Eliciting community‐preferred policy alternatives for achieving workable coexistence in a human‐dominated landscape: Insights from Chitwan National Park, Nepal

Human‐wildlife conflict is a global issue that poses a major threat to biodiversity conservation and sustainable development. Wildlife adapted to human‐modified landscapes in developing countries can threaten the safety and livelihood of local communities. In such landscapes, where people and wildlife compete for resources, community involvement is promoted to achieve workable coexistence. We used a choice experiment methodology approach to identify community‐preferred policy alternatives in Chitwan National Park in south‐central Nepal. The results indicated that conflicts between the local community and wildlife are on the rise, crop damage being the highest form of damage and most perceived compensation and fencing as conflict‐mitigation tools. The choice experiment showed that residents have a positive preference for enhancing livelihood diversification skills, cultivating buffer crops, promoting alternative livelihoods, and strengthening rapid response teams. We identified concerned mitigators who are younger age group, encountered and concurred that HWC is increasing, and agreed that fencing and compensation are conflict‐mitigation tools. We designed scenario planning using community‐preferred policy alternatives to assist the park management. This is the first‐choice experiment study to design an evaluation framework under community resilience perspectives to achieve workable coexistence. Our study has potential implications for promoting coexistence in a human‐dominated landscape.

Human-wildlife conflict (HWC) is a global issue that poses a major threat to biodiversity conservation and sustainable development (Braczkowski et al., 2023;Gross et al., 2021;IUCN, 2023).HWC is a global concern (Gross et al., 2021;Madden, 2004;Nyhus, 2016).However, the extent and severity of conflicts between people and wildlife differ greatly between developed and developing regions (Anand & Radhakrishna, 2017;Manfredo & Dayer, 2004).In particular, the prevalence of HWC in developing countries is expected to escalate near protected areas as both human and wildlife populations grow (Manfredo, 2015).The competition for resources intensifies with the modified landscape providing a new form of habitat for wildlife (König et al., 2020;Manfredo, 2015).As human settlements expand into wildlife habitats and wildlife increasingly adapts to human-modified landscapes, the challenge of managing wildlife that is adapted to human-modified landscapes is expected to intensify (König et al., 2020;Manfredo, 2015).
Developing strategies for coexistence between local communities and adapted wildlife populations is an urgent necessity that is mutually beneficial to humans and wildlife (Gross et al., 2021;IUCN SSC HWCTF, 2020;Nyhus, 2016).However, international policymakers have repeatedly overlooked the issue of HWC and the crucial role of understanding the perspectives of local communities in HWC management (Dickman, 2010;Gross et al., 2021).The most fundamental requirement for coexistence is that HWCs are well understood and managed and that there is acceptable persistence and tolerance between humans and wildlife that share a common landscape (Carter & Linnell, 2016;IUCN SSC HWCTF, 2020).
The prevalence of HWCs in Nepal is a reflection of the conflicts occurring in South Asia, which is renowned for its rich biodiversity (Hasnat et al., 2018;Reddy et al., 2019).Nepal, located in the central Himalayas, is a mountainous country that covers less than 0.1% of the total global land area (Paudel et al., 2012).Due to the substantial variation in altitude, Nepal accounts for 3.2% of the world's flora and 1.1% of the world's fauna (NTNC, 2020;Paudel et al., 2012).However, Nepal's biodiversity faces numerous threats, such as climate change, poaching, illicit wildlife trade, habitat degradation, HWC, overexploitation of natural resources, agricultural and infrastructure expansion, and invasive alien plant species (MoFSC, 2014;NTNC, 2020).HWC, specifically crop raiding and livestock depredation by wild animals, are prevalent issues in all protected areas, including the iconic Chitwan National Park (CNP) (CNP, 2018;MoFSC, 2014).
In such cases, wildlife poses a threat not only to food security, livelihoods, and safety of local communities while also to increasing uncertainty about the future in an already volatile environment (Khumalo & Yung, 2015).HWCs are unpredictable events as they depend on the behavior of both humans and wildlife (Wieczorek Hudenko, 2012).The strategies employed to address conflicts frequently fail to consider the uncertainties and complexity associated with such conflicts (Carter & Linnell, 2016).Although the application of scenario planning in the human-wildlife context is still in its infancy, developing it can assist protected area managers in making decisions in the midst of rapid change and uncertainty (Beach & Clark, 2015;Neff, 2007;Prato & Fagre, 2020).Developing scenario planning based on communitypreferred conservation policies provides an opportunity to strengthen community resilience in the face of uncertainties, such as those associated with HWC (Marshall & Marshall, 2007;Neff, 2007;Peterson et al., 2003).
This highlights the need for developing an evaluation framework at the landscape level under the perspectives of community resilience to achieve workable coexistence.The aim of our study is twofold.First, we aim to identify community-preferred policy alternatives for dealing with HWC using the choice experiment (CE) methodology.This methodology has been successfully employed in the field of wildlife management (Harihar et al., 2015;Nguyen et al., 2022;Subroy et al., 2018) and is particularly well-suited for studies related to HWC management practices and policy perspectives (Hanley et al., 2010;Kubo & Shoji, 2014;Sherchan et al., 2022).Second, we used the identified policy alternatives to create scenario planning to assist park management in decision-making.This study addresses the gap in incorporating community resilience perspectives into CE method, which has beneficial policy implications for wildlife management.
The remainder of the paper is organized as follows: theoretical framework, materials and methods, results, discussion, and conclusion.

