WRF QNSE test and evaluation

There has been a need within the numerical weather prediction community for improvements in the characterization of the planetary boundary layer (PBL) and surface layer. The Quasi-Normal Scale Elimination (QNSE) PBL and surface layer schemes were developed by Sukoriansky, Galperin and Perov, (Sukoriansky et. al, 2005) with the goal of addressing these issues. To assess the performance of the newly available QNSE schemes in the Weather Research and Forecasting (WRF) model, the Developmental Testbed Center (DTC) performed extensive testing and evaluation with the Advanced Research WRF (ARW) dynamic core for two physics suite configurations. One configuration was based on the Air Force Weather Agency (AFWA) Operational Configuration, which utilizes the Yonsei University (YSU) PBL scheme and the MM5 similarity surface layer scheme. This configuration now provides a baseline for testing and evaluating new options available in the WRF system. The second configuration substituted AFWA's current operational PBL and surface layer schemes with the newly available QNSE PBL and surface layer schemes. The initial intercomparison test utilized WRF version 3.1.1, while a retest utilizing WRF version 3.2.1 was also conducted to evaluate the impacts of the newest release of the code.

Objective model verification statistics for both intercomparison tests were generated using the Model Evaluation Tools (MET) software package, for surface and upper air temperature, dew point temperature, and wind, along with precipitation. Each type of verification metric is accompanied by confidence intervals (CIs), at the 99% level, computed using the appropriate statistical method.

This paper will focus on the standard verification metrics of each configuration, as well as the pair-wise differences between the two configurations from the same version of WRF. An assessment of the statistical significance (SS) is also included. Differences between the two versions of the WRF code will also be discussed.