Uncertainty in satellite sea surface temperature with respect to air temperature, dust level, wind speed and solar position

Sea surface temperature (SST) plays a significant role in studying climate change and oceanic environmental processes. SST has always been measured with buoys and thermometers in the field and by remote sensing-based techniques. The use of remotely sensed SST is significantly more common for mapping SST in wide areas and at high temporal resolution. Remotely sensed SST, however, may reveal uncertainties due to factors such as water depth and turbidity. This is in addition to the atmospheric effects, such as dust, fog, wind, solar angles, and humidity. Thus, in this work, SST estimation by the Moderate Resolution Imaging Spectroradiometer (MODIS) is compared with the measured in-situ SST in the Arabian Gulf with respect to the aforementioned atmospheric and oceanic variables. Results show that SST estimation is affected by air temperature, resulting in a higher SST bias in the summer than in winter. Further, the SST estimation is affected by water depth (i.e. bathymetry). As the water depth decreases below 30 m, the SST bias increases due to the sea bottom effect. SST estimates are also influenced by the instantaneous changes in dust levels, particularly when the aerosol optical thickness exceeds 0.3. However, when SST is averaged over each month, this impact is not prominent because the dust events are occurring occasionally, so they do not impact the average. On the other hand, wind speed and solar angles do not contribute to SST residuals, as the bias is caused predominantly by air temperature and shallow water depth.