| Theoretical framework
Humans and wildlife are an integral part of the socialecological system (SES), where they interact and impact each other within the system (Carter et al., 2014;Liu et al., 2007).HWC is a complex conservation problem that requires transdisciplinary and multidisciplinary approaches when framed within the SES (Game et al., 2014).It is crucial to achieve resilient human-wildlife relationships when global, environmental, and social changes are in a state of rapid change (Jochum et al., 2014).Therefore, adopting a resilience-based approach to human-wildlife coexistence strategies can help achieve this goal.
The concept of resilience has been explored in broad perspectives by various disciplines such as psychology, engineering, ecology, and disaster studies (Madsen & Chesham, 2015).Resilience is the ability of a system to adapt to change and absorb disturbance while maintaining essential functions, structure, identity, and feedbacks (Walker et al., 2004).Resilience has emerged as a new development paradigm, offering fresh perspectives to communities to cope and adapt to ever-changing circumstances and uncertainties (Adger et al., 2005;Lew, 2014;Walker et al., 2004).One way to achieve this is by enhancing community resilience, which enables communities to cope with uncertainty (Berkes & Ross, 2013).
Community resilience was built upon socio-ecological studies and systems theories that are concerned with system change (Magis, 2010).There are linkages between social and ecological resilience, especially for communities that rely on ecological and environmental resources for a living (Adger, 2000).Community resilience is perceived as a level of social resilience and is well-defined as the capacity of communities to withstand external pressures and disturbances brought on by socio-political, and environmental change (Adger, 2000).Research that focuses on community resilience should actively involve community residents, and their perceptions must be taken into consideration (Janssen & Ostrom, 2006;Magis, 2010).Involving communities that share the landscape with wildlife and their preferences on HWC management policies is key for building community resilience for achieving workable coexistence.

| Study area
Despite its successes in biodiversity conservation, CNP in Nepal has been subject to frequent and severe HWCs over the years (Nyaupane et al., 2022;Silwal et al., 2017).It was established as Nepal's first protected area in 1973 and was recognized internationally by UNESCO as a world heritage site in 1984.It covers a core area of 953 km 2 while its buffer zone covers an area of 750 km 2 (DNPWC, 2017;DNPWC, 2022).The buffer zone comprises both forest and private lands, including agricultural lands (Figure 1).This pristine habitat harbors flagship fauna of CNP, including elephants (Elephas maximus indicus), tigers (Panthera tigris tigris), rhinoceros (Rhinoceros unicornis), and gharial (Gavialis gangeticus).The park forms a contiguous landscape with Parsa National Park in the east, India's Valmiki Tiger Reserve in the south, Daunne forest links it to Churia in the west, and the mid-hills in the north through the Barandabhar Corridor.To protect CNPs unique ecosystem and to mitigate HWC, the CNP management has been endlessly working to improve its relationship with local communities living in the buffer zone (Sapkota et al., 2014).However, HWC remains a major challenge for the park management (DNPWC and DFSC, 2022;MoFSC, 2014).

