Spatial Forecast Verification Methods Inter-Comparison Project

About the ICP

Recent advancements in weather forecasting and observational systems have created great improvements in resolution and prediction. However, use of standard verification practices often indicate poorer performance because, among other things, they are unable to account for small-scale noise or discriminate types of errors such as displacement in time and/or space (see papers in the references section). This issue has motivated recent research and development of many new verification techniques for handling spatial forecasts.

The intent of this project is to compare the various newly proposed methods to give the user information about which methods are appropriate for which types of data, forecasts and desired forecast utility. The point is not to determine which method is best. However, it is hoped that advantages and disadvantages of the methods will become clear.

Most new methods are concerned with gridded verification sets, and that has been (so far) the thrust of the ICP.

Use the links to the left to explore the most up-to-date information about this meta-verification project, including papers from the recent special collection of Weather and Forecasting, test cases used or being used, past and upcoming meetings, available software, references of other relavent papers not part of the special collection, and initial results for some methods.

News

Check here for the latest ICP news.

*Any information collected is used solely to determine the legitimacy of subscription requests (e.g., to protect against spam). Email addresses are added to a controlled list (only subscribers may send messages). The list is intended primarily for participants in the ICP, but others wishing to keep updated on the progress of the project may also subscribe. If you have any trouble subscribing (or unsubscribing) contact the webmaster (Eric Gilleland).

The National Center for Atmospheric Research is sponsored by the National Science Foundation. Any opinions, findings, and conclusions or recommendations expressed in this publication are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.