Groundwater Potential Assessment of Penang Island, Malaysia, Through Integration of Remote Sensing and GIS with Validation by 2D ERT

Penang Island is the most water-vulnerable state in Malaysia. Groundwater is an alternative water source that has not been thoroughly explored within the state. This paper aims to delineate areas of high and low groundwater potential on the island via an integrated approach of remote sensing and geographical information system techniques, with validation using 2D resistivity data. Groundwater potential map was generated using weighted sum of 7 thematic layers: lineament density, geomorphology, rainfall, lithology, slope, drainage density, and land use. Model was validated by evaluating the (a) predictive accuracy of the map, and (b) Spearman's correlation between groundwater potential score and the potential groundwater (10-100 Ohm-m) within inverted resistivity model. Regions classified as Low on groundwater potential were predominantly hilly, forested areas with granitic lithology (e.g., Taman Negara). Regions classified as Very High were mainly coastal urban and agricultural areas with alluvial lithology (e.g., Georgetown). Inland Georgetown, inland Balik Pulau, and Ayer Itam were identified as regions of hard rock terrain with High groundwater potential. Strong relationship was found between groundwater potential score and the percentage of potential groundwater in inverted resistivity models, rho(5) = 0.72, 90% CI [0.09, 0.94]. Furthermore, the map was 83% accurate in identifying regions with suitable groundwater abstraction zones. Findings corroborated with meta-analysis of five studies, where mean correlation of rho = 0.77, 90% CI [0.67, 0.87] was found between groundwater potential score/classification and validation measure. This paper demonstrates the potential in utilizing geoelectrical data for groundwater potential map validation, especially for regions with limited well yield data.

Automated summary

This study generated a groundwater potential map for Penang Island using remote sensing and geographical information system techniques. The map was validated using 2D resistivity data and showed a strong correlation between groundwater potential score and potential groundwater. The results identified regions with high and low groundwater potential, providing valuable information for water resource management.