Comparison of precipitable water vapor observations by spaceborne radar interferometry and meteosat 6.7-μm radiometry

Satellite radar interferometry (InSAR) can be applied to study vertically integrated atmospheric refractivity variations with a spatial resolution of 20 m and an accuracy of 2 mm, irrespective of cloud cover or solar illumination. The data are derived from the difference between the radar signal delay variations within the imaged area during two acquisitions with a temporal separation of one or more days. Hence, they reflect the superposition of the refractivity distribution during these two acquisitions. On short spatial scales, integrated refractivity variations are dominantly caused by spatial heterogeneities in the water vapor distribution. Validation of the radar interferometric results can be difficult, since conventional imaging radiometers do not provide quantitative measures for water vapor content over the entire tropospheric column and are lacking in spatial resolution. Moreover, comparable quantitative data such as signal delay observed by Global Positioning System (GPS) receivers are only available as time series at a fixed position. In this study, the technique of lnSAR-integrated refraetivily mapping is discussed and validated for a specific atmospheric situation where brightness temperature variations in Meteosat 6.7-μm radiometer data could be mapped to precipitab]e water vapor to validate the InSAR data. The parameterization of the radiometer data is obtained by using a series of 27 hourly GPS signal delay observations at a fixed location and corresponding Meteosat observations at the location of the GPS receiver. Although this methodology for validating the InSAR results is not generally applicable, the results for this specific situation show that the precipitable water vapor observations in both datasets agree to an accuracy of 1.23 kg m^-2, supporting the interpretation of the InSAR data in terms of water vapor distribution.