T109 Groundwater Pollution with Arsenic and Other Geogenic Legacy Contaminants, Related Health Risks and Management: Advancing Learning through Geospatial Tools, Machine Learning, Data Science, and Artificial Intelligence (Posters)

Monday, 23 September 2024
8:00 AM - 5:30 PM
Hall D

MULTIVARIATE REGRESSION MODEL EVALUATION OF THE CONTRIBUTING FACTORS TO GROUNDWATER SALINITY IN THE SOUTHWEST COASTAL REGIONS OF BANGLADESH

Paper number: T109-3
Program: Technical Sessions
Time: 8:00 AM – 5:30 PM

Authors

Abstract

This study aimed to find a properly specified regression model to visualize the correlations between salinity contamination and vegetation dynamics in the southwestern coastal regions of Bangladesh. We extracted the Enhanced Vegetation Index (EVI) from Landsat imageries using Google Earth Engine and calculated the vegetation loss or gain percentage, which affects saltwater intrusion in the study area. Additionally, we employed multivariate statistical methods to analyze the relationship between salinity (dependent variable) and vegetation index (independent variables) for informed decision-making.Results from the Ordinary Least Squares (OLS) regression model indicate that the negative relationship between salinity and EVI variables is statistically significant (p-value is < 0.01). The spatial autocorrelation (Global Moran's I) result shows the spatial distribution trend of high or low salinity values in the dataset. Geographically Weighted Regression (GWR) was applied to resolve the drawbacks of OLS. The adjusted R-squared value shows that around 30% of the dependent variable is explained by EVI. The Akaike's Information Criterion (AIC) was 3298.54 for OLS and needs to be lower. Geographically Weighted Regression (GWR) increased the adjusted R-squared value to 95.22% and decreased the AIC to 2849.32, proving the GWR is a better model. Overall, this study observed a negative correlation between salinity and vegetation trends, underscoring the intricate interplay between salinity contamination and vegetation dynamics in the study area.

Keywords: Groundwater Salinity, Vegetation Dynamics, Spatial Regression, Spatial Correlations, Machine Learning

Geological Society of America Abstracts with Programs. Vol. 56, No. 5

doi: 10.1130/abs/2024AM-404952

© Copyright 2024 The Geological Society of America (GSA), all rights reserved. Permission is hereby granted to the author(s) of this abstract to reproduce and distribute it freely, for noncommercial purposes. Permission is hereby granted to any individual scientist to download a single copy of this electronic file and reproduce up to 20 paper copies for noncommercial purposes advancing science and education, including classroom use, providing all reproductions include the complete content shown here, including the author information. All other forms of reproduction and/or transmittal are prohibited without written permission from GSA Copyright Permissions.