Abstract
Pest risk assessment is typically performed by expert taxonomists using a pest's biological data. However, the biological data or expert taxonomists may be difficult to find. Here, we used species distribution modelling to predict potential invasion in which phytophagous quarantine pests survive in Taiwan; the pests (unrecorded yet in Taiwan) included were three notorious quarantine whiteflies (Crenidorsum aroidephagus, Aleurothrixus trachoides, and Paraleyrodes minei) and three aphids (Nasonovia ribisnigri, Macrosiphum euphorbiae, and Viteus vitifoliae). In brief, maximum entropy modelling (MaxEnt) was used to predict the suitability of the pests' habitats under certain climatic conditions, and then receiver operating characteristic curve analysis was performed (to verify the prediction result). We then analysed environmental variables affecting the habitat suitability and matched them with Taiwan's crop cultivation areas for the assessment of potential invasion. We observed that the habitat suitability of the cultivation areas of host plants was low for C. aroidephagus, A. trachoides, and N. ribisnigri but was high for the remaining three species. Moreover, precipitation of coldest quarter negatively affected habitat suitability for C. aroidephagus, P. minei, N. ribisnigri, and M. euphorbiae. Seasonal temperature changes also negatively affected the habitat suitability for A. trachoides. This is the first study to demonstrate the use of species distribution modelling as the preliminary step for the pest risk assessment of these emerging pests with limited biological data before their invasion.