Study of Lake Eutrophication Tendency Based on Gray-Markov Forecast Model

In order to analyze the lake eutrophication tendency in China efficiently, Grey-Markov forecast model is used to forecast the eutrophication tendency of three lakes, which are Tai Lake, Dian Lake and Chao Lake. The original time-series data have the characteristics of tendency and random fluctuation. In order to increase the predicting precision, this method not only increases the procedure of abnormal data testing to the grey forecasting model but also simplifies Markov forecast process. The forecast results show that using the Grey-Markov forecasting model, the Absolute Mean Relative Percentage Error(AMRPE) of three lakes are 3.59%, 1.73%, 2.2% respectively, and the forecasting precision can reach 97.5% , which is 0.32% higher than GM (1,1) Grey forecasting model. The forecasting values from July 2005 to Dec 2005 show that the eutrophic state of Dian lake have transferred to moderate from severe, Chao lake keeps low eutrophic state and Tai lake keeps moderate, both keep invariably in two years.