Spatial Evaluation of Machine Learning-Based Species Distribution Models for Prediction of Invasive Ant Species Distribution

Abstract

Recent advances in species distribution models (SDMs) associated with artificial intelligence (AI) and increased volumes of available data for model variables have allowed reliable evaluation of the potential distribution of any species. A reliable SDM requires suitable occurrence records and variables with optimal model structures. In this study, we developed three different machine learning-based SDMs [MaxEnt, random forest (RF), and multi-layer perceptron (MLP)] to predict the global potential distribution of two invasive ants under current and future climates. These SDMs showed that the potential distribution of Solenopsis invicta would be expanded by climatic change, whereas it would not significantly change for Anoplolepis gracilipes. The models were compared using model performance metrics, and the optimal model structure and spatial projection were selected. The MaxEnt exhibited high performance, while the MLP model exhibited low performance, with the largest variation by climate change. Random forest showed the smallest potential distribution area, but it was robust considering the number of occurrence records and changes in model variables. All the models showed reliable performance, but the difference in performance and projection size suggested that optimal model selection based on data availability, model variables, study objectives, or an ensemble approach was necessary to develop a comprehensive SDM to minimize modeling uncertainty. We expect that this study will help with the use of AI-based SDMs for the evaluation and risk assessment of invasive ant species.

Modelling Distribution of an Endangered Longhorn Beetle, Callipogon relictus (Coleoptera: Cerambycidae), in Northeast Asia

“…Habitats determined with the highest probability of species detection are considered the most suitable. The effectiveness of this method in predicting the potential habitat of alien and rare insect species has been confirmed by the results of numerous studies, including Anoplophora glabripennis (Motschulsky 1853), Rosalia alpina (Linnaeus 1758) (Coleoptera: Cerambycidae), Paracyphoderris erebeus Storozhenko, 1980 (Orthoptera: Prophalangopsidae), and others [17][18][19][20][21][22][23][24][25][26][27][28].…”

Comparison of different variable selection methods for predicting the occurrence of Metisa Plana in oil palm plantation using machine learning

“…ANN is a fundamental deep learning algorithm that mimics the functionality of the brain through interconnected layers and neurons, it has been used to predict the potential presence of insect species [30]. This study employed Multi-Layer Perceptron (MLP) models and three main layers characterize the architecture of neural network: the input, output, and hidden layers (see figure 6).…”

Estimation of Potential Suitable Habitats for the Relict Plant Euptelea pleiosperma in China via Comparison of Three Niche Models

“…Scholars have developed multiple models based on different algorithms [44,45], which are extensively applied in numerous fields including species protection, natural environment monitoring, evaluation of ecological situation, protection against invasive species, and estimation of the risk of infection [44,[46][47][48][49]. This study used three models (BIOCLIM, DOMAIN, and MaxEnt) to estimate the suitable habitats of species [50,51]. The BIOCLIM model is a framework niche model based on bioclimatic data, which can extract various environmental data from the known distribution areas to form fixed and complex rectangle packing.…”