Impacts of future land use/land cover (LULC) and climate change on runoff of the teesta basin

Abstract

Teesta basin is one of the most important transboundary river basins in South Asia, shared by Bangladesh and India. Last few decades, the basin has exhibited significant land-use, and land cover (LULC) changes, and these LULC changes with climate change are the main factor in the water balance of the basin in the future. In the past, several studies have been conducted that either assessed the hydrological parameter affected by the LULC or climate change impact on the runoff of the Brahmaputra-Teesta basin. This study will attempt to analyze the combined effects of future LULC change and climate change on the runoff of the Teesta basin. To reach the goal, GIS and RS applied for the LULC assessment, and LULC was predicted based on the widely used Cellular Automata-Markov (CA-Markov) for two future periods: 2050s (2035–2064) and 2080s (2071-2100). A semi-distributed hydrological model, Soil Water Assessment Tools (SWAT), is set up for the baseline period (1993–2013) and driven primarily by outputs from eight General Circulation Models (GCMs) under two Sheared Socioeconomic Pathways (SSPs) (i.e., SSP245 and SSP585) for the 2050s and 2080s. A significant LULC change happens in the basin area. About 0.49% of agriculture, 1.67% of settlement and 4.24% of water/river area increased over the basin from 2000 to 2010. On the other hand, 1.45% forest and 5.49% sand bar decreased from 2000 to 2010. However, from 2010 to 2020, 2.77% of agriculture, 8.59% of settlement and 7.7% of water/river area increased while 1.36% of forest and 0.78% of the sand bar decreased. The CA-Markov prediction show settlements will be increased up to 42.9% in 2050 and 293% in 2080 from the base period. The agricultural land area will also be increased by 76.2% in 2050 and 121% in 2080. The forest, water and snow cover will be decreased by 13%, 36% and 5% in 2050 and 33%, 38% and 8% in 2080. The SWAT hydrological model was calibrated and validated with satisfactory results, with NSE values of 0.81 and 0.81, PBIAS values of 18 and 16.9, and R2 values of 0.85 and 0.84, respectively. The monsoon may see as much as a 26% (-40% to +120%) and 38% (-36% to +151%) increase in discharge in the 2050s and a 30% (-42% to +144%) and 45% (-30% to +167%) increase in the 2080s for SSP245 and SSP585 scenarios, respectively. At the same time, the winter/dry period shows a 46% (-84% to +24%) and 49% (-84% to +27%) decrease in the 2050s and a 40% (-84% to +51%) and 38% (-86% to +48%) decrease in the 2080s for SSP245 and SSP585 scenarios, respectively. Considering the LULC change, during the monsoon, the discharge would increase additionally by another 4% (-8% to 5%) and 5% (-8% to 10%) in the 2050s and 13% (0.4% to 23%) and 52% (-29% to 151%) in 2080s for the SSP245 and SSP585 scenarios. Interestingly, the dry period flow would increase additionally by 0.3% (-1% -3%) and 0.8% (-1% -5%) in the 2050s and decrease by 6% (-16% and -4%) and 13% (-64% and 63%) in the 2080s for the SSP245 and SSP585 scenarios, respectively. In general, an increasing trend is found in the peak flow during the monsoon and a decreasing trend in mean flow during the winter season without further upstream water control. This study concludes that the average discharge on the basin will be increased in the near and far future, which may impact lives and livelihoods.