Analyzing the Role of Salinity in the Chlorophyll Prediction in the Ashtamudi Estuary, India

Chlorophyll-a is often used as an indicator of eutrophication in water bodies and hence its prediction is highly essential. Several models have proposed for predicting chlorophyll-a in lentic ecosystems. Unlike lakes and reservoirs, estuaries are affected by salinity that varies over a spatial and temporal scale. However, the influence of salinity on the chlorophyll-a concentration in estuaries is not well explored. Therefore, the present study evaluates the role of salinity on the chlorophyll-a dynamics in the Ashtamudi estuary, India. For this, two sets of models for chlorophyll-a were developed using Artificial Neural Network (ANN) and Adaptive Neuro-Fuzzy Inference System (ANFIS) tools: the first set of models considered salinity as one of the input parameters along with other significant parameters namely, pH, Dissolved Oxygen, turbidity and phosphates, while the other predicted Chlorophyll-a concentrations without considering salinity. The monthly data of parameters collected during 2011 to 2018 have been employed for training and validating the models, and the data collected during 2022 were used for testing the models. Results suggest that salinity is a major influencing factor in Chlorophyll-a prediction. The predictability (R2 value) of the ANN model enhanced from 93.63% to 98.96% and the ANFIS model with four membership functions yield an R2 value of 1.