Open AccessArticle

Community-Based Farming Water Resource Management and Important Factors for Adaptation Practices in Terai, Nepal

Sharada Karki

^1,* and

Shigehiro Yokota

Graduate School of Environmental and Information Studies, Tokyo City University, 3-3-1 Ushikubo-nishi, Tsuzuki Ward, Yokohama 224-8551, Kanagawa, Japan

Faculty of Environmental Studies, Tokyo City University, 3-3-1 Ushikubo-nishi, Tsuzuki Ward, Yokohama 224-8551, Kanagawa, Japan

Author to whom correspondence should be addressed.

Water 2025, 17(1), 47; https://doi.org/10.3390/w17010047

Submission received: 20 October 2024 / Revised: 14 December 2024 / Accepted: 23 December 2024 / Published: 27 December 2024

Download

Browse Figures

Figure 1
Study and data collection area within the Kawasoti Municipality farming area with administration and elevation distribution (a) and sub-streamlines with the mainstream river, Narayani (b). "> Figure 2
(a) Irrigation distance from farmland and water accessibility within community groups. (b) Irrigation water intake methods within community groups (%, N = 200). "> Figure 3
River distance from farmland and flood impact within community groups (%, N = 200). "> Figure 4
Farmers’ community-based integrated farming water resource management in total (%, N = 200). Structural measures for (a) irrigation system and (b) riverside and non-structural measures or strategies for (c) ecosystem services and (d) farming continuity. * Water flow management (widening, deepening, and cleaning). ** Unused land (public land uses for water storage, paddy farming, or vegetation). "> Figure 5
Farmers’ community-based structural measures (a) for irrigation systems and (b) for riversides within community groups (%, N = sample size of each community groups). * Water flow management (widening, deepening, and cleaning). ** Unused land use (public land uses for water storage, paddy farming, or vegetation). "> Figure 6
Farmers’ community-based non-structural measures for (a) irrigation systems and (b) riverside community groups (%, N = sample size of each community groups). "> Figure 7
Decision tree (CHAID) for explaining farmers’ total adaptation behavior for community-based water flow management in irrigation channels and most important associated factors. "> Figure 8
Decision tree (CHAID) for explaining farmers’ total adaptation behavior for community-based buffer zone vegetation on riverside areas and most important associated factors. "> Figure A1
Access to irrigation channels, their water accessibility, and management across community groups (A–G, accordingly). Groups (A,E,G) had better facilities for irrigation channels and water and better management. Groups (B,C) had access to channels but poor management. The majority of farming areas in group (D) had a higher irrigation distance. Group (F) had access to channels; and both groups (D,F) depended on groundwater or rainwater for farming. "> Figure A2
Community participation in irrigation channel management within riverside areas and river management (gray embankment measures). ">

Versions Notes

Abstract

Driven by the growing frequency of flood risks, this study focused on farming water resource management (FWRM) as an ecosystem-based solution. Despite its significance, there are limited studies investigating paddy farmers’ community-based adaptations (CBAs) for managing diverse farming water resources at a micro-spatial level, particularly within multidimensional communities. This study aims to bridge this gap and focuses on how community diversity and household characteristics impact farmers’ adaptation to different CBA methods. We conducted a household questionnaire survey in floodplain paddy farming communities in Kawasoti Municipality, Nepal, based on cultural, socioeconomic, and settlement diversity. The questionnaire was subjected to farmers’ CBAs for integrated FWRM and multiple structural and nonstructural adaptation measures for irrigation and rivers. The results showed that farmer participation varied across community groups. To understand the most important associated factors within community diversity and household attributes to adopt different water resources, the most adopted structural measures of water flow management (54%) from irrigation and buffer zone vegetation (54%) from rivers were analyzed. We used the Chi-squared Automatic Interaction Detector model, which suggests that water accessibility associated with community diversity, landholding, and water intake is important to improve farmers’ participation in irrigation management. However, for river management, community diversity, which relates to location in relation to a river and is associated with household income and farmland distance, is an important factor.

Keywords:

flood risk mitigation; ecosystem-based disaster risk reduction; paddy farming water resources; community-based adaptation; chi-squared automatic interaction detector model

1. Introduction

There is an emerging interest in the role of ecosystem services in reducing natural disaster risks, such as floods, and the subsequent benefits, such as lower maintenance costs, social equitability, and environmental sustainability. It is a substitute for typical engineering solutions and is called ecosystem-based disaster risk reduction (Eco-DRR) [1,2]. In Eco-DRR, farming water resources, such as irrigation systems and rivers or streams, are essential to mitigate flood risks, especially in areas with potential need and vulnerability, to improve river flow, agricultural stability, soil fertility, conservation of water and soil resources, and livelihoods for communities around the watershed area [3,4]. It has huge potential as a major green infrastructure to cope with flood risk, not only for farming areas, but also on a large scale, such as urban or total watersheds [5,6].

Farming water resource management (FWRM) is an integrated approach for managing the relationships among hydrological, biophysical, and socioeconomic systems to enhance the use of water resources for productivity, conservation, and sustainability [7,8]. From an integrated perspective, water resource management (as an Eco-DRR) covers different mitigation measures or adaptation practices regarding access to water resources and different stakeholders. Integrated water resource management is a crucial activity that is widely implemented, and successful lessons have been learned. It combines watershed resources, water resource management, and typical nature-based and hybrid solutions with different strategies, considering long-term conservation initiatives and resilience developed by different participants with different needs and capabilities [5,8,9]. In FWRM, there is strong advocacy for a micro-level management model. This model serves as a robust tool for coordinating local institutions within cohesive ecosystem units that encompass villages, subdistricts, and primary river tributaries. These tributaries play a critical role in collecting, storing, and channeling rainwater to subsequent orders, and their distribution is naturally constrained by topography [10].

At the micro-spatial level of FWRM, farming communities and their integrated adaptation practices for water resources are crucial in a different manner. For instance, sustainability, equity, livelihood resilience, local empowerment, and cost–benefit ensure continuous maintenance and improvement of local ecosystems with extensive knowledge and experience of the local climate [11,12]. Furthermore, they have more power and responsibility, which can drive adaptation to anticipate change [13,14]. Their participation and adaptation capabilities depend on multidimensional socioeconomic and environmental aspects [15,16]. The effectiveness of such a bottom-up level of Eco-DRR adaptation practices has often failed because of the policy negligence of local communities [17,18] and their different needs [19]. However, to ensure sustainable and effective implementation, it is important to increase community resilience by empowering local actors, strengthening their existence, addressing their needs, and building adaptive capabilities. This is possible by investigating farmers’ current existence and positions, their concrete Eco-DRR practices, and how their decision to adopt is influenced by their different needs and circumstances [17,18,19,20].

The importance of major FWRM practices, such as irrigation and rivers, as an Eco-DRR practice has been explained in previous studies in terms of community-based macro- or micro-level [10,21,22], integrated-based [23], and climate change adaptation [24]. Collectively, these studies include both public and private resource management. Participation, methodologies, and capabilities can differ significantly, depending on whether actions and goals are defined and executed by communities or individuals [25,26] or by communities through indigenous or voluntary initiatives [27]. Hence, when adopting community-based farming water resource adaptation practices, it is important to include related mutual resources and actual community-based actions to identify different approaches to different resources. It is essential to explore the current adaptation situation and relate it to different needs or possibilities to improve community-based Eco-DRR practices.

Regarding the position of key actors and their impact on community-based adaptation, this study analyzed both community diversity and participants in community-based actions. In community-based adaptation, the character of the community itself has been found to be critical because it creates an individual’s belief system, norms, values, needs, and perceptions and guides a group of individuals to make decisions that can shape different livelihood settlements and decision making for common threads and the value of ecosystem services [28,29]. Hence, community-based adaptation was found to vary with community diversity, such as urban or rural [30], socioeconomically advantaged or disadvantaged [31], location [32], and culture [33]. In addition, owing to the socioeconomic or cultural diversity in the community (mixed community), farmers’ mutual activities were also found to be obstacles to consuming and conserving local resources [34]. In particular, countries or regions with great diversity, shaped by different cultures, socioeconomic development, and settlements, can be found even at a micro-spatial level [17,18,29]. This diversity can also shape mutual Eco-DRR practices.

