Sensitivity analysis of standardized precipitation index to climate state selection in China

In the calculation of the standardized precipitation index (SPI) index, it is necessary to select a certain period of precipitation samples as the reference climate state, and the SPI obtained by different reference climate states have different size. Therefore, the influence of different reference climate states on the accuracy of SPI calculation is worth analyzing. Based on the monthly precipitation data of 1184 stations in China from 1961 to 2010, the influence of the selection of the reference climatic state in the calculation of SPI was analyzed. Using 30 consecutive years as the duration of the reference climatic state, 1961-2010 is divided into three periods 1961-1990, 1971-2000, 1981-2010. Taking the SPI obtained from the entire period as the standard value, the spatial distribution of SPI error and the accuracy of SPI classification based on each reference period were analyzed. Then, the resampling method was used to analyze the influence of time-continuous precipitation samples on the size of SPI. The results show that the SPI error of most sites is less than 0.2, and the accuracy of SPI classification is more than 80%. Although the errors of SPI mostly come from extreme drought and extremely wet, this does not affect the accuracy of the recognition of extreme drought and extremely wet. In most regions, it is reliable to calculate SPI based on the precipitation data of continuous 30 years, but the reliability of SPI is relatively low in areas with frequent drought. The results of the resampling analysis and 30-year sliding analysis show that the distribution parameters have noticeable turning characteristics, and the precipitation distribution parameters of nearly 85% stations had noticeable turning point before 1985, which led to the precipitation data of continuous 30 years easily overestimate the dry/wet.