VAWS: A Kriging-Based Quantitative Precipitation Estimate for the Alaska Region With Confidence Bounds

Speaker: Brett Hoover, University of Wisconsin-Madison

In collaboration with NWS Alaska and the Alaska-Pacific River Forecast Center, we have developed a quantitative precipitation estimate (QPE) for 6-hour accumulated precipitation over the Alaska region by merging rain gauge data with high-resolution numerical weather prediction forecast data. Merging is achieved through kriging, a geostatistical interpolation technique that interpolates corrections at rain gauge sites to the QPE grid as well as an estimate of interpolation error. This error estimate is then refined by cross-validation to provide 5% and 95% confidence bounds on the QPE, providing uncertainty information as a potential range of QPE values at each grid point. The resulting QPE product can be utilized for probabilistic or ensemble-based forecasting products of precipitation hazards including fire-weather risk and hydrological model forecasts of flooding.

The Virtual Alaska Weather Symposium (VAWS) is a collaboration between ACCAP, the Geographic Information Network of Alaska, and the NOAA National Weather Service. We present cutting-edge technologies in satellite remote sensing, forecasting, and modeling to a statewide audience through this webinar series.

If you have questions about the presentations, please email satellite@gina.alaska.edu.