HYDROLOGICAL MODELLING OF A TROPICAL WATERSHED UNDER LAND USE AND CLIMATE CHANGE SCENARIOS

Abstract

Water is one of the essential components of our environment. Therefore, proper planning and management are essential to achieve sustainable utilization. Changes in climate and land use have significantly altered the hydrological cycle which in turn has affected the water resources. Due to increased uncertainty in both climate and land-use change projections, improved knowledge of watershed hydrology and resource availability are indispensable for current and future policy formulation and sustainable development of the water sector. The present study has been carried out to ascertain the availability of water and its distribution under the impact of climate change projection and anthropogenic intervention in the Kharun watershed, India. This study investigated the changes in water balance components under varied land use and climate change projections over the Kharun watershed. Kharun watershed lies in the tropical region of central India. Trend changes in meteorological parameters of the past and the future constituted the climate change aspect of the study. The land use land cover (LULC) change dynamics constituted the anthropogenic intervention aspect of the study. Keeping into account the changes in climatic conditions and land change patterns, a hydrological impact assessment was carried out over the study area. Trend analysis is one of the most significant tools to analyze the global warming problem as it quantifies the past and future changes in meteorological and hydro-climatological parameters. In the present study, trend detection was carried out for two metrological parameters namely, long term temperature (maximum, minimum and mean) and precipitation using regression analysis and Modified Mann-Kendall (MMK) test. The magnitude of change was estimated using the Sen’s slope estimator over 22 grids in and around the study area. Cumulative sum (Cusum) and sequential Mann-Kendall (SQMK) test was used to identify the climatic shift (change per year) over the meteorological time series. Significant findings of the study stated an increase in average maximum temperature during summer (0.19⁰C), post-monsoon (0.21⁰C), and winter (0.61⁰C) seasons. A significant reduction in average yearly minimum temperature (-0.68⁰C) was also observed. The annual precipitation decreased by almost 210 mm over 115 years. Similar statistics were computed over 23 indices of meteorological extremes derived from long term precipitation and temperature time series. Out of these 23 indices, five were proposed in the study ii based on the precipitation intensity indices suggested by India Meteorological Department (IMD). Long term trend changes in these indices were computed for both historical as well as future periods. For reproduction of meteorological parameters in order to study changes in extreme value indices in the future, regional climate model (RCMs) were evaluated. Four RCMs were identified as the most suitable models to determine future times series data of precipitation and temperature (maximum and minimum) for the study viz. CCCma, CSIRO, MIROC5 and NorESM. The distribution mapping technique was used to remove systematic biases present in the data. MMK test statistic was used to evaluate the presence of any trend while the magnitude of the trend was quantified using Sen’s slope estimator over the entire period (2011-2100) and for three climate periods, namely CC1 (2011-2041), CC2 (2041-2070) and CC3 (2071-2100). These tests were applied over two scenarios viz. RCP 4.5 and RCP 8.5. After the computation of long term variation in meteorological extremes, it can be inferred that the gap between the minimum and maximum temperature is increasing over the study period at an average rate of 0.09⁰C/decade (4.6%), which explains the increasing trend in Diurnal Temperature Range (DTR). This precisely precedes the fact that the days are getting hotter, and the nights are getting colder and its effects can be seen over the rainfall intensities in the region. As per the results obtained, there is a reduction observed in the number of light rainy days (-10.2%), moderate rainy days (-17.8%) in contrast to heavy and heavy rainy days (-25.5 and -18.4%). The number of cumulative dry days in the study area has also increased by 19.5%, which explains the reduction in rainy days. The overall result indicates an increase in DTR in the future along with an increase in days with heavy rainfalls in the case of both scenarios for the study area. Evaluation of land use land cover is critical and must be monitored to assess the impact on the environment. For this purpose, LULC mapping was carried out for the region using satellite imageries (LANDSAT 5, 7, and 8), remote sensing (RS), and geographical information system (GIS) tools. The LULC maps were classified into six different classes namely water bodies, urban areas, agricultural land, barren land, mixed forest, and sand/open rocks. Significant findings in the study state a decrease in vegetation (agricultural land and mixed forest) in the region due to the rise in the urban area and barren land. After the analysis of historical trend patterns in LULC, the land use land cover map for the near future (2030) was projected using the CA-Markov model. The model was validated and simulated with the classified LULC map of 2015. The projected LULC map of 2030 iii indicated the continuation of the same trend of the past. These future projections indicate the expected changes in the near future. Therefore, the LULC changes concerning different classes in the near future will help in cautioning the concerned authorities for proper planning and management of the study area. In order to investigate the effect of land use land cover change and historical and future climate variability on water availability of Kharun watershed, Soil and Water Assessment Tool (SWAT), a semi-distributed hydrological model was calibrated and validated for the area. Parameters namely Baseflow Alpha Factor (ALPHA_BF), Plant uptake compensation factor (EPCO), and Deep aquifer percolation fraction (RCHRG_DP), were found to be the most sensitive parameters for the Kharun watershed. For monthly simulations, the values of Coefficient of determination (R2), Nash-Sutcliffe efficiency (NSE), and Percent bias (PBIAS) were found to be 0.84, 0.8, and -9.4% during calibration, and 0.85, 0.79 and -9.2% during validation respectively. The results indicated a very good model performance for Kharun watershed. Based on these results, it is concluded that the SWAT model can be successfully employed for the hydrological simulation purposes over Kharun watershed. In order to compute the hydrological components under the dynamics of land use land cover and climate change, 29 simulations were carried out under different variations of land use and climate parameters. Results indicated that the increase in settlement (urban and barren land) for real estate development, accompanied by a decrease in vegetation (agricultural land and mixed forest), has resulted in an increased water yield but the evapotranspiration (ET) reduced due to reduction of vegetation. It is observed that ET reduced with time due to a decrease in vegetation, earlier it used to be 326.71 mm in 1990 but it declined to 298.39 mm during the projected the year of 2030. Due to an increase in overland flow, the water yield increased from 781.58 mm in 1990 to 881.84 mm in the projected the year of 2030. During the last two decades (2010-2030), LULC change increased water yield by 45.88 mm and accounted for 5.48% of the total change (881.84 mm). Moreover, ET decreased by 4.19% in the same duration. Reduction in precipitation was observed for both RCP scenarios in the period CC1 (2011-2040) by -16.83% for NorESM and by -16.29% for MIROC5. The simulation result suggests that the evapotranspiration (ET) in the region is going to increase between 2011 and 2100 but when compared to IMD simulation as a reference, it was observed that the ET has decreased. The maximum change in ET was obtained in CC3. For RCP 4.5, it was 3.99% (MIROC5) and for RCP 8.5, it was 7.26% (MIROC5). While the minimum change iv in ET was observed in CC1. The maximum increase in water yield was observed in CC3, 37.36% for CSIRO (RCP 4.5), and 77.10% for CCCma (RCP 8.5). In summary, the study provided a scientifically essential and practically relevant approach towards identifying the historical climate variability and hydrological assessment under land use and climate change scenarios considering representative climate models output, in contributing to water resources planning and management in the context of a small tropical watershed.