Estimating radar beam blockage from thermal noise with verification by precipitation coverage

Estimating the noise floor of radar data on a beam-to-beam basis is very useful for noise removal estimation and more accurate computation of radar moments/products at low signal-to-noise ratios. Thermal back ground blackbody noise usually dominates and the main contributors are the radar antenna and the external environment (i.e., ground, trees, atmosphere, precipitation, etc.). This paper will first describe a robust technique for estimating the noise floor on a beam-by-beam basis. In clear air situations, this noise estimate can also be used to estimate the fraction of beam-blockage, an important consideration in radar data quality, precipitation estimation, and the interpretation of radar returns.

Another proxy for percent beam blockage is derived from precipitation signatures integrated over an extended period of time. The idea is that for long integration periods, the amount of precipitation over a radar scene should be about equal for all locations. Areas with partial beam blockage will show measurably less precipitation. Beam blockage can also be estimated from the physical environment (visually or photgrammetrically ), as well from direct reflectivity measurements.

This paper examines all of these beam-blockage estimators and inter-compares them. Experimental data from NCAR's S-Pol radar taken during the DYNAMO field experiment in the Indian Ocean are used to illustrate the techniques.