Search

Streamflow modelling plays a critical role in water resource management activities. The “physically based" models require high computation resources and large amounts of input meteorological data which results in high operating costs and longer running times. On the other hand, with advancements in deep-learning techniques, data-driven models such as long short-term memory (LSTM) networks have been shown to successfully model non-linear rainfall-runoff relationships through historically observed data at a fraction of computation cost. Moreover, using physics-informed machine learning techniques, the physical consistency of data-driven models can be further improved. In this study, one such method is applied where we trained a physics-informed LSTM network model over 278 Indian catchments to simulate streamflow at a daily timestep using historically observed precipitation and streamflow data. The ancillary data included meteorological forcings, static catchment attributes, and Noah-MP simulated land surface states and fluxes such as soil moisture, latent heat, and total evapotranspiration. The LSTM model's performance was evaluated using error metrics such as Nash-Sutcliffe Efficiency (NSE), Kling-Gupta Efficiency (KGE) and its components, along with skill scores based on 2x2 contingency matrix for hydrological extremes. The trained LSTM model shows improved performance in simulating streamflow over the catchments compared to the physically based model. This will be the first study over India to generate reliable streamflow simulations using a hybrid state-of-the-art approach, which will be beneficial to policy makers for effective water resource management in India.

A land data assimilation system plays a pivotal role in effective water resources management by providing reliable estimates of various land surface states and fluxes including water balance components. An Indian Land Data Assimilation System (ILDAS) has been setup that aims to provide spatially consistent, high resolution and reliable estimates of land surface states, water balance and energy fluxes over the Indian mainland. The ILDAS is built on NASA’s Land Information System Framework (LISF), which is an open-source framework that enables a multi-model, multi-data approach to terrestrial modeling. In this study, we evaluated the water balance components simulated using ILDAS while driven by three reanalysis meteorological forcing datasets: Modern-Era Retrospective analysis for Research and Applications, Version 2 (MERRA-2), and Climate Hazards Group InfraRed Precipitation with Station data (CHIRPS) and ECMWF's ERA5. The multi-site autocalibration is performed on multiple catchments to further improve the performance of ILDAS. We have used Kling-Gupta Efficiency (KGE) and its components to evaluate the major components of the terrestrial water balance: surface runoff, soil moisture, terrestrial water storage anomalies, evapotranspiration, and monthly mean streamflow, against a combination of satellite and in-situ observation datasets. We also assessed the uncertainty and bias due to spatio-temporal heterogeneity in the forcing precipitation by validating against gauge-based gridded precipitation provided by Indian Meteorological Department (IMD). Overall, considering nationwide median KGE scores, CHIRPS showed the highest performance for surface runoff and streamflow. However, MERRA-2 performed the highest in soil moisture simulation whereas ERA5 showed best results in total evapotranspiration.

In India, one of the major sources of water is the groundwater. Currently, its quality is being degraded due to many reasons such as intrusion of waste from various chemical factories and other industries. The objective of the study was to analyze the groundwater quality in the vicinity of the Kala Sanghian drain, Jalandhar. The drain carries waste from various industries in Jalandhar and is unlined which might affect the quality of the nearby groundwater. The physiochemical and heavy analysis was carried out on 30 samples collected from different locations of varying depths along the drain. It was observed that the values for some of the parameters are exceeding the permissible limit. The reason for dwindling quality of water in the area may be due to the discharge of the industrial waste into the Kala Sanghian drain. Hence, quick and reliable monitoring measures are essential for safeguarding the water quality. Geographic Information System was used to map the parameters values in terms of WQI in spatial domain which would help in better visualization of the problem and improved decision making.