Microphysics Schemes Based on Dsd-parameter Constraints and Their Impact on Convective Storm Forecasts

It is clear that in order to accurately model convective scale events, we require the most accurate microphysical schemes possible. However this need for detailed microphysical parameterization must be balanced with the computational expense that such detailed parameterizations require. A new way of calculating the drop-size distribution (DSD) parameters in the single moment (hereafter SM) and double moment (hereafter DM) microphysical scheme available in the Advanced Regional Prediction System (ARPS; Xue et al. 2003) has been implemented and is tested here using the ARPS test case of the May 20, 1977 Del City, Oklahoma tornadic supercell storm. This case has been examined in detail many times previously using a variety of platforms. Recent work by Zhang et al. (2008) used a method of deriving relationships between the moments of the gamma DSD in order to develop a diagnostic relation between the intercept parameter and the water content. The relationships that they derived were based on TwoDimensional Video Disdrometer (2DVD) measurements taken in Oklahoma during the summer seasons of 2005, 2006 and 2007. These data are expected to be representative of the kind of severe convective events that are often witnessed in the Southern Great Plains region. Further work needs to be done to establish whether these derived moment relations are valid for other areas and types of rainfall event. It is well known that the DSD parameters can vary significantly between stratiform and convective events (e.g. Tokay et al. 1995) and Munchak and Tokay (2008) found that the shape parameter of the gamma DSD varies regionally. This reinforces the need to move towards a tunable DSD parameter model, since modeling the governing parameters as constants is clearly making an approximation without good physical basis. It can also be seen in multi-moment model simulations that the DSD parameters can vary widely within a single storm. For example, we get very large drops on the leading edge of strong convective events, as has been witnessed observationally. Model simulations of the May 3 1999 tornadic supercell event performed using the Advanced Regional Prediction System (ARPS, Xue et al. 2003) and the Milbrandt and Yau (2005b) three-moment microphysics scheme illustrate this variation of the DSD parameters within the storm (Dawson et al. 2008). Figure 1 shows the variation of the gamma DSD shape parameter α across the domain one hour into the simulation at 277 m height. It is clear that the shape parameter varies widely across the storm domain. This again reinforces the need for appropriately complex microphysics options, as the assumption of a constant shape parameter (as in the exponential DSD) clearly cannot capture the full complexity of the DSD, which will have impacts for many microphysical processes.