This presentation will focus on improvements to forecast verification visualization and how verification results can be used. The QPFVS verifies forecast grids at grid points by using the Unrestricted Mesoscale Analysis (URMA), as well as at station observing points by using METeorological Aerodrome Reports (METARs). The QPFVS utilizes multiple open-source data display technologies to show verification metrics in several forms, including traditional line and scatter plots, and maps of metrics. The verification plots and maps returned are fully interactive, allowing users to quickly and effectively interrogate returned statistics. The flexible and dynamic nature of QPFVS allows for NWS forecasters, managers, and model developers alike to quickly assess short and long term trends in forecast and model errors.
Further, this talk will discuss how the QPFVS suite is designed to become the backbone for MDL’s new national forecast verification process that will facilitate improvements to the quality and quantity of verification metrics as the NWS ventures into probabilistic forecasting. It will also enable the production of forecast verification metrics during impactful weather events. The new verification suite will enable monthly verification statistics to be available in hours instead of weeks, freeing MDL developers to continue to improve verification visualization.