Forecasting green sea turtle (Chelonia mydas) landing in Sarawak using grey model

Abstract

Green sea turtles, known scientifically as Chelonia mydas, prefer to nest on specific sandy beaches in Sarawak, particularly within the Sarawak Turtle Islands (STI). The number of turtles landing, among other variables (number of eggs collected, eggs incubated, and eggs hatched) is an important element in assessing the population size in Sarawak. However, modeling and predicting the number of turtles landing presents challenges due to limited data availability, resulting in less accurate forecasts for medium and long-term periods. To overcome this problem, this study presents a Grey Model (GM) approach, leveraging its capacity to effectively model systems with limited data, irregular patterns, and a lack of prior knowledge. Using data from 1949 to 2016, GM (1,1) was found to be the most suitable model for the given dataset, exhibiting the lowest Root Mean Square Error (RMSE) and Mean Absolute Error (MAE) as compared to other statistical models such as Autoregressive Integrated Moving Average (ARIMA), Long Short-Term Memory (LSTM) and Exponential Smoothing. The model also suggested that the current conditions will likely increase turtle landings. This approach will find useful applications in evaluating the conservation status of the species.