To overcome the
areas, based on the different combinations of the Landsat bands model's limitations, the experts consulted for this research came
displayed in red, green, and blue (RGB). A signature file for a su- from diverse backgrounds and affiliations, thus minimizing
pervise classification was created, based on the training areas. To subjectivity, and a consensus rule, using the average (mode) of their
examine classification accuracy, 300 reference points were created scores, was used to produce the final matrix. Their opinions were
randomly across the study area. The classification of each random complemented by the perceptions of people in the local commu-
point was validated by using Google Earth. Values were calculated nities, collected though household surveys.
for overall accuracy and kappa value. The classification had an
overall accuracy of 94.3%, with a kappa value of 0.923. This level of 2.4. Household survey
accuracy was considered acceptable.
Household surveys were conducted among 238 (5.16% of CFUGs
2.3. Indicators selection for ecosystem services affiliated households) randomly chosen respondents from house-
holds affiliated with the 11 CFUGs mentioned earlier. These
The potential indicators of ecosystem services for the six households comprised 96 upstream land managers (from Gadi and
different LULC classes of the study area were identified through an Jarbuta VDCs) and 142 downstream beneficiaries (from Birendra-
opinion survey with experts who were familiar with the concept of nagar municipality). Interviews were conducted either on the farm
ecosystem services and field conditions. Twenty local experts (11 or at the respondent's residence during October and November
officers from the Ministry of Forest and Soil Conservation of Nepal, 2014, using structured and semi-structured questionnaires. The
three from local non-governmental organizations, four from the interviews began with an explanation of the objectives of this study
institute of Forestry, Tribhuvan University, and two from the Con- and some background information about ecosystem services.
servation Unit of the President Chure Conservation Program) were The interview questionnaire was in three parts. The first part
asked to score the capacity of the different land covers identified on focused on information related to the socio-economic status (age,
our map to provide various ecosystem services on a 0e5 relative gender, education, occupation, and income) of the households (see
scale (0 ¼ no capacity, 1 ¼ very low capacity, 2 ¼ low capacity, Table 1). Questions in the second part measured the respondents'
3 ¼ medium capacity, 4 ¼ high capacity, and 5 ¼ very high capacity) familiarity with the concept of ecosystem services and devised
(Burkhard et al., 2009, 2012). This expert-based assessment of rankings based on the importance to them of the various ecosystem
ecosystem service, called ‘the matrix model’, is one of the most services. The third part dealt with their WTP for the ecosystem
popular ecosystem services assessment techniques (Burkhard et al., services provided by the watersheddonly the 142 respondents
2012; Jacobs, Burkhard, Van Daele, Staes, & Schneiders, 2015). from the downstream communities were asked the questions in
Following the evaluation framework of the matrix model, a matrix this part.
of LULC types on the y-axis and ecosystem services on the x-axis To estimate the amount of money that downstream benefi-
were created, based on the experts' scores, to determine the ca- ciaries would be willing to pay for ecosystem services or watershed
pacity of the different LULC classes to provide various ecosystem services, the contingent valuation method (CVM) was used. This is
services. Their responses were grouped into four categories: pro- an interview technique widely used to estimate the monetary value
visioning, regulating, supportive, and cultural services, based on that people ascribe to certain environmental or public goods
the Millennium Ecosystem Assessment classification (MEA, 2005). (Lankia, Neuvonen, Pouta, & Sieva €nen, 2014; Mitchell & Carson,
Indicators of ecological integrity, which is defined as ‘preser- 2013).
vation against non-specific ecological risks that are general dis- To generate as much realistic information as possible and to
turbances of the self-organizing capacity of ecological systems’ verify the participants' responses, three types of questions were
(Barkmann, Baumann, Meyer, Müller, & Windhorst, 2001; Müller, used to collect data on WTP: ‘open-ended’, ‘bidding game’, and
2005), include abiotic heterogeneity, biodiversity, biotic water ‘payment card’ (Loomis, Brown, Lucero, & Peterson, 1996; William,
flows, metabolic efficiency, energy capture, reduction of nutrient Russell, Rodriguez, & Darling, 1999). Open-ended questions were
loss, and storage capacity. According to Burkhard et al. (2012), these used to enumerate the various types of ecosystem services ob-
indicators are useful in determining whether the condition of an tained from the NW and to document the beneficiaries' WTP for
ecosystem is improving or being degraded over time. In this study, specific ecosystem services (e.g., drinking water). With regard to
there was no response from the local experts regarding the the bidding game questions, they were first asked in the general
ecological integrity of the ecosystemdthey may have had limited questionnaire whether they knew the specific ecosystem services
understanding of those indicators. However, the objective of this (e.g., drinking water) they were getting from the NW and whether
research was to assess the capacity of various landscapes to provide or not the supply/quality of some ecosystem services had declined
ecosystem services based on the LULC data with expert estimates, (e.g., ‘Is there enough drinking water?’). If they responded that the
rather than to examine the temporal dynamics of ecosystem. water supply was insufficient, they were asked how they managed
The matrix model was first proposed by Burkhard et al. (2009) this problem.
and significantly modified and improved later (Burkhard et al., To understand the user's maximum WTP for sufficient and
2012; Stoll et al., 2015). Customized versions of this model have sustainable ecosystem services, a hypothetical situation was
been used in various geographic locations and different socio- created (e.g., ‘If you got sufficient drinking water, how much would
ecological and political contexts (Jacobs et al., 2015). The subjec- you be willing to pay for it?’). A list of the ecosystem services ob-
tive judgment of expertsdand a priori chosen sets of ecosystem tained from the watershed was developed and WTP values for
servicesdused in this method have often been criticized (Stoll those services were fixed, based on their current payment situation
et al., 2015). However, the inclusion of multiple opinions from a (e.g., monthly payment for drinking water). Finally, downstream
broad range of experts with different professional backgrounds, users were asked to choose a WTP point estimated from a list of
skills, affiliations, and experience could reduce uncertainties by values (payment cards of different values). The data collected from
 P. Bhandari et al. / Applied Geography 69 (2016) 25e34 29

