Control Mechanism of Phytoplankton Biomass Explaining by Data Mining Method of Partial Least-Squares Regression

摘要

Due to many environmental and biology factors influened the phytoplankton biomass, there are multiple predictor variables often themselves related in cases of severe multicolinearity. The paper introduced a new iterative dimensionality reduction data mining method based on partial least-squares regression to studys the control mechanism of phytoplankton biomass. Results showed that control mechnism changed from bottom-up in winter to top-down in summer. There were always negative relations between phytoplankton and phosphate, and phytoplankton was negative correlated with NH~+_4 and NO~-_3 but positve correlated with NO_2 in two seasons.