Scrutinizing Relationships between Submarine Groundwater Discharge and Upstream Areas Using Thermal Remote Sensing: A Case Study in the Northern Persian Gulf

Nutrient input through submarine groundwater discharge (SGD) often plays a significant role in primary productivity and nutrient cycling in the coastal areas. Understanding relationships between SGD and topo-hydrological and geo-environmental characteristics of upstream zones is essential for sustainable development in these areas. However, these important relationships have not yet been completely explored using data-mining approaches, especially in arid and semi-arid coastal lands. Here, Landsat 8 thermal sensor data were used to identify potential sites of SGD at a regional scale. Relationships between the remotely-sensed sea surface temperature (SST) patterns and geo-environmental variables of upland watersheds were analyzed using logistic regression model for the first time. The accuracy of the predictions was evaluated using the area under the receiver operating characteristic curve (AUC-ROC) metric. A highly accurate model, with the AUC-ROC of 96.6%, was generated. Moreover, the results indicated that the percentage of karstic lithological formation and topographic wetness index were key variables influencing SGD phenomenon and spatial distribution in the northern coastal areas of the Persian Gulf. The adopted methodology and applied metrics can be transferred to other coastal regions as a rapid assessment procedure for SGD site detection. Moreover, the results can help planners and decision-makers to develop efficient environmental management strategies and the design of comprehensive sustainable development policies.

Automated summary

This study utilized Landsat 8 thermal sensor data to identify potential sites of SGD and analyzed the relationship between remotely-sensed sea surface temperature patterns and geo-environmental variables using a logistic regression model for the first time, resulting in a highly accurate model with an AUC-ROC of 96.6%. The results indicated that the percentage of karstic lithological formation and topographic wetness index were key variables influencing SGD phenomenon and spatial distribution.