| Methods-discrete choice experiment
To conduct the discrete choice modeling process, this study followed the steps outlined by Hoyos (2010).Figure 2 shows a flowchart of the steps involved in our study.The first step involved (1) identification of suitable attributes and levels, followed by (2) experimental design, (3) questionnaire development, and (4) household survey.

| Selection of attributes and levels
The first crucial step in designing a CE is to determine specific attributes and their corresponding levels that will be included in the survey.To identify the most relevant attributes for our choice experiment, we conducted a literature review, and in the end, we selected four community resilience dimensions developed by Marshall and Marshall (2007).The four identified attributes are: (R1) residents' perception of risk associated with HWC, (R2) planning, learning, and reorganizing for managing HWC (R3) residents' ability to cope with HWC and, (R4) residents' interest in adapting to future HWC.
After selecting the attributes, the levels were identified based on a literature review of HWC management.They were further refined through qualitative interviews with experts from the International Union for Conservation of Nature (IUCN), the Department of National Parks and Wildlife Conservation, non-governmental organizations, protected area officials, and program managers involved in HWC management in South Asia.Diverse stakeholder consultation was included to ensure that the HWC management strategies are effective and sustainable.To achieve workable coexistence between wildlife and communities that share the same landscape, it is necessary to include the latter in management efforts.Therefore, we acquired and finalized the willingness-to-participate as the fifth attribute following a pilot study with 50 respondents from the buffer zone.Once the attributes and levels were selected, we consulted with the park management about the selected attributes and levels.The levels were further refined and finalized to be used in the CE and are presented in Table 1.

| Experimental design
Once the attributes and levels have been determined, the subsequent step involves developing the choice experiment.Too many scenarios (3 4 Â 5 ¼ 405) can be constructed from the combination of all attributes and levels if we utilize a full-factor design.The fractional factorial design can help us to reduce the number of possible scenarios, which is widely employed in CE studies.The IBM SPSS Statistics for Windows, Version 26.0, were used to obtain an orthogonal main effect design.
The obtained design consisted of 25 alternatives which is too much of a task for the respondents.These combinations with high levels of attributes but low WTP and those with low levels of attributes with high WTP were discarded.The final design had three choice sets with 12 alternatives, each of which had two recommended alternatives in addition to the status quo or current situation as a baseline (Figure 3).To reduce respondent burden and cognitive fatigue, each respondent faced three choice sets with illustrations as a guideline as shown in Figure 3.

| Questionnaire development
We used a questionnaire comprising of three sections.The first section centerd on residents' experiences with T A B L E 1 Description of attributes and levels used in CE design, and socioeconomic data.

Attribute Levels and description Variables
Residents' perception of risk associated with HWC HWC, including questions related to compensation and recent occurrences of HWC.The second part of the questionnaire featured three choice sets accompanied by illustrations as a guide (Figure 3).The third section assessed the socio-economic characteristics of the respondents and their families.The content of the first and third sections of the questionnaire was formulated based on existing studies related to HWC.A pilot study was conducted with 50 residents residing in the buffer zone of CNP to assess their understanding of the choice tasks.Based on feedback received from the residents, the survey questionnaire (Appendix S1) was refined and subsequently validated with the park management.The questionnaire was initially developed in English and later translated into Nepali.

| Household survey
We Using stratified random sampling, the enumerator, who is native and familiar with the park's landscape, interviewed a total of 506 households.The population was divided into strata based on the available sociodemographic information obtained from the BZUC.The sampled households provided a sufficient sample size to generalize the findings in the study area with a 95% confidence level and a precision of less than 5% (Israel, 1992).The head of the household or, in their absence, respondents aged 18 years or older were selected for the survey.
F I G U R E 3 An example of choice set.

