An artificial neural network approach for studying phytoplankton succession

Artificial neural networks are used to model phytoplankton succession and g ain insight into the relative strengths of bottom-up and top-down forces sh aping seasonal patterns in phytoplankton biomass and community composition. Model comparisons indicate that patterns in chlorophyll a concentrations r esponse instantaneously to patterns in nutrient concentrations (phosphorous (P), nitrite and nitrate (NO2/NO3-N) and ammonium (NH4-H) concentrations) and zooplankton biomass (daphnid cladocera and copepoda biomass); whereas l agged responses in an index of algal community composition are evident. A r andomization approach to neural networks is employed to reveal individual a nd interacting contributions of nutrient concentrations and zooplankton bio mass to predictions of phytoplankton biomass and community composition. The results show that patterns in chlorophyll a concentrations are directly as sociated with P, NO2/NO3-N and daphnid cladocera biomass, as well as relate d to interactions between daphnid cladocera biomass, and NO2/NO3-N and P. S imilarly, patterns in phytoplankton community composition are associated wi th NO2/NO3-N and daphnid cladocera biomass; however show contrasting patter ns in nutrient- zooplankton and zooplankton-zooplankton interactions. Toget her, the results provide correlative evidence for the importance of nutrien t limitation, zooplankton grazing and nutrient regeneration in shaping phyt oplankton community dynamics. This study shows that artificial neural netwo rks can provide a powerful tool for studying phytoplankton succession by ai ding in the quantification and interpretation of the individual and interac ting contributions of nutrient limitation and zooplankton herbivory on phyt oplankton biomass and community composition under natural conditions.