Analysis of the climate change impact on water availability and the links between water pollution and economy for sustainable water resource management in Kaski District, Nepal

Water is vital for humans, and for economic development. Continued population and economic growth means an increased demand for water worldwide. The global water demand is projected to double to 6,900 billion m³ in 2030 compared with 2005, over 40% higher than the existing reliable, sustainable water supply (Addams et al. 2009). The global water supply is further constrained by increasing pollution. United Nations (UN)-Water (2021) reported that nearly 2.3 billion people live in water-stressed countries, of which 733 million people are in high and critically water-stressed countries. Furthermore, 24–700 million people in some arid and semi-arid places are likely to be displaced due to intense water scarcity by 2030 (UN 2014a).

Water availability in the Asia-Pacific region varies greatly due to the wide range of climates, and water scarcity is a critical issue as this region holds approximately 36% of global water resources but accommodates 60% of the population (UN 2014b). The situation is further worsened by the rapidly growing economy, where excessive withdrawal of water resources, environmentally unfriendly production, and low levels of wastewater treatment are being practised. Consequently, nine out of the 10 most polluted rivers are found in the Asia-Pacific region (Asian Development Bank: ADB 2020). The declining water quality and quantity call into question the continued growth of this region.

Nepal, in South Asia, receives an ample amount of precipitation (average 1,500 mm annually), but its distribution greatly varies with season and location. The majority (80%) of the total rainfall occurs within 4 months (June–September), resulting in a temporary excess of water, causing flooding and water-induced disasters. Rainfall in the remaining 8 months is insufficient and leads to drought and water scarcity in many parts of the country (Merz et al. 2003). As part of the Himalayan range, Nepal is often referred to as the water tower of South Asia; however, the Himalayas have undergone rapid deglaciation with climate change. Glaciers serve as important water sources for rivers in Nepal and India, especially during the dry season and rapid deglaciation is likely to have serious consequences on a regional scale on water resources (Shrestha & Aryal 2011). Concurrently, the changes in the rainfall pattern and the spatial variation, which is more pronounced in Nepal due to diverse topography and altitude, have been linked to the potential impacts on socio-economic activities (Panthi et al. 2015).

Climate change has made water resource management more challenging in this region (Aryal et al. 2019), confirming the need to assess the physical and socio-economic risks (Shrestha & Aryal 2011). Aside from climate change risks, water quality is deteriorating with the growth of socio-economic activities. This has reduced clean water availability, with a significant population forced to use contaminated water (Merz et al. 2003). The large population of this country is confronted with a poor water environment, and major cities, including the capital, face periodic or chronic water scarcity (Pandey 2021).

Increasing environmental stress is expected with rising socio-economic activities. Comprehensive policies that create a balance between socio-economic growth and the environment are essential. However, there is growing awareness of resource and impact decoupling, aimed at reducing environmental stress while increasing economic growth for sustainable development (United Nations Environment Programme: UNEP 2011). Despite considerable efforts, Nepal's attempts to manage water resources have not been effective, mainly due to a lack of integrated and coordinated approaches, linked to the fragmented and weakened institutions with overlapping scopes (Upadhyay & Gaudel 2018).

Nepal recently enacted the National Water Resources Policy (Ministry of Energy Water Resources and Irrigation MoEWRI 2020), emphasising integrated water resource management (IWRM), overcoming the failure of earlier water resources management plans to undertake coordinated and holistic approaches. This policy implements multiple integrated strategies for water resource management and recognises the pivotal role of water in socio-economic development. Water is considered a foundation for national prosperity, and its management is key to achieving several interconnected sustainable development goals (SDGs). Furthermore, the policy emphasises the need to address the challenges posed by increasing socio-economic growth and climate change, while ensuring a reliable supply of good quality water to all sectors and households.

However, several questions remain about the successful implementation of these policies under a paucity of data, especially for water availability, given the limited research and knowledge on the dynamics of water resources, climate change, and socio-economic activities. For instance, the current water policy undertakes water allocation strategies to satisfy the sectoral water demands. Effective implementation of the policy requires information on the local water availability, which usually varies with space and time. An earlier study reported seasonal and temporal changes in rainfall, with potential impacts on agriculture and livestock in the Gandaki River basin (Panthi et al. 2015). No research has yet described how changes in precipitation affect water availability. However, national water policy is concerned with the climate-induced challenges and highlights the need for further research.

