Algae bloom has become a serious problem of global concern. Scientists have managed to study it using various mathematical models with different degrees of complexity. However, these conventional modelling approaches are limited due to the complexity of the processes involved, the scarcity of data and spatial heterogeneity. In this study, hybrid soft computing (SC) algorithms, including support vector machine (SVM), genetic algorithms (GA) and cellular automata (CA) are illustrated and then employed to model chlorophyll-a (Chl-a) in Bohai Bay, China, with two models. In the first model, SVM tuned by GA (GA-SVM) is developed to model Chl-a for its greater capacity in dealing with nonlinear complex relationships. Then, in order to take into account the spatial heterogeneity and local interaction of the blooms, an integrated model of CA and SVM (CA-SVM) has been developed to model Chl-a in Bohai Bay using remote sensing data. Through this study, it can be observed that the hybrid SC approach is the prefered tool to predict the concentration of Chl-a in Bohai Bay, and capture the non-linear information in ecological processes. This work should be helpful in understanding marine ecological processes, protecting coastal aquatic ecosystem and marine management.