The subgrid variability of surface precipitation in synoptic systems
Szeto, K.K., P.M.K. Yau, and B. Proctor
J. Geophys. Res.
(Submitted)
ABSTRACT
Several mid- and high-latitude synoptic systems were simulated by using a cloud resolving model to provide synthetic precipitation datasets for investigating the subgrid variability of precipitation in different storm regions. Despite the widely different storm characteristics of the test cases, characteristic probability density functions (PDFs) were found for precipitation fields associated with different storm regions. Cold-frontal precipitation is typically characterized by the lack of a local peak in the PDF and the presence of a long tail in the high precipitation rate limit. On the other hand, precipitation in the occluded and warm-frontal region is characterized by a bimodal PDF with two distinct peaks in the distribution. These different characteristics of the PDFs can be interpreted qualitatively from the distinct storm dynamics, and thus the subsequently different precipitation structures, in the various storm regions. Fitting of the model data with three theoretical PDFs (the exponential, Gamma and log-normal PDFs) were examined. All three theoretical PDFs have some difficulties in representing the extreme convective precipitation events in the cold-frontal cases and the bimodal distribution in the warm-frontal cases. In all the meteorological situations considered, the Gamma PDF is found to be a better model of the subgrid variability of precipitation than either the exponential or log-normal PDFs. These results will be valuable in the development of improved precipitation downscaling schemes for use in climate models.