Table 1
Socio-economic characteristics of the respondents.

Socio-economic characteristics of the households Total number (%) Upstream respondents (%) Downstream respondents (%)

Sex of the respondent

Male 98 (41.2) 47 (49.0) 51 (35.9)
Female 140 (58.8) 49 (51.0) 91 (64.1)
Age category (years)
18-36 90 (37.8) 29 (30.2) 61 (43.0)
37-51 99 (41.6) 32 (33.3) 67 (47.2)
52 and above 49 (20.6) 35 (36.5) 14 (9.9)
Education
Illiterate 85 (35.7) 49 (51.0) 36 (25.4)
Primary 61 (25.6) 29 (30.2) 32 (22.5)
Secondary 60 (25.2) 15 (15.6) 45 (31.7)
College and higher 32 (13.4) 3 (3.1) 29 (20.4)
Occupation
Farmer 180 (75.6) 88 (91.7) 92 (64.8)
Services 45 (18.9) 3 (3.1) 42 (29.6)
Seasonal labor 6 (2.5) 4 (4.2) 2 (1.4)
Others 7 (2.9) 1 (1.0) 6 (4.2)
Monthly reported income (NRs)
1001e5000 29 (12.2) 22 (22.9) 7 (4.9)
5001e10000 70 (29.4) 36 (37.5) 34 (23.9)
10,001e20,000 101 (42.4) 35 (36.5) 66 (46.5)
20,001e50,000 38 (16) 3 (3.1) 35 (24.6)

