Intercomparison of bulk microphysics schemes in simulations of polar lows

In this study, we investigated the performance of five mixed-phase bulk microphysics schemes (BMS) on the simulations of polar lows. Four spiraliform polar lows, two over Japan Sea and two over Nordic Seas, were simulated with WRF model. Five different BMSs provided with WRF were run respectively for each case. The precipitation intensity products from the University of Wisconsin Precipitation Algorithm (UWPA) for the Advanced Microwave Scanning Radiometer for EOS (AMSR-E) and surface radar data around Japan were used for validation. It was found that BMS didn't have much effect on the location of polar low development. However, the pattern, intensity and structure of precipitation were quite different. Purdue Lin scheme produced more light precipitation coverage and larger precipitation rate than others. The primary hydrometeor in the Lin scheme was graupel while snow was dominated in other BMSs.

The domain-averaged precipitation rate were computed around the regions of polar lows. It was shown that Lin scheme had always the largest average precipitation rate while in others the average precipitation rate were close. For two Japan Sea cases, Lin scheme was consitently the best one close to the validation data. Other BMSs produced lower precipitation for both of the Japan Sea cases. For two Nordic Seas cases, all the 5 BMSs produced higher precipitation than the UWPA products.

The difference of model performace between Japan Sea and Nordic Seas could be induced by three reasons. First, UWPA products were likely to overlook the light precipitation. Over Nordic Seas, it was colder and less moist than over Japan Sea. Also the two cases over Nordic Seas were shallower than the cases over Japan Sea. Most of the precipitation intensity were smaller than 2 mm/hr so that they may be misinterpreted by the UWPA algorithm. Second, the initial data over Nordic Seas was not so representative as over Japan Sea. The model may overestimate the precipitation rate over Nordic Seas. The third reason may be that the BMSs didn't represent this kind of cold precipitation well at high latitude.