1. Introduction
Urbanization has increased significantly across the globe over the last two centuries. More than 50% of the world’s population lived in urban areas in 2008 and that population is expected to increase to 75% by 2030 [
1]. Currently many regions of the world have clearly experienced the significant land use/cover (LULC) change and this is already creating environmental, social and economic problems [
2]. Hydrological impacts of LULC changes (e.g., [
3]) as well as climate change (e.g., [
4]) are evident in different parts of the globe. Many studies have assessed impacts of climate change and LULC changes separately, however, very few (e.g., [
5]) have attempted to assess integrated impacts of both climate and LULC changes on hydrology. Therefore, there is a need to enrich the literature in this area with evidence from watersheds in different regions of the world.
The LULC change is becoming a common problem for developing countries that have economies basically dependent on agriculture [
6]. Conversion of the LULC type due to urbanization has affected sustainable management of water resources [
7]. The urbanization process is transforming permeable land surfaces into impervious surfaces and ultimately changing regional hydrological characteristics [
3]. Furthermore, LULC plays a significant role in the hydrological process by directly influencing the surface runoff, infiltration, groundwater flow, and interflow. These hydrological phenomena affect the hydrology of the river in terms of runoff volume, flood frequency, and base flow [
8]. Various techniques have been developed over the years and applied in various parts of the world for assessing LULC change impacts on hydrology. There are various models available to project LULC. Some examples include Conversion of Land Use and its Effects at Small regional contest (CLUE-S) [
9], Dinamica EGO [
10], GEOMOD [
11], Landuse Sim [
12], and Land Change Modeler (LCM) [
13], among others. The land use projection models are usually combined with hydrological models to evaluate LULC changes on hydrology. For example, CLUE-S and WetSpa models were used to evaluate LULC change impacts on river hydrology of Baghsalian watershed in the north of Iran [
7]. Nowadays, the hydrological models are coupled with ArcGIS for better spatial analysis of hydro-climatic variability in a basin [
6]. The soil water assessment tool (SWAT), a semi-distributed hydrological model, has been widely used in previous studies to simulate hydrology and evaluate impacts [
14,
15,
16]. Wang et al. [
17] applied the SWAT and geographically weighted regression (GWR) models in the Xitiaoxi river basin to evaluate impacts of LULC change on the spatial hydrological responses. Change in climatic parameters such as temperature and precipitation alter the river discharge, regional and local water availability, and water supply [
4]. As per the IPCC [
18], the average temperature and precipitation in Asia by the end of the century is projected to increasefrom1.8 °C to 3.9 °C and from 1% to 12%, respectively. In the case of the Hindu Kush Himalayan region, (Indus, Ganges, and Brahmaputra river basins), the mean temperature is projected to rise up to 3.5 °C and 6.3 °C for representative concentration pathway (RCP) 4.5 and RCP8.5 scenarios, respectively [
19] and precipitation is projected to change from 3% to 37% under the RCP4.5 and RCP8.5 scenarios, but with a higher level of uncertainty. Earlier studies have reported that future precipitation is projected to increase in the wet season but decline in the dry season, which causes the arid area to be further drier in dry season (see e.g., [
7,
15,
19,
20]).
Most of the previous studies have focused either on impacts of LULC or climate change on hydrology/water resources. However, limited studies (e.g., [
5,
19,
21]) have assessed impacts of both LULC and climate change. Other than Shrestha et al. [
5], none of the studies have used projected future LULC data for assessing the impacts of both LULC and climate change. A study carried out by Shi et al. [
21] in the Heihe river basin in northwest China showed that climate change, LULC change, and combined changes result in a change in river discharge by 107%, 7.3% and 3.2%, respectively. As the more realistic scenario is the one where both changes happen simultaneously, it is imperative to evaluate impacts of separate as well as combined impacts on hydrology.
Climate change impacts have been observed in several sectors in Nepal and the water resources sector is not an exception. Sharma and Shakya [
20] studied the climate change impact in the Bagmati river basin and concluded that monsoon flood magnitude has a negative trend, but that the duration and frequency have a positive trend. Similarly, a study by Pokhrel [
6] has assessed impacts of LULC change on sediment concentration in the Bagmati basin. However, there are no such studies in Nepali basins, which assess the impacts of both LULC and climate change (CC) on river basin hydrology. Even though LULC and CC affects the water balance in a river basin [
22] and provides better evidence for policy and practice, integrated quantitative assessment of projected changes in future LULC and climate and their hydrological impacts is still an area that is less explored in many watersheds across the world. The objectives of this study, therefore, is to project future climate and LULC, and assess current and future hydrology and water balance under separate and combined scenarios of both LULC and climate changes in the Bagmati river basin, central Nepal. The basin has an irrigation project (i.e., Bagmati Irrigation Project with a command area of 122,000 ha) and a hydropower plant (i.e., Bagmati Hydropower Project with 22 MW capacity) in the downstream. Furthermore, the government has also planned for a Bagmati high dam as a multi-purpose project in the same basin. Therefore, creating a knowledgebase in the area of hydrology, and the impacts of LULC and climate change, is useful for better water resources planning, development and management in the basin.
5. Conclusions
This study projected future LULC and climate changes in the Kathmandu Valley watershed, the upper catchment of the Bagmati river basin, central Nepal, and then evaluated the impacts of individual as well as integrated changes in hydrological regime and water balance. A hydrological model in Soil and Water Assessment Tool (SWAT) was developed for the hydrological simulation and scenario analysis. Future climate (temperature and precipitation) were projected based on three regional climate models (RCMs) (i.e., ACCESS-CSIRO-CCAM, CNRM-CM5 and CCSM4), and two representative concentration pathways (RCP) scenarios (i.e., RCP4.5 and RCP8.5). Linear scaling and quantile mapping techniques were used for bias correction of the RCM data. The performance of the ACCESS-1, bias-corrected with quantile mapping method, gave a robust output when compared with the observed data, and therefore, the same was used for further evaluation of climate change impacts. Similarly, future land use/cover (LULC) data were simulated in the CLUE-S model with five different LULC projection scenarios, out of which, the normal LULC change rate gave a better performance than the others, and was therefore selected for impact evaluation. The SWAT model performance was evaluated for annual, seasonal, and monthly time-scales and found that the calibrated and validated hydrological model is capable to reproduce hydrological regime with reasonably good model performance indicators. Three scenarios were simulated with calibrated and validated model, namely, only climate change, only LULC change, and both LULC and climate changes.
The maximum and minimum temperatures (Tmax and Tmin) and precipitation in the Kathmandu Valley watershed for the period of 2010 to 2050 are projected to increase by 0.66 °C, 0.6 °C, and 23%, respectively, for RCP4.5 scenario and 1.21 °C, 1.04 °C and decrease by 9% respectively for RCP8.5 scenarios. In the same period, the agricultural land and forest area within the basin was converted into the built-up area by 21.4% (130.7 km2). As the result, the river discharge due to climate change alone, LULC alone, and combination of LULC and CC scenarios was increased by 37%, 21%, and 12%, respectively, for RCP4.5 scenario. The decadal average rate of groundwater contribution to the river discharge for the RCP4.5 and RCP8.5 scenarios declined with 58% and 68%, respectively for the integrated scenario due to less infiltration by the expansion of the built-up area.
For all the scenarios, future temperature is projected to increase, summer runoff volume is projected to increase, and winter runoff is projected to decrease. These indicate that the dry and wet seasons are projected to be dryer and wetter, respectively. LULC change alone is projected to increase average annual discharge but decrease the lean period flow. In case of integrating both changes together, it is projected to force the decline in river discharge. Response of the river system to projected changes in climate and LULC is not simply a superposition of response due to individual changes. For this specific case study, the combined change has smoothed the shape of the monthly hydrograph. Here we can conclude that change in the water balance and hydrological process of the Bagmati basin was mainly due to changes in climatic variable (precipitation and temperature) and the LULC change gave the counteractive role toward the urbanized section of the river basin area. The increase in rainfall and river discharge as a result of projected change in climate and LULC; it could be used beneficially by storing excess rainwater and river discharge, the people and ecosystem can harness benefits of climate change/variability. The research results are expected to be useful for the planning and management of water supply, irrigation, hydropower, river basin management projects in the study area. Furthermore, this study has contributed in the body of knowledge on integrated impacts of climate and LULC changes on hydrology.