CLIMATE-INDUCED POTENTIAL DISTRIBUTION AREAS OF SOME INVASIVE INSECT SPECIES WITH RESTRICTED OCCURRENCES IN TÜRKIYE

Abstract

Invasive insect species pose significant ecological and economic threats globally under changing climate. Türkiye sits at a diverse biogeographical setting spanning three phytogeographical regions., enabling the country to host a rich agricultural heritage and serve as a biodiversity hotspot. Nevertheless, increasing trade and travel activities, and climate change are making the country increasingly susceptible to biological invasions. Predicting the potential distribution areas of insect species helps to halt their spread through quarantine enforcements and rapid response activities. Species distribution models are reliable tools to predict the potential spread of target species by correlating their known occurrence with climate and other variables. Therefore, this study used Maximum Entropy (MaxEnt) species distribution model to predict potential distribution of five invasive insect species (Anoplophora chinensis, Dryocosmus kuriphilus, Halyomorpha halys, Garella musculana, and Corythucha arcuata) under current and future climatic conditions corresponding to two Coupled Model Intercomparison Project Phase 6 (CMIP6) climate change scenarios (SSP1-2.6 and SSP5-8.5). Occurrence records of all species were retrieved from global biodiversity databases and peer-reviewed literature. These records were subsequently georeferenced and cleaned. Highresolution bioclimatic variables (bio1-bio19) from the WorldClim database were utilized to predict potential distribution of all insect species across Türkiye. MaxEnt model was optimized by using ‘ENMeval’ package in R statistical environment, whereas environmental variables to be included in the modeling exercise were decided using correlation matrix and their contribution towards model training. The variable selection from the highly correlated pairs was based on their contribution towards model, i.e., variable with low contribution to model was deleted. The model was trained with 75% of the occurrence data and tested on the remaining 25%. Predictive performance of the model was evaluated by are under the receiver operating characteristic curve (AUC). The model indicated strong predictive performance in estimating potential distribution of all species (AUC > 0.90). The results demonstrate that bio4 (temperature seasonality) and bio14 (precipitation of driest month) emerged as the most consistently significant variables influencing potential distribution of all invasive insect species, serving as common climate drivers. All five species demonstrated potential range expansions under future climate scenarios, with higher expansion under high-emission scenario (SSP5-8.5). The species were predicted to extend their distribution range towards north and high elevation regions. The potential distribution maps identify areas under invasion risk and stress the immediate necessity for biosecurity measures, early-warning systems, and climate-adaptive pest management solutions. The study integrated climate-informed spatial modeling into invasive insect species surveillance, providing essential evidence for proactive policy and sustainable agriculture and forestry planning in Türkiye.