Longitudinal analyses of hydro-meteorological variables are extremely important for climate studies and water resources planning. Precipitation across the most severely eroded areas of Huangfuchuan Watershed in the Loess Plateau, China was analysed using data from 10 rainfall stations during the period 1954-2010. The ensemble empirical mode decomposition (EEMD), Hurst exponent, and Mann-Kendall methods were utilized to detect periodicities, discontinuities as well as long-term persistence of precipitation. The results show abrupt changes (i.e. discontinuities) in spring during the period 1963-1969 and in 1975, the summer period of 1962 and 1986-1994, the autumn of 1978, and the winter of 1964. These abrupt changes were determined to be statistically significant at the P = 0.05 level. There were inter-annual periods of quasi-3- and quasi-6-year for annual and the four seasons, and decadal periods of quasi-10-, quasi-15-, quasi-25- and quasi-50-year for different seasons. However, periodical features in inter-annual periods were not statistically noticeable. Moreover, Hurst exponent analysis indicated that the current trends of precipitation over the four seasons would continue in the future. The results also indicate that the EEMD method is able to effectively reveal deviations in longterm precipitation series at various timescales and could be utilized for complex analysis of nonstationary and nonlinear signal change. These findings could provide important information for ecological restoration and farming operations across the study region.
Keywords
Ensemble Empirical Mode Decomposition, Periodicities and Discontinuities, Precipitation Variability, Watershed.
User
Font Size
Information