Comparative study of statistical and machine learning techniques for fish production forecasting in Andhra Pradesh under climate change scenario

The present study emphasizes the forecast of Andhra Pradesh's total marine fish production and catch commercially important fishes, viz., Indian Mackerel, Oil Sardine, Horse Lesser Sardines for next 5 years by different statistical machine learning approaches under climate change scenario. Forecasting is done with without inclusion climatic environmental parameters in models. Exogenous variables, i.e., such as Sea Surface Temperature (SST), wind speed, Chlorophyll-a, diffusion attenuation coefficient, Photo-synthetically Active Radiation (PAR), were used model. following models like Non-linear Autoregressive (NAR) Artificial Neural Network (ANN) (NNAR-ANN), Auto-Regressive Integrated Moving Average (ARIMA), Empirical Mode Decomposition based (EMD-ANN), are to predict data using time series quarterly fishes variables. Auto Regressive method exogenous variables (ARIMAX) Non-Linear Regression (NARX) along average model developed predicts also 20 quarters. forecasts compared on error measure, MAPE (Mean Absolute Percentage Error), results showed that NARX outperformed other ARIMAX, ARIMA, NNAR-ANN, EMD-ANN. Implementation management strategies considering impact fisheries will enhance sustainable pave a pathway mitigation change.