Lake Dianchi, one of the main water sources for Kunming, China, experiences severe cyanobacterial blooms due to rapid urbanization and local industrial development. Scientific interest in the mechanisms that cause blooms has been increasing. An integrated model combining rough set theory with binary logistic regression was used to examine the correlation between weather conditions and cyanobacterial blooms in Lake Dianchi based on daily monitoring data. The binary logistic regression yielded quantitative correlations between cyanobacterial blooms and the assessed meteorological variables, including temperature, wind velocity, and wind direction. The rough decision process connected the weather conditions and cyanobacterial blooms, which were used to verify the binary regression model results. It was shown that by comparing the methods, the rough decision-adjusted binary logistic regression model significantly improved model accuracy. The integrated model of cyanobacterial blooms in Lake Dianchi may inform decision-makers at local water purification plants of the water quality in the lake and assist them in making more cost-effective decisions.
