基于SPEI的北京低频干旱与气候指数关系

Abstract

干旱是北京地区发生最频繁、波及面积最大、持续时间最长的一种自然灾害。基于1868—2010 年每月的降水和平均气温数据，应用综合了降水和气温变化共同效应的新的干旱指标标准化降水蒸散指数（SPEI）定量描述北京地区的干湿状况，并利用历史旱灾记录对其进行验证；采用连续小波转换(CWT)分析近150 a来的干旱振荡特征，并利用交叉小波变换(XWT)探论了干旱与大尺度气候因子之间的关系。结果表明：1）SPEI揭示的干旱与历史记录比较吻合，证明该指数可以在多时间尺度上有效地反映北京地区旱涝程度及其持续时间；2）北京地区干旱具有80—120个月年际尺度和250个月、480个月年代际尺度的周期振荡，呈现了同大尺度气候因子相似的变化特征；3）北京干旱变化与四大气候因子存在着多时间尺度的显著相关性，SPEI和北大西洋涛动（NAO）、北极涛动(AO)、太平洋涛动（PDO）都具有100—120个月和250个月的年代际主共振周期，而SPEI和厄尔尼诺-南方涛动（ENSO）在整个研究期内都表现出极显著的32—64个月年际主共振周期，同时SPEI与4个气候因子在共振周期上均体现出比较明确的时滞特征（2—6月不等）。因此，可以基于大尺度气象因子结合SPEI预测北京地区未来的干旱变化。;Drought occurs in nature when precipitation is significantly lower than normal. When lasting many months or even years in a large area, drought will develop into a natural hazard that permanently damages the environment and causes great economic losses. Thus, improving our knowledge about the variability and impacts of drought is fundamental to quantify the drought hazard and improve the prediction and drought mitigation. Beijing is located in the middle and lower reaches of Haihe River Basin, which belong to a temperate continental monsoon climate zone. The precipitation distribution is very uneven, and often accompanied by high temperature. So drought is one of the most frequently and enduring natural hazard that influences most area in Beijing, north China. In this paper, we analyzed the variability and possible teleconnections between drought occurrences and large-scale climate indices between 1868—2010 in Beijng, such as El Niño-Southern Oscillation (ENSO), North Atlantic Oscillation (NAO), Arctic Oscillation (AO), and Pacific Decadal Oscillation (PDO). The drought occurrences were quantified by a new drought index, Standardized Precipitation Evapotranspiration Index (SPEI) based on the data of monthly mean temperature and precipitation. The SPEI considers not only precipitation but also temperature data by means of evapotranspiration in calculation, allowing for a more complete approach to explore the effects of climate changes on drought occurrences under global warming. The SPEI can also be calculated at several time scales to adapt to the critical times of responses to drought in target natural and economic systems and to determine their resistance to drought. Local historical drought hazard records in Beijing since 1868 were used to improve the validation of SPEI. We then used the method of continuous wavelet transform (CWT) to analyze inter-decadal and decadal oscillation in the time and frequency of drought. Finally, we analyzed the correlations between SPEI and four large scale climate indices through the cross wavelet transform (XWT). The good agreement between SPEI and historical drought records proves that SPEI can effectively reflect the intensity and duration of drought in multi-temporal dimension in this region. SPEI of Beijing had 80—120 month, 250 month, and 480 month oscillation circles, which was similar to the pattern of the four large-scale climate indices. The significant coherence was found between SPEI and the four large-scale climate indices. There were the common patterns of 100—120 month decadal and 250 month inter-decadal oscillation circles between SPEI and NAO, AO, PDO, as well as a common pattern of 32—64 month inter-decadal oscillation circle between SPEI and ENSO during the whole period. There was a clear lag time (2—6 months) during the coherence circle. Therefore, we can forecast the future drought variations in Beijing based on the data of large scale climate indices and SPEI, which is useful for water resources management and agriculture. This article is an initial step to application of the new multi-scalar SPEI drought index in studying the drought variability and impacts in China.