APPLICATION OF SATELLITE PRECIPITATION ESTIMATES AND CLIMATIC PROJECTIONS THROUGH A HYDROLOGICAL FRAMEWORK

Abstract

Watershed management is essential for sustainable development, socio-economic growth, and the protection of the environment. Due to the complexities of hydrological processes and lack of in-situ data availability, hydrological modelling is challenging, especially in data-sparse regions of developing countries. Satellite precipitation estimates (SPEs) offer higher spatial and temporal resolution and wider areal coverage than ground-based observations, making them a potentially valuable alternative data source for watersheds with sparse or no ground-based data. Hydrological models driven by SPEs, which are more widely available, can be helpful in generating high-resolution, watershed-wide hydrological datasets and making effective decisions on watershed management. Considering this and the availability of hydro-meteorological data, the present research has focused on the Simga watershed. This watershed is an agriculturedominant watershed and faces major challenges of land degradation. Furthermore, The Simga watershed is particularly vulnerable to the impact of climate change. For the sustainable management of water resources over this watershed, understanding its hydrological processes is required. Keeping the aforementioned issues, a comprehensive spatio-temporal evaluation of various SPEs (i.e. PERSIANN, MSWEP, CHIRPS, CMORPH, TRMM, and IMERG) was carried out at daily and monthly time steps against the ground-based observed rainfall. SPEs showed the overestimation of the daily rainfall, indicating that the accurate and precise estimation of rainfall by SPEs is still challenging. The results suggest that there is a need for an update in the precipitation algorithm to detect the different rainfall intensities more precisely over the study areas. Statistical analysis showed that the IMERG dataset had a higher degree of accuracy in predicting average monthly precipitation at the watershed level among the selected SPEs. Furthermore, hydrological modelling of the Simga Watershed was carried out using the SWAT model. Monthly observed streamflow and sediment data were used to calibrate (1980-2000) and validate (2001-2015) the SWAT model to ensure that the model can accurately simulate the streamflow and sediment yield of the watershed. Monthly simulation, results were very good for runoff and good for total sediment load. Hence, the SWAT model could be used for further hydrological applications over the study area. For the hydrological evaluation of IMERG dataset, the SWAT model was simulated for the monsoon season (2000 -2017) and compared against the observed dataset. IMERG precipitation has the potential for hydrological simulations over the Simga Watershed; however, the accuracy needs to be improved, especially at daily scales.Furthermore, to assess the impact of climate change on the hydrologic regime of the Simga Watershed; SWAT hydrologic model was employed for the assessment of the past and future periods. To determine the effect of projected climate change on water resources in the Simga Watershed, five bias-corrected GCMs (CanESM5, ACCESS-CM2, BCC-CSM2-MR, NorESM2-LM, MPI-ESM1-2-HR) from the CMIP6 experiment under SSP2-4.5 and SSP5-8.5 scenarios were used. The duration of 1980-2015 is considered as the historical period, whereas the future period is divided into three phases, i.e., near-future (2020-2046), mid-future (2047- 2073), and far-future (2074-2100). Surface runoff forced by three GCM runs (ACCESS-CM2, CanESM5 NorESM2-LM) showed increasing trends, while the BCC-CSM2-MR and MPIESM1- 2-HR driven runs showed a decreasing trend in the near future (2020 -2046) under both the scenarios. For any given period, the model driven by the SSP5-8.5 scenario predicts more lateral flow than the SSP 2-4.5 scenario. For the far future (2076-2100), each model driven by the SSP5-8.5 scenario predicts more sediment load than the SSP 2-4.5 scenario. These findings can be used to devise better planning and management strategies for the Simga Watershed. The overall hydrological assessment of the Simga Watershed indicates several areas that face soil erosion problems. These critical areas were prioritized for the BMP treatments at the subwatershed level. The SWAT was used for evaluating the effectiveness of different overland and channel BMPs in terms of controlling soil erosion. Strip cropping could be the most effective treatment of agricultural land, reducing streamflow by 9.78% to 13.36% and sediment yield by 19.48% to 26.23%. Analysis of the in-stream BMPs shows that both grassed waterways and streambank stabilization practices can be effective river channel treatments for sediment yield reduction.