Annual precipitation analysis and forecasting – taking Zhengzhou as an example

Abstract

Abstract Precipitation is greatly affected by natural conditions and human activities, and its sequence has the characteristics of periodic variation and non-stationarity. In order to study the periodic characteristics of regional precipitation, reduce the non-stationarity of the water sequence and improve the forecasting accuracy, the Morlet wavelet model and CEEMD − Elman + ARIMA model are introduced, and applied to the precipitation sequence of Zhengzhou. The results show that the annual precipitation in Zhengzhou is mainly affected by periodic fluctuations of 55a, 40a and 15a, and the mean absolute error of the CEEMD − Elman + ARIMA model in 2013–2017 is 14.1%. Based on the analysis of precipitation period and forecast of future precipitation, the characteristics and evolution of precipitation in Zhengzhou are revealed. This provides a theoretical basis for rational utilization of water resources, prevention, industrial and agricultural production in Zhengzhou.