近57年来黄土高原干旱特征及其与大气环流的关系

通过黄土高原地区52个气象站点1961-2017年的气象资料，利用不同尺度的标准化降水指数（Standardized Precipitation Index，SPI）和标准化降水蒸散指数（Standardized Precipitation Evapotranspiration Index，SPEI）对该区57年来干旱的时空变化特征进行了分析，并利用交叉小波变换探讨了干旱指标与大气环流的遥相关分析，得到了以下结论：（1）时间变化上，黄土高原57年来不同干旱指标均呈下降趋势，整体逐渐变干旱。但SPEI6指标较SPI6指标相比，干旱年份更多，干旱特征更明显，说明SPEI的计算由于考虑了蒸散发输入因此结果偏重.不同指标均显示，1999年以来，黄土高原地区干旱时有发生，但整体有降低的趋势。（2） SPEI6和SPI6的站次比和干旱强度最高点都出现在1999年，但SPEI6的站次比和干旱强度的变化幅度更剧烈，且出现全域性干旱的年份（5年）也多于SPI6（3年）；SPEI12相较于SPI12，站次比和干旱强度较为相似，都在1966年达到顶峰，虽然出现全域性干旱的年份SPEI12（9年）多于SPI12（3年），但SPI12的干旱强度更高。（3）平原区的汾渭平原是轻旱多发区，河套平原、宁夏平原易发生中旱，同时宁夏平原还是重旱多发区。丘陵区西部的中宁、同心两地易发生重旱，乌审旗出现特旱。山地区干旱频率普遍较高，尤其是西部山地区的乌鞘岭重旱、特旱频发。（4） SPEI指数对环流指数的变化更敏感。AMO对区域各干旱指标的影响较小，ENSO、WPI对SPI6、SPEI6有显著的响应；而PNA对6个月尺度的干旱指标（SPI6、SPEI6）影响较小，对12个月尺度的干旱指标（SPI12、SPEI12）影响较大。区域干旱是一个复杂的自然现象，为了进一步探索不同干旱指标在不同区域的运用，必要时可采用多种指标，从不同角度比较多种干旱指标的相似性，从而避免单一指标对结果的局限性。

Drought characteristics of the Loess Plateau in the past 60 years and its relationship with changes in atmospheric circulation

Based on the long time series (during the period of 1961-2017) of meteorological data from 52 meteorological stations of the Loess Plateau, the spatiotemporal characteristics of reginal drought were analyzed by the standardized precipitation indicate (SPI) and standardized precipitation evapotranspiration indices (SPEI) with 2 time scales. The results showed that (1) during past 57 a, the drought of the study area has presented a significant decrease trend, which indicated the study area being more and more drier. Comparing the result of SPI6 and SPEI6, the SPEI6 index has more drought years than the SPI6 index. This difference is caused by taking the input of evapotranspiration as a variable in SPEI calculation. Despite the characteristics of drought occurred frequently in the Loess Plateau since 1999, the overall trend of drought is still decreasing. (2) The station ratio and drought intensity of SPEI6 and SPI6 both appeared in 1999, but the variation of the station ratio and drought intensity of SPEI6 was more dramatic comparing to SPI6, and there were more years (5 years) of global drought than SPI6 (3 years). Comparing to SPI12, the station number ratio and drought intensity of SPEI12 are more strongly correlated, and both reached their peak in 1966. Although SPEI12 (9 years) has more years of global drought than SPI12 (3 years), the drought intensity of SPI12 higher. (3) In the plain area, the Fenwei Plain is a light drought-prone area, the Hetao Plain and the Ningxia Plain are prone to moderate drought, while the Ningxia Plain is a drought-prone area. Zhongning and Tongxin in the west of the hilly area are prone to severe droughts. There is an extreme drought in Wushen Banner. The frequency of drought is generally higher in mountainous areas, especially in the western mountain area of Wushaoling. (4) The SPEI index is more sensitive to changes in the circulation index and Atlantic multidecadal oscillation (AMO) is more sensitive to regional droughts. The impact of the indicators is small. El Niño-Southern Oscillation (ENSO) and Western Pacific Index (WPI) have direct responses to SPI6 and SPEI6; while Pacific North America Index (PNA) has a small impact on the 6-month-scale drought indicators (SPI6, SPEI6), and has a greater impact on the 12-month-scale drought indicators (SPI12, SPEI12). Regional drought is a complex natural phenomenon. In order to further explore the application of different drought indicators in different regions, the multiple indicators can be used when necessary, and the similarity of multiple drought indicators can be compared from different angles, so as to avoid the limitation of a single indicator to the results.