Precipitation Variation in Association with Climate Indices in Loess Plateau of Gansu Province, Northwest China
ZHAO Yi-fei,ZOU Xin-qing,ZHANG Bo,ZHANG Duo-yong,XU Xin-wang-hao
DOI: https://doi.org/10.13249/j.cnki.sgs.2015.10.016
2015-01-01
Abstract:Based on the daily precipitation data of 30 stations in Loess Plateau of Gansu Province from 1961 to 2010, the spatial and temporal changes of precipitation were analyzed by the methods of liner trend, Mann-Kendall method and inverse distance weighted interpolation methods, Mann-Kendall method and IDW. ana-lyzed the cycle characteristics and the relationship between the precipitation and PDO and ENSO by the contin-uous wavelet transform, cross wavelet and wavelet coherence techniques. Then, based on the above analysis, the future trends of precipitation were forecast by the method of rescaled range analysis. The results indicated that the average annual rainfall decreased by-1.54 mm per year, especially since the 1990s precipitation de-crease obviously. In terms of season, In addition to winter precipitation showed a slight increase, the other quar-ter showed a decreasing trend, which decreased the most in autumn, followed by spring, summer. Spatial distri-bution, annual precipitation decreased in most areas in the Loess Plateau of Gansu Province, decrease in precip-itation in the southeast region above the northwest. Precipitation had 2-3 year, 4.3-5 year oscillation circles, which was similar to the pattern of the two large-climate indices. There were the common patterns of 5-5.6 year oscillation circles between precipitation and PDO, ENSO, phase spectrum month precipitation change compared to the PDO and ENSO in advance. Precipitation had 3-3.5 year significant resonance circles with PDO, and close to the same phase shift, as well as has 0-3 year and 3-6 year showed a negative phase of the resonance cycles with ENSO. The Hurst indices of the precipitation are all bigger than 0.5 for all the stations, the fractal dimensions, meanwhile, are all less than 1.5, which indicate that precipitation would keep the same trend with the past in a period of time in the future.