Xiulin Wang, Ocean University of Qingdao, Peiyan Sun Monitoring Center for North Seas Environments, SOA

Abstract

On the basis of research trend of harmful algal blooms (HABs), it is described several methods for predicting and forecasting occurrence of HABs, including experiential predication methods, statistic prediction methods, and numeric prediction methods. The experiential predication methods are based on the experiential analysis for environmental factors influencing or resulting from formation and collapse of HABs, and currently cumulative water temperature method and seawater transparency method have been developed. The statistic prediction methods are mainly based on multi-variance statistics to statically analyze the monitoring data during the formation and collapse of HABs. For example, discriminant analysis, principal component analysis, etc., have been applied. The numeric model prediction methods are constituted of the marine ecosystem dynamics models in which species of HABs is involved, such as nutrient dynamics model and the population dynamics model, etc. From the comments on the prediction methods, it is concluded that in situ monitoring data for the occurrence of HABs, especially during the whole process, and mechanism for the formation and collapse of HABs are necessary and key precondition for establishing the various prediction methods. Furthermore, the in situ monitoring is basic data, and the mechanism can provides theoretic basis for establishing the prediction methods, even for both the experiential and the statistic prediction methods.