广西极端降水事件气候态及其对ENSO的潜在响应

本研究利用日本人文与自然研究所和日本气象厅气象研究所发布的0.25°×0.25°亚洲区域高解析度观测集成降水数据（APHRODITE)，并结合NOAA发布的Niño3.4区海温指数和美国国家环境预报中心NCEP再分析数据，分析了广西地区不同极端降水事件的气候平均态和气候倾向率的时空分布，并通过快速傅里叶变换方法分析了不同极端降水逐月区域平均值的气候变率。在此基础上，对ENSO不同位相的周期交替对广西不同极端降水的逐月区域平均值周期的影响及其滞后性进行交叉相关分析，并得出不同极端指标对不同ENSO位相的潜在响应及其显著性特征。结果表明1951-2007年广西全区除无雨日数呈现增加趋势外，暴雨日数、连续干期和湿期均呈减少趋势。不同极端指标在3~7 a的傅里叶谱周期内均呈现显著的周期性变率，说明其与ENSO的潜在联系。交叉相关分析表明Niño3.4指数与无雨日数（暴雨日数）和连续干（湿）期具有显著的负（正）相关特征，其滞后期约为6~18个月，但不同极端指标与ENSO相关程度具有差异。通过滞后6~18个月的平均Ni?o3.4指数所确定的不同ENSO位相，发现相对于拉尼娜冷位相，广西无雨日数（暴雨日数）和连续干（湿）期在厄尔尼诺暖位相下总体呈现减少（增多)，但该响应的统计显著性在全区分布不普遍。这可能与湿静力稳定度、低层温度露点差与中层湿位涡正压项等热力因子的响应不利于极端降水增加有关。

Climatology of the Extreme Precipitation Events over Guangxi Province andIts Potential Responses to ENSO

The climatological distributions and the long-term trends in the extreme precipitation events were investigated over Guangxi province, China. The extreme precipitation events were diagnosed using the 0.25°×0.25° APHRODITE’s (Asian Precipitation-Highly Resolved Observational Data Integration Towards Evaluation) daily gridded precipitation from the Research Institute for Humanity and Nature and Meteorological Research Institute of Japan Meteorological Agency. The Fast Fourier Transform (FFT) was used to identify the periodicities of the temporal variations in the different extreme precipitation events. A two-dimension Cross-Correlation Analysis (CCA) was performed to examine the lag-correlations between the Niño3.4 index from the CPC/NOAA and the different extreme precipitation indicators. Based on the lagged periods indicated by the CCA, the potential responses of the different extreme indicators to the different phases of ENSO were analyzed. The result shows a slight increasing trend in dry days (DD) and slight decreasing trends in heavy-rain days (HRD), consecutive dry days (CDD) and consecutive wet days (CWD) during 1951-2007. The FFT power spectrums of the different extreme precipitation indicators exhibit significant spectrum power with periods around 3-7 years, implying a potential association with ENSO. The CCA indicates a positive (negative) correlation between the wet (drought) events and ENSO with a lagged period around 6-18 months. The mean Niño3.4 index averaged during the 6-18 months, as indicated by the CCA, was used to identify the corresponding phases of ENSO for the extreme precipitation in each month. The potential responses in the El Niño phases relative to the La Niña phases show a general decrease (increase) in the DD (HRD) and CDD (CWD). However, these potential responses are not statistically significant in most of the region over Guangxi province. As shown by the NCEP/NCAR 40-year reanalysis data, the responses of moist static stability, the lower-level difference of temperature and dew-point, and the mid-tropospheric barotropic moist potential vorticity tend to be unfavorable for the increase in extreme rainfall in the El Niño phases, which may limit statistical significance of the responses of the different extreme precipitation indicators.