A machine learning algorithm-based approach (MaxEnt) for predicting invasive potential of Trioza erytreae on a global scale

The African citrus triozid (ACT) Trioza erytreae (Del Guercio) (Hemiptera: Triozidae), is a principal vector of “Candidatus Liberibacter species”, the pathogens implicated in citrus greening disease, infecting millions of citrus trees annually. Unfortunately, present control measures are insufficient, necessitating the development of novel climate-smart pest management strategies. Maximum entropy model (MaxEnt) was employed to assess habitat suitability for ACT. The area under the curve values for the initial and final models were 0.968 and 0.962, respectively. The model predicts an expansion of suitable areas outside the pest's known historical records. The environmental variables that most influenced ACT's distribution were isothermality, average temperature of coldest quarter, average temperature of driest quarter, and highest temperature of warmest month, with contributions of 37.5%, 16.4%, 13.2%, and 12.7%, respectively. The major citrus-producing countries, such as China, Brazil, and the USA, would have suitable areas for ACT until 2050. The risk maps created in this study could be used in the field to prevent further ACT invasions, thereby contributing to sustainable management of citrus greening disease.