88th Annual Meeting (20-24 January 2008)

Tuesday, 22 January 2008: 1:30 PM
Use of a Geographic Information System to assess gridded weather forecasts
206 (Ernest N. Morial Convention Center)
John B. Settelmaier, NOAA/NWS, Fort Worth, TX
Poster PDF (955.8 kB)
In recent years, the NWS has begun issuing forecasts in gridded form. These forecasts are available through the National Digital Forecast Database and the National Digital Guidance Database. In contrast to the traditional NWS point- or zone-based forecasts, these gridded products can be evaluated using new geospatial verification methods and techniques.

An automated process has been developed to gather, convert, process, and display gridded hydrometeorological datasets to assess their accuracy and value. The use of a Geographic Information System (GIS) to compare various gridded forecasts with observations can enable forecasters and their customers to determine the utility of forecasts that, while still not perfect, contain useful information when interpreted properly.

Gridded verification of temperature, probability of precipitation, and quantitative precipitation forecasts from 2007 will be shown. The gridded verification data will be compared with official point verification statistics to illustrate the added value of incorporating geospatial information as part of the NWS verification program. For example, GIS-based assessment tools provide users with a variety of analysis, query, and display options to better understand the temporal and geospatial impacts on forecast verification. Including GIS in the verification process acknowledges the inherent geospatial nature of the gridded forecasts. Statistics grounded in GIS can assist forecasters in better understanding the geospatial nature of their forecast performance.

The spatial relationship between locations of climatic anomalies and the verification of gridded temperature forecasts will be shown. Developing an understanding of the spatial patterns common to both verification data and climatic anomalies can lead to use of forecast anomalies to not only improve forecasts, but also optimized future forecast processes.

Supplementary URL: http://www.srh.noaa.gov/srh/ssd/assess/assessment.html