SSM/I integrated water vapor content measurements in coastal regions: A comparison with shipborne and airborne remote sensing measurements, radiosonde measurements, and NWP model retrievals

[1] The Special Sensor Microwave Imager (SSM/I) integrated water vapor content (IWVC) estimates in a coastal region during nonstationary cold-air outbreak conditions are analyzed using IWVC retrievals obtained from shipborne microwave radiometry (DRAKKAR), water vapor differential absorption laser remote sensing (LEANDRE 2), radiosonde measurements, and numerical weather prediction (NWP) model ALADIN of Meteo-France. The study focuses on the 24 March 1998 Tramontane/Mistral event, which occurred over the Gulf of Lion (GoL) (western Mediterranean) in the framework of the flux, etat de mer et teledetection en condition de fetch variable (FETCH) experiment. Comparisons of SSM/I and ALADIN IWVCs with collocated shipborne microwave radiometry, at distances from the coast between 50 and 100 km, were carried out on a full diurnal cycle. In the morning, when moister conditions prevailed, the agreement between SSM/I (one passage) and DRAKKAR was good while the bias observed in the afternoon (three passages) was larger (i.e., 2 kg m(2)). Reasons for this are investigated. We show that this is due to the fact that the sensitivity limit of the algorithm has been reached during very dry Mistral conditions. This bias is not explained by effects of land contamination, which is shown to remain below 0.2 kg m(-2). No systematic bias was observed between ALADIN and DRAKKAR. Nevertheless, the trend observed on the DRAKKAR data between 1430 and 2030 UT was best captured by SSM/I despite the sensitivity issue. SSM/I and ALADIN IWVC estimates were also found to be in good agreement with collocated airborne lidar measurements, thereby reproducing the spatial evolution of the IWVC with the distance to the coast.