SST is an ocean variable that is readily measured by satellites and in situ sensors, and it is needed as a key input to forecasting systems to constrain the modeled upperocean circulation and thermal structure, and for the exchange of energy • AMERICAN METEOROLOGICAL SOCIETY

The measurements made by the Mong-track scanning radiometer are now converted routinely into sea surface temperature (SST). The detMls of the atmospheric model which had been used for deriving the SST a•gorithms are given, together with tables of the coefficients in the a•gorithms for the different SST products. The accuracy of the retrieva• under norma• conditions and the effect of errors in the model on the retrieved SST are briefly discussed. IntroductionThe growing demand for more and more accurate global measurement of sea surface temperature (SST) has led to the development of the along-track scanning radiometer (ATSR), launched on the European Space Agency's (ESA) first European Remote Sensing Satellite (ERS 1) in July 1991. Before the instrument could be built or even before a proposal could be submitted to ESA for acceptance it was necessary to have a more or less complete design of the instrument and a prediction of its anticipated performance. The ATSR is fully described by, for example, Edwards et al. [1990]. One of its most important novel features is the measurement of the upwelling radiation from every picture element, not only in two or more different wavelength channels, but also at tw0different zenith angles. The larger the difference between the angles, and hence the difference in atmospheric pathlength from the satellite to the surface, the larger the difference in atmospheric effect, hence the easier the correction for the atmospheric absorption and emission. For obvious reasons, the optimum angle for one of the measurements is an angle as close to the vertical as the width of the swath permits. The second view must not be too oblique, however, and one of the reasons for this is that in this case the signal would have its dominant contribution from the middle atmosphere, and hence the compensation for the lower-tropospheric effects dominating the atmospheric signal in the nadir view would be poor. Although, in principle, by flying a prototype instrument with a range of viewing angles, the effect on the SST accuracy could be determined, the cost and time of build-x Now at ing the instrument, the range of atmospheric conditions over which the measurement campaigns would have had to be performed, and the very vigorous quality checking of the data would have made this method impractical. We have chosen the method of simulating the brightness temperatures that would be measured under different conditions through the writing of a computer code for a radiative transfer model. The brightness temperatures generated were then used to determine the optimum instrument parameters, to predict the performance of the instrument, and, finally, for calculating SST retrieval coefficients to be used with the ATSR measurements.Before the model could be used in this way, however, it was imperative to check its accuracy. Owing to the unavailability of accurate coincident atmospheric and infrared brightness temperature data measured from space, only an indirect validation was possible by using the infrared brightnes...

No abstract

No abstract

The along-track scanning radiometer (ATSR) was launched in July 1991 on the European Space Agency's first remote sensing satellite, ERS 1. An initial analysis of ATSR data demonstrates that the sea surface temperature (SST) can be measured from space with very high accuracy. Comparison of simultaneous measurements of SST made from ATSR and from a ship-borne radiometer show that they agree to within 0.3øC. To assess data consistency, a complementary analysis of SST data from ATSR was also carded out. The ATSR global SST field was compared on a daily basis with daily SST analysis of the United Kingdom Meteorological Office (UKMO). The ATSR global field is consistently within 1.0øC of the UKMO analysis. Also, to demonstrate the benefits of along-track scanning SST determination, the ATSR SST data were compared with high-quality bulk temperature observations from drifting buoys. The likely causes of the differences between ATSR and the bulk temperature data are briefly discussed. These results provide early confidence in the quantitative benefit of ATSR's two-angle view of the Earth and its high radiometric performance and show a significant advance on the data obtained from other spaceborne sensors. It should be noted that these measurements were made at a time when the atmosphere was severely contaminated with volcanic aerosol particles, which degrade infrared measurements of the Earth's surface made from space. 22,575 22,576 MUTLOW ET AL.: SEA SURFACE TEMPERATURE MEASUREMENTS FROM ERS 1 teorological Office daily bulk SST analysis, and (3) comparison of the ATSR spatially averaged SST products with SST measurements from the drifting buoy network. Brief Description of the ATSR InstrumentThe ATSR instrument and its unique features are described in detail elsewhere [Delderfield et al., 1985; Edwards et al., 1990]. In short, ATSR is a four-channel infrared imaging radiometer with spatially coregistered spectral channels centered at 1.6, 3.7, 10.8, and 12.0 prn. It has been designed for exceptional sensitivity and stability of calibra-MEASUREMENTS FROM ERS 1 22,577 the cloud-free scene brightness temperature observed by ATSR. The coefficient a i is evaluated by linear regression of T s with T i calculated from a representative set of atmospheres using modeled radiation transfer of the atmosphere. The atmospheric data set was obtained from meteorological data provided by the United Kingdom Meteorological Office (UKMO). The comprehensive radiative transfer calculations are performed using a line-by-line model of molecular absorption. Spectral line parameters were taken from the Air Force Geophysical Laboratory line compilation [Rothman et al., 1987], which has been augmented by experimental measurements of the chlorofluorocarbons CFC-11 and CFC-12.