Previous studies on the adaptation of mutual resources to disaster risk have primarily considered household-level attributes. The studies found that farmers’ participation and adaptation capabilities varied mainly because of gender, education [35], income level [36], age [37], farming experience, and income source [37,38]. In farming situations, the size of landholding [38], location of farmland or access to local resources [39,40], and the impact of flood risk [41] were found to impact participation in adaptation. The aforementioned studies showed that for community-based adaptation, household-level circumstances are as important as community characteristics. Nevertheless, studies including both important factors, community diversity and household attributes, of local society and analyzing how they impact different mutual adaptation practices have not yet been conducted.

Based on this study gap, our main research questions are as follows:

What are the community diversities at a micro-spatial level in a country or area with multidimensional socioeconomic and cultural diversity?
What are the integrated community-based Eco-DRR practices for different water resources and how might these practices vary across diverse community groups?
How do multidimensional socioeconomic factors, community diversity, and household attributes affect the participation of farmers in different community-based adaptation practices?
In what ways are these factors associated with and what are the targets for improving farmers’ participation in Eco-DRR practices to mitigate flood risk in a multidimensional society?

In terms of community-based water resource management and its influencing factors, this study aimed to identify the following key components: farmers’ community-based water resource management, the position of key actors, and their impact on different community-based adaptation practices. The study findings are expected to provide valuable insights for bottom-up Eco-DRR policies to mitigate flood risk by improving farmers’ community-based adaptation capabilities for farming water resources.

2. Materials and Methods

2.1. Study Area

Nepal is rich in water resources, among the top six globally, and has the second highest susceptibility to flood hazards in South Asia [42]. It has socioeconomic diversity and is a cultural mosaic comprising multiple castes and ethnic groups. This determines their settlement, development, access to, and control over socioeconomic or political resources and even the interactions between different groups [43]. The floodplain area (Terai region) of the country covers approximately 14% of the total land area and is home to more than half of the population. In this region, more than 84% of the households are actively engaged in rice production and are among the most vulnerable to flood risks [44]. Apart from topography, poor land use, the government’s extensive focus on engineering infrastructure and its failure, and insufficient drainage systems are the main issues in flood risk mitigation measures [45]. In addition, the way policies are implemented is not in favor of mainstream community-based adaptation due to ignorance, lack of evidence of the local circumstances and community diversity, adaptation practices, and participation [46]. Thus, both the government and farmers’ low adaptation capacity contribute to the greater impact of flood disasters [16,45,47].

This study was conducted within paddy farming floodplain areas (Terai) in the Kawasoti Municipality in the Nawalpur District of Gandaki Province, Nepal (Figure 1). This study took place between 27°35′ and 27°45′ EW and 84° and 84°15′ NS, elevated between 124 and 1309 m above mean sea level. The total study area was 114 km². The monsoon period is from June to August, and the average monthly precipitation is above 550–700 mm. The area is located in a narrow valley of the Shiwalik Mountain (a young mountain with fragile soil and sedimentary rocks). The total population was 62,421, with a density of 550/km². Due to the narrow valley and mainstream river (Narayani), this area experiences flash floods with high-speed runoff and riverine floods. Within a short distance of floodplain areas, other than heavy river cutting, different impacts of flooding can be seen, such as upstream paddy fields covered by sand and pebbles and long duration water-logging downstream but improved soil quality. The impact of floods on food grains has increased over the years, primarily due to the significant loss of agricultural land.

Agriculture is a major industry in this area, and paddy is the main crop. It is also well known for farmers’ cultural-based homestay tourism and is a Tharu ethnic heritage site [48]. As in other Terai regions, there are three main types of cultural communities that can be found based on language and social settlement. These include (a) Pahadi ethnicities or communities that migrated from the hills, (b) Terai ethnicities or the Tharu community, and (c) Musahar (Musahar, Majhi, or Bote) communities or marginal communities of Terai ethnicity. The Musahar community, one of the most marginalized communities, is found in some parts of Terai, including the Kawasoti Municipality [34,48]. The communities are diverse, with different languages and cultures, norms and values, socioeconomic and political development, approaches to local resources, and relationships and dependencies on local resources, particularly water resources. Especially in rural areas, settlements can vary within cultural groups, such as the typical Pahadi, Tharu, or Musahar villages. With the rapid increase in development and infrastructure in this area, the structure of the community is also diversifying owing to cultural mix and socioeconomic development. Hence, choosing this area for our study, we explored community diversity, such as culture, socioeconomic development, and landscape at a micro-spatial level; farmers’ community-based adaptation measures; and the important factors for farmers to participate in different adaptation methods.

2.2. Methods

2.2.1. Sampling Size and Questionnaire Survey

The survey was conducted in two phases in June and July 2022, comprising a pre-processing questionnaire and a final questionnaire. The pre-processing questionnaire primarily involved understanding the socioeconomic distribution, local farming water resources, flood risk, and community-based adaptation measures. We selected data collection areas considering community diversity at a microspatial level with common water resources [30,31,32,33,34]. We observed and participated in local community-based adaptation practices, discussions, and consultations with focus groups, including municipality and ward members. Based on this, the first major flood risk and farming water resource we selected was the Kerung River (sub-watershed).

Irrigation systems from common farming resources in Nepal often use indigenous methods, which are locally negotiated, constructed, and collectively managed by farmers with their indigenous knowledge and experience [49], while rivers have more volunteer-based management [50]. This is similar to the study area (see Appendix A). Hence, for farmers’ community-based adaptation practices for FWRM using different methods [25,26,27], the questionnaire subjects were related to irrigation systems as farmer-managed and rivers as government-managed water resource adaptation practices.

For fine data collection, locations were selected through a carefully designed combination of cultural, socioeconomic, and landscape settlements such as stream location, river distance, and city location within the same territory and stream catchment (Table 1). The total sample size was 200, and the number of farmers in each community group was slightly higher or lower than the predetermined target owing to the ward location.

The questionnaire language was Nepali (the national language) or Tharu (the local ethnic language), and respondent farmers were mobilized by farming leaders and social workers in each community. This enabled us to share information on Eco-DRR for flood mitigation and the survey’s purpose, address misconceptions, and build trust in the survey before information collection. We also employed two local college students from Musahar and another from the Tharu community to assist with data collection, interpret the local language, and provide in-depth local information. The questionnaire was developed in a semi-structured manner and divided into two parts. The first part of the questionnaire included farming households’ socioeconomic and farming information to analyze how the different circumstances impact the decision making of participation in community-based adaptation [35,36,37,38,39,40,41] and rely on local or Terai farmers’ circumstances, such as main income sources [51] and how they access irrigation [52].

To obtain information on integrated Eco-DRR practices (at a micro-spatial level) essential for sustainable development [5], the questionnaires for adaptation measures for services were subject to both structural and non-structural measures, and participants were asked to choose multiple options. Accordingly, the content included multiple structural measures that considered combining various alternatives and strategies for improving adaptation capability and resilience to continue farming and farming resources [5,8,9] (see Appendix B).

We excluded plantations and grassing measures from irrigation system management because they belonged to the farmers. For river management, voluntary participation was exchanged for economic compensation lower than the market price. In terms of non-structural measures, the contents were related to farmers’ participation in local, national, or international organization programs for developing awareness or conducting conservation activities. Accordingly, strategies for farming continuity were included based on local farmers’ involvement in local organizations or committees (Tole sudhar samiti), microfinance institutions, and different sharing activities among farmers to ease farming continuity.

2.2.2. Statistical Model and Data Processing

The goal of our statistical model was to identify the most important factors for community participation in different adaptation methods and to analyze the relationships and interactions between them. This could provide in-depth information on the target factors to improve farmers’ adaptation or participation in community Eco-DRR practices.

