img

Wechat

Adv search

JOURNAL OF DESERT RESEARCH ›› 2016, Vol. 36 ›› Issue (2): 441-448.DOI: 10.7522/j.issn.1000-694X.2015.00024

Previous Articles     Next Articles

Objective Verification of GRAPES_SDM Model in Xinjiang, China

Duan Haixia1, Li Yaohui1, Huo Wen2, Qin He3, Ma Yufen2   

  1. 1. Key Laboratory of Arid Climatic Change and Reducing Disaster of Gansu Province/Key Open Laboratory of Climatic Change and Disaster Reduction of CMA, Institute of Arid Meteorology, China Meteorological Administration, Lanzhou 730020, China;
    2. Institute of Desert Meteorology, China Meteorological Administration, Urumqi 830002, China;
    3. Xinjiang Meteorological Observatory, Urumqi 830002, China
  • Received:2014-11-02 Revised:2015-01-12 Online:2016-03-20 Published:2016-03-20

Abstract: By comparing the forecast product by the GRAPES_SDM model and the observed data of dust-storm weather in Xinjiang in the springs of 2008-2013, we tested the forecast product with synoptic verification method. By analyzing mean error, error standard deviation and root-mean-square error, the characteristics of error distribution of numerical forecast in Xinjiang are shown, and the possible sources of the systematic and non-systematic error are analyzed. The results showed that the 2-m height air temperature and 10-m height wind speed error mainly came from the uncertainty of the initial conditions and the inconsistencies of the resolution of observation and prediction. We also proposed some suggestions on improving the GRAPES_SDM model forecast accuracy.

Key words: GRAPES_SDM Model, TS score, mean error, root-mean-square error

CLC Number: