Surface water is crucial for understanding and predicting disease spread, especially in monsoon-affected regions like India’s Western Ghats. Before the arrival of Sentinel-1 Synthetic Aperture Radar (SAR) missions, optical remote sensing was used to map water availability. However, cloud cover often hindered these methods. This study explored the impact of cloud cover on mapping surface water in three districts of the Western Ghats.

These districts experience heavy monsoon rains which contribute to water-related diseases. The study compared Sentinel-1A SAR data with the JRC Global Surface Water product, derived from Landsat’s optical data. Both methods were accurate when clouds were scarce, but heavy monsoon clouds posed a significant challenge for optical methods.

During the monsoon months of July-August in 2017 and 2018, cloud cover was extremely high at 92% and 90% respectively. This led to a substantial 25% and 23% of the surface water area being unmapped in the JRC product. In contrast, Sentinel-1A SAR’s 10m resolution could detect small water features missed by the 30m resolution of the JRC product.

The study found that Sentinel-1A SAR was more effective for detecting small water features and predicting disease risks during heavy rains. Additionally, automatic backscatter thresholding can be effective if adapted to regional variations in backscatter.

Water and Health: A Tale of Radar vs. Optical in India's Monsoon

Actions