Weather Forecast Verification

A collection of ideas, references, and resources for forecasters, researchers, teachers, and students.

***Note that this page is constantly available for new construction. If you know of something that should be here and isn't, please don't be offended. Send e-mail to Harold Brooks (Harold.Brooks@noaa.gov). I'll be glad to put a link to it if it's available on the Web or put a reference to it.***



(Updated 31 October 1997. Maintained by Harold Brooks)

Motivation

I started this page because of my need for information.

First, I became the head of the Mesoscale Applications Group at the National Severe Storms Laboratory. We do research that is relevant to the Storm Prediction Center. One of our big tasks is to create a flexible and useful system for verifying SPC forecasts.

Second, I'm putting together a course on the topic of measuring forecast quality and value. It's first scheduled incarnation will be in the fall of 1998 at the University of Oklahoma School of Meteorology. The course is being designed so that it can be taught as a short course (e.g., two hard-core weeks) or as a semester-long offering. I'd be delighted to talk to anyone about collaborating on teaching or providing materials for teaching a similar course at other institutions. I'd really love it if the course could take on a life of its own, such as the CHANCE course has done with a number of people working on it and teaching slightly different versions of it at different places.

Importance of Forecast Verification

Verification allows a forecaster (or the institution) to identify strengths and weaknesses of a forecast system. It helps identify areas where the application of effort and resources can improve the forecasts. Issuing a forecast without verifying it is an implicit admission that the quality and value of the forecast are not important.

New Items

Glossary

Bibliography

Data Sets

Other Links