An experiment to evaluate the use of quantitative precipitation forecasts from numerical guidance by operational forecasters

Operational forecasters have been using numerically generated model output to assist in making forecasts since the late 1960s. Over the years increased computational speed of computers has led to the development and use of high resolution models in the operational forecast setting. Included in these new data sets is high-resolution precipitation output. The increasingly higher resolution of operational models has out paced the development of training to incorporate this increased information into a daily forecast. In addition, these rapid changes in models have not allowed for forecasters to develop an experience level to identify model biases, strengths, and weaknesses.

This study utilizes the National Weather Service's (NWS) Warning Event Simulator (WES) to create a virtual forecasting environment. NWS meteorologists from several field offices used the WES to make a snowfall forecast, as well as a warning decision for two different winter weather events across the Central United States. Meteorologists first developed their forecast and warning decisions without use of any numerically derived precipitation output. They were then allowed to revise their forecast using the precipitation output. Statistics were calculated to compare both forecasts and warnings decisions. A forecast philosophy is developed to provide guidance for forecasters to incorporate this high resolution data into the operational setting.