the household surveys were analyzed in Statistical Package for the environment as well as supporting the livelihoods of the local
Social Sciences (SPSS) 22 and R. people by providing wide range of provisioning services.
Agricultural and agroforestry areas had the second highest score
3. Results and discussions (61), and the highest overall score in its capacity to produce food
and horticultural crops, including vegetables. The agricultural area
3.1. Land use/land cover classification in the study was not purely an agricultural system; it also had some
remnant trees at the edge of a terrace, which provided fodder and
The major LULC types in the study area included forest, bush, played a role in erosion control. Thus, this category had plant
settlement, agriculture, grassland, and water body (see Fig. 2). species other than crops, as reflected in the responses of the
Forest was the most predominant LULC type, covering 3,406 ha experts.
(36.13% of the area). Bush was the second most dominant LULC Grasslands were open areas near the forests or agricultural lands
type, covering 2,427 ha (25.75% of the area), followed by settlement on which the local communities grazed their livestock. Grassland
(1,390 hae14.74%). Agriculture and agroforestry, grassland, and received only one high score, for livestock grazing. Overall, human-
water body were the least dominant LULC types, covering 1,092 ha dominated landscapes and settlements received a very low score
(11.58%), 940 ha (9.97%), and 172 ha (1.82%) respectively. In the except for cultural services such as religious shrines, temples, and
study area, forests were dominated by Shorea robusta, Pinus rox- playgrounds.
burghii, Syzygium cuminii, Lagerstroemia parviflora, and Bauhinia Birch et al. (2014) used a rapid assessment tool to study
variegate, and were managed by the local CFUGs. Bush areas were ecosystem services in the Phulchoki Mountain Forest of central
degraded forested areas with many shrubby species and scattered Nepal and also found that forests provided maximum ecosystem
trees, whereas grasslands were normally the open areas used for benefits, compared with other land uses. Similar results have been
livestock grazing. found in Germany (Kroll, Müller, Haase, & Fohrer, 2012) and
Bangladesh (Sohel, Mukul, & Burkhard, 2014).
3.2. Assessment of ecosystem services
3.3. Local perceptions of ecosystem services
The ecosystem services that are provided by an ecosystem
depend on the ecosystem structure and function, represented by The local communities' perceptions of ecosystem services,
the different LULC types (Burkhard et al., 2012). The matrix of the particularly the watershed services provided by the NW, were
different land covers' capacity to provide ecosystem goods and analyzed (see Fig. 4). The responses of the local people were placed
services, based on the experts' opinion, is shown in Fig. 3. The rows in the 10 different ecosystem services categories used in the MEA:
of the matrix had six LULC classes, and the columns had 26 different namely, forest resources, air quality regulation, landscape beauty,
ecosystem service indicators linked with LULC types. Forests, with a biodiversity protection, water purification, erosion/landslide con-
total score of 80, offered the highest number of ecosystem goods trol, local climate regulation, carbon sequestration and storage,
and services, as compared with other LULC types in the study area, pollination service to agriculture, and nutrient cycling. All of the
followed by agricultural and agroforestry areas. Of the 26 different respondents were familiar with the role of the watershed in
indicators of ecosystem services, forests received the highest score providing water and forest products, as those services and goods
(5) in six ecosystem services (two in provisioning, three in regu- were vital for their subsistence livelihoods. Every respondent
lating, and one in supportive). Those services were timber, local mentioned that they brought at least one forest product, such as
climate regulation, carbon sequestration, air quality regulation, timber, fuel wood, fodder, wild fruits, mushrooms, medicinal
erosion control, and biodiversity. Forests also received the second plants, poles, and leaf litter, from the NW. The diverse responses
highest score (4) in nine other ecosystem services. The experts' related to forest products were placed under a single categor-
scores suggested that forests played an important role in regulating ydforest resources. Three major products derived from forests
30 P. Bhandari et al. / Applied Geography 69 (2016) 25e34

Fig. 2. Land use/land cover types of the Northern watershed.

were fuel wood (mentioned by 85% of respondents), fodder (84%), respondent (out of 238) mentioned nutrient cycling as one of the
and timber (26%). roles of the NW. The majority of the respondents who expressed
After forest resources, the majority of respondents were familiar their familiarity with the role of the watershed in local climate
with air quality regulation (89%), landscape beauty (84%), and regulation and carbon sequestration and storage were the mem-
biodiversity protection (60%). Their responses, which included bers of CFUG's executive committee, who had had opportunities to
‘purify air’ and ‘cleanse ambient environment’, were placed under participate in awareness training and workshops run by govern-
‘air quality regulation’. Responses that were placed under ‘biodi- mental and non-governmental agencies.
versity protection’ included ‘habitat for wild plants and animals’, Although the local peoples' perceptions of ecosystem services
‘habitat for trees and grass’, and ‘habitat for birds’. Only one provided by the NW could be categorized under 10 different
 P. Bhandari et al. / Applied Geography 69 (2016) 25e34 31

Fig. 3. Matrix for the assessment of the various land use and land cover types' capacities to provide selected ecosystem goods and services identified by local experts.