We adopted the Chi-squared Automatic Interaction Detection (CHAID) decision tree model, which reflects the relationships between different explanatory variables and identifies the possible interaction effects between them. Unlike other decision tree models, this model provides a high visual output, can detect feature interactions, make multiway splits, and be used for both binary or categorical dependent and independent variables [53,54]. The non-parametric method efficiently handles datasets in which groups differ in size without compromising the validity of the results. The CHAID method, using IBM SPSS, was originally developed by Kass [55]. This model builds a predictive tree to determine how the variables merge and explains the outcomes of the dependent variables. In the tree-shaped diagram, the CHAID analysis splits the dependent variables into two or more categories, called initial or parent nodes, and the nodes are then split using statistical algorithms into child nodes. The last category of the tree is called a terminal node. The category that has a major influence on the dependent variable comes first, followed by the less important category. Therefore, it is referred to as a terminal node [56].

The binary variables were assigned to multiple-choice options, which means that the set of observations was split into two options: consistent codes (Y) = {1 if the farmers adapt, 0 if the farmers do not adapt or become involved}. The purpose of this was to explore the maximum and most concrete adaptation information. In addition, this study assumed that farmers decide to adapt measures under different circumstances; meanwhile, we also presumed that non-adapting farmers do so because of barriers [57]. Once completed, the questionnaires were edited, integrated, and generated using a Microsoft Excel spreadsheet. Descriptive statistics of the collected variables were summarized as percentages of the sample size and graphs.

To analyze the collected quantitative data for small and unequal sample size distributions, the multivariate inferential statistical analysis technique MANNOVA was used. Appendix C presents the results. Data were edited, coded, and assigned values for statistical analysis (Table 2). We used farmers’ households and farming capital, including community groups, as independent variables. The most adopted measures from each structural method were taken as dependent variables to comparatively analyze the impact factors influencing different water resource adaptation methods. The independent variables were converted into nominal, continuous, and ordinal variables based on their natural characters [53,54]. We used SPSS, and the minimum number of cases was 10 for the parent node and 5 for the child node.

3. Results

3.1. Farmers’ Personal Attributes

The collected questionnaire data on household attributes are summarized in Table 3. Of the 200 respondents, 47% were female, and 53% were male. The age of the participants ranged from 18 to 76 years, with almost half of them being middle-aged (36–55 = 48%), and the maximum farming experience was approximately 11–20 and then 21–30 years (56%). This indicates that most participants started farming at a young age (less than 15 years old). The education of the family head was particularly low (above secondary = 15.5%); in contrast, the education of family members was comparatively higher (43.5%). The landholding distribution revealed that almost all farmers were in the marginal category, which is less than 1 hectare (<25 Kattha = 94%). Surprisingly, most farmers (63.5%) depended on non-farm financial resources, and more than half of farmers’ total household’s monthly income was no more than NPR 20,000.

3.2. Farming and Environmental Characteristics

The results in Table 4 show respondents’ farming circumstances. The results show that 71% of farmland had a lower distance from main irrigation channel. In contrast, only 35% of farmers had better water accessibility to the irrigation channel, and almost half of the farmers (54%) used artificial methods to transport water from irrigation channels to their farmland. A total of 95.5% of farmers had at least a 100-m distance from their farms to the river, which also highlights buffer zone areas. We found that more than half of the respondents’ land faced a medium to very high flood risk.

We compared the farming circumstances between the community groups. The purpose was to demonstrate access to common resources, their facilitation, and the probability of common disaster risk within community groups. Figure 2a shows that only rural groups A, B, E, F, and G had better access to irrigation channels (relatively high and very high). However, Figure 2b shows that only groups A, E, G, and B had better water accessibility. This result shows the unequal distribution of irrigation infrastructure and water accessibility, even within the same territory, as well as differences in infrastructure and service consumption. The result of the intake method also shows a similar situation, where only farmers in groups A, B, E, and G had better facilities for the natural intake of irrigation water. In contrast, farmers in the other groups mostly used artificial intake methods. Figure 3 shows that almost all farmers in all groups maintained a minimum river distance from their farmland. This result also shows that groups A, C, G, and F had comparatively shorter distances, indicating greater access to river services or a higher possibility of flood risk. However, respondents in groups A, F, and G (riverside groups) had a higher flood impact.

3.3. Community-Based Farming Water Resources Management

We inventoried farmers’ community-based adaptation measures (multiple choice) to describe the overall trends in which adaptation measures are favored or less adopted by the majority (Figure 4). Regarding structural measures, farmers adopted different methods such as typical natural-based, hybrid, and engineer-based (grey) methods. Figure 4a shows that communities were prioritized over typical natural-based (water flow management and natural dam) and hybrid (sandbag embankment) methods, with much less priority over grey methods for irrigation system management. Most participants adopted water flow management (53%) followed by rehabilitation (44.5%). For riverside farms, buffer zone vegetation and grey embankments were adopted by most farmers (54% and 46%, respectively). Very few farmers reported participating in strategies for ecosystem services, with most focusing on buffer zone protection (19.5%) and watershed ecosystem conservation (16.5%). This shows that most farmers engaged in manpower sharing (73.5%), farming (71%), and seeds and fertilizers (53.5%); however, participation in agricultural training remained relatively low.

Farmers’ participation behavior and consistency in adaptation measures across community groups differed. Regarding irrigation measures (Figure 5a), the upstream Pahadi (A) and downstream Tharu (G and E) rural groups were highly adapted to various measures, using both natural and hybrid methods. This also shows that the downstream groups had a higher level of agreement with various measures than the upstream group. In contrast, diverse cultural communities, such as group B, had great involvement in only a few measures, and group C had very low involvement and consistency in all adaptation measures, similar to cities located in groups D and F, showing the potential for lower consent for any adaptation measures.

Figure 5b shows farmers’ participation in riverside measures within community groups. The riverside rural communities A, F, and G highly adopted buffer zone vegetation with consistency of agreement, and similarly, groups F and G also gave high priority to the adoption of grey embankments. Interestingly, none of the respondents from group F were involved in either adaptation measures. In contrast, groups B, C, and D had very low participation in all adaptation measure options, similar to the irrigation system results.

The results in Figure 6a show that the downstream riverside groups (F and G) participated in developing strategies for watershed ecosystem conservation and buffer zone protection. The upstream group A participated in buffer zone protection and flood risk measures. Respondents of other community groups B, C, D, and E had almost no participation in any given option. Interestingly, among the various adaptation measures, the results in Figure 6b indicate that farmers actively engaged in several strategies to ensure farming continuity. Among the given strategies, most groups adopted manpower sharing and farming loans. We also observed that agricultural training was the least adopted by all community groups.

3.4. Factors Influencing Variables of Farmers’ Community-Based Adaptation Practices

We employed the most widely adopted structural flood risk mitigation measures from each structural method to represent farmers’ participation in community-based adaptation for different water resources: water flow management (from irrigation systems as an indigenous method), buffer zone vegetation (from riverside management as a non-indigenous method), and a decision tree (CHAID) model for comparative analysis of the most important impact factors.

The overall percentages of correct predictions of water flow management were 85.5% and 92.5% for buffer zone vegetation. The model’s robust statistical capabilities and holistic approach validated the findings and contributed significantly to our understanding of adaptation within these communities.

In the classification tree for water flow management (Figure 7), we found a total of 10 nodes and 7 terminal nodes at the three depths, and the overall correct percentage was 85.5%. Among the total independent variables, only four were found to be significantly associated with farmers’ community-based adaptation behaviors, as shown in the tree: irrigation water accessibility, community diversity or groups, landholding, and irrigation water intake methods. The hierarchical tree began with a root node, showing that 53% of the 200 farmers had adopted water flow management. The first influencing factor was water accessibility to the irrigation system, and farmers’ adaptation behavior was split according to the level of water accessibility, such that the more access to water, the higher the involvement (Nodes 1, 2, 3, and 4). At Node 1, farmers whose paddy farming mostly depended on rainwater or underground water had the lowest participation in adaptation measures (14.9%). Only 37.2% of farmers whose land had low and medium irrigation water accessibility participated in adaptation (Node 2). Among them, farmers belonging to community groups B, E, and G had higher involvement (Node 5, 68.4%) than those belonging to community groups C, D, and F (Node 6, 12.5%). Farmers who had access to water for more than one monsoon period (medium to relatively high) had 73.9% participation. Node 4 showed farmers’ higher adaptation behavior with higher water accessibility (94%). However, participation varied according to land ownership and water intake. The results show that other (>2.5 Kattha) landowners had higher participation (Node 8, 98.2%) than the small ones (Node 7, 70%), and those who could consume water naturally had more participation (Node 9, 100%) than those who used artificial intake methods (Node 10, 80%).

