Into the Simulator: Development of a New Suite of Training Tools to Address the Needs of the Modern Weather Forecaster

The availability of weather data online has increased dramatically in recent years, changing the way students learn about and apply techniques in weather forecasting. The goal of this project is to transform the way forecasting techniques are taught and applied by addressing the needs of the modern weather forecaster. Two such needs in particular – the over-reliance on raw forecast model guidance, and the issue of data inequality that has been introduced by paywall-protected forecasting data sources – are addressed through the development of a new forecasting tool: The Penn State University ReForecast Simulator (PSU-RFS). This tool was developed to create a ‘real-time’ forecasting simulation using events from the climatological period. By choosing events that have already occurred, we are able to control the amount of model guidance given to students, ensuring each student has access to the same data while also achieving desired teaching outcomes. Furthermore, the system can provide a near-instantaneous assessment of each forecast using Brier skill scores, allowing for the class to discuss the outcome of the forecast shortly after preforming it. The PSU-RFS was tested in an upper-level weather forecasting class at Penn State in Fall 2019, and student feedback was collected. The results of this feedback and improvements to the PSU-RFS will be discussed along with future plans for implementing this tool to address future challenges facing forecasters and automation in the weather industry.