| Data analysis
The descriptive statistics from the first and third parts of the questionnaire were analyzed using Microsoft Excel 365 for Windows and IBM SPSS Statistics for Windows, Version 26.0.The choice responses were analyzed using LIMDEP NLOGIT 5.0 to construct both the random parameter logit (RPL) and latent class model (LCM).The RPL model is employed to identify respondents' preferences for different attributes, while the LCM is used to identify class membership based on respondents' characteristics (Ferdin et al., 2022;Juutinen et al., 2011;Lin et al., 2020).The RPL and LCM models were constructed to obtain a more detailed insight of our respondents' choice preferences.Both of these models are commonly used in the field of conservation (Hanley et al., 2018;Montero-Botey et al., 2021).Based on the log-likelihood ratio (LLR), both models exhibited a high level of goodness-of-fit.Our study combined the attributes and levels, and this empirical utility function can be expressed as: where V ni is the utility function created by the researcher for the respondent n who is presented with the alternative option set i;β 1À8 are the coefficients of each level and β 9 is the coefficient of WTP (for detailed description of each level, refer Table 1).ASC SQ is an alternative-specific constant that highlights whether our respondents prefer policy alternatives "alternative 1" or "alternative 2" instead of the status quo.The respondent's marginal willingness to participate, can be estimated by the ratio of two parameters for an attribute, written as: where β i and β WTP are the coefficients of attribute i and WTP.

| Wildlife perceptions and experience
Our data shows that 63% of the residents agreed that HWC had been increasing in recent years.Additionally, 66% of the residents agreed that the compensation process for HWC incidents is lengthy.Furthermore, most residents (76%) agreed that installing fences effectively reduces HWC (Figure 4).Over the last years, 89% of the residents have encountered instances of HWC, indicating a significant impact on the local community.Moreover, a high percentage (96%) of the respondents believed that compensation for HWC incidents would help reduce the suffering caused by such conflicts.

| Damages
Out of 506 respondents, 88% reported facing crop damage in the four sectors of the park.Large herbivores, such as elephants and rhinoceros, as well as deer and wild boars, caused the highest damage among wildlife in our study.
In Madi, elephants were responsible for the highest crop damage, while in the Eastern sector, rhinoceros and deer caused crop damage.In the Western sector, wild boars were responsible for the highest crop damage (Figure 5).

| Choice responses
As expected, the respondents exhibit clear evidence against the current situation or status quo.The ASC SQ was negative and strongly significant.This indicates that the respondents highly prefer the presented policy alternatives against the current situation.The respondents prefer enhancing livelihood diversification skills, cultivating buffer crops, promoting alternative livelihoods, and strengthening RRTs (Table 2).The positive and significant coefficients at the 1% level indicate that residents have a keen interest in these policy alternatives.Respondents showed a significant but negative preference for strengthening HWC awareness workshops, indicating that they favored the existing HWC awareness workshops by the government.The ability to secure resources, fast-tracked compensation scheme, and improving parkpeople relationships were not statistically significant, indicating that respondents did not express concerns about improving these policy alternatives.Based on the negative and significant coefficient of WTP, it appears that the respondents have expressed a preference against increased participation hours for the positive and significant policy alternatives.
The final selected LCM model was constructed based on several factors, including age, income, past encounters with HWC, perception of HWC trends, and opinions on the effectiveness of fencing and compensation as mitigation measures.The LCM model identified two heterogeneous groups, which are divided into two segments, as displayed on the right side of Table 2.The concerned mitigators 72%, n ¼ 364 ð Þ exhibited alignment with the policy alternatives identified by the RPL model.They were more likely to have a higher income and belong to a younger age group.In addition, they reported encountering HWC in the last 5 years and concurred that HWC is increasing in recent years.Furthermore, they agreed that fencing and compensation are effective ways to mitigate HWC.Interestingly, the Concerned mitigators exhibited a negative and insignificant ASC SQ , indicating that their decision-making process was influenced by these class membership parameters.
The comfortable traditionalists 28%,n ¼ 142 ð Þ were more likely to have a lower income and belong to an older age group.None of the attributes were statistically significant, despite representing a substantial proportion of the respondents.The ASC SQ was positive but insignificant.This suggests that the Comfortable traditionalists are content with the current situation and do not feel the need for any significant changes or interventions, in contrast to the Concerned mitigators.

