关键词: 极端降水, 持续干旱日数, Morlet小波分析, 河东地区

Abstract:

Based on daily precipitation data collected in 13 meteorological stations over Hedong area of Gansu Province, China during 1960-2011, the method of linear regression, inverse distance weighted interpolation, Mann-Kendall test, wavelet analysis and rescaled range analysis are employed to analyze the spatial and temporal variability of extreme precipitation, and the trend of extreme precipitation is also predicted. The results are as follows: (1) During 1960-2011, consecutive dry days have significantly increasing trend. The other indices decrease and only heavy precipitation days decrease significantly at the 0.05 level. There are obvious differences between monsoon sub-area and plateau sub-area. In monsoon sub-area, heavy precipitation days, very wet day precipitation and precipitation on wet days have significantly decreasing trends, whereas consecutive dry days increase significantly. In plateau sub-area, consecutive wet days have asignificantly decreasing trend. (2) Spatially, for consecutive dry days, most of places display increasing trends. The other indices show decreasing trends, mainly concentred in the middle and southeast of Hedong. (3) The departure sequence vibration of the extreme precipitation indices is quasi-20 years and 5-8 years. At the same time, the extreme precipitation indices reflect different structures which are more or less in different time scales. (4) For heavy precipitation days, consecutive dry days, very wet day precipitation and precipitation on wet days, the abrupt change is discovered in the 1970s, whreras the abrupt change of consecutive wet days and maximum 5-day precipitation happen is in the 1980s. (5) In the plateau sub-area, the variation of heavy precipitation days and consecutive wet days is not stable in future. Compared to the past, the other indices are consistent with patterns in future.

Key words: extreme precipitation, consecutive dry days, Morlet analysis, Hedong area