No abstract

Groundwater depletion has been an emerging crisis in recent years, especially in highly urbanized areas as a result of unregulated exploitation, thus leaving behind an insufficient volume of usable freshwater. Presently Ganges river basin, the sixth largest prolific fluvial system and sustaining a huge population in South Asia, is witnessed to face (i) aquifer vulnerability through surface waterborne pollutant and (ii) groundwater stress due to summer drying of river as a result of indiscriminate groundwater abstraction. The present study focuses on a detailed sub-hourly to seasonally varying interaction study and flux quantification between river Ganges and groundwater in the Indian subcontinent which is one of the first documentations done on a drying perennial river system that feeds an enormous population. Contributing parameters to the total discharge of a river at its middle course on both temporal and spatial scale is estimated through three-component hydrograph separation and end-member mixing analysis using high-resolution water isotope (δ 18 O and δ 2 H) and electrical conductivity data. Results from this model report groundwater discharge in river to be the highest in pre-monsoon, that is, 30%, whereas, during postmonsoon the contribution lowers to 25%; on the contrary, during peak monsoon, the flow direction reverses thus recharging the groundwater which is also justified using annual piezometric hydrographs of both river water and groundwater. River watergroundwater interaction also shows quantitative variability depending on river morphometry. The current study also provides insight on aquifer vulnerability as a result of pollutant mixing through interaction and plausible attempts towards groundwater management. The present study is one of the first in South Asian countries that provides temporally and spatially variable detailed quantification of baseflow and estimates contributing parameters to the river for a drying mega fluvial system.

No abstract

Forward modeling of ground penetrating radar (GPR) is an important part to the inversion/modeling of the observed data. The aim of this study is to establish specific numerical schemes for forward modeling of GPR data by finite difference frequency domain (FDFD) method which were originally developed for seismic or finite difference time domain (FDTD) method. A total number of six modified and improved FDFD techniques have been used to discretize the two-dimensional (2D) transverse electric (TE)-mode scalar wave equation in order to find the suitable method for this. These techniques include five-point classical to nine-point mixed unstaggered-grid configurations. The numerical schemes for three unsplit perfectly matched layer (PML) for nine-point mixed unstaggered-grid configurations are also presented. The applicability of these techniques is tested by using the underground models of relative permittivity and conductivity for the two cases of homogeneous and 2-cross models. GPR shot gather data for these two models are also produced for this study. The relative reflection errors of the numerical schemes are also estimated for the homogeneous model to comprehend the appropriate method for the modeling. The algorithm for complex-frequency shifted PML (CFSPML) gives the least error in case of the forward modeling of the GPR data.

No abstract