The new algorithm is based on the empirical fact that thunderstorms are often associated with strong radar echoes above the height of -10 degrees C. In addition, we use the property that the dual-frequency ratio DFRm which is defined as DFRm = dBZm(Ku) – dBZm(Ka) where dBZm(ku) and dBZm(Ka) are measured radar reflectivity factors at the Ku and Ka bands respectively, becomes large when precipitation particles are large. Since the ice precipitation particles are generally larger than rain drops, we can expect that DFRm are larger in snowing regions than in raining regions as far as the attenuation can be ignored. Since spaceborne radar measures precipitation from above the storm, we can generally assume that we can ignore liquid water attenuation above -10 degrees C level.
The algorithm will set a flag in both dual-frequency and single-frequency products when a relatively strong echo is measured above the -10 dgrees C isotherm height. In the dual-frequency products, the flag is also set when a large DFRm larger than 7 dB with KuPR's Zm larger than 27 dB appears above the -10 degrees C isotherm height. Note that DFRm is examined only when KuPR's Zm is larger than 27 dBZ in order to avoid misjudgment due to noisy signal of KaPR when its echo is weak.
The performance of the algorithm is tested in typical hail cases and in a relatively heavy snow storm. In the hail cases in Italy and the US, both the single-frequency and dual-frequency algorithms give very reasonable results compared with the precipitation classification by a ground-based polarimetric radar. In the wide spread snowfall case over Japan, the dual-frequency method that utilizes the DFRm identifies the heavy snowfall regions well. These examples indicate that the new flag is useful to identify heavy ice precipitations, even though it may miss many light snowfalls.
A month of statistics of the flag is taken to see the global distribution of heavy ice precipitations. The result indicate that the dual-frequency algorithm flags more pixels than the single-frequency algorithm with the conditions that are often used to identify the thunder storms. It detects heavy ice precipitations not only in strong tropical convections, but also in winter storms at high latitudes.
In order to see the fraction that ice precipitations actually reach the surface without melting, only the cases with this flag on and with the surface air temperature below 0 degrees C are counted. The result shows that ice precipitation reaches the surface without melting in quite a few cases even over a relatively warm ocean surface. Since the detection of very shallow ice precipitations over ocean is very difficult with passive microwave measurement, the new flag must be very valuable in the study of snowfalls over ocean.
This algorithm is not intended to identify all solid precipitation regions, but to detect heavy ice precipitations. Even if this algorithm misses many light ice precipitations, information about the distribution of heavy ice precipitation should be useful to understand the global climate related to snowfall.
This paper describes this simple algorithm and shows some of its test results.
- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner