Data-Driven Assessment of Contaminant Vulnerability in River Ganga Coastal Aquifers

Groundwater of coastal aquifers have strong hy-draulic connections with the coastal hydrodynamics in a various spatiotemporal scale. But the along with the sub daily tidal influence the Tropical cyclone dependent acute groundwater level response has rarely been studied. These acute and instantaneous groundwater level (GWL) fluctuations during tropical cyclones (atmospheric depression) have a direct impact on groundwater flow dynamics and have an impact on instantaneous solute and contaminant mobilization. These groundwater dynamics include influx of sweater and submarine groundwater discharge over the period each atmospheric low-pressure events simultaneously. The dynamic state of GWL fluctuations get stabilized along with decay of atmospheric low-pressure events but the change in solute concentration may stay for a month to seasonal recharge of groundwater and is proportional to the level of acute GWL fluctuations. Thus, this monitoring and prediction of atmospheric low- pressure event dependent GWL fluctuations is a possible indicator of groundwater vulnerability, especially in coastal aquifers. This study conducts a comparative analysis of machine learning and deep learning models to predict groundwater level fluctuations. The models include a Fully Connected Neural Network (FCNN), Artificial Neural Network (ANN), MLPRe-gressor, Support Vector Regressor (SVR), and Random Forest Regressor (RFR). Each model is evaluated using metrics such as Mean Squared Error (MSE), R2 score, Normalized Root Mean Squared Error (NRMSE), Root Mean Squared Error (RMSE), and standard deviation. The results underscore the capability of deep learning techniques to capture nonlinear features in hydrological data, thereby enhancing the understanding and prediction of groundwater levels. Such predictions are crucial for assessing seawater infiltration, which is a potential threat to the available drinking water sources and nutrient and contaminant fluxes within coastal aquifers. Submarine groundwater discharge, facilitated by these aquifers, plays a pivotal role in nutrient transfer from terrestrial to marine ecosystems, thereby influencing marine health and primary productivity.

Authors

Institutions

Publication Details

Published
2024-09-23
DOI
https://doi.org/10.1109/oceans55160.2024.10754075
Primary Topic
Advanced Data Processing Techniques
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Data-Driven Assessment of Contaminant Vulnerability in River Ganga Coastal Aquifers

Farheen Fauziya, Chandranath Banerjee, Ananya Sen Gupta, Kousik Das et al.
Advanced Data Processing Techniques
article

Data-Driven Assessment of Contaminant Vulnerability in River Ganga Coastal Aquifers

Farheen Fauziya, Chandranath Banerjee, Ananya Sen Gupta, Kousik Das, Abhijit Mukherjee, Somdipta Sen, Prakrity Majumder
article en

Abstract

Groundwater of coastal aquifers have strong hy-draulic connections with the coastal hydrodynamics in a various spatiotemporal scale. But the along with the sub daily tidal influence the Tropical cyclone dependent acute groundwater level response has rarely been studied. These acute and instantaneous groundwater level (GWL) fluctuations during tropical cyclones (atmospheric depression) have a direct impact on groundwater flow dynamics and have an impact on instantaneous solute and contaminant mobilization. These groundwater dynamics include influx of sweater and submarine groundwater discharge over the period each atmospheric low-pressure events simultaneously. The dynamic state of GWL fluctuations get stabilized along with decay of atmospheric low-pressure events but the change in solute concentration may stay for a month to seasonal recharge of groundwater and is proportional to the level of acute GWL fluctuations. Thus, this monitoring and prediction of atmospheric low- pressure event dependent GWL fluctuations is a possible indicator of groundwater vulnerability, especially in coastal aquifers. This study conducts a comparative analysis of machine learning and deep learning models to predict groundwater level fluctuations. The models include a Fully Connected Neural Network (FCNN), Artificial Neural Network (ANN), MLPRe-gressor, Support Vector Regressor (SVR), and Random Forest Regressor (RFR). Each model is evaluated using metrics such as Mean Squared Error (MSE), R2 score, Normalized Root Mean Squared Error (NRMSE), Root Mean Squared Error (RMSE), and standard deviation. The results underscore the capability of deep learning techniques to capture nonlinear features in hydrological data, thereby enhancing the understanding and prediction of groundwater levels. Such predictions are crucial for assessing seawater infiltration, which is a potential threat to the available drinking water sources and nutrient and contaminant fluxes within coastal aquifers. Submarine groundwater discharge, facilitated by these aquifers, plays a pivotal role in nutrient transfer from terrestrial to marine ecosystems, thereby influencing marine health and primary productivity.

University of Iowa (US), Indian Institute of Technology Kharagpur (IN), SRM University, Andhra Pradesh (IN)
Life below water
Openalex Percentile: Top 75%
Advanced Data Processing Techniques
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.