Accelerating the Evaluation of Experimental NWP Forecasts via Crowdsourcing

In order to accelerate the evaluation of experimental model runs at NCEP/EMC, students at Purdue University will coordinate with EMC's Model Evaluation Group (MEG) to assist with their evaluation efforts and provide detailed assessment through classroom exercises. Purdue undergraduate and graduate students will perform detailed analysis and comparison of the performance of experimental and operational NWP forecasts during a pre-implementation evaluation period. This project is intended to be a prototype of a larger-scale crowd-sourced NWP model evaluation project that may be pursued in the near future. During the Fall 2019 semester, Purdue students in the “Weather Analysis and Forecasting” class will examine the operational and experimental RAP/HRRR forecasts on a regular basis, using these to perform routine forecasting tasks that will focus on the Midwest as well as areas over CONUS where significant weather events are predicted. Students will create forecast products and also evaluate how the operational/experimental guidance performed once observations are available. This group will work with EMC/MEG to communicate the results of these evaluations and will also use METplus analysis tools as part of their formal assessments.