engleski

Atmospheric influence on satellite derived sea surface temperature: A case of the Adriatic Sea

Accurate SST data are of great importance for monitoring and modelling the global ocean as well as coastal seas. In situ SST measurements are often too sparse to provide adequate data sets whereas satellite measurements provide synoptic view and repetitive coverage of vast areas. In this paper the RTTOV 7.1 radiative transfer model was used to simulate brightness temperatures (BTs) in the infrared bands corresponding to channels 3, 4 and 5 of the AVHRR sensor on NOAA 15, 16 and 17 satellites. The ECMWF 40-year reanalysis (ERA-40) dataset has been employed to provide the necessary temperature, humidity and ozone profiles as well as related SST data. Out of total ERA-40 global grid only 10 grid points over the Adriatic Sea were used in the analysis. Such a sampling provided 61991 cloudless profiles of temperature, humidity and ozone. The ERA-40 profiles were further interpolated to 43 RTTOV pressure levels. The ERA-40 SST and the RTTOV-derived TOA BT were then used in a multivariate analysis to obtain the SST retrieval coefficients. The calculation was done for 7 zenith angles between 0O and 60O, separately for daytime (12 UTC) and nighttime (00 UTC) using 29592 ERA-40 00/12 UTC cloudless profiles. To address the extent of spatial and temporal variability separate analysis was done for different parts of the Adriatic and different seasons. The analysis has clearly shown pronounced seasonal variability in the modelled state of atmosphere, and less marked differences between Adriatic regions. Determination of several sets of coefficients and partial in situ validation suggest improved SST estimation and reinforce the relevance of regional and seasonal adjustment of retrieval coefficients.

atmospheric influence; Adriatic; SST

nije evidentirano