Establishing SMMS approach to accurately mine the characteristics of regional precipitation trends

Abstract

Under the joint influence of climate change and human activities, the temporal and spatial variability and trend of precipitation have changed. The trend of precipitation investigated using traditional methods is quite different, which may be opposite of that at adjacent stations. This contradiction may lead to erroneous decision making in regional water resources planning and management, and may cause safety hazard. In order to evaluate the variation in temporal and spatial characteristics of regional precipitation and avoid different trends of regional precipitation time series, this study developed System clustering - Maximum entropy - Multi window sliding trend - Same frequency trend (SMMS) approach, which leads to the method for analyzing regional precipitation changes, trends, and laws. The local microclimate zones (LMZs) are redefined, so that the general law of precipitation variation will not be significantly different from the actual situation and the SMMS method is more accurate and comprehensive in identifying regional precipitation trends compared to traditional methods. Making use of the SMMS approach, the trend analysis of regional representative stations method can not only reasonably divide the study area and reduce computational complexity, but also analyze it according to the typical representative station with the maximum regional information. Furthermore, the impact caused by the length of data can be avoided, and the trend of precipitation intensity can be identified. It is necessary to fully understand the new characteristics of temporal and spatial changes of precipitation and make early preparations to mitigate the potential impact of climate or weather disasters.