基于多元回归理论的太湖湖泛预警模型研究

在太湖宜兴段藻源性湖泛高发区设立4个监测点,以湖泛发生的物质基础"藻类生物量"为研究对象,运用数据分析软件SPSS对监测点的藻类生物量、水质、气温等数据进行相关分析,建立了以藻密度为因变量的多元逐步回归模型。结合往年太湖藻源性湖泛发生时的气象条件等历史资料以及相关藻密度阈值的报道,构建了太湖宜兴段藻源性湖泛高发区监测预警模型系统,该模型能够基于监测点的实时水质数据和气象预报数据,对监控区域湖水在未来某时间段内发生湖泛风险的可能性进行分级预警。

Study on the Early Warning Model of Feculent and Anaerobic Water Aggregation in Taihu Lake Using the Multiple Regression Theory

Feculent and anaerobic water aggregation (FAWA) is a type of specific environmental disaster with characteristics of occurring abruptly,lasting in short duration,and causing serious consequences on the environment and ecosystems.Experiments choosing algal density as the investigated object were carried out to analyze the parameters using data analysis software SPSS,including the water quality,temperature,and algal density,in four monitoring sites of the Yixing part where FAWA happened historically.An early warning model using the multiple regression theory was constructed based on the algal cell density together with weather conditions and related algal density threshold when FAWA happened in the past years.Based on the current weather data and real time water quality parameters in the monitoring site,the model could be used to evaluate the risk grade of FAWA occurring in the monitoring region.