The Automation of Nowcast Model Assessment Processes

Abstract

Nowcast model assessment involves applying model verification techniques to generate error statistics to improve model performance and the accuracy of forecasts produced by US Army Research Laboratory's nowcast model, Weather Running Estimate-Nowcast (WRE-N). This report documents the design and implementation of the automated process of generating domain-level error statistics that can be used by modelers to improve the accuracy of WRE-N model forecasts. This process allows multiple user configurations, and produces a controlled data structure that could be easily used in data analysis and the evaluation of model improvements.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Sep 01, 2016
Accession Number
AD1017503

Entities

People

  • Jeffrey A. Smith
  • John W. Raby
  • Leelinda P. Dawson

Organizations

  • United States Army Research Laboratory

Tags

Communities of Interest

  • Materials and Manufacturing Processes
  • Weapons Technologies

DTIC Thesaurus Topics

  • Accuracy
  • Air Force
  • Automation
  • Case Studies
  • Data Analysis
  • Data Science
  • Directories
  • Errors
  • High Resolution
  • Information Science
  • Military Research
  • Standards
  • Statistics
  • Test And Evaluation
  • United States
  • Verification
  • Weather Forecasting

Readers

  • Atmospheric Science/Meteorology
  • Software Engineering.
  • Team-Based Human-Centered Cognitive Task Decision Making and Information Performance.