Projection Pursuit-Based Microcystis Bloom Warning in a Riverside Lake

Abstract

A high-dimensional driving function for Microcystis bloom warning was developed, in which both the inhibition and promotion impacts on Microcystis growth activation energy are integrally considered. Five factors, including flow disturbance, temperature, light intensity, nutrient concentration, and biological inhibition, are embedded in the equation, which results in a high-dimensional problem. The projection pursuit principle was applied for dimension reduction to resolve the numerical problem, and an integrated hydro-environmental model was established. Jinshan Lake, a typical riverside lake, was selected as the research area, and six bloom grades were determined for warning analysis. Based on the established model, the processes of Microcystis growth under varied hydrodynamic conditions were simulated. It was found that the established warning model could well reveal the Microcystis bloom processes in Jinshan Lake. The low-water year was characterized by the largest number of days on which Microcystis bloom might occur for its poor water exchange frequency; The areas where Microcystis bloom might occur in the flood seasons of high-water year, common-water year, and low-water year varied with the uneven-distributed dynamic conditions, which were respectively 0.15, 0.91, and 1.26 km2.