Besides water availability, preserving the water environment and implementation of water pollution control strategies are equally important for ensuring the supply of good water quality, and the current water policy priorities polluter pay policies to water pollution controls. The polluter pay model, although a well-known method, has some practical limitations, such as bringing non-point pollution (agriculture) into the levying system. Furthermore, this system requires strict monitoring and enforcement for effective implementation (ADB 2008), which has yet to be developed.

Therefore, all adopted strategies must be carefully analysed and supported by detailed studies. A case study was carried out in Kaski District, Gandaki Province, Nepal, with the aim of providing inputs for decision makers to implement policies for water security at the local level. This work specifically focuses on (i) estimating water availability and demands, (ii) determining the water carrying capacity, expressed as the extent of socio-economic growth supported by the available water, (iii) establishing sectoral roles in water pollution, and (iv) providing policy measures for sustainable water supply and pollution control.

Kaski District is one of 77 districts of Nepal, situated in the Gandaki Province. It lies in central Nepal, part of the Gandaki River Basin (GRB). The basin is party snow fed, contains all agro-ecological zones of the country, and has a very diverse climate (Panthi et al. 2015), making it vulnerable to the impacts of climate change. Water resources management is crucial for a reliable water supply as this district is a popular tourist destination and has the largest metropolitan city in Nepal.

Water is abundant in this area; however, accessibility for domestic and agricultural use is still limited. For instance, Pokhara Metropolitan City (PMC), faces water supply shortages: as of 1971, PMC had a water supply of 6.5 and 3.4 million litres per day (MLD) in wet and dry seasons, respectively, against a demand for domestic water of 8.0 MLD. Similarly, in 2011, the average daily demand was 42.0 MLD, of which the Nepal Water Supply Corporation (NWSC) could supply 24.5 MLD in the wet season and 21.3 MLD in the dry season (NWSC 2011).

Agriculture dominates the local gross domestic product (GDP), contributing nearly 23% of the district GDP and providing 40.6% of the local employment (District Statistics Office: DSO 2017). Rice, maize, millet, wheat, and barley are the major cereal crops and apple, potato, bean, oil seed, and herbal products are the main cash crops. About 52% of the total cultivated land has access to irrigation, which is predominantly fed by surface water (rivers, lakes, and ponds), as per the national census of agriculture 2011–2012.

According to the manufacturing census 2011–2012, the district has 193 industrial establishments, engaging 4,205 employees and contributing 4.6% of the district GDP (DSO 2017). The district is popular with domestic and foreign tourists, and the tourism sector significantly boosts local socio-economic activities. It has a population of 492,098, per the 2011 Nepal census (DSO 2017). The district has experienced rapid urbanisation and attracts inter-district migration due to its abundant water resources, pleasant climate, better job opportunities, and easy access to services (Rimal et al. 2018).

To sustain the increasing socio-economic growth of the district, a reliable water supply must be ensured. Due to topographic, climatic, and socio-economic disparities, the ability of the water supply to sustain communities should be examined. The new Water Policy of Nepal promotes water resource management plans to be implemented at all levels of government: local, state, and federal. The Kaski District is used as a case study in this context, while the methods are applicable to other areas to analyse local water availability and pollution.

The total available water, water demand, and water carrying capacity, and their calculation methods are discussed in the subsequent sections.

The Soil and Water Assessment Tool (SWAT) hydrological model can simulate a long period with varying watershed sizes, and has been widely used across Nepal by several researchers (e.g., Aryal et al. 2019; Pandey et al. 2020). The model was applied to estimate the total water availability for the GRB and then transferred to the entire Kaski District (study area) using Equation (1). The model was applied for the GRB since no hydrological station exists within the Kaski District, and this area has similar hydro-meteorological characteristics (Panthi et al. 2015). The input variables: rainfall, temperature, relative humidity, wind, digital elevation model (DEM), soil map, land use, and land cover were fed into the SWAT model to simulate the water availability for the baseline period (1992–2010). Water availability is expressed as an average monthly and annual basis. The average monthly was calculated in two steps, by first averaging each month over the baseline and projected years, and then, again, calculating the average over a year (i.e., 12 months). In contrast, the average annual availability of water was determined by calculating the average annual availability over the baseline year and future years.

