The Loess Plateau of China (CLP) is located in the transition zone from a semi-humid climate zone to semi-arid and arid climate zones. It is influenced by the westerly circulation, plateau monsoon, and East Asian monsoon circulation, and the drought disasters across the CLP have obvious regional characteristics. In this study, climate regionalization was performed by a spatial hierarchical cluster approach based on the gridded datasets of monthly precipitation across the CLP from 1961 to 2017. Then, the standardized precipitation index (SPI) was used to explore the temporal evolution of regional meteorological droughts. Finally, wavelet methods were used to investigate the drought cycles in each homogeneous subregion and the linkages between SPI and the Southern Oscillation Index (SOI). The results show that: (1) Spatially, the CLP can be divided into four homogeneous regions, namely, Ordos Plateau semi-arid area (Region I), Northern Shanxi hilly semi-humid area (Region II), Longzhong plateau cold-arid area (Region III), and Fenwei Plain and Shaanxi-Shanxi hilly semi-humid area (Region IV). (2) There are apparent differences in the temporal evolution of meteorological droughts in different subregions, but two wet periods from the 1960s to 1980s and 2010s, and a drought period in the 1990s, can be found in each subregion. (3) There is a significant drought cycle of 3–8 years in the four subregions, and the first main cycles of drought variation are not completely consistent. (4) The linkages between SPI and SOI are time- and space-dependent and the phase differences are dominated by in-phase. The strongest correlations between the two time series occur in the 1980s in the four subregions. The results of this research have important implications for the establishment of drought monitoring programs in homogeneous climate regions, and informed decision making in water resource management.
Water shortages have always been the primary bottleneck for the healthy and sustainable development of the ecological environment on the Loess Plateau (LP). Proper water resource management requires knowledge of the spatiotemporal characteristics of precipitation frequency. This paper employed the gridded precipitation dataset obtained from the China Meteorological Data Service Centre to present a spatially explicit characterization of precipitation frequencies in tandem with their return periods on the LP based on the L-moment method. The 60% and 80% of the mean annual precipitation from 1981 to 2010 were synonymous with severe and moderate droughts, respectively. Droughts occurred more frequently in the northwest than in the southeast of the LP. Moreover, the frequencies of moderate drought showed a slight difference throughout the area, while those of severe droughts demonstrated considerable differences between the northwestern arid zone and the southeastern semi-humid zone. The maps associated with various return periods of precipitation deficits can be used to produce drought risk maps together with drought vulnerability maps. These findings could also provide useful information for drought management, water resource management and the development of food security policies.