Fig. 4. Local people's familiarity with various ecosystem services provided by the watershed.

ecosystem services, the perceived importance of those ecosystem services was varied. The respondents ranked the individual
32 P. Bhandari et al. / Applied Geography 69 (2016) 25e34

ecosystem goods and services that they were familiar with, starting downstream respondents. According to them, if upstream dwellers
with 1 for their first choice or highest rank and numbering the had not protected the watershed, the major impacts on down-
goods and services down to their last choice or lowest rank, ac- stream areas that could have occurred were natural disasters such
cording to their importance to them. Due to incomplete responses as landslides (31%), scarcity of drinking water (29%), increased
caused by the respondents' unfamiliarity with every ecosystem sedimentation on agricultural lands (22%), irrigation water scarcity
service category, only the frequency scores from 1 to 6 were (9%), negative effects on local tourism (6%), and reduced livelihood
counted. The respondents did not provide a score for nutrient options (3%). Due to the protection of forest in the NW, a greater
cycling, pollination service, carbon sequestration and storage, and number of birds and wildife was visible now. Locals perceived that
local climate regulation. touristic activities such as bird watching and wildlife viewing,
The average score of the perceived importance of the various which provides livelihood benefits, would not be possible if the
ecosystem services is shown in Fig. 5. Drinking water was the first forest had not been protected. At the same time, locals also
choice (highest score), with 62% of respondents ranking drinking expressed concerns with regard to the diminishing quality of the
water at 1. Water for agriculture was ranked the second highest, ecosystem services due to the degradation of forests outside the
followed by forest products, land erosion, biodiversity protection, community forests, although community forests had played a sig-
and landscape beauty. In other watersheds of Nepal that are nificant role in increasing the forest cover in the watershed.
managed by CFUGs, water has also been identified by local people The PES scheme is a new market-based mechanism that has
as the most important ecosystem service (Oort et al., 2014). Unlike produced successful outcomes and had considerable appeal with,
the ratings for drinking water, high variation was observed in re- and acceptance by, local communities elsewhere (Schomers &
spondents' ranking for ‘forest products’ and ‘biodiversity protec- Matzdorf, 2013). During the interview to collect their WTP for the
tion’. Forest products could be critical for the communities who ecosystem services derived from the watershed, the respondents
used fuel wood for cooking, collected fodder for livestock, and were asked about their familiarity with the concepts and termi-
gathered herbs for medicinal purpose, but less important for the nologies of PES given in the scholarly literature. Nearly all of the
communities who were less dependent on forest products for their respondents (99.6%) said they had not heard about PES, even
livelihoods. Often, watershed services with both market and non- though the locals were currently paying for drinking water. The
market values were more numerous than had been appreciated basic concepts of PES were explained, giving the example of the
by local communities (Dyson, Bergkamp, & Scanlon, 2003). In the monthly payment for drinking water, and they were asked if they
case of NW, water services had a market value for designing and were potentially interested in a PES program. Surprisingly, all 142
initiating a payment for an ecosystem services scheme. downstream interviewees said they were willing to pay for
ecosystem services if payment would ensure the quality and long-
term supply of ecosystem services. The interviewees also respon-
3.4. Downstream communities' willingness to pay for ecosystem ded the tentative amounts of willingness to pay for various
services ecosystem services.
The participants' responses on WTP for ecosystem services were
Three types of users of ecosystem services were seen in the grouped into four categories: WTP for drinking water; overall
study area: upstream land holders, who were conservators and watershed service (e.g., forest resource, air quality regulation, water
users of selected provisioning services from the NW (timber, fuel purification, and biodiversity protection); erosion/landslide con-
wood, and fodder, but not drinking water); buffer zone dwellers, trol; and landscape beauty (see Fig. 6). All of the respondents said
who were both conservators and service users; and downstream they were willing to pay for drinking water, whereas very few (only
beneficiaries, who acquired benefits only (drinking water, water for 12.7% of respondents) were willing to pay for landscape beauty.
irrigation, and soil erosion prevention) without making a signifi- Similarly, all of the respondents were willing to pay for erosion/
cant contribution to the conservation of the NW. landslide control service, followed by WTP for overall watershed
Due to the wide variety of ecosystem services provided by the service (98.6%). Likewise, the amount they were willing to pay for
watershed, its sustainability was a prime concern for the drinking water was high, with a median value of NRs 250/month
(NRs is Nepalese rupees; US $1~NRs 99.00) as compared with the
amount they were willing to pay for erosion/landslide control (NRs
100/month), overall watershed service (NRs 100/month), and
landscape beauty (no median value). At the time of the survey,
locals were paying an average NRs 150/month/household for
drinking water.
When the effects of socio-economic factors (e.g., age, education,
occupation, and income) on respondents' WTP for ecosystem ser-
vices were examined using a multiple linear regression model, it
was found that the amount of payment was correlated with the
local's socio-economic status (see Table 2). A significant correlation
was observed between the amount they were willing to pay with
the monthly income of a household (t ¼ 9.5, P < 0.00). On average,
the lowest income family, with an income of NRs 1001e5000/
month, said they were willing to pay NRs 300/month for various
ecosystem services. The average total amounts that households
were willing to pay for various ecosystem services were NRs 365 in
households with a low income of NRs 5001e10,000; NRs 450 in
households with an income of NRs 10,001e20,000; and NRs 600 in
households with an income of NRs 20,001e50,000. This result was
consistent with the widespread assumption of economic theory
Fig. 5. Perceived importance of ecosystem services by the local people. that income has a positive impact on WTP (Lankia et al., 2014).
 P. Bhandari et al. / Applied Geography 69 (2016) 25e34 33