Calibration and validation of the model were performed at the Devghat Station (Figure 1) in 1992–2007 and 2008–2010, respectively. Three statistical indicators: coefficient of determination (R²), percentage of volumetric bias (PBIAS), and Nash–Sutcliffe efficiency (NSE) were examined for this purpose. Details of the calibrated parameter are provided in Table S1 of Supplementary Materials.

The total water demand is estimated from the domestic water demand and productive water demand.

Domestic water use varies between rural and urban residents. In this study, domestic water consumption at 100 per capita per day was used for urban residents, following the minimum water requirement of the 2003 national building code (Ministry of Physical Planning and Works: MoPPW 2003), and rural domestic water consumption of 45 L per capita per day, as per the rural water supply and sanitation national policy (MoPPW 2004). Higher daily water use for the urban areas is estimated due to additional use for sanitation such as toilets and gardening. The total population of 492,098 and national average urban and rural populations of 17.1 and 82.9%, respectively, were used to calculate the residents for each category (Central Bureau of Statistics: CBS 2014).

Water used for major economic activities is expressed as productive water use. This includes agriculture and livestock (primary industries), manufacturing (secondary industries), and hotel and restaurants (tertiary industries). Agricultural water use (irrigated water) was estimated for the major crops (rice, wheat, maize, millet, and barley) as well as cash crops (potatoes, oilseeds, and beans). The water use (m³/year) for major crops was determined by multiplying the blue water footprint (m³ per ton) for each crop (Shrestha et al. 2013) and the respective crop yield (tons/year). The blue water footprint refers to the amount of irrigated water used in a unit of crop yield production (ton). These values are taken for the district level from Shrestha et al. (2013), and the total crop yield of the district was from district agriculture statistics. Total water use by livestock was calculated by taking the per capita cattle water use (daily drinking water and service water) (Mekonnen & Hoekstra 2010) and the total livestock population in the district. The number of beef cattle, buffalo, goats, pigs, fowl, and ducks, obtained from the district agriculture and livestock statistics, were included in the calculation.

The manufacturing sectors comprised 198 establishments and provided a total gross output value of 5,893 million Nepalese Rupee (NPR) in 2011. The sector engaged 4,268 employees, with 1,992 employed in the food industries, as per the manufacturing census of 2011–2012. Water use in this sector was determined based on water use coefficients per employee (Malla et al. 2019). Hotels and restaurants are the major water users in the emerging tourism sector. Water use values for hotels and restaurants were obtained by taking water use coefficients as 100 L per bed, guided by the national building code (MoPPW 2003), and 50 L per restaurant seat.

The ecological water demand refers to the quantity of water needed to sustain freshwater ecosystems. This is usually performed by analysing environmental flow requirements and a range of methods are available including hydrological, hydraulic rating, habitat simulation, and holistic methods, at a regional, national, or continental level. These methods all require a long-term data set and flow velocity. Pastor et al. (2014) identified that 37% of the annual discharge is a good representative to estimate the quantity of water needed to meet environmental flow requirements. In the absence of data, the global average of 37% of annual discharge was adopted to estimate the ecological water demand. This amount likely varies within the seasons but is an important basis for the estimation.

Water resources are a limiting factor for socio-economic development. Water resources are under immense pressure with increasing socio-economic growth. Therefore, it is important to ascertain the maximum socio-economic growth that can be sustained by the available water resources. The quantitative analysis of water resources is often expressed as the water resources carrying capacity (WRCC), which serves as a useful tool for sustainable development planning in relation to available water resources (Yang et al. 2015; Yan & Xu 2022). Several definitions and methods are employed in WRCC analyses, which are broadly grouped into three categories: background analysis, supply and demand balance analysis, and recursive dynamic simulation (Li et al. 2010).

Background analysis estimates the carrying capacity of a region by using a comparable area with similar natural and social background. The approach is limited to a static historical background analysis and cannot be used for a dynamic process analysis as it is too simplistic and unable to represent complex natural social–economic systems. The ‘supply–demand balance analysis’ approach is used to determine the water-use quotas based on the total amount of regional water resources, the available resources, and the demand for water. A recursive dynamic simulation method is employed to reveal the WRCC analysis through computer simulation with dynamic simulation and mathematical economic analysis (Li et al. 2010). The requirements and performance of the methods increase from the first to the third methods, that is, (i) background analysis method, (ii) supply–demand balance method, and (iii) recursive dynamic simulation method

