First assessment of potential distribution and dispersal capacity of the emerging invasive mosquito Aedes koreicus in Northeast Italy

Background Invasive alien species represent a growing threat for natural systems, economy and human health. Active surveillance and responses that readily suppress newly established colonies are effective actions to mitigate the noxious consequences of biological invasions. However, when an exotic species establishes a viable population in a new area, predicting its potential spread is the most effective way to implement adequate control actions. Emerging invasive species, despite monitoring efforts, are poorly known in terms of behaviour and capacity to adapt to the new invaded range. Therefore, tools that provide information on their spread by maximising the available data, are critical. Methods We apply three different approaches to model the potential distribution of an emerging invasive mosquito, Aedes koreicus, in Northeast Italy: 1) an automatic statistical approach based on information theory, 2) a statistical approach integrated with prior knowledge, and 3) a GIS physiology-based approach. Each approach possessed benefits and limitations, and the required ecological information increases on a scale from 1 to 3. We validated the model outputs using the only other known invaded area in Europe. Finally, we applied a road network analysis to the suitability surface with the highest prediction power to highlight those areas with the highest likelihood of invasion. Results The GIS physiological-based model had the highest prediction power. It showed that localities currently occupied by Aedes koreicus represent only a small fraction of the potentially suitable area. Furthermore, the modelled niche included areas as high as 1500 m a.s.l., only partially overlapping with Aedes albopictus distribution. Conclusions The simulated spread indicated that all of the suitable portion of the study area is at risk of invasion in a relatively short period of time if no control policies are implemented. Stochastic events may further boost the invasion process, whereas competition with Aedes albopictus may limit it. According to our analysis, some of the major cities in the study area may have already been invaded. Further monitoring is needed to confirm this finding. The developed models and maps represent valuable tools to inform policies aimed at eradicating or mitigating Aedes koreicus invasion in Northeast Italy and Central Europe.