Satellite-derived sea surface temperature fronts in a river-influenced coastal upwelling area off central-southern Chile

The variability of thermal fronts in eastern ocean boundaries has received increased attention because of being active regions of vertical fluxes of tracers and biological activity. Sea Surface Temperature (SST) images from three distinct satellite products are used to identify areas with enhanced surface thermal gradients (i.e. SST fronts) in the coastal ocean off central-southern Chile. The main objective is to evaluate their use in the study of SST frontal variability in a river-influenced continental shelf. In contrast with previous studies focused on the mesoscale structure of the upwelling front, this study highlights the importance of using high spatial resolution (i.e. 1 km) satellite products to resolve the small-scale surface thermal gradients on a relatively narrow continental shelf impacted by freshwater river outflows. An improved approach, merging Moderate Resolution Imaging Spectroradiometer (MODIS) imagery from Aqua and Terra satellites (MODIS-AT), increases the percentage of available SST data for the computation of SST gradients and frontal probability over these shelf waters. Overall, SST data from all three sources (Pathfinder, Geostationary Operational Environmental Satellite (GOES), and MODIS) resolve the major mesoscale frontal features along the offshore limit of the continental shelf. However, MODIS-AT considerably improves the detection of SST fronts over the continental shelf, especially during winter and spring when river outflows are important on the dynamics of coastal flows. A case study off the Itata River mouth reveals high spatio-temporal variability of thermal fronts over the continental shelf, which is not well detected from GOES and Pathfinder data. The analysis of MODIS-AT images is highly recommended for studies of thermal fronts over shelf waters. In contrast, the use of GOES imagery improves the monitoring of the mesoscale frontal activity farther offshore. (c) 2020 Elsevier B.V. All rights reserved.