Besides availability, preserving water quality is becoming more challenging with the rise in socio-economic activities. Increasing water pollution has further challenged water scarcity issues. Water pollution has traditionally been investigated from a single category approach, which focuses on pollution from discrete economies, for instance, the textile industry, the food industry, hotels, and restaurants. However, this has not been effective in pollution control. The economy of a region consists of multiple sectors, interlinked through supply chains. Any changes in the demand of a sector (outputs) will directly influence other sectors’ economic demands. Thus, the environmental performance of sectors and water pollution needs to be analysed by going beyond a traditional approach, where the indirect roles of sectors, besides their direct pollution, should be considered. An environmentally extended input method was employed to establish water pollution and local economic activities for effective pollution control.

Leontief's input–output (I–O) model is conventionally used to describe the interconnections between sectors Sánchez-Chóliz & Duarte (2003). The I–O model extended with environmental variables is increasingly used to couple the environmental and economy (Sánchez-Chóliz & Duarte 2005; Nguyen et al. 2018). In this study, the environmentally extended input–output (EEIO) method was used to analyse sectoral water pollution using biological oxygen demand (BOD) as a proxy water quality parameter. This method describes the dual roles of a sector as input supplier and receiver in the production process of an economy. Based on these roles, sectoral pollution is also distinguished into two categories: direct (i.e., a source of pollution) and indirect (i.e., a cause of pollution). Direct pollution is the pollutant load directly discharged by a sector in the production of total outputs that satisfy all demands, that is, the intermediate and final demands of an economy. In contrast, indirect pollution is the amount of pollutants discharged by a sector and other sectors to fulfill the inputs it requires. Unlike direct pollution, a sector's indirect pollution is mainly caused by the inputs that they receive from the other sectors. This relationship is captured by the EEIO and assists decision makers in sustainable water pollution control policies.

Two sets of data, the I–O table and BOD discharge coefficient of sectors are required for the EEIO. The 2010–2011 national I–O table (26 × 26 sectors), was derived from the supply and use table (SUT) that originally consisted of 81 × 60 economic sectors from the ADB. The national I–O table was downscaled to the district level I–O table by employing the cross-industry location coefficient (CILC) method (Flegg et al. 1995). The revised I–O table was extended by adding the sectoral BOD discharge coefficient (kg/million NPR). The coefficient for manufacturing sectors was obtained from Chapagain et al. (2020), whereas other sectors were estimated indirectly by taking multiple pollution discharge variables. For example, the cultivated area and average BOD export coefficients from agricultural land were used to determine BOD loads from the agricultural sector, while BOD discharge of cattle farming (kg/day/cattle) and slaughtering activities (kg/tons of meat) from total cattle heads and meat quantity were used to determine the total BOD discharge from the livestock sector. For hotels and restaurants, total visitors and restaurant seats, together with BOD load per capita (visitor) per day and per seat were used. The computation of EEIO and extension of pollution loads in the conventional I–O is described in Chapagain et al. (2022).

A significant challenge of EEIO is the availability of data. This analysis is based on 2010–2011, I–O data, and the latest data will provide more accurate results. The extension of the I–O table was done by adding the BOD discharge intensity of the selected sector due to the widely used water quality indicator and relatively easy to estimate. However, it can be extended with other water quality parameters for characterising sectoral specific pollution. Therefore, developing pollution intensities for the large sector and the sub-sectoral levels and new water quality parameters will further strengthen such analyses.

The spatial change in future water availability (provided in Figure S1 and Figure S2 of Supplementary Materials) shows a reduction between November and February under the RCP4.5 emission scenario and between October and November under the RCP8.5 emission scenario. This change is projected to alter by −301.2 to +3,886.9 MCM under the RCP4.5 emission scenario and by −432.8 to +5,746.1 MCM under the RCP8.5 emission scenario. The maximum reduction is projected to occur in December in Machhapuchhre rural municipality, while the maximum increase is projected in June in Madi rural municipality under the RCP4.5 emission scenario. Similarly, under the RCP8.5 emission scenario, the maximum reduction and increase are projected to occur in Madi rural municipality in November and June, respectively. On average, the change in future water availability is projected to range from +962.6 to +1,296.4 MCM under the RCP4.5 emission scenario and from +235.1 to +1,396.3 MCM under the RCP8.5 emission scenario. The increase in the future water availability in the district might be attributed to the melting of ice and glaciers. Further, the future change in land use and the land cover pattern will also have a significant impact on water availability. Similar findings were found in a snow fed region of the Thuli Bheri River basin in Northwest of Nepal (Maharjan et al. 2021). The research clearly shows the spatio-temporal variability in total water availability under climate change scenarios in the study site. Together with the national level, an appropriate water policy needs to be envisioned to cope with the challenges exerted by climate change at a local level.

