An Evaluation of the Unrestricted Mesoscale Analysis as Gridded Observations for Spatial Model Verification
Abstract
The Army requires weather knowledge products, which include tactically significant meteorological variables and decision aid products impacting Army missions and systems. The Army's Weather Running Estimate Nowcast (WRE-N) model provides forecasts that are ingested into such decision aids. To verify the accuracy of high-resolution WRE-N forecasts, advanced spatial methods of model verification are needed, which require high-resolution gridded observations. An evaluation was conducted to determine the accuracy of a gridded observation product called the Unrestricted Mesoscale Analysis (URMA). This novel evaluation involved applying a grid-to-point verification technique to compare the URMA observations to point ground-truth measurements, and the computation of error statistics for several meteorological variables. The evaluation used hourly URMA observations and point observations collected over a continuous 99-day period from 11 November 2016 to 17 February 2017, characterized as a winter season. The domain of the evaluation was over a large portion of the Western United States and Northern Mexico and included a wide range of terrain types associated with complex mountain-desert-basins, the central and southern plains, and coastal plains, as well as oceanic areas. The results of the study show that the URMA gridded observations can reliably be used for assessment of the WRE-N.
Document Details
- Document Type
- Technical Report
- Publication Date
- Nov 18, 2020
- Accession Number
- AD1115820
Entities
People
- Huaqing Cai
- John W. Raby
- Leelinda Dawson
- Robert Dumais
Organizations
- United States Army Research Laboratory