Vicarious calibration of an ocean salinity radiometer from low earth orbit

Abstract

Statistical properties of the brightness temperature (TB) measured by a low-earth-orbiting radiometer operating at 1.4 GHz are considered as a means of calibrating and validating the sensor. Mapping of ocean salinity by such an instrument requires that its calibration be extremely stable over time. Whether certain statistical properties of the measurements are stationary (time invariant) enough to be of value as benchmarks to which the calibration can be referenced is considered. The global minimum, maximum, and average TB are considered, together with a vicarious cold TB statistic that makes use of a sharp lower bound on naturally occurring values for TB. Examination of simulated global distributions of the TB measurements suggests several things about the stationarity (or lack thereof) of the statistics in question. Global minima can vary widely due to instrument noise and are not a reliable calibration reference. Global maxima are strongly influenced by a number of environmental factors as well as by instrument noise and are even less stationary than the minima. Global averages are largely insensitive to instrument noise and, in most cases, to environmental conditions as well. The global average TB varies at only the 0.1-K rms level except in cases of anomalously high winds, when it can increase considerably more. The vicarious cold TB is similarly insensitive to instrument effects and most environmental factors. It is not significantly affected by high wind conditions. The stability of the vicarious cold TB is, however, found to be sensitive at the several tenths of a kelvin level to variations in the background cold space brightness. The global average is much less sensitive to this parameter, so using the two approaches together should be mutually beneficial.