Freshwater demand for various sectoral activities and domestic use in Kaski District is presented in Table 3. The agricultural sector is a major user that requires about 158.28 MCM, corresponding to 90.8% of the total water use. Agriculture requires a huge amount of water, and more than 80% of the population engages in agriculture in Nepal, accounting for 35% of the country's gross domestic product (Kong et al. 2019). The agricultural water demand will potentially increase with the expansion of irrigated land since the present coverage of irrigated land is about 52% of the total cultivated land (DSO 2017). At the same time, there is also increasing competition for water and the diversion of water for agriculture to domestic water use. Therefore, water allocation and management are crucial in this area.

Water use by livestock is estimated as 2.72 MCM. The manufacturing withdraws, which is about 0.33 MCM. The district contains 198 industrial establishments, with 4,268 employees, contributing 4.6% of the district's GDP in 2011–2012 (DSO 2017). The food and beverage industries are the major water-consuming industries, accounting (0.19 MCM) for more than half of the total water consumption of all the sectors. In the service sector, about 0.2 MCM water is utilised by hotels and restaurants annually. With rising socio-economic activities in the district, hotels, restaurants, and manufacturing sectors are growing rapidly, which is likely to increase the demand for water. Besides water quantity, there will be an increasing concern for maintaining high quality water while increasing the supply for domestic and industrial purposes.

The WRCC was determined by Equation (3), using average annual water availability for the baseline period as input. The analysis determined the WRCC population to be 127 million, 254 times higher than the existing population of the Kaski District. This clearly shows that water is not a limiting factor for the socio-economic development of this district. The findings confirm that water is abundant in this region, and quantifies the extent of socio-economic growth that can be supported by the available water resources. These results can guide local water policies and provide an example of how a similar approach can be adopted in other parts of the country.

Despite the large WRCC, the current water supply for domestic and productive use is still limited in this case study. This situation is likely to be more pronounced with diverse seasonal and spatial variations in rainfall, and the wide range of inter-seasonal discharges, that is, high flow and low flow in the future. To ensure a consistent and reliable supply of water throughout the areas and seasons, water storage facilities are necessary. The Global Water Partnership: GWP (n.d.) reported that Nepal has one of the lowest dam storage capacities in Asia, and the country's water services are hindered by inadequate infrastructure and low investments. This study emphasises that investment in sustainable water infrastructure is necessary to ensure an adequate supply of water in the future.

A total of 30,049 tons of BOD was produced by economic activity in Kaski in 2011. Agriculture and livestock contributed the largest amount of direct BOD discharge (29,930 tons); 99.6% of the total discharge. The huge contribution of this sector is mainly due to its high BOD loading coefficient, and as a dominant sector in the local economy. The significant role of livestock is globally recognised; however, pollution levels vary with livestock farming methods and treatment practices (Wen et al. 2017). Hotels and restaurants discharged 74.3 tons of BOD, the second largest BOD loading sector. Manufacturing sectors released 41.6 tons of BOD, among which, the food, textile, and paper industries produced a dominant share.

Apart from direct pollution, the indirect role of each sector in BOD production, either from itself or other sectors in acquiring its inputs was analysed. BOD distribution among major sectors in Kaski District is provided in Table S2 of the Supplementary Materials. Indirect pollution is often overlooked in conventional studies, however, evaluating a sectoral performance considering the supply chain is increasing in environmental planning. In Kaski District, agriculture had high indirect pollution, with 58.7% of the total BOD discharge (30,049 tons), followed by the manufacturing sector, and hotels and restaurants, responsible for 27.6 and 3.3% of total BOD, respectively.

The agriculture and livestock sector has a major role in pollution, being the highest contributor to both direct and indirect BOD discharge. The high indirect BOD discharge of the sector reflects its heavy reliance on its own sector, characterising it as a self-polluting sector. On the other hand, the manufacturing sectors had a much greater role in BOD discharge indirectly, which was linked to the high dependence on other sectors for raw materials. Similarly, hotels and restaurants released 3.3% of the total BOD indirectly, much higher than the amount of BOD released directly. Since the sector relies on the agriculture and livestock sector and manufacturing industries for its inputs. Other service sectors such as transportation, finance, and administrative, contributed 10.4% of total BOD discharge indirectly, while their direct BOD discharge was much lower (0.002%).

