Detection of Thermal Fronts in the Arabian sea through SAR (Synthetic Aperture Radar) imagery

During the summer (southwest) monsoon (SWM), the Arabian Sea’s surface circulation is clockwise, and heavy upwelling happens along the coasts of Oman and Somalia, resulting in high chlorophyll productivity which forms a thin biogenic slick over the sea surface. Satellite remote sensing observations of these features through optical and infrared wavelengths are confined to low resolution, and are observed to have data gaps due to cloud cover and dust storms. Space-borne Synthetic Aperture Radar (SAR) offers nearly all-weather day-night observation capabilities at a higher resolution. In this study, Advanced Land Observation Satellite (ALOS)-1/2 Phased Array L-band Synthetic Aperture Radar (PALSAR) datasets are used for the detection and extraction of physical oceanographic features of temperature fronts in the Arabian Sea during Southwest monsoon season (SWM). More than 100 HH-polarized ALOS PALSAR 1/2 images for the years 2007, 2010, 2014, and 2015 were acquired from JAXA during Southwest monsoon season. These datasets were pre-processed and Canny edge detection was implemented to extract temperature frontal features. For further analysis of the results, three length scales for the fronts are chosen by selecting length threshold according to the ocean dynamics of the study area. A few cases of the detected fronts are then validated against MODIS SST imagery. Validation shows that fronts of greater length are validated but some fronts of smaller length are not validated because of unavailability of data at their corresponding locations and also due to low spatial resolution of SST images.