On the use of NLSST and MCSST for the study of spatio-temporal trends in SST gradients

Author(s)

Bouali, M. ^[1] ; Polito, P. S. ^[1] ; Sato, O. T. ^[1] ; Vazquez-Cuervo, J. ^[2]

Affiliation

^[1] IOUSP, Lab Oceanog Satelites, Sao Paulo - Brazil

^[2] CALTECH, NASA, Jet Prop Lab, Pasadena, CA - USA

• Document type: Journal article
• Source: REMOTE SENSING LETTERS; v. 10, n. 12, p. 1163-1171, DEC 2 2019.
• Web of Science Citations: 2

Abstract

Ongoing efforts are dedicated by meteorological and oceanographic agencies to improve the accuracy of Sea Surface Temperature (SST) estimates from satellite observations via improved retrieval algorithms and validation data. An important application of satellite-based SST observations is the analysis of the spatio-temporal characteristics of ocean fronts, which depend on several parameters including the SST retrieval scheme from Top-of-Atmosphere Brightness Temperatures. In this study, we focus on two widely used SST retrieval algorithms, namely the Multichannel SST (MCSST) and the Nonlinear SST (NLSST). Using night-time Level 2 SST derived from the Moderate Resolution Imaging Spectroradiometer (MODIS) onboard Aqua from 2005 to 2015 over the Southwestern Atlantic Ocean (SAO), we show that 1) the spatial distribution and temporal variability of SST gradient magnitudes derived from these two SST retrieval schemes are different despite statistical consistency of SST fields 2) the widely used NLSST formulation introduces a correlation between SST gradient magnitudes and SST values. This correlation, likely due to the use of a first guess SST in the NLSST formulation, is not observed in the MCSST data and may affect the study of long-term changes in ocean dynamics. (AU)

• FAPESP's process: 17/04887-0 - Estimation of high resolution Sea Surface Temperature (SST) fields using multi-sensor satellite observations
• Grantee: Marouan Bouali
• Support Opportunities: Research Grants - Young Investigators Grants