In order to provide technical bases for blue green alga treatment, according to the data on the degree of blue algal occurrence in Dianshan Lake and the observation data from 1998 to 2009, an analysis is made of meteorological conditions for the occurrence of blue green algae by adopting the statistical method, and the prediction model is devised. The meteorological indexes of blue green alga bloom are: the mean temperature in July and August greater than 29℃; the number of hot days with daily maximum temperature above 35℃ is great than 16 days from June to September; the number of monthly sunlight hours is greater than 420 hours in July and August; the monthly rainfall is less than 420 mm from June to August. Based on the inspection of historical data, the result shows that the accuracy of Model Ⅰ (prediction starts in early June) and Model Ⅱ (prediction starts in early August) are 91.7% and 100%, respectively. The trial running in 2010 shows that the prediction result of Model Ⅰis correct and that of Model Ⅱ is 1 grade lower than the real, which can be applied in operational meteorological service.