| Overall conflict pattern and mitigation interventions
This study reinforces that residents of the buffer zone have been experiencing frequent, severe, and widespread HWCs since the establishment of the park, as reported elsewhere in other global studies (CNP, 2018;Gross et al., 2021;IUCN, 2023).A high percentage of residents expressed agreement with the assertion that the nature of the compensation scheme is time-consuming, indicating potential dissatisfaction with the process (Lamichhane et al., 2019).The relief guideline for wildlife damage in Nepal is provisioned to expedite the relief process as early as possible.Our findings are consistent with other studies conducted in the region.For instance, a study conducted in CNP in Nepal revealed that claimants had to wait, on average, 6.6 months to receive compensation for their losses (Lamichhane et al., 2019).A study done in India showed that the compensation given to claimants affected by conflicts was considered too little, and the process of applying for compensation was seen as difficult and time-consuming (Ogra & Badola, 2008).
Interestingly, a significant percentage of respondents believed that compensation for HWC incidents would help reduce the suffering caused by such conflicts, indicating the perceived value of compensation as a tool for conflict mitigation (Ravenelle & Nyhus, 2017).However, the effectiveness of compensation as a wildlife-conflict mitigation tool is a topic of widespread debate and often faces criticism (Ravenelle & Nyhus, 2017).Involving residents in the early stages of designing a compensation scheme is crucial for ensuring what losses should be covered and that the appropriate amount of compensation to be provided (IUCN, 2023;Ravenelle & Nyhus,2017).In addition, it is essential to establish effective systems and institutional arrangements to ensure a smooth and efficient process, from reporting wildlife damages to receiving compensation (IUCN, 2023).Furthermore, our results revealed that residents perceive the installation of fences as an effective measure to reduce conflicts.
A recent study by Dahal et al. (2022), found that 38% of respondents believed that electric fencing would reduce crop depredation in Eastern Nepal.This underscores the importance of physical barriers as a mitigation measure.However, during our field survey, we observed that residents sometimes break the mesh wire fences to collect resources from the core area for their day-to-day living, which leads to encounters with wildlife, consistent with observations made by Lamichhane et al. (2019).This also results in wildlife, such as chital (Axis axis) and Indian muntjac (Muntiacus muntjak), entering through these broken fences and often being unable to return to the wild, resulting in deaths due to stray dog attacks, as we observed during discussions with our respondents.The fences are not uniformly maintained among all communities to minimize wildlife impacts, as noted by Lamichhane et al. (2019).
Fencing the park to decrease conflicts may unintentionally increase HWC incidents if it restricts residents from accessing forest products for their livelihoods (IUCN, 2023).Traditional barrier approaches such as biofencing, developed over millennia, have often been highly successful in reducing conflicts (Woodroffe et al., 2005).Considering the urgent need for interventions, the park management could conduct a study on communitypreferred fencing and its potential negative impacts on wildlife.Involving the local community in the design and implementation of fencing measures can minimize unintended consequences and ensure a sustainable and effective approach to reducing human-wildlife conflicts (IUCN, 2023;Woodroffe et al., 2005).
Crop depredation was the most frequently reported form of damage in our study area, accounting for 88% of all reported cases.These results align with similar studies in the region, which have consistently shown that crop losses are the most significant type of damage caused by wildlife.For example, a study in different protected areas in Bangladesh found that 64% of respondents reported crop losses due to elephant and wild boar attacks (Raihan Sarker & Røskaft, 2010), while a study conducted across four Indian states reported that 71% of households claimed crop damages were the highest (Karanth & Kudalkar, 2017).Similarly, a study in Eastern Nepal showed that 77% of the respondents reported crop depredation by wildlife attacks (Dahal et al., 2022).
The modified landscape in the park has created new habitats for large herbivores and other wildlife, resulting in increased crop raiding.However, the park management has not been able to effectively address losses due to crop depredation caused by wildlife such as wild boars, rhinoceroses, elephants, and other wildlife (CNP, 2018).Our observations show that these species cause the most damage to crops, and the extent of depredation varies across the park's four sectors.Due to the park's large size and diverse socio-economic and cultural characteristics, it has been divided into four sectors to enhance management effectiveness and efficiency (CNP, 2018).To develop effective mitigation strategies, it is essential to understand the ecological factors that influence the movement and foraging behavior of wildlife species in each sector (Fehlmann et al., 2021).
One strategy that works in one sector may not be effective in another due to its unique socio-economic, and cultural characteristics, and wildlife foraging behavior (Fehlmann et al., 2021;Gross et al., 2021;IUCN, 2023).Based on our survey results, most of the respondents favored the implementation of fencing and compensation measures for HWC management.This suggests that park management should work with local residents using a bottom-up approach to develop customized interventions for each sector instead of imposing a uniform strategy for long-term conservation success and community support (Brooks et al., 2013).

