Prediction of Algal Bloom Using Genetic Programming

In this study an attempt is made to mathematically model and predict algal blooms in Tolo Harbour (Hong Kong) using Genetic Programming (GP). Chlorophyll plays a vital role and is taken as a measure of algal bloom biomass and 8 other variables are taken as input for its prediction. It is observed that GP evolves multiple models with almost same values of errors – of – measure. Previous studies on GP modeling primarily focused on comparing the GP results with actual values. In contrast, in this study, the main aim is to propose a systematic procedure for identifying the most appropriate GP model from a list of feasible models (with almost same errorof measure) using physical understanding of the process aided by data interpretation. The study of the GP-evolved equations shows that they correctly identify the ecologically significant variables. Analysis of final GP evolved mathematical model indicates that of the 8 variables assumed to affect the algal bloom, the most significant effect is due to chlorophyll, total inorganic nitrogen and dissolved oxygen, as far as one week prediction is concerned. For higher lead prediction (biweekly), secchi disc depth and temperature appears as significant variables in addition to chlorophyll.