The CHAID analysis results of buffer zone vegetation measures (Figure 8) showed that the tree had nine nodes, two branches, and six terminal nodes. The tree was permed at two depths, and the overall correct percentage was 92.5%. The results for Node 0 showed that 54% of the farmers were associated with the local government for vegetation measures in buffer zone areas. Of the 14 variables, we found only three significant and associated variables: community diversity, average monthly income, and river distance. The total tree was distributed by community diversity based on the landscape as per the similarities in each node, such as riverside (Node 1), river distant (Node 2), and city located along the river (Node 3). Farmers of riverside communities from upstream (A) and downstream (F and G) participated more in buffer zone vegetation (Node 1 = 98.8%) than in other communities (Node 2 = 30.4% and Node 3 = 0%). However, the terminal decisions were split by income level (nodes 4 and 5), where farmers who had total participation had lower incomes (≤65,000) (Node 4, 100%), and those who had lower participation had higher incomes (>65,000) (Node 5, 85.7%). The river-distant community groups (B, C, and E) had much lower participation (Node 2, 30.4%) than the riverside groups, whose decisions to participate were influenced by their farming circumstances or by farmland distance from the river (Nodes 6, 7, and 8). It was found that 71.9% of the lower river distance farmers participated in adaptation measures (Node 6), which was the opposite for medium river distance (Node 7, 28.6%), whereas almost none of the farmers had land located at a higher distance (Node 8, 5.7%).

4. Discussion

4.1. Farming Scenario and Adaptation Practices of Terai Farmers

This study explored a multidimensional society in Terai, Nepal, at a micro-spatial level, which can deprive farmers of leveraging not only mutual activities [17,18,38] but also different natural resources in different manners. Considering the entire approach of farmers to irrigation systems, water intake methods, and river management, we found less utilization of unused land and did not find measures that adjusted water resources to create more water spaces around the riverside, channels, and paddy farmland to decrease water runoff and flood risk in human settlements by improving water retention capacity [58,59]. In addition, almost no one participated in strategies or programs to update information on the local ecosystem and flood risk. Additionally, there was a notable lack of community participation in programs addressing ecosystems and flood risk awareness, contributing to prolonged floodwater stagnation due to inadequate drainage facilities [46,47], not only in farming areas but also in the entire catchment area [58,59]. We observed significant disparities in irrigation distribution, water accessibility, adaptation practices, and participation among community groups even within the same area. River management practices also varied, with higher engagement observed in rural riverside communities involved in ecosystem service strategies.

4.2. Important Factors for Community-Based Eco-DRR

This study used the CHAID decision tree model to analyze the key factors influencing farmers’ participation in community-based adaptation, considering community diversity and household attributes. Despite the challenges of small and unequal sample sizes, the model effectively highlighted how various groups engage in adaptation strategies, emphasizing the significance of farming circumstances and community characteristics over personal traits.

Variables such as farming condition and community diversity are crucial for irrigation management. Farms with consistent water availability were more involved in community efforts, which aligns with previous findings that greater access to water resources is more likely to stimulate innovation and investment patterns in conservation practices that augment and enhance the functions of irrigation resources [39]. Larger landholders using natural water intake methods were more engaged as land size affected livelihood dependency [38,39] and adaptation capabilities [57]. Natural intake methods provided access to optimum facilities for farming continuity. Poor water access was correlated with lower involvement, revealing a cyclical issue: a lack of resources reduced participation, which in turn limited resource improvement [38,60]. Community diversity was a critical factor, particularly in areas with limited access to water. Urbanized, culturally mixed communities and marginalized rural groups showed less participation, likely because of cultural differences and socioeconomic disparities that hindered collaboration [33,34]. These findings suggest that the impact of cultural diversity in communities on adaptation depends on the variability of diversity.

For buffer zone vegetation, we found different impact factors, such as farmers’ different needs and capabilities within different adaptation methods [27]. This finding shows that community diversity is the most crucial factor in adapting measures for river management. Riverside communities participated more frequently than distant communities. Among river-distant community groups, farmers whose land was closer to the river had higher participation in adaptation practices compared to those farther away. This output clarified that, for accessibility, farmers need direct relationships with local resources as close inhabitants or landowners to increase their participation in community-based activities. Further, river distance was also associated with the possibility of flood risk: “The more threatening the perceived risk by authorities, the greater the effort towards reducing the risk” [61]. However, we also found that income level or economic heterogeneity created by higher incomes [36] motivated farmers to have lower participation, even with high potential flood risk or when close to local resources. The participation patterns of marginalized groups, such as the Musahar community, differed, highlighting how adaptation capabilities and infrastructure access influence involvement. This suggests that vulnerable groups, who often rely on government-managed resources, require targeted support to enhance their participation [62]. This study also acknowledges that urbanization can detach farmers from local resources [30] and can be more affected by resources that are directly related to the local government.

4.3. Policy Implications

This study offers valuable insight into specific adaptation practices and reflects a diverse range of community experiences, particularly in Nepal, which is significant for developing long-term policies to mitigate flood risk and enhance Eco-DRR practice.

The updated policies on “River and Water-Induced Disaster Management” and “National Irrigation Policy” [63] in Nepal emphasize equitable water distribution, flood mitigation through nature-based solutions, and the active participation of marginalized groups. However, like previous policies [64,65], they fall short of recognizing farming ecosystems (paddy farmland and water resources) as key green infrastructure [58,59] for sustainable and cost-effective flood management solutions, instead focusing on more immediate, rehabilitation-driven responses. Indigenous knowledge and local practices are acknowledged but treated as supplementary, not integral, limiting their impact. While local participation is encouraged, it remains procedural and overlooking deeper sociocultural dimensions, such as mutual decision-making processes. Given the diverse ecological, social, and cultural contexts across Nepal, a one-size-fits-all approach to policy formulation is unlikely to be effective.

This study also advocates for tailored policies that ensure equitable water distribution and equal access to natural resources, which are crucial for enhancing farmers’ participation in adaptation practices. In the context of Eco-DRR, sustainability is a primary goal, focusing on local development, ongoing maintenance, and the conservation of resources. Greater emphasis should be placed on non-structural measures such as providing information, raising awareness, and creating equal opportunities to improve traditional farming water management systems—considered Eco-DRR for mitigating flood risks [5]. Additionally, socioeconomically disadvantaged community groups need empowerment to enhance their adaptive capacities through direct benefits that create “win–win” situations. Local governments should collaborate with communities or landowners exposed to higher flood risks, as they can serve as both primary conservators and active participants in disaster risk mitigation.

5. Conclusions

This study used multiple scales of socioeconomic background at a micro-spatial level to explore the actual socioeconomic circumstances of community groups in Terai, Nepal, and concrete community-based integrated adaptation measures for different water resources and their decision making for different Eco-DRR or adaptation practices under the given circumstances. This study explored the community diversities that exist due to the various cultures and socioeconomic factors and found different approaches to water resources to continue farming, community-based adaptation, and participation in flood risk mitigation in similar territories.

The CHAID method was adopted to highlight important factors associated with farmers’ participation in adaptation practices. Despite the definite relationship between farmers’ participation in adaptation practices and their socioeconomic circumstances, this study only identified important variables with unique relationships. Our results show that irrigation water accessibility, landholding level, water intake methods, alongside community groups diversified by socioeconomic and cultural factors, are key factors influencing farmers’ decisions to participate in the community for water flow management. For buffer zone vegetation, community diversity around the river, household income level, and farm–river distance were the main influencing factors. Notably, it showed different characteristics for community diversity than water flow management, such as riverside versus river-distant communities and rural versus urban communities.