| Leveraging residents' preferences for scenario planning
Engaging with the local community and incorporating their preferences can help in constructing scenarios using the RPL model (Ferdin et al., 2022;Nguyen et al., 2022).This approach can provide unique insights into possible future scenarios by taking into account the perspectives of those who live in close proximity to wildlife and share the same landscape (Schoemaker, 1995).With this in mind, we designed three scenarios (Table 3) that may help park managers make informed decisions to address the ongoing conflict: i. Mitigation and response scenario: Effective mitigation strategies are urgently needed to reduce the conflict between people and wildlife, and cultivating buffer crops can help significantly in reducing crop depredation (Breitenmoser et al., 2005;Mukeka et al., 2019;Thirgood & Woodroffe, 2005).By addressing the conflict at the crop level and cultivating buffer crops with economic value, residents can reduce the impacts on their livelihoods (Lamichhane et al., 2019;Mukeka et al., 2019).These buffer crops could be cash crops that mature early or non-palatable to wildlife (Mukeka et al., 2019;NPPC and WWF-Bhutan, 2016).
The role of RRTs is very important in responding to wildlife attacks and rescuing wildlife (Barlow & Brooks, 2019).They are the first to respond during an HWC incident and are an essential part of HWC management.The RRTs provide affected communities with necessary training and resources to better handle future incidents (Barlow & Brooks, 2019;Gross et al., 2021).In CNP, community-based anti-poaching units (CBAPU) were established with the aim of engaging local youths in conservation programs (Mahatara et al., 2018).These programs were designed to accomplish various objectives, including raising awareness about biodiversity conservation, reporting illegal activities to park authorities, mitigating HWCs, and rescuing and rehabilitating problematic, orphaned, and injured wild animals.As a crucial component of CBAPU, RRTs were formed to ensure prompt and effective conservation action within the local area (Mahatara et al., 2018).The work of RRTs is valuable to both communities and wildlife.However, RRT members are often poor, at risk of danger, and have limited time to devote to their work (Barlow & Brooks, 2019).It is important to strengthen RRTs in the long run to ensure their sustainability by involving motivated members to deal with HWC (Barlow & Brooks, 2019).This scenario, focused on mitigation and response, can help park management plan and mitigate conflicts.
ii. Livelihood diversification scenario: Improving the livelihoods of local people is a key objective of park management at CNP (CNP, 2018).As previously mentioned, the local communities in the park's buffer zone are particularly vulnerable to the impacts of wildlife.As they rely primarily on subsistence agriculture for their livelihoods, which makes them highly susceptible to crop depredation (Lamichhane et al., 2018).However, the popularity of wildlife tourism is growing, particularly among local youth, which presents an opportunity for livelihood diversification (Lamichhane et al., 2018;Nyaupane et al., 2022).
Finding effective and practical ways to mitigate the impacts of wildlife on communities' livelihoods can be challenging (IUCN SSC HWCTF, 2020).To address this challenge, the park management can adopt various strategies.One approach is to support alternative livelihoods by promoting ecotourism and facilitating the growth of local product-consuming enterprises.The local community in the park acknowledges the advantages associated with the national park and wildlife tourism (Nepal & Spiteri, 2011).CNP is one of the most popular and visited protected areas in Nepal, attracting both domestic and international tourists.This influx of tourists not only generates foreign currency but also contributes to economic development and promotes local employment and livelihoods (CNP, 2018;Nyaupane et al., 2022;Pandit et al., 2015).
Additionally, engaging in buffer zone community forest management can provide sustainable economic opportunities for the local community.The park management can also play a crucial role in offering skill development training and education to enhance employment prospects.It is important to tailor these support measures to the specific needs and preferences of the local community to ensure their effectiveness.This scenario highlights the potential benefits of enhancing residents' livelihood opportunities and could guide park management in planning for livelihood diversification.
iii.Human-wildlife Coexistence scenario: Conflicts between wildlife and humans are an ongoing concern in conservation and agricultural landscapes, especially as human-modified landscapes in the park have created new habitats for wildlife (König et al., 2020).To address this issue, integrative and holistic strategies, preferred by local communities, are essential for promoting sustainable coexistence (Gross et  Note: The 95% confidence intervals are provided in parentheses.König et al., 2020).These strategies are often tailored to specific sites and require the development of durable and feasible coexistence scenarios at a landscape level (IUCN, 2023).By integrating enhanced livelihood diversification skills, cultivation of buffer crops, promotion of alternative livelihoods, and strengthening RRTs, park management can enhance local community resilience to achieve workable coexistence.Additionally, the park management should pay close attention to the concerned mitigators as they prefer these policy alternatives.The park management should also consider the diverse socio-economic and cultural characteristics of the park to tailor these strategies for different sectors to achieve a sustainable coexistence.

