In the data-sparse region of northern Nebraska, convective precipitation forecasting and validation is a difficult task. Operational meteorologists must often rely on numerical output of quantitative precipitation forecasts (QPF) when forecasting in this region during the warm season. Improvement of QPF accuracy was labeled a priority by the National Center for Environmental Prediction (NCEP) as well as the general meteorological research community (Fritsch et al., 1998). Moreover, research has shown that MCS-type complexes can be responsible for a significant portion of the Central Plains’ yearly warm season precipitation total, adding to the importance of improving accurate prediction. Thus, a quantitative assessment of model forecasted precipitation is conducted. The motivation for this experiment is the thorough testing and evaluation of model output over a relatively small domain to assess QPF performance over this geographically limited area. An initial step to evaluate this issue is explored with a numerical simulation of an MCS that occurred over north central Nebraska during the overnight hours of 11 June 2000. In this study, the PSU/NCAR MM5 is used to conduct a 48-hour simulation of this event with a triple-nested mesh and horizontal resolutions of 36, 12 and 4 km.

A matrix of sensitivity studies of model physics is performed in an attempt to identify statistically the optimal model configuration for this MCS event. In particular, model microphysics, radiation scheme and cumulus parameterization are varied in a number of experiments to explore differences in precipitation timing and spatial pattern. Differences between the 12 and 4 km model runs are also examined. Forecast validation statistics, such as threat score (TS), bias (BIA), mean error (ME) and other nonprobabilistic measures of model predictive ability are compiled and analyzed. Comparisons are made between MM5 output and Level III NEXRAD derived storm total precipitation estimates from the North Platte, NE (KLNX) NWS radar. It is important to note, however, that additional simulations and sensitivity experiments conducted under a wider array of warm season meteorological conditions, including linear and non-linear type MCSs, supercell events, etc., need to be assessed. These tests must be completed in order to determine the most accurate model configuration for these deep moist convective events. This study merely serves as an initial investigation of the topic.