Identification of rainy periods from ground based microwave radiometry

In this paper the authors present the results of a study aiming at detecting rainy data in measurements collected by a dual band ground-based radiometer. The proposed criterion is based on the ratio of the brightness temperatures observed in the 20-30 GHz band without need of any ancillary information. A major result obtained from the probability density of the ratio computed over one month of data is the identification of threshold values between clear sky, cloudy sky and rainy sky, respectively. A linear fit performed by using radiometric data and concurrent rain gauge measurements shows a correlation coefficient equal to 0.56 between the temperature ratio and the observed precipitation.