| CONCLUSION
Our study documented the widespread occurrence of human-wildlife conflicts in Chitwan National Park, with crop depredation being the most common form of damage experienced by local communities.Our findings conclude that fencing is perceived as a mitigation tool by most respondents.However, further research is needed to understand its actual effectiveness and to develop best practices for fencing installation and maintenance.It is important to note that the frequency and intensity of crop damage vary across the four sectors of the park.Additionally, the prevailing compensation mechanism poses challenges in terms of its time-consuming nature.By using a choice experiment, we identified the community's preferred policy alternatives and targeted segments.Additionally, we developed three scenarios to assist park management in making decisions when faced with uncertainty.
From the empirical results of this study, we developed an evaluation framework to enhance community resilience for coexistence (Figure 6).While our results come from Chitwan National Park in Nepal, they serve as an important reference point for other South Asian countries in the region that face frequent and severe conflicts.On one hand, it is important to acknowledge that humanwildlife conflicts are often site and context-specific.Additionally, the cost of protecting threatened species can be a significant challenge, particularly in developing countries.However, by following our example, researchers and policymakers in the region can design choice experiments to tailor community-preferred interventions that promote a workable coexistence.

AUTHOR CONTRIBUTIONS
Arockia E J Ferdin and Chun-Hung Lee conceived the ideas for this manuscript.Arockia E J Ferdin and Jia Wei Chook analyzed the data.Arockia E J Ferdin led the writing of the manuscript and wrote the first draft of the paper.Arockia E J Ferdin created all the tables, and Nabin Dhungana created the study area map Figure 1.Arockia E J Ferdin and Jia Wei Chook designed the Figures 2-6; all co-authors participated in the writing of the manuscript and gave final approval for publication.

F
I G U R E 2 Flowchart showing steps involved in conducting a discrete choice experiment.F I G U R E 1 Map showing the study area of Chitwan National Park, Nepal.
conducted a quantitative survey of households in the 21 Buffer Zone User Committees (BZUC) of Chitwan National Park, located in the subtropical south-central part of Nepal.This study was approved by the Department of National Parks and Wildlife Conservation (Letter No: 131/078/79) and the CNP Headquarters, Kasara Sector, Chitwan, Nepal (Letter No: 2003/078/79).We adhered strictly to the guidelines and conducted interviews from sunrise to sunset on each day.Only the household members who had provided verbal consent and expressed interest were interviewed.

F
I G U R E 4 Residents' perceptions of HWC and its mitigation measures.F I G U R E 5 Crop damages caused by wildlife in different sectors of Chitwan National Park.

F
I G U R E 6 Evaluation framework to enhance community resilience for coexistence.
Estimation results for RPL and LCM models.
Scenario planning.
al., 2021; T A B L E 3