Hence, this study suggests that even at a microspatial level, exhaustive circumstances, such as community diversity, personal attributes, and farming scenarios, are important to comprehensively understand farmers’ participation in adapting to different water resources, as they all predict unique variance. In addition, comparative analyses of different farming approaches to local water resources, such as indigenous and associated stakeholder methods or different ecosystem services, are also important for understanding the relationships between variables for various decision-making processes.

Author Contributions

Conceptualization, Investigation, Methodology, Data Curation, Writing-Original Draft Preparation, Review, Editing, and Validation, S.K.; Conceptualization, Supervision, Writing-Review, and Editing, S.Y. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Data Availability Statement

The data presented in this study are available upon request from the corresponding author. These data are not publicly available because of privacy concerns.

Acknowledgments

We would like to thank Kawasoti Municipality member Simon Mahato and ward office members for facilitating data collection and local information and local social worker Prem Bahadur Mardania of Bagkhor, who arranged the stakeholders for the focus group discussion and continued assisting and providing local information. We also thank Kul Prasad Bhusal from Lumbini Printing and Upset, local students Sarita Majhi and Prem Tharu, and Green Village resort family members for their assistance, feedback, and contributions to the field survey. Finally, our heartfelt thanks go to all the local farmers who mobilized and participated as respondents in the questionnaire survey and included us in their adaptation practices despite their busy schedules.

Conflicts of Interest

The authors declare no conflicts of interest.

Appendix A. Additional Figures

Figure A1. Access to irrigation channels, their water accessibility, and management across community groups (A–G, accordingly). Groups (A,E,G) had better facilities for irrigation channels and water and better management. Groups (B,C) had access to channels but poor management. The majority of farming areas in group (D) had a higher irrigation distance. Group (F) had access to channels; and both groups (D,F) depended on groundwater or rainwater for farming.

Figure A2. Community participation in irrigation channel management within riverside areas and river management (gray embankment measures).

Appendix B. Additional Information

Questionnaire survey for “Community-based Farming Water Resource Management and Important Factors for Adaptation Practices in Terai, Nepal” across the Kawasoti municipality.

1. Farmers’ household socio-economic and farming capital (Q1–13, please select one option).

A. Farmers’ personal attributes:

(Q1) Your gender:
	☐ Male ☐ Female
(Q2) Your age (years):
	☐ 18–25 ☐ 26–35 ☐ 36–45 ☐ 46–55 ☐ 56–65 ☐ 66–75 ☐ 76 or above
(Q3) Your farming experience (in years):
	☐ ≤5 ☐ 6–10 ☐ 11–20 ☐ 21–30 ☐ 31–40 ☐ >40
(Q4) Your household’s average monthly income (in Nepali Rupees):
	☐ <20,000 ☐ 20,000–40,000 ☐ 40,000–60,000 ☐ >60,000
(Q5) Your educational background (education level):
	☐ Illiterate ☐ Primary ☐ Secondary ☐ High school ☐ College
(Q6) Higher education level in your family member:
	☐ Illiterate ☐ Primary ☐ Secondary ☐ High school ☐ College
(Q7) Main income source for your household:
	☐ Farming ☐ Non-farming (salary, remittances, business, wages, etc.)
(Q8) How much is your total landholding for paddy farming (in Kattha)?
	(………..) Kattha.

B. Farming Circumstances

(Q9) What is the distance of main irrigation channels from your farmland?
	☐ None (attached to farmland)	☐ Very low (≤100 m)	☐ Low (100–300 m)
	☐ Medium (300–500 m)	☐ Relatively high (500–1000 m)	☐ Very high (>1000 m)
(Q10) What is the water accessibility to main irrigation channels for your farming continuity (based on paddy farming)?
	☐ Very low (mostly depends on rain or underground water)	☐ Low (only access after heavy rain)
	☐ Medium (access during the monsoon; June, July, and August)	☐ Relatively high (access for more than 6 months)
	☐ Very high (continuous or access for almost a whole season)
(Q11) How do you intake water from the main irrigation channels?
	☐ Natural (direct from irrigation or use small canals)	☐ Artificial (use motor pipes or from (others’) land)
(Q12) What is the river distance from your farmland (in meters)?
	☐ None (attached to farmland)	☐ Very low (≤100 m)	☐ Low (100–300 m)
	☐ Medium (300–500 m)	☐ Relatively high (500–1000 m)	☐ Very high (>1000 m)
(Q13) What is the flood impact level on your farmland?
	☐ None/no impact	☐ Relatively high (high level of waterlogging, infrastructure damages, and replantation or delayed plantation)
	☐ Very low (effects to plantation or cultivation sometimes)
	☐ Low (effects to plantation or cultivation sometimes)	☐ Very high (speed run-off, soil quality damages, infrastructure damages, and replantation or delayed plantation)
	☐ Medium (waterlogged and replantation or delayed plantation)

2. Farmers’ community-based adaptation measures for farming water resources to mitigate flood risk (Q14–17 Yes/No choice. Yes = if participation (multiple choices), No = if no participation).

A. Structural measure

(Q14) Do you participate in any community-based actions to manage your mutual or main irrigation channels to mitigate flood risks? If yes, in which measures do you participate?
	☐ Water flow management	☐ Sandbag dam	☐ Gray embankment	☐ None
	☐ Natural dam	☐ Sandbag embankment	☐ Rehabilitation
(Q15) Do you participate in any community-based actions to manage rivers and mitigate flood risk? If yes, in which measures do you participate?
	☐ Buffer zone vegetation	☐ Unused land use (water storage, paddy farming, or wet land)	☐ None
	☐ Soil/sand protection	☐ Gray embankment

B. Non-structural measures

(Q16) Do you participate in any organization (I/NGOs or local governments) or programs regarding watershed buffer zones or flood risk management? If yes, in which programs do you participate?
	☐ Watershed ecosystem conservation	☐ Buffer zone protection
	☐ Flood risk mitigation	☐ None
(Q17) For your farming continuity, for which purposes do you participate or engage in local organizations and cooperate with other farmers (such as financial institutes, farmers’ development committees, and mutual actions or cooperation with other farmers for paddy plantation and cultivation)?
	☐ Farming loan	☐ Agricultural trainings	☐ Tools sharing
	☐ Seeds and fertilizers	☐ Manpower sharing	☐ None

Appendix C. Additional Table

Table A1. Multivariate (MANOVA) tests (after post hoc test).

Effect		Value	F	Hypothesis df	Error df
Intercept	Pillai’s Trace	0.935	133.171 ^b	19.000	175.000
	Wilks’ Lambda	0.065	133.171 ^b	19.000	175.000
	Hotelling’s Trace	14.459	133.171 ^b	19.000	175.000
	Roy’s Largest Root	14.459	133.171 ^b	19.000	175.000
Community	Pillai’s Trace	2.829	8.451	114.000	1080.000
	Wilks’ Lambda	0.012	10.348	114.000	1014.799
	Hotelling’s Trace	8.180	12.437	114.000	1040.000
	Roy’s Largest Root	3.862	36.591 ^c	19.000	180.000

The MANOVA results indicated significant differences in the total adaptation measures across community groups (Pillai’s Trace = 2.829, F(114, 1080) = 8.451, p < 0.001), confirming the validity of the analysis despite the unequal sample sizes of the community groups. F-value ^b: Indicates that the F-statistic has been adjusted for unequal sample sizes across community groups (based on Pillai’s Trace, Wilks’ Lambda, Hotelling’s Trace, and Roy’s Largest Root). This adjustment ensures the robustness of the MANOVA results. F-value ^c: Highlights that Roy’s Largest Root has the largest F-value among the multivariate tests, suggesting a pronounced effect size for specific dimensions in the dataset.