Fig. 6. Willingness to pay of downstream communities for various ecosystem services based on the monthly income.

Table 2
Multiple linear models of total amount of willingness to pay for the ecosystem services.

Socio-economic variables Coefficients estimate Standard error t p value

(Constant) 120.1 49.3 2.4 0.02

Age (years) 7.1 13.7 0.5 0.61
Level of education 6.0 8.1 0.7 0.46
Occupation 20.1 12.0 1.7 0.09
Monthly income (NRs) 102.2 10.7 9.5 0.00

R2 ¼ 0.40; R2 adj ¼ 0.38; F ¼ 23.03; P ¼ 0.000.

Further, the average amount of WTP was 4.89% of respondents' by local communities. This study has also provided a case for up-
reported monthly income. Although the percentage was higher for stream and downstream linkages in terms of the supply and de-
the poorest household (10%) as compared with the richest house- mand of ecosystem services. This could be useful elsewhere,
hold (1.7%), this is an internationally acceptable affordability figure particularly in developing country such as Nepal where the

squez, 2014). There was no significant correlation between the
(Va upstreamedownstream linkage is very common and upstream
amount of WTP for ecosystem services with age, education, and communities play a crucial role in maintaining ecosystem functions
occupation. and the resulting supply of ecosystem services to the downstream
communities.
The results of this study showed that the NW currently offers
4. Conclusion ecosystem services that have market value, such as water for
drinking and irrigation, and non-market value, such as protection
This study has highlighted the importance of a local watershed from soil erosion and landslide. The results with regard to the
by documenting experts' opinions and local perceptions as well as downstream communities' WTP for water services have provided
the preferences of the local communities using the ecosystem valuable insights for designing and initiating a potential PES
services framework. This study successfully used the matrix model scheme, which may improve current forest management practices
of ecosystem services in a localized context in a data-deficient re- as well as support upstream land holders by creating economic
gion, where knowledge of ecosystem services among stakeholders opportunities.
was weak. The results indicated that drinking water was the most
important ecosystem service of the area and local communities Acknowledgments
were willing to pay a higher amount than they were currently
paying if the quality and sustainability of the ecosystem services We are grateful to the local communities of the study area and
were assured. the local experts who have helped us to carry out this study.
This result can help decision makers to make appropriate land
and forest management decisions on a landscape that is undergo- References
ing a rapid change in LULC due to urbanization and population
growth as it highlighted views of local users on various ecosystem Ananda, J., & Herath, G. (2009). A critical review of multi-criteria decision making
services and their importance to them. Decision makers can take an methods with special reference to forest management and planning. Ecological
Economics, 68(10), 2535e2548.
appropriate policy decision to manage the Northern watershed in Arriagada, R. A., Ferraro, P. J., Sills, E. O., Pattanayak, S. K., & Cordero-Sancho, S.
order to optimize certain ecosystem service that was highly valued (2012). Do payments for environmental services affect forest cover? A farm-
34 P. Bhandari et al. / Applied Geography 69 (2016) 25e34

