A unified verification system for operational models from Regional Meteorological Centres of China Meteorological Administration. (10th June 2013)

• Record Type: Journal Article
• Authors: Chen, Jing; Wang, Yu; Li, Li; Zhao, Bin; Chen, Fajing; Li, Yinglin; Cui, Yingjie
• Is Part Of: Meteorological applications. Volume 20:Number 2(2013:Jun.)
• Page Start: 140
• Page End: 149
• Publication Date: 2013-06-10
• Subjects: Regional Meteorological Centers of China Meteorological Administration -- operational limited‐area models -- unified verification system -- functionality and flow chart -- traditional verification measures -- relative improvement ratio -- comparison of model performance; Meteorology -- Periodicals; Meteorological services -- Periodicals; 551.5
• DOI: 10.1002/met.1406
• Languages: English
• ISSNs: 1350-4827

Abstract

ABSTRACT: A new unified verification system was set up for evaluating operational models from Regional Meteorological Centres (RMC) of the China Meteorological Administration (CMA) using traditional verification methods. The functionality of the system includes data transmission and processing, forecast verification, new product validation and operational monitoring. Eight limited‐area models from different regions of China were verified in this study, including Weather Research and Forecasting (WRF) models from four RMCs. Global/Regional Assimilation and Prediction System (GRAPES) models from two RMCs, Mesoscale Model version 5 (MM5) and Advanced Regional Eta‐co‐ordinate Model (AREM) from the remaining two RMCs. Real time ground observations and model analysis are used as truth for surface and upper‐air variables, respectively. The variables verified are 500 hPa geopotential height, wind and temperature at 850 hPa, wind at 250 hPa, 24 and 3 h accumulated precipitation, and 2 m temperature. The results of the RMCs' models were also compared with those from a limited‐area version of the GRAPES model (GRAPES‐MESO) run operationally at the Numerical Weather Prediction Center (NWPC)/CMA. A new score called 'relative improvement ratio' was defined for continuous variables in the comparison. The preliminary results for the year 2011 showed that forecast error varied from one RMC to another, e.g., performance is more stable with relatively lower error for the regional centres of Shanghai, Beijing, Guangzhou and Shenyang than others. These centres provide also more accurate predictions of the rain‐shine category and of precipitation amount, especially the moderate rain forecast in 3 h accumulation. Copyright © 2013 Royal Meteorological Society