Uncertainty in the Absolute Calibration of Differential Reflectivity

Confidence in measurement accuracy of dual pol radars requires the selection of radar calibration targets whose differential reflectivity (ZDR) is more accurately known than the desired error bar in that measurement. If that error bar is 0.1 dB, it has been previously established that any carefully machined metal sphere or a single drizzle drop both satisfy this condition. And yet any single-pulse radar measurement of an isolated metal sphere or a population of drizzle drops shows a variability in ZDR that substantially exceeds 0.1 dB, for physical reasons that differ between the two cases. When the Sun and Bragg backscatter are used as calibration targets for ZDR, one gives up the guarantee for 0 dB in ZDR in a single realization or single pulse, but retains the expectation for 0 dB in the mean, attainable (with some uncertainty) by a long dwell time. This study seeks to gather together the distributions for ZDR on all known targets expected to show 0 dB in the mean, but with a quantification of the spread of values about this mean. Implications for a hard limit on the uncertainty in the calibration of ZDR will be addressed.