Cardiff University | Prifysgol Caerdydd ORCA
Online Research @ Cardiff 
WelshClear Cookie - decide language by browser settings

A seasonal forecast scheme for the Inner Mongolia spring drought part-II: a logical reasoning evidence-based method for spring predictions

Gao, Tao, Wulan, Wulan, Yang, Zelong, Gao, Jing, Hua, Weiqi, Yang, Peng and Si, Yaobing 2019. A seasonal forecast scheme for the Inner Mongolia spring drought part-II: a logical reasoning evidence-based method for spring predictions. Theoretical and Applied Climatology 136 (1-2) , pp. 703-715. 10.1007/s00704-018-2517-2

Full text not available from this repository.

Abstract

Spring precipitation is the predominant factor that controls meteorological drought in Inner Mongolia (IM), China. This study used the anomaly percentage of spring precipitation (PAP) as a drought index to measure spring drought. A scheme for forecasting seasonal drought was designed based on evidence of spring drought occurrence and speculative reasoning methods introduced in computer artificial intelligence theory. Forecast signals with sufficient lead-time for predictions of spring drought were extracted from eight crucial areas of oceans and 500-hPa geopotential height. Using standardized values, these signals were synthesized into three examples of spring drought evidence (SDE) depending on their primary effects on three major atmospheric circulation components of spring precipitation in IM: the western Pacific subtropical high, North Polar vortex, and East Asian trough. Thresholds for the SDE were determined following numerical analyses of the influential factors. Furthermore, five logical reasoning rules for distinguishing the occurrence of SDE were designed after examining all possible combined cases. The degree of confidence in the rules was determined based on estimations of their prior probabilities. Then, an optimized logical reasoning scheme was identified for judging the possibility of spring drought. The scheme was successful in hindcast predictions of 11 of the 16 (accuracy: 68.8%) spring droughts that have occurred during 1960–2009. Moreover, the accuracy ratio for the same period was 82.0% for drought (PAP ≤ −20%) or not (PAP > −20%). Predictions for the recent 6-year period (2010–2015) demonstrated successful outcomes.

Item Type: Article
Date Type: Publication
Status: Published
Schools: Engineering
Publisher: Springer
ISSN: 0177-798X
Date of Acceptance: 18 May 2018
Last Modified: 04 Jan 2021 12:15
URI: https://orca.cardiff.ac.uk/id/eprint/137119

Citation Data

Cited 1 time in Scopus. View in Scopus. Powered By Scopus® Data

Actions (repository staff only)

Edit Item Edit Item