Estimation of Ground Water Level (GWL) for Tropical Peat land Forest using Machine Learning

Water scarcity becoming major problem in many countries like India, Recent problems of critical shortage of water in so many parts of our country due to population, lack of water harvesting plans, excess wastage of water and extreme pollution of water bodies, bringing inaccurate estimation of ground water available in these parts. Still we facing challenge in our country for prediction of ground level water based on the various factors like rain fall, in order to address this issue in this work developed analysis of various machine learning models to predict the ground level water for the data set available from the government side. This work deals with dataset considered from India official site and hydrological parameters consideration for different states of India. We conducted a laboratory experiment to test the ground penetrating radar (GPR) full-waveform forward and inverse modeling approach for electromagnetic wave propagation in water. The GPR system consisted of a vector network analyser combined with an air-launched, 0.8-2.2 GHz horn antenna, thereby setting up an ultra-wideband stepped frequency continuous-wave radar. The apparent frequency-, salinity-, and temperature-dependent dielectric permittivity and electrical conductivity of water were estimated by using existing electrical models. Using these models, the radar data could be simulated and a remarkable agreement was obtained with the laboratory measurements. Neglecting the frequency-, salinity-, and temperature-effects led to less satisfactory results, especially regarding signal amplitude. Inversion of the radar data permitted to reconstruct the air and water layer thicknesses, and to some extent, the water electrical properties. This analysis particularly showed the benefit of using proper water electrical models compared to commonly used simplified approaches in GPR forward and inverse modeling. Index Terms—Ground penetrating radar, full-waveform inversion, multi-layered media, and complex permittivity of water.