基于SPEI-PM指数的黄淮海平原干旱特征分析

Analysis of drought characters based on the SPEI-PM index in Huang-Huai-Hai Plain

利用黄淮海平原45个气象站点1961—2014年月值气象数据,基于Penman-Monteith蒸散模型计算了标准化降水蒸散指数(SPEI),对黄淮海平原近54年干旱变化趋势、发生频率和持续性特征进行了分析,并探讨了SPEI指数与河南、河北和山东省农业干旱面积的关系,结果表明:(1)改用Penman-Monteith蒸散公式后,SPEI干旱指数在黄淮海平原呈整体上升趋势,即趋于湿润;(2)近54年干旱演变具有明显的年代际差异,20世纪60年代干旱频率最高,而21世纪初(2000—2014)干旱频率整体偏低;(3)黄淮海平原干旱发生具有持续性的特点,20世纪60年代遭受的持续性干旱最为严重,平均干旱持续时长约2.6个月,21世纪初下降到1.5个月;(4)河南、河北和山东省的农业干旱面积年际变化表明,干旱面积呈减少趋势,2000年以后年均受灾面积、成灾面积和绝收面积比2000年之前分别下降了58.0%、44.4%和49.1%;(5)农业干旱面积与SPEI具有中等以上的相关强度,其中对山东省受灾、成灾和绝收面积相关系数r达到-0.7以上,表明基于Penman-Monteith蒸散模型的SPEI指数在黄淮海平原具有良好的适用性。

The quantitative analysis of drought is crucial essential for drought risk assessment. The standardized precipitation evapotranspiration index （SPEI） has been widely used as an effective approach to quantitatively analyze the trends, duration, frequency, and severity of drought. However, how to calculate the evapotranspiration is big challenge for the reliable and accuracy of SPEI results. The most previous studies calculated SPEI based on an empirical evapotranspiration equation, such as the Thomthwaite method, rather than the physical Penman-Monteith equation. Moreover, majority of the studies analyzed the spatio-temporal pattern of the index without linkage of the relationship between the climactic drought index and actual cropland drought areas. This study utilized SPEI based on the PenmanMonteith equation to explore the variations of drought frequency and duration during 1963-2014 in 3H Plain, and established the relationship between SPEI and the actual field drought areas by Pearson correlation in Henan, Hebei, and Shandong Province separately, which aims to understand the trends of drought during the past 50 years and to explore the relationship between the climatic drought index and actual field drought areas. The results showed that the SPEI had upward trend for 1, 3, 6, and 12-month scales in most areas of the 3H Plain while as the 1-month and 3-month scales, the wetting trend was significant （P〈0.1） in the northern regions of 3H Plain. Drought frequency in the 1960s was the highest during the past 54 years, but lowest in 2000-2014. The longest duration was occurred in the 1960s by comparing the decadal spatial distribution of drought duration. Moreover, the mean drought duration declined from 2.6 months in the 1960s to 1.5 months during 2000-2014. The changes of annual drought areas decreased in Henan, Hebei, and Shandong Province, which could be partly attributed to the lower frequency and shorter duration of drought detected by SPEI. Correlation analyses indicated that the index series of Dec-SPEI- 12 had medium to high correlation with observed cropland drought areas. For example, Pearson＇ s r between Dec-SPEI- 12 and the drought area of Shandong Province were -0.7, -0.7, and -0.8 for affected, disaster and no harvest drought areas, respectively. The results of the study suggest a wetting trend in the 3H Plain during the past 54 years, and the chosen of Penman-Monteith evapotranspiration model with decreasing evapotranspiration over the past years in the 3H Plain attributes to this wetting trend. The high correlation between the climatic drought index and actual drought area indicated that SPEI could be used as a reference or trigger for establishing a drought warning system in the 3H Plain.