Global warming has led to the increase in average temperature and annual rainfall in most parts of India. The country has also faced a remarkable change in land cover in the past decades. These anthropogenic impacts can potentially have a serious impact on hydrology. The Mahanadi river basin is a large-scale river basin located in the eastern part of India. The basin has undergone severe environmental changes during the last decades resulting in serious threat of increased flows. Therefore, understanding how these environmental impacts affect the hydrological behaviour of the basin, especially on a regional scale, forms an important step towards water resources planning and management. Such impacts on the hydrological components are sometimes predicted using a single model realization in conjunction with different land use or climate scenarios. However, such impacts are associated with considerable uncertainties which can arise from model parameterisation, calibration procedures, and due to data availability at local to regional scale as opposed to global products. Little attention has been directed towards understanding these uncertainties while assessing the hydrological impacts of climate and land cover changes in India. In this thesis, we use the most recently released land cover and climate scenarios from Land Use Harmonisation, version 2 (LUH2) database and Climate Model Intercomparison Project, version 6 (CMIP6), respectively, to predict the hydrological responses of Mahanadi river basin to changing land cover and climate conditions. We accounted for the uncertainties associated with modelling those responses, through the sensitivity-guided model calibration performed within a Monte Carlo Framework. This will likely make the predictions more robust and reliable. We identified the important parameters that would have a major control in simulating the hydrological components thereby yielding good simulations on a daily scale for all subbasins, with median KGE ranging between 0.63 to 0.86 in calibration and 0.59 to 0.82 in validation across subcatchments. With regards to predicting the hydrological system in the Mahanadi basin, our findings suggests that a noticeable increase in the cropland at the expense of forest would cause a percent increase in the extreme flows of upto 347 m3s-1. The effects of projected increase in temperature and precipitation in the basin is however more pronounced, resulting in a significant increase in mean annual discharge and peak river discharge upto 29,776 and 2849 m3s-1 respectively. Further, modelling hydrological responses in developing countries like India face additional challenges, because of acute shortages of in-situ hydro-meteorological data. In this respect, we seek to understand how reliable the hydrological model predictions in the region are when combinations of global datasets are used instead of locally available observations. Our results suggest, some global datasets (such as precipitation from Global Precipitation Measurements and soil from SoilGrids) could be used as a viable alternative to local observations in this river basin. Our modelled hydrological responses will be useful for water resource managers to mitigate future risks associated with climate and land use changes, and also would help in selection of the most robust combination of input datasets for the basin.
Modelling hydrological responses of a large-scale river basin in India
Naha, S. (Author). 22 Mar 2022
Student thesis: Doctoral Thesis › Doctor of Philosophy (PhD)