The agriculture and livestock sector is the major water polluter in the district; however, most efforts concentrate on wastewater discharge from other sectors such as manufacturing industries, hotels, and restaurants. These results indicate that targeting manufacturing sectors by introducing stringent standards will not be sufficient for achieving a good water environment. Therefore, all economic activities, including both point and non-point polluters, should be targeted. Nepal's recent water resource policy (MoEWRI 2020) emphasises that the polluter pay policy complies with the discharge standards, while polluters are subjected to fees and penalties for not meeting these standards. The policy is widely used in wastewater discharge, however, it is not equally fair for all economic sectors. The study results show that agriculture and livestock are major polluters, but the polluter pay model might be difficult to apply to this sector due to its non-point pollution nature. If the sector is brought under the criteria, and pollution fees are determined by estimating BOD loads based on their BOD discharge coefficient, it would not be fair and competitive due to the much higher cost per unit sectoral output compared with the manufacturing and service sectors. Therefore, alternative strategies should be sought that encourage polluters to reduce their pollution loads. For agriculture and livestock, good agricultural practices and proper waste handling should be targeted. We recommend that policy should focus on turning livestock waste into compost or biogas, for which technical and financial assistance should be given. This EEIO analysis provides a backdrop for decision makers to consider several measures, being aware of the close link between agriculture and livestock to other sectors. For indirect water pollution, all economic activities should be looked at from a broad perspective and multiple measures should be opted in pollution control policies.

This study explores current and future water availability and analyses the relationship between water pollution and the economy for sustainable water management in Kaski District, Nepal. Several new insights are presented which will contribute to the effective management of water resources. The average annual water availability for baseline periods was 11,030.7 MCM, while future predictions are 7,677.4 and 7,674.2 MCM under RCP4.5 and 8.5 emission scenarios, respectively. Annual water availability is projected to be reduced in the future, while the difference between high and low flows will be more pronounced. Water availability data were used for determining WRCC as a decision-making tool, in conjugation with water demands and socio-economic variables. The high WRCC demonstrates that water is not a limiting factor in this area in the near-future. However, due to declining water availability and more pronounced high and low flows in the future, investment in water storage facilities and adequate supply systems are recommended to ensure a reliable water supply.

Water pollution was investigated by coupling BOD discharge and sectoral activities, for both direct and indirect discharge. The agriculture and livestock sector was a major BOD polluter, accounting for 99.6% of the total BOD discharge, followed by hotels and restaurants, and manufacturing industries. The manufacturing sector and hotels and restaurants caused greater indirect pollution discharge compared with their direct BOD discharge, due to their high dependence on agriculture and livestock. Pollution controls must focus on the agriculture and livestock sector (i.e., major polluters), which is often ignored. The polluter pay policy is difficult to apply to major polluters as non-point pollution is harder to track and quantify. If the sector is brought into polluter pay criteria, then, considering their pollution per unit outputs, the fee will be very high, and may not be affordable. Better consideration of appropriate strategies for all sectors and sub-sectors in controlling water pollution is necessary, whereby policymakers must consider multiple control measures for specific sectors. For example, technological solutions such as turning livestock waste into biogas and animal feedstuffs, and providing technical and financial support to facilitate this, are necessary. Moreover, providing awareness and training to local farmers to opt for the right dose and mode of fertiliser application is helpful. In the case of wastewater from manufacturing industries and hotels and restaurants, clean technologies should be encouraged, and polluter pay policies with strict wastewater discharge standards could be employed. Sectors such as construction, transportation, and finance do not cause pollution directly, however, they cause significant water pollution indirectly. This provides a new avenue for decision makers to reconsider and expand pollution control measures, however, the addition of more water quality parameters and sub-sectoral levels will further enhance the analysis. We recommend similar studies representing large water basins and socio-economic activities to complement and foster the implementation of sustainable water resource management policies.

We thank The Ministry of Environment of Japan (MoEJ) for funding the project and the Central Bureau of Statistics (CBS), Nepal for providing the necessary data.

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.