Visualizing the Verification of Quantitative Precipitation Forecasts using PyQGIS

The Southeast River Forecast Center produces 48-hour precipitation forecasts twice daily, in addition to quality controlling hourly precipitation estimates. While forecast uncertainty naturally leads to errors, persistent biases can be identified and mitigated by verifying the office's precipitation forecasts against its estimates. In order to investigate the presence of persistent forecast biases, monthly gridded mean errors were calculated across the Southeast U.S for a six year period (2008 to 2014). The flexible, open-source nature of QGIS makes it especially useful in an operational environment, and PyQGIS (Python scripting within QGIS) allows users to quickly create GIS graphics. PyQGIS was used in this study to visualize the statistics across the Southeast and expedite the creation of graphics. In addition to the raw gridded statistics, zonal statistics were calculated and displayed across states and groups of river basins.

This presentation will include a methodology for how the statistics were visualized using PyQGIS and a summary of the results.