The development and operational application of nonlinear algorithms for the measurement of sea surface temperatures with the NOAA polar‐orbiting environmental satellites

Abstract

Abstract. Since 1990, the NOAA National Environmental Satellite Data and Information Service (NESDIS) has provided satellite-derived sea surface temperature (SST) measurements based on nonlinear SST algorithms, using advanced very high resolution radiometer (AVHRR) multiple-infrared window channel data. This paper develops linear and nonlinear SST algorithms from the radiative transfer equation. It is shown that the nonlinear algorithms are more accurate than linear algorithms but that the functional dependence of the nonlinearity is data dependent. This theoretical discussion (sections 2-4) is followed with a discussion in section 5 of the accuracy over a 9-year period of the satellite-derived SST measurements provided by NOAA NESDIS when compared with coincident drifting buoys. Between 1989 and 1998 the global scatter of the daytime satellite SST against drifting buoy measurements has decreased from -0.8 ø to 0.5øC, while the nighttime scatter has remained fairly constant at 0.5øC. An exception to these accuracy measurements occurred after the eruption of Mount Pinatubo in June 1991.

Accuracy Assessment of Sea Surface Temperature from NOAA/AVHRR Data in the Seas around Korea and Error Characteristics

“…Korean Journal of Remote Sensing, Vol.27, No.6, 2011 -672- MCSST algorithm without any significant improvement in the study region, which was contrast with the previous researches concerning its improved accuracy (Walton, 1988;Walton et al, 1998).…”

“…This may be partly related to difficulty to eliminate cloudcontaminated pixels because of unavailability of visible channel data at night. NLSST algorithm was evaluated to be slightly more accurate than MCSST algorithm as presumed from the rms errors in Table 4, but no significant improvement unlike the previous literature (e.g., Walton et al, 1998). Although NOAA-19 matchups could not include the first half of 2009, the errors satisfied the reasonable limit of less than 1˚C.…”

“…Since the study region included the Okhotsk Sea and the Russian coast in the East Sea with extremely low temperatures below 0˚C in winter, it would be more important to deploy the drifters in the seas. et al, 1985;Walton, 1988;Vazquez et al, 1994;Walton et al, 1998). However, the individual points of the matchup database indicated that many of individual SSTs from satellite observations were deviated from oceanic insitu measurements by greater than 1˚C.…”

“…We selected two representative algorithms, MCSST and NLSST, developed by NOAA/NESDIS (McClain et al, 1985;Walton, 1988;Walton et al, 1998) and used their coefficients to assess the accuracy of SSTs in the study region. The MCSST algorithm has a fundamental equation as the following form:…”

Lake surface water temperatures of European Alpine lakes (1989–2013) based on the Advanced Very High Resolution Radiometer (AVHRR) 1 km data set

“…Results from several studies (e.g. Li et al, 2001;Oesch et al, 2005) have shown that the linear version of the multi-channel splitwindow equation gives slightly better results (lower biases) than the non-linear equation of Walton et al (1998). Thus, we use similar to Hulley et al (2011) the linear multi-channel equation…”