Fully polarimetric radiometer system for airborne remote sensing

As a result of the increasing interest in polarimetric microwave radiometer measurements, a fully polarimetric radiometer system was developed and thoroughly tested in this work. The system is primarily designed for airborne remote sensing; however, ground-based and laboratory measurements are possible as well. The feasibility of the system for remote sensing was demonstrated in laboratory measurements and in an airborne experiment. The system comprises a polarimetric radiometer at 36.5 GHz (the Fully Polarimetric Radiometer) and a passive calibration standard (the Fully Polarimetric Calibration Standard). This work is the first successful demonstration of a polarimetric remote sensing radiometer that is based on analog direct cross-correlating architecture; analog multipliers were used as correlating elements. Compared to other possible polarimetric radiometer topologies, the applied solution possesses several advantages, e.g., the simple structure of both the receiver and the correlator, which is beneficial in terms of reliability, costs, mass, size, and power consumption. A novel fully polarimetric end-to-end calibration technique was developed. The theoretical background of the technique is thoroughly studied, the mathematics necessary to determine the a priori Stokes vectors and error analysis are derived, and a calibration matrix inversion technique is presented. Using the principles presented, fully polarimetric calibration standards were developed at 10.6 GHz and 36.5 GHz and their function demonstrated. The uncertainties associated with the practical implementation of fully polarimetric calibration standards for wind vector remote sensing are discussed. Using the developed radiometer system, an airborne experiment was conducted over the Gulf of Finland. Polarimetric dataset obtained for a semi-enclosed sea composed of brackish water is presented. The results suggest that the relationship between the wind vector and the Stokes parameters may vary locally.reviewe