References

Rogelj, J.; Meinshausen, M.; Knutti, R. Global warming under old and new scenarios using IPCC climate sensitivity range estimates. Nat. Clim. Change 2012, 2, 248–253. [Google Scholar] [CrossRef]
Renaud, F.G.; Sudmeier-Rieux, K.; Estrella, M.; Nehren, U. (Eds.) Ecosystem-Based Disaster Risk Reduction and Adaptation in Practice; Springer International Publishing: Cham, Germany, 2016; Volume 42. [Google Scholar]
Natuhara, Y. Ecosystem services by paddy fields as substitutes of natural wetlands in Japan. Ecol. Eng. 2013, 56, 97–106. [Google Scholar] [CrossRef]
Ohgushi, K.; Nakashima, H.; Hino, T.; Morita, T.; Jansen, T. A Study on Jobaru river basin management by numerical simulations of flooding and sediment deposition with field survey. Lowl. Technol. Int. 2016, 18, 23–30. [Google Scholar] [CrossRef] [PubMed]
Estrella, M.; Saalismaa, N.; Renaud, F.G. Reduction (Eco-DRR): An Overview. The Role of Ecosystems in Disaster Risk Reduction; UNU Press: Tokyo, Japan, 2013; p. 26. [Google Scholar]
Osawa, T.; Nishida, T.; Oka, T. Potential of mitigating floodwater damage to residential areas using paddy fields in water storage zones. Int. J. Disaster Risk Reduct. 2021, 62, 102410. [Google Scholar] [CrossRef]
Nelson, E.; Mendoza, G.; Regetz, J.; Polasky, S.; Tallis, H.; Cameron, D.; Chan, K.M.; Daily, G.C.; Goldstein, J.; Kareiva, P.M.; et al. Modeling multiple ecosystem services, biodiversity conservation, commodity production, and tradeoffs at landscape scales. Front. Ecol. Environ. 2009, 7, 4–11. [Google Scholar] [CrossRef]
Reddy, V.R.; Saharawat, Y.S.; George, B. Watershed management in South Asia: A synoptic review. J. Hydrol. 2017, 551, 4–13. [Google Scholar] [CrossRef]
Wolka, K.; Uma, T.; Tofu, D.A. The role of integrated watershed management in climate change adaptation for small-scale farmers in Southwest Ethiopia. Environ. Sustain. Indic. 2023, 19, 100260. [Google Scholar] [CrossRef]
Sriyana, I.; De Gijt, J.G.; Parahyangsari, S.K.; Niyomukiza, J.B. Watershed management index based on the village watershed model (VWM) approach towards sustainability. Int. Soil Water Conserv. Res. 2020, 8, 35–46. [Google Scholar] [CrossRef]
Mekonnen, M.; Keesstra, S.D.; Baartman, J.E.; Ritsema, C.J.; Melesse, A.M. Evaluating sediment storage dams: Structural off-site sediment trapping measures in Northwest Ethiopia. Cuad. Investig. Geogr. 2015, 41, 7–22. [Google Scholar] [CrossRef]
Dalimunthe, S.A. Who manages space? Eco-DRR and the local community. Sustainability 2018, 10, 1705. [Google Scholar] [CrossRef]
Fanzo, J. From big to small: The significance of smallholder farms in the global food system. Lancet Planet. Health 2017, 1, e15–e16. [Google Scholar] [CrossRef] [PubMed]
Posthumus, H.; Hewett, C.J.M.; Morris, J.; Quinn, P.F. Agricultural land use and flood frisk management: Engaging with stakeholders in North Yorkshire. Agric. Water Manag. 2008, 95, 787–798. [Google Scholar] [CrossRef]
IPCC. Climate Change 2014: Impacts, adaptation, and vulnerability. Summary for policymakers. In IPCC WGII AR5; IPCC: Geneva, Switzerland, 2014. [Google Scholar]
Devkota, R.P.; Maraseni, T. Flood risk management under climate change: A hydro-economic perspective. Water Sci. Technol. Water Supply 2018, 18, 1832–1840. [Google Scholar] [CrossRef]
Fekete, A.; Aslam, A.B.; de Brito, M.M.; Dominguez, I.; Fernando, N.; Illing, C.J.; Apil, K.K.; Mahdavian, F.; Norf, C.; Platt, S.; et al. Increasing flood risk awareness and warning readiness by participation–but who understands what under ‘participation’? Int. J. Disaster Risk Reduct. 2021, 57, 102157. [Google Scholar] [CrossRef]
Sudmeier-Rieux, K.; Paleo, U.F.; Garschagen, M.; Estrella, M.; Renaud, F.G.; Jaboyedoff, M. Opportunities, incentives and challenges to risk sensitive land use planning: Lessons from Nepal, Spain, and Vietnam. Int. J. Disaster Risk Reduct. 2015, 14, 205–224. [Google Scholar] [CrossRef]
Yawson, D.O.; Adu, M.O.; Armah, F.A.; Kusi, J.; Ansah, I.G.; Chiroro, C. A needs-based approach for exploring vulnerability and response to disaster risk in rural communities in low income countries. Int. J. Disaster Risk Reduct. 2015, 14, 347–356. [Google Scholar]
Rizvi, A.R.; Baig, S.; Verdone, M. Ecosystems Based Adaptation: Knowledge Gaps in Making an Economic Case for Investing in Nature Based Solutions for Climate Change; IUCN: Gland, Switzerland, 2015; p. 48. [Google Scholar]
Fenta, A.A.; Tsunekawa, A.; Haregeweyn, N.; Tsubo, M.; Yasuda, H.; Kawai, T.; Berihun, M.L.; Ebabu, K.; Sultan, D.; Mekuriaw, S. An integrated framework for improving watershed management planning. Environ. Res. 2023, 236, 116872. [Google Scholar] [CrossRef]
Mekonnen, M.; Fisseh, G.; Mulatie, N. Integrated watershed management approach after three decades, Northwest Ethiopia. Northwest Ethiop. 2022. [Google Scholar] [CrossRef]
Chuenchum, P.; Meneesrikum, C.; Teerapanuchaikul, C.; Sriariyawat, A. Community participation and effective water management: A study on water user organizations (WUOs) in Thailand. World Dev. Perspect. 2024, 34, 100589. [Google Scholar] [CrossRef]
Chen, J.; Yin, S.; Yang, X. The impact of adaptive management on community resilience in arid rural areas facing environmental change: An integrated analytical framework. Environ. Sci. Policy 2023, 150, 103589. [Google Scholar] [CrossRef]
Ambuehl, B.; Kunwar, B.M.; Schertenleib, A.; Marks, S.J.; Inauen, J. Can participation promote psychological ownership of a shared resource? An intervention study of community-based safe water infrastructure. J. Environ. Psychol. 2022, 81, 101818. [Google Scholar] [CrossRef]
Sloot, D.; Jans, L.; Steg, L. Can community energy initiatives motivate sustainable energy behaviours? The role of initiative involvement and personal pro-environmental motivation. J. Environ. Psychol. 2018, 57, 99–106. [Google Scholar] [CrossRef]
Murty, M.N. Management of common property resources: Limits to voluntary collective action. Environ. Resour. Econ. 1994, 4, 581–594. [Google Scholar] [CrossRef]
Karlsson, M.; Hovelsrud, G.K. Local collective action: Adaptation to coastal erosion in the Monkey River Village, Belize. Glob. Environ. Change 2015, 32, 96–107. [Google Scholar] [CrossRef]
Mengistu, F.; Assefa, E. Towards sustaining watershed management practices in Ethiopia: A synthesis of local perception, community participation, adoption, and livelihoods. Environ. Sci. Policy 2020, 112, 414–430. [Google Scholar] [CrossRef]
Buchori, I.; Zaki, A.; Pangi, P.; Sejati, A.W.; Pramitasari, A.; Liu, Y. Adaptation strategies and community participation in government-led mitigation projects: A comparison between urban and suburban communities in Pekalongan, Indonesia. Int. J. Disaster Risk Reduct. 2022, 81, 103271. [Google Scholar] [CrossRef]
Banks, G. Understanding ‘resource’ conflicts in Papua New Guinea. Asia Pac. Viewp. 2008, 49, 23–34. [Google Scholar] [CrossRef]
Sahoo, R.S.; Swain, M.S. Contribution of common property resources for sustainable rural livelihoods in Odisha: Prospects and constraints. J. Rural Dev. 2013, 32, 245–262. [Google Scholar]
Raheem, H.M.J.; Mayo, S.M.; Kamel, A.Y.; Maqbool, R.; Mohamed, M.M.A.; Maraqa, M.A.; Gouda, H.M.; Hamdani, F.K.; Butt, T.E. Assessing the relationship between cultural diversity and disaster preparedness: A case study of flood hazard for Northern and Southern Punjab. Int. J. Disaster Risk Reduct. 2023, 84, 103452. [Google Scholar] [CrossRef]