level evaluation from Costa Rica. Land Economics, 88(2), 382e399. Environment, 7(1), 4e11.
Barkmann, J., Baumann, R., Meyer, U., Müller, F., & Windhorst, W. (2001). Oort, B., Bhatta, L. D., Baral, H., Rai, R. K., Dhakal, M., Rucevska, I., et al. (2014).
€
Okologische Integrit€
at: Risikovorsorge im Nachhaltigen Landschaftsmanage- Assessing community values to support mapping of ecosystem services in the
ment. Gaia, 10(2), 97e108. Koshi river basin, Nepal. Ecosystem Services, 13, 70e80.
Birch, J. C., Thapa, I., Balmford, A., Bradbury, R. B., Brown, C., Butchart, S. H., et al. Pandey, S. S., Cockfield, G., & Maraseni, T. N. (2016). Assessing the roles of com-
(2014). What benefits do community forests provide, and to whom? A rapid munity forestry in climate change mitigation and adaptation: a case study from
assessment of ecosystem services from a Himalayan forest, Nepal. Ecosystem Nepal. Forest Ecology and Management, 360, 400e407. http://dx.doi.org/10.1016/
Services, 8, 118e127. j.foreco.2015.09.040.
Burkhard, B., Kroll, F., Müller, F., & Windhorst, W. (2009). Landscapes' capacities to Pattanayak, S. K., Wunder, S., & Ferraro, P. J. (2010). Show me the money: do pay-
provide ecosystem servicesea concept for land-cover based assessments. ments supply environmental services in developing countries? Review of
Landscape Online, 15(1), 22. Environmental Economics and Policy, 4(2), 254e274.
Burkhard, B., Kroll, F., Nedkov, S., & Müller, F. (2012). Mapping ecosystem service Paudyal, K., Baral, H., Burkhard, B., Bhandari, S. P., & Keenan, R. J. (2015). Partici-
supply, demand and budgets. Ecological Indicators, 21, 17e29. patory assessment and mapping of ecosystem services in a data-poor region:
Central Bureau of Statistics (CBS). (2014). National population and housing census case study of community-managed forests in central Nepal. Ecosystem Services,
2011 (Village development committee/municipality), surkhet. Kathmandu, Nepal: 13, 81e92. http://dx.doi.org/10.1016/j.ecoser.2015.01.007.
Government of Nepal, National Planning Commission Secretariat. Perrings, C., Duraiappah, A., Larigauderie, A., & Mooney, H. (2011). The biodiversity
Charnley, S., & Poe, M. R. (2007). Community forestry in theory and practice: where and ecosystem services science-policy interface. Science, 331(6021), 1139e1140.
are we now? Annual Review of Anthropology, 36(1), 301e336. Peterson, M. J., Hall, D. M., Feldpausch-Parker, A. M., & Peterson, T. R. (2010).
Convention on Biological Diversity (CBD). (2010). Aichi biodiversity targets. Available Obscuring ecosystem function with application of the ecosystem services
http://www.cbd.int/sp/targets/ Accessed 15.04.15. concept. Conservation Biology, 24(1), 113e119.
Corbera, E., Soberanis, C. G., & Brown, K. (2009). Institutional dimensions of pay- Quintero, M., Wunder, S., & Estrada, R. (2009). For services rendered? Modelling
ments for ecosystem services: an analysis of Mexico's carbon forestry pro- hydrology and livelihoods in Andean payments for environmental services
gramme. Ecological Economics, 68(3), 743e761. schemes. Forest Ecology and Management, 258(9), 1871e1880.
Daily, G. (1997). Nature's services: Societal dependence on natural ecosystems. Island Redford, K. H., & Adams, W. M. (2009). Payment for ecosystem services and the
Press. challenge of saving nature. Conservation Biology, 23(4), 785e787.
Department of Forest (DoF). (2015). Community forest users group database, man- Regmi, B. R., Kafle, G., Adhikari, A., Subedi, A., Suwal, R., & Paudel, I. (2009). Towards
