水体富营养化响应规律与水华预警方法研究进展  

Progress of Research on Eutrophication Response Law andEarly Warning Method of Water Bloom in Water Bodies

作　　者：陈弘 周庆琛[1] 佟光远 郭冰冰 CHEN Hong;ZHOU Qing-chen;TONG Guang-yuan;GUO Bing-bing(College of Hydraulic Engineering,Tianjin Agricultural University,Tianjin 300392,China;Tianjin Agricultural University-China Agricultural University Joint Smart Water Conservancy Research Center,Tianjin 300392,China;College of Computer and Information Engineering,Tianjin Agricultural University,Tianjin 300392,China)

机构地区：[1]天津农学院水利工程学院,天津300392 [2]天津农学院-中国农业大学智慧水利研究中心,天津300392 [3]天津农学院计算机与信息工程学院,天津300392

出　　处：《中国农村水利水电》2024年第7期117-125,134,共10页China Rural Water and Hydropower

基　　金：水体污染控制与治理国家科技重大专项子课题(2017ZX07106003)。

摘　　要：水体富营养化问题日益严峻,这种现象导致了浮游植物的异常增长,释放有害的藻毒素,造成了水体的污染,并对动植物乃至人类造成了危害。与之相关的,水体通常流动缓慢,连通性差,水流停滞时间较长,使得这些水体更容易受到富营养化的影响,也使水华风险升高。针对水华的产生机理,深入探讨了水动力因素、水环境因素和气象因素的影响。水动力因素包括水体的流速、流量和水体扰动等情况,它们共同决定了藻类的积聚程度;水环境因素包括水温、溶解氧、营养盐浓度以及酸碱度等,它们将直接影响浮游藻类的生长;气象因素如气温、风力等会改变水体的温度,从而对水华的形成和发展产生影响。还依据江河湖库水体的特点,探讨了数值模型和数据驱动模型在水华预测中的应用,数值模型在水华预测中的应用是通过模拟水体中的环境因素,来预测水华的发生和发展趋势;数据驱动模型以数据挖掘算法和统计技术为基础,利用机器学习方法,通过分析和识别监测数据、图片与水华的对应关系,进行迭代学习,创建其与水华爆发的预测联系。它们为探究水体状况、预测水华事件提供了有效工具,二者相耦合的方法对水华的精准预测有着重要意义,也是未来研究的发展趋势,有助于水资源管理者更好地对其进行调控,提高预测精度和水质管理的效率以减少水华的发生。The problem of eutrophication of water bodies is becoming increasingly serious,a phenomenon that leads to an abnormal growth of phytoplankton and the release of harmful algal toxins,which contribute to the pollution of water bodies and pose a risk to flora,fauna and even humans.Related to this,water bodies are often slow-moving,poorly connected and have long stagnation times,making them more vul⁃nerable to eutrophication and increasing the risk of blooms.In this paper,the impacts of hydrodynamic factors,water environment factors and meteorological factors are discussed in depth. Hydrodynamic factors include water velocity, flow rate and water disturbance, which to⁃gether determine the degree of algae accumulation;water environment factors include water temperature, dissolved oxygen, nutrient salt con⁃centration and pH, which directly affect the growth of planktonic algae;meteorological factors such as air temperature and wind will changethe temperature of the water body, which will have an impact on the formation and development of the bloom. This paper also discusses theapplication of numerical model and data-driven model in the prediction of water bloom based on the characteristics of water bodies of rivers,lakes and reservoirs. Numerical model is used in the prediction of water bloom to predict the occurrence and development trend of waterbloom by simulating the environmental factors in the water body;the data-driven model is based on data mining algorithms and statisticaltechniques, and uses the machine learning method to iteratively learn the correspondence relationship between the monitoring data and pic⁃tures and the bloom by analysing and identifying the monitoring data and pictures. correspondence, and iterative learning to create their pre⁃dictive links with the outbreak of waterwaters. They provide an effective tool for exploring the condition of water bodies and predicting waterbloom events, and their coupled approach is of great significance for the accurate prediction of water bloom, which

关 键 词：富营养化 蓝藻水华 水华预警 数值模型 数据驱动模型 

分 类 号：X524[环境科学与工程—环境工程]