Skip to main content
U.S. flag

An official website of the United States government

Dot Gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

HTTPS

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

GSD'S Short-term GFE Point Blender

Abstract

The National Weather Service Weather Ready Nation Roadmap within the Forecast Decision Support Environment section calls for better forecaster efficiency when updating the short-term forecast (0-12 hours). In response, the Global Systems Division (GSD) has developed a gridpoint-based model guidance blending technique to improve short-term forecast quality. GSD's Short-term Point Blender archives model guidance and observational grids out of the AWIPS Graphical Forecast Editor (GFE) database and assesses the performance of each model at each gridpoint over the NWS Weather Forecast Office (WFO) domain. Short-term forecast grids are then generated using a weighting scheme derived from that model's performance at each gridpoint. By blending point-by-point, this technique has shown to produce more accurate forecasts than other methods that measure performance as a single value over the entire domain. The GFE Point Blender compliments the the National Blend of Global Models Project being undertaken at the National Centers for Environmental Prediction which plans to provides forecasts for the 3-8 day period. Ongoing testing shows promise with the new technique generally providing a more accurate forecast than any of its model components. Furthermore, it is very configurable with regard to what models are archived and used in the blend as well as the analysis or ?truth? data set which ultimately determines accuracy. Most recently this technique has been extended into a framework that allows for many more guidance combinations as well as a range of methods to correct bias and calculate error. Using this framework users will be able to explore additional blending and verification techniques.

Article / Publication Data
Active/Online
YES
Available Metadata
Fiscal Year
Published On
January 01, 2016
Type
Event

This publication was presented at the following:

Title
2016 - 96th AMS Annual Meeting
Sponsor
Amer. Meteor. Soc.
Type
Conference presentation

Institutions

Not available

Authors

Authors who have authored or contributed to this publication.

Not available