agement information system. Kathmandu, Nepal: Community Forestry Division, an innovative approach to integrated wetland management in Rupa Lake Area
Department of Forest. of Nepal. Journal of Geography and Regional Planning, 2(4), 80e85.
Dyson, M., Bergkamp, G., & Scanlon, J. (2003). Flow: The essentials of environmental Rio García-Amado, L., Pe
rez, M. R., & García, S. B. (2013). Motivation for conserva-
flows. Gland, Switzerland and Cambridge, UK: IUCN. tion: assessing integrated conservation and development projects and pay-
Ferraro, P. J., & Kiss, A. (2002). Direct payments to conserve biodiversity. Science, ments for environmental services in La Sepultura Biosphere Reserve, Chiapas,
298(5599), 1718e1719. Mexico. Ecological Economics, 89, 92e100.
Fisher, B., & Turner, R. K. (2008). Ecosystem services: classification for valuation. Sanchez-Azofeifa, G. A., Pfaff, A., Robalino, J. A., & Boomhower, J. P. (2006). Costa
Biological Conservation, 141(5), 1167e1169. Rica's payment for environmental services program: intention, implementation
Jacobs, S., Burkhard, B., Van Daele, T., Staes, J., & Schneiders, A. (2015). ‘The matrix and impact. Conservation Biology, 21(5), 1165e1173.
reloaded’: a review of expert knowledge use for mapping ecosystem services. Schomers, S., & Matzdorf, B. (2013). Payments for ecosystem services: a review and
Ecological Modelling, 295, 21e30. comparison of developing and industrialized countries. Ecosystem Services, 6,
Khatri, D. B., Paudel, N. S., Bista, R., & Bhandari, K. (2013). REDDþ financing: What 16e30.
can we learn from the piloting of forest carbon trust fund in Nepal?. Discussion Seppelt, R., Dormann, C. F., Eppink, F. V., Lautenbach, S., & Schmidt, S. (2011).
Paper Series 12.1. Kathmandu: Forest Action Nepal. A quantitative review of ecosystem service studies: approaches, shortcomings
Kinzig, A. P., Perrings, C., Chapin, F. S., Polasky, S., Smith, V. K., Tilman, D., et al. and the road ahead. Journal of Applied Ecology, 48(3), 630e636.
(2011). Paying for ecosystem servicesdpromise and peril. Science, 334(6056), Shrestha, U. B., & Bawa, K. S. (2014). Economic contribution of Chinese caterpillar
603e604. fungus to the livelihoods of mountain communities in Nepal. Biological Con-
Koschke, L., Fuerst, C., Frank, S., & Makeschin, F. (2012). A multi-criteria approach for servation, 177, 194e202.
an integrated land-cover-based assessment of ecosystem services provision to Shrestha, U. B., Shrestha, B. B., & Shrestha, S. (2010). Biodiversity conservation in
support landscape planning. Ecological Indicators, 21, 54e66. community forests of Nepal: rhetoric and reality. International Journal of
Kroll, F., Müller, F., Haase, D., & Fohrer, N. (2012). Ruraleurban gradient analysis of Biodiversity and Conservation, 2(5), 98e104.
ecosystem services supply and demand dynamics. Land Use Policy, 29(3), Simelton, E., & Dam, B. V. (2014). Farmers in NE Vietnam rank values of ecosystems
521e535. from seven land uses. Ecosystem Services, 9, 133e138.
Lankia, T., Neuvonen, M., Pouta, E., & Siev€ anen, T. (2014). Willingness to contribute Sohel, M. S. I., Mukul, S. A., & Burkhard, B. (2014). Landscape‫ ׳‬s capacities to supply
to the management of recreational quality on private lands in Finland. Journal of ecosystem services in Bangladesh: a mapping assessment for Lawachara Na-
Forest Economics, 20(2), 141e160. tional Park. Ecosystem Services, 12, 128e135.
Loomis, J., Brown, T., Lucero, B., & Peterson, G. (1996). Improving validity experi- Stoll, S., Frenzel, M., Burkhard, B., Adamescu, M., Augustaitis, A., Baeßler, C., et al.
