Impact of rain cell on scatterometer data: 1. Theory and modeling

The two scatterometers currently in operation, the Ku-band NASA Seawinds on the QuikScat satellite and the C-band AMI-Wind on the ERS-2 satellite, are designed to infer the ocean wind vectors from sea surface radar backscatter measurements. They provide excellent coverage of the ocean, and their wind products are of great value for ocean and meteorological communities. However, the presence of rain within scatterometer cells can significantly modify the sea surface backscatter coefficient and hence alter the wind vector retrieval. These perturbations can hamper the analysis of wind fields within atmospheric low-pressure systems or tropical cyclones. Rain perturbations result from volume scattering and attenuation by raindrops in the atmosphere as well as changes of sea surface roughness by impinging drops. For scatterometers operating at Ku-Band, attenuation and volume scattering are strong and one order of magnitude larger than at C-band. The wind retrieval will thus be less affected for the C-band AMI-Wind instrument than for the Ku-band Seawinds. A theoretical model, based on radiative transfer formulation including rain attenuation and scattering, has been developed to quantify the modification by rain of the measured backscatter and of the retrieved wind vectors. Changes in surface roughness, a complex phenomenon not yet fully understood and parameterized, is not considered here although it could be of importance for high rain rates. As a scatterometer cell covers several hundred square kilometers, inhomogeneities of rain within the cell will further modify the measured backscatter, particularly in case of small, intense precipitating rain cells. Using analytical rain cell models and constant wind fields, the effects of partial beam filling by rain is investigated. The model results show that Ku-band scatterometer data are greatly affected by rain and are extremely sensitive to the distribution of rain within scatterometer cells, i.e., to the distance between the rain cell center and the scatterometer resolution cell center. When the scatter from the sea surface is low, the additional volume scattering from rain will have a marked effect leading to an overestimation of the low wind speed actually present. Conversely, when the backscatter is already high (at high winds), attenuation by rain will reduce the signal causing an underestimation of the wind speed. The wind direction is modified in a complex manner and mainly depends on the rain distribution within the scatterometer cell. These results show that, especially at low and moderate wind speed, rain data such as the Special Sensor Microwave/Imager (SSM/I) rain fields are too coarse for correction of Normalized Radar Cross Section (NRCS) and that high-resolution rain data (such as the Tropical Rainfall Mapping Mission (TRMM) ones) are necessary. They also show that a good rain flagging is still an important issue for the operational use of Ku-band scatterometer data. A succeeding paper will present an example of application of the model for the correction of QuikScat data using TRMM rain data within a tropical cyclone.