Pine Wilt Disease (PWD) is a disease causing mass deaths of pine trees in South Korea, and the dead trees serve as breeding grounds for insect vectors responsible for spreading the disease to other host trees. Because the PWD requires early monitoring to minimize its damage on domestic forestry, this study aims to develop a species distribution model for predicting the potential distribution of PWD by using artificial neural network (ANN) with time-series data. Among the architectures, the Convolutional Neural Network exhibited the highest performance, achieving a validation accuracy of 0.854 and a cross-entropy loss of 0.401, and the InceptionTime model emerged as the second-best performer. This study identified the best-performing ANN architecture for a spatiotemporal evaluation of PWD occurrence, emphasizing the importance for determining hyperparameters with ecological characteristics and data types to apply deep learning into SDMs.

A machine learning-based algorithms have used for constructing species distribution models (SDMs), but their performances depend on the selection of backgrounds. This study attempted to develop a noble method for selecting backgrounds in machine-learning SDMs. Two machine-learning based SDMs (MaxEnt, and Random Forest) were employed with an example species (Spodoptera litura), and different background selection methods (random sampling, biased sampling, and ensemble sampling by using CLIMEX) were tested with multiple performance metrics (TSS, Kappa, F1-score). As a result, the model with ensemble sampling predicted the widest occurrence areas with the highest performance, suggesting the potential application of the developed method for enhancing a machine-learning SDM.