OS001-0003
Comparison of Saildrone observations with satellite- and model-derived sea surface salinity and temperature in the Bering Strait

Monday, 7 December 2020
Poster
Jorge Vazquez1, Chelle Gentemann2, Dustin Carroll3, Dimitris Menemenlis1, Hong Zhang1, Wenquing Tang4, Severine Fournier1, Jose Gomez-Valdes5 and Marouan Bouali6, (1)NASA Jet Propulsion Laboratory, Pasadena, CA, United States, (2)Farallon Institute, Petaluma, CA, United States, (3)Moss Landing Marine Laboratories, Moss Landing, CA, United States, (4)NASA Jet Propulsion Laboratory, Pasadena, United States, (5)CICESE National Center for Scientific Research and Higher Education of Mexico, Ensenada, BJ, Mexico, (6)USP University of Sao Paulo, IOUSP, São Paulo, Brazil
Abstract:
The Arctic is a key region for understanding global climate processes. At the same time, due to complex sea-ice processes and the sensitivity of geophysical models to changes in salinity, comparisons of satellite-derived products with in-situ observations in this region often contain large differences. The 2019 Saildrone deployments present an opportunity for a direct comparison in the Bering Strait region off the Alaskan Coast.

Here we compare Saildrone observations with the Group for High Resolution Sea Surface Temperature (GHRSST) Level 4 products that include the UK MetOffice Operational Sea Surface Temperature and Sea Ice Analysis (OSTIA), the Canadian Meteorological Center (CMC), the NAVOCEAN K10 product, the Remote Sensing Systems (REMSS) merged SST product, and the NOAA/NCEI AVHRR_OI product. The SMAP JPL Captive Active Passive Microwave (CAP) SSS and REMSS 40 km and 70 km products were also compared. Finally, we compare these results with the lower resolution (1/3°–18km) ECCO LLC270 ocean state estimate and higher-resolution (~1–2 km) ECCO LLC4320 simulation.

For SST, the REMSS product had the highest correlations, approximately 0.80, while for SSS the REMSS 70 km had correlations over 0.90. The ECCO LLC270 exhibits SST correlations of approximately 0.70, while the LLC4320 has weaker correlations of 0.60. For SSS, the LLC270 correlation was approximately 0.90 while the LLC4320 was 0.80. The ECCO model results and their spectra indicate that there is significant variability at scales <25km, which provides incentive for improved remote sensing products with <25km resolution.

<< Previous Abstract | Next Abstract >>