A statistical prediction model for summer extreme precipitation days over the northern central China. (3rd January 2020)

• Record Type: Journal Article
• Authors: Han, Tingting; Li, Shangfeng; Hao, Xin; Guo, Xinyi
• Is Part Of: International journal of climatology. Volume 40:Number 9(2020)
• Page Start: 4189
• Page End: 4202
• Publication Date: 2020-01-03
• Subjects: extreme precipitation days -- northern central China -- seasonal prediction -- year‐to‐year increment approach; Climatology -- Periodicals; Climat -- Périodiques; Climatologie -- Périodiques; 551.605
• DOI: 10.1002/joc.6451
• Languages: English
• ISSNs: 0899-8418

Abstract

Abstract: The characteristics of summer precipitation over northern central China (NC China) and its associated atmospheric circulation have been documented in many previous studies. However, the prediction of extreme precipitation events, which is vital for water resource management and agricultural yield, has received little attention. In this study, we attempted to establish a statistical model to predict mid‐summer extreme precipitation days (EPDs) over NC China, based on a year‐to‐year increment (DY) approach. Three variables—preceding early‐spring snow cover DY over western China, previous early‐summer sea surface temperature (SST) DY in the North Pacific, and SST DY differences between the equatorial western and eastern Pacific—were selected as potential predictors. The first predictor exerted an impact on precipitable water content anomalies, South Asian high and vertical movement anomalies over NC China through anomalous soil moisture over northern China. The latter two predictors affected the EPDs over NC China via modulating moisture transport originating from tropical oceans. Further results showed that the prediction model performed well in predicting the inter‐annual variation of EPDs, with respective correlation coefficients of .79/.67 between the predicted and observed EPDs DY/EPDs for 1980–2014. The agreement index of the prediction model was 0.67 and the Nash–Sutcliffe efficiency was 0.37, suggesting good capabilities of the prediction model. Both the one‐year‐out cross‐validation test for 1980–2014 and independent hindcast for 1995–2014 demonstrated that these three predictors showed skilful predictions of EPDs over NC China. The cross‐validation (hindcast) simulated EPDs DY achieved a significant correlation coefficient of .74 (.78) with the observations, along with positive Nash–Sutcliffe efficiency values. Notably, the decadal shift of EPDs around the 2000s was successfully replicated in both cross‐validation and hindcast results. Abstract : Three predictors—preceding‐spring snow area cover year‐to‐year increment (referred to as DY) over western China, previous early‐summer sea surface temperature (SST) DY in the North Pacific, and SST DY differences between equatorial western and eastern Pacific—were used to establish a statistical prediction model for mid‐summer extreme precipitation days (EPDs) over northern central China, based on the year‐to‐year increment approach. The predictive model performs well in predicting the inter‐annual and inter‐decadal variations of EPDs in cross‐validation and hindcast results. In the figure, the 139 stations located over northern central China (32.5°–38°N, 105°–110°E) were selected.