CAPS Realtime Storm-scale Ensemble and High-resolution Forecasts as Part of the NOAA Hazardous Weather Testbed 2008 Spring Experiment

In the spring of 2008, the Center for the Analysis and Prediction of Storms (CAPS) continued its participation in the NOAA Hazardous Weather Testbed (HWT) Spring Experiment by providing 4-km storm-scale ensemble and 2-km high-resolution numerical forecasts for evaluation by HWT. Improving upon the configurations used in the spring of 2007, the 10-member 4-km ensemble included initial and boundary condition perturbations as well as physics perturbations in each member. Furthermore, level-2 radial velocity and reflectivity data from over 120 operational WSR-88D radars were analyzed into the initial condition (IC) using the ARPS 3DVAR system and cloud analysis package, and the analysis is performed for both 4 and 2 km horizontal resolution grids. Initial and boundary condition (BC) perturbations were constructed from NCEP Short-Range Ensemble Forecast of 00 UTC cycle each day, and added to the control member's IC (and BC) that included radar data. One of the ensemble members differs from the control member only in the exclusion of radar data in the IC. The domains of both grids cover almost the entire continental U.S, and the daily 30-hour forecasts started at 00 UTC during an 8 week period. The WRF ARW model and its physics options were used. About 1500 CPUS of a Cray XT-3 supercomputer at the Pittsburgh Supercomputing Center (PSC) were used for up to 8 hours each night for the forecasts, and pre- and post-processing.

This paper will report on the experimental design and discuss certain logistic issues. It will present examples of ensemble and deterministic forecasts, and verifications against radar observations. Particular emphasis will be given to the value of radar data assimilation and the effect of resolution on short- and longer-range forecasts. A comparison paper will focus on the evaluation of the ensemble forecasts and products.