ments of contingent valuation methods: results of efforts to reduce and (2015). Assessment of ecosystem integrity and service gradients across Europe
disparity of hypothetical and actual willingness to pay. Land Economics, 72(4), using the LTER Europe network. Ecological Modelling, 295, 75e87.
450e461. TEEB. (2010). The economics of ecosystems and biodiversity: Mainstreaming the eco-
Lopa, D., Mwanyoka, I., Jambiya, G., Massoud, T., Harrison, P., Ellis-Jones, M., et al. nomics of nature: A synthesis of the approach, conclusions and recommendations
(2012). Towards operational payments for water ecosystem services in of TEEB, ISBN 978-3-9813410-3-4.
Tanzania: a case study from the Uluguru Mountains. Oryx, 46(01), 34e44. Turner, R., & Daily, G. (2008). The ecosystem services framework and natural capital
Milder, J. C., Scherr, S. J., & Bracer, C. (2010). Trends and future potential of payment conservation. Environmental and Resource Economics, 39, 25e35.
for ecosystem services to alleviate rural poverty in developing countries. Ecol- Turpie, J. K., Marais, C., & Blignaut, J. N. (2008). The working for water programme:
ogy and Society, 15(2), 4. evolution of a payments for ecosystem services mechanism that addresses both
Millennium Ecosystem Assessment (MEA). (2005). Ecosystems and human well- poverty and ecosystem service delivery in South Africa. Ecological Economics,
being: Biodiversity synthesis. Island Press. 65(4), 788e798.
Mitchell, R. C., & Carson, R. T. (2013). Using surveys to value public goods: The Van Hecken, G., Bastiaensen, J., & V
asquez, W. F. (2012). The viability of local pay-
contingent valuation method. Routledge. ments for watershed services: empirical evidence from Matigua
s, Nicaragua.
Molnar, K., Scherr, S. J., & Khare, A. (2004). Who conserves the world's forests? Ecological Economics, 74, 169e176.
Community-driven strategies to protect forests and respect rights. Washington,
squez, W. F. (2014). Willingness to pay and willingness to work for improvements
Va
D.C., USA: Forest Trends and Ecoagriculture Partners [online] URL http://www. of municipal and community-managed water services. Water Resources
rightsandresources.org/documents/files/doc_136.pdf. Research, 50(10), 8002e8014.
Müller, F. (2005). Indicating ecosystem and landscape organisation. Ecological In- White, A., & Martin, A. (2002). Who owns the world's forests? Forest tenure and public
dicators, 5(4), 280e294. forests in transition. Forest Trends. Washington, D.C.: Center for International
Mun ~ oz-Pin~ a, C., Guevara, A., Torres, J. M., & Bran~ a, J. (2008). Paying for the hydro- Environmental Law.
logical services of Mexico's forests: analysis, negotiations and results. Ecological William, J. V., Russell, C. S., Rodriguez, D. J., & Darling, A. H. (1999). Willingness to
Economics, 65(4), 725e736. pay: Referendum contingent valuation and uncertain project benefits. Sustainable
Muradian, R., Corbera, E., Pascual, U., Kosoy, N., & May, P. H. (2010). Reconciling development Department Technical papers series; ENV-130 (pp. 9e11). 1300 New
theory and practice: an alternative conceptual framework for understanding York Avenue, N.W. Washington, D.C: Publications, Environment Division, Inter-
payments for environmental services. Ecological Economics, 69(6), 1202e1208. American Development Bank, 20577.
Nelson, E., Mendoza, G., Regetz, J., Polasky, S., Tallis, H., Cameron, D., et al. (2009). Wunder, S. (2005). Payment for ecosystem services: Some nuts and bolts. CIFOR Oc-
Modeling multiple ecosystem services, biodiversity conservation, commodity casional Paper, 42.
production, and tradeoffs at landscape scales. Frontiers in Ecology and the