No abstract

No abstract

No abstract

No abstract

Data measured by microwave radiometers are brightness temperatures, from which, using algorithms, geophysical parameters arc retrieved. Errors in these parameters can be due either to the algorithms or to instrumental deficiencies affecting the brightness temperatures. Only geophysical data can be validated directly since dedicated meteorological experiments cannot be performed frequently enough or in a sufficient number of places during the satellites lifetime. To achieve such a global validation, we propose to use global meteorological model analyses, which contain the information of all routine observations after checking their quality. In this article, we test this approach by comparing Nimbus-7/SMMR geophysical data with ECMWF model analyses. Data from five months in 1979 are examined, and deviations between both data sets are analysed in terms of satellite or model errors. This comparison reveals SMMR problems in SSTand surface wind data, and local model errors in the precipitable water. Therefore, a good knowledge of model behaviour is necessary for using its analyses in satellite data validation, yet it appears as a useful tool, working at scales ranging from regional to global, and in time, from a particular event to a season or more.