Pudasaini, N. Community Based Climate Change Vulnerability Assessment: A Case of Tharu Community. Ph.D. Thesis, Tribhuvan University, Nawalparasi, Nepal, 2015. [Google Scholar]
Tiwari, K.R.; Sitaula, B.K.; Nyborg, I.L.; Paudel, G.S. Determinants of farmers’ adoption of improved soil conservation technology in a middle mountain watershed of Central Nepal. Environ. Manag. 2008, 42, 210–222. [Google Scholar] [CrossRef]
Dhakal, T.R.; Davidson, B.; Farquharson, B. Factors affecting collective actions in farmer-managed irrigation systems of Nepal. Agriculture 2018, 8, 77. [Google Scholar] [CrossRef]
Bahinipati, C.S.; Venkatachalam, L. What drives farmers to adopt farm-level adaptation practices to climate extremes: Empirical evidence from Odisha, India. Int. J. Disaster Risk Reduct. 2015, 14, 347–356. [Google Scholar] [CrossRef]
Below, T.B.; Mutabazi, K.D.; Kirschke, D.; Franke, C.; Sieber, S.; Siebert, R.; Tscherning, K. Can farmers’ adaptation to climate change be explained by socio-economic household-level variables? Glob. Environ. Change 2012, 22, 223–235. [Google Scholar] [CrossRef]
Maya, K.A.; Sarker, M.A.R.; Gow, J. Factors influencing rice farmers’ adaptation strategies to climate change and extreme weather event impacts in Bangladesh. Clim. Change Econ. 2019, 10, 1950012. [Google Scholar] [CrossRef]
Sjöberg, L.; Moen, B.E.; Rundmo, T. Explaining risk perception: An evaluation of the psychometric paradigm in risk perception research. Rotunde Publ. Rotunde. 2004, 84, 55–76. [Google Scholar]
Arbuckle, J.G.; Morton, L.W.; Hobbs, J. Farmer beliefs and concerns about climate change and attitudes toward adaptation and mitigation: Evidence from Iowa. Clim. Change 2013, 118, 551–563. [Google Scholar] [CrossRef]
Bajracharya, S.R.; Mool, P.K.; Shrestha, B.R. Global climate change and melting of Himalayan glaciers. In Melting Glaciers and Rising Sea Levels: Impacts Implications; ICFAI University Press: Hyderabad, India, 2008; pp. 28–46. [Google Scholar]
Bhattarai, K.; Conway, D.; Bhattarai, K.; Conway, D. Demography, caste/ethnicity, federalism, and socioeconomic conditions in relation to contemporary Environment. Contemporary Environmental Problems in Nepal: Geographic Perspectives; Springer: Berlin/Heidelberg, Germany, 2021; pp. 37–114. [Google Scholar]
Malla, G. Climate change and its impact on Nepalese agriculture. J. Agric. Environ. 2008, 9, 62–71. [Google Scholar] [CrossRef]
Dewan, T.H. Societal impacts and vulnerability to floods in Bangladesh and Nepal. Weather Clim. Extrem. 2015, 7, 36–42. [Google Scholar] [CrossRef]
Regmi, B.R.; Star, C. Exploring the policy environment for mainstreaming community-based adaptation (CBA) in Nepal. Int. J. Clim. Change Strat. Manag. 2015, 7, 423–441. [Google Scholar] [CrossRef]
Devkota, R.P.; Maraseni, T.N.; Cockfield, G.; Devkota, L.P. Flood vulnerability through the eyes of vulnerable people in Mid-Western Terai of Nepal. J. Earth Sci. Clim. Change 2013, 4, 1–7. [Google Scholar] [CrossRef]
Saru, B.D.; Adhikari, J.N.; Bhattarai, B.P. Medicoethnobiology of Musahar community in Nawalpur District, Nepal. ZOO-J. 2021, 6, 9–23. [Google Scholar] [CrossRef]
Tiwari, K.R.; Rayamajhi, S.; Pokharel, R.K.; Balla, M.K. Determinants of the climate change adaptation in rural farming in Nepal Himalaya. Int. J. Multidiscip. Curr. Res. 2014, 2, 2321–3124. [Google Scholar]
Sharma, R. Community-based flood risk management: Local knowledge and actor’s involvement approach from lower Karnali River Basin of Nepal. J. Geosci. Environ. Protect. 2021, 9, 35–65. [Google Scholar] [CrossRef]
Rapsomanikis, G. The Economic Lives of Smallholder Farmers: An Analysis Based on Household Data from Nine Countries; Food and Agriculture Organization of the United Nations: Rome, Italy, 2015. [Google Scholar]
Pandey, V.P.; Dhaubanjar, S.; Bharati, L.; Thapa, B.R. Spatio-temporal distribution of water availability in Karnali-Mohana Basin, Western Nepal: Hydrological model development using multi-site calibration approach (Part-A). J. Hydrol. Reg. Stud. 2020, 29, 100690. [Google Scholar] [CrossRef]
Önder, E.; Uyar, S. CHAID analysis to determine socioeconomic variables that explain students’ academic success. Univ. J. Educ. Res. 2017, 5, 608–619. [Google Scholar] [CrossRef]
Strzelecka, A.; Zawadzka, D. The use of Chi-Squared Automatic Interaction Detector (CHAID) analysis to identify characteristics of agricultural households at risk of financial self-exclusions. Procedia Comput. Sci. 2023, 225, 4443–4452. [Google Scholar] [CrossRef]
Kass, G.V. An exploratory technique for investigating large quantities of categorical data. J. Roy. Stat. Soc. Ser. C Appl. Stat. 1980, 29, 119–127. [Google Scholar] [CrossRef]
Milanović, M.; Stamenković, M. CHAID Decision Tree: Methodological Frame and Application. Econ. Themes 2016, 54, 563–586. [Google Scholar] [CrossRef]
Arunrat, N.; Wang, C.; Pumijumnong, N.; Sereenonchai, S.; Cai, W. Farmers’ intention and decision to adapt to climate change: A case study in the Yom and Nan basins, Phichit Province of Thailand. J. Clean. Prod. 2017, 143, 672–685. [Google Scholar] [CrossRef]
Senoo, H.; Ishikawa, T. Estimation of flood control function of Kasumi levee system on the Kurobe alluvial fan in the Edo era by numerical flow simulation. J. Jpn. Soc. Civ. Eng. Ser. B1 Hydraul. Eng. 2018, 74, I_1411–I_1416. [Google Scholar] [CrossRef]
Kuriqi, A.; Hysa, A. Multidimensional aspects of floods: Nature-based mitigation measures from basin to river reach scale. In Nature-Based Solutions for Flood Mitigation: Environmental and Socio-Economic Aspects; Springer International Publishing: Cham, Germany, 2021; pp. 11–33. [Google Scholar]
Shiferaw, B.; Bantilan, M.C.S. Agriculture, rural poverty, and natural resource management in less-favored environments: Revisiting challenges and conceptual issues. Food Agric. Environ. 2004, 2, 328–339. [Google Scholar]
Slovic, P. The risk game. Reliab. Eng. Syst. Saf. 1998, 59, 73–77. [Google Scholar] [CrossRef]
Sujakhu, N.M.; Ranjitkar, S.; He, J.; Schmidt-Vogt, D.; Su, Y.; Xu, J. Assessing the livelihood vulnerability of rural indigenous households to climate changes in Central Nepal, Himalaya. Sustainability 2019, 11, 2977. [Google Scholar] [CrossRef]
Department of Water Resources and Irrigation (DWRI). River and Water-Induced Disaster Management National Policy 2080 and National Irrigation Policy 2080. Available online: https://dwri.gov.np/document/polices (accessed on 11 December 2024).
Sharma, L.; Gupta, N.; Basnayake, S. Assessment of Water Sector Policies and Guidelines of Nepal: Identifying Gaps and Addressing Needs; Asian Disaster Preparedness Center: Bangkok, Thailand, 2023. [Google Scholar]
Paudel, P.K.; Parajuli, S.; Sinha, R.; Bohara, M.; Abedin, A.; Adhikari, B.R.; Gautam, S.; Bastola, R.; Pal, I.; Huntington, H.P. Integrating traditional and local knowledge into disaster risk reduction policies: Insights from Nepal, India, and Bangladesh. Environ. Sci. Policy 2024, 159, 103825. [Google Scholar] [CrossRef]

Figure 1. Study and data collection area within the Kawasoti Municipality farming area with administration and elevation distribution (a) and sub-streamlines with the mainstream river, Narayani (b).

Figure 2. (a) Irrigation distance from farmland and water accessibility within community groups. (b) Irrigation water intake methods within community groups (%, N = 200).

Figure 3. River distance from farmland and flood impact within community groups (%, N = 200).

Figure 4. Farmers’ community-based integrated farming water resource management in total (%, N = 200). Structural measures for (a) irrigation system and (b) riverside and non-structural measures or strategies for (c) ecosystem services and (d) farming continuity. * Water flow management (widening, deepening, and cleaning). ** Unused land (public land uses for water storage, paddy farming, or vegetation).

Figure 5. Farmers’ community-based structural measures (a) for irrigation systems and (b) for riversides within community groups (%, N = sample size of each community groups). * Water flow management (widening, deepening, and cleaning). ** Unused land use (public land uses for water storage, paddy farming, or vegetation).

Figure 6. Farmers’ community-based non-structural measures for (a) irrigation systems and (b) riverside community groups (%, N = sample size of each community groups).

Figure 7. Decision tree (CHAID) for explaining farmers’ total adaptation behavior for community-based water flow management in irrigation channels and most important associated factors.

Figure 8. Decision tree (CHAID) for explaining farmers’ total adaptation behavior for community-based buffer zone vegetation on riverside areas and most important associated factors.

Table 1. Community diversity and data collection areas in Kawasoti Municipality.

Community Groups	Number of Farmers	Ward Numbers	Cultural Distribution	Landscape	Flood Type
A	27	4 and 5	Typically migrant or Pahadi (mountainous)	Upstream of flat, rural area attached to sub-stream river	Speed run-off, pebble, sandy
B	32	9	Mixed (Tharu dominated)	Mid-stream, rural area	Sandy, waterlogged
C	32	7 and 10	Mixed (Migrant dominated)	Middle downstream, urban area	Sandy, waterlogged
D	27	14	Typically migrant	Downstream, urban area	Waterlogged
E	28	14 and 15	Typically Tharu	Downstream, rural area	Waterlogged
F	28	13 and 15	Marginal groups of Terai ethnicity (Mushar, Bote, Musahar, Majhi)	Downstream, attach to both mainstream and sub-stream rivers	High, waterlogged
G	26	15	Madhesi	Downstream, attach to both mainstream and sub-stream rivers	High, waterlogged

Table 2. Description of variables for statistical analysis.

Variables	Measures	Description
Community groups	Nominal	A, B, C, D, E, F, G
Gender	Nominal	Male, Female
Age	Continuous	Years
Farming experience	Continuous	Years
Household’s average monthly income	Continuous	NPR (Nepalese rupee, NPR 1 = USD 0.0077 during the survey)
Education level of head of the family	Ordinal	1 = Illiterate, 2 = Primary, 3 = Secondary, 4 = High school, 5 = College
Higher education in family member	Ordinal	1 = Illiterate, 2 = Primary, 3 = Secondary, 4 = High school, 5 = College
Main income source	Nominal	Farming, Off-farming
Land holding	Continuous	(Kattha, 1 Kattha = 0.0126 hectare)
Irrigation distance with farmland	Ordinal	1 = None, 2 = Very low, 3 = Low, 4 = Medium, 5 = Relatively high, 6 = Very high
Irrigation water accessibility	Ordinal	1 = Very low, 2 = Low, 3 = Medium, 4 = Relatively high, 5 = Very high
Irrigational water intake method	Nominal	Natural, Artificial
River distance	Ordinal	1 = None, 2 = Very low, 3 = Low, 4 = Medium, 5 = Relatively high, 6 = Very high
Level of flood impact	Ordinal	1 = None, 2 = Low, 3 = Medium, 4 = Relatively high, 5 = Very high

Table 3. Farmers’ personal attributes.

Socioeconomic Characteristics	Description	Number of Respondents	Proportion (%)
Gender	Male	94	47.0
Gender	Female	106	53.0
Age	18–35	45	22.5
	36–55	96	48.0
	56–65	37	18.5
	>65	22	11.0
Farming experience	≤10	39	19.5
	11–20	64	32.0
	21–30	48	24.0
	31–40	34	17.0
	>40	15	7.5
Household’s average monthly income (Nepalese Rupee, NPR) (NPR 1 = USD 0.0077 during the survey period)	<20,000	112	56.0
	20,000–40,000	53	26.5
	40,000–60,000	23	11.5
	>60,000	12	6.0
Education level of head of the family	Illiterate	55	27.5
	Primary	57	28.5
	Secondary	58	29.0
	High school	16	8.0
	College	14	7.0
Higher education level of family members	Illiterate	8	4.0
	Primary	24	12.0
	Secondary	81	40.5
	High school	50	25.0
	College	37	18.5
Main income source	Farming	73	36.5
Main income source	Non-farming	127	63.5
Land holding (Kattha, 1 Kattha = 0.0126 hectare)	≤5	46	22.9
	>5–10	51	25.4
	>10–20	71	35.3
	>20–25	20	10.0
	>25	12	6.0

Table 4. Farming scenario (N = 200).

Farming Characteristics	Description	Number of Respondents	Proportion (%)
Main irrigation channel’s distance from farmland (based on meters)	None (attached)	82	41.0
	Very low (<100)	60	30.0
	Low (100–300)	24	12.0
	Medium (300–500)	23	11.5
	Relatively high (500–1000)	3	1.5
	Very high (>1000)	8	4.0
Irrigation water accessibility	Very low (mostly depends on rain or underground water)	67	33.5
	Low (only access after heavy rain)	32	16.0
	Medium (access during the monsoon; June–August)	31	15.5
	Relatively high (access for more than 6 months)	24	12.0
	Very high (continuous or access for almost a whole season)	46	23.0
Irrigation water intake method	Natural	112	56.0
Irrigation water intake method	Artificial	88	54.0
River distance from farmland (based on meters)	None (attached to farmland)	1	0.5
	Very low (<100)	53	26.5
	Low (100–300)	33	16.5
	Medium (300–500)	31	15.5
	Relatively high 500–1000)	26	13.0
	Very high (>1000)	56	28.0
Level of flood impact	Very low	36	18.0
	Low	32	16.0
	Medium	57	28.5
	Relatively high	56	28.0
	Very high	19	9.5

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

© 2024 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).

Share and Cite

MDPI and ACS Style

Karki, S.; Yokota, S. Community-Based Farming Water Resource Management and Important Factors for Adaptation Practices in Terai, Nepal. Water 2025, 17, 47. https://doi.org/10.3390/w17010047

AMA Style

Karki S, Yokota S. Community-Based Farming Water Resource Management and Important Factors for Adaptation Practices in Terai, Nepal. Water. 2025; 17(1):47. https://doi.org/10.3390/w17010047

Chicago/Turabian Style

Karki, Sharada, and Shigehiro Yokota. 2025. "Community-Based Farming Water Resource Management and Important Factors for Adaptation Practices in Terai, Nepal" Water 17, no. 1: 47. https://doi.org/10.3390/w17010047

APA Style

Karki, S., & Yokota, S. (2025). Community-Based Farming Water Resource Management and Important Factors for Adaptation Practices in Terai, Nepal. Water, 17(1), 47. https://doi.org/10.3390/w17010047

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Menu