Talk, IECID2015 Impact of Environmental Changes on Infectious Diseases, Sitges, Barcelona: 2015-03-22 - 2015-03-25
The geographical range of areas where dengue epidemics occur has expanded dramatically during the last decades. Up to now, Europeans tend to exclusively consider dengue as a travel-related disease. However, first European cases of autochthonous dengue were reported in the Mediterranean and Madeira experienced an epidemic in 2012/2013. As arthropod vectors are ectothermic species, the environmental temperature directly determines the amplification of the virus inside the vector (extrinsic incubation period EIP). Thus warmer temperatures will shorten the EIP. If minimum temperature thresholds for the EIP are not exceeded the virus cannot accomplish its amplification and transmission to humans can be excluded. To identify European current and future areas at risk for dengue transmission we combine ecological knowledge on vector and virus and statistical species distribution models. First, we developed maps with areas at risk for a long-term establishment of Aedes albopictus – vector for dengue – with special focus on the cold tolerance of mosquito’s eggs. Secondly, we detect areas where the temperature requirements of dengue EIP are fulfilled. In a final step areas with possible transmission of dengue are detected by combining the results for vector and virus. Species distribution modelling was performed with R and the package biomod2, which allows the use of an ensemble of different modelling algorithms. All future risk maps are based on five different global climate models and the two representative concentration pathways RCP4.5 and RCP8.5 (IPCC AR5). The European distribution of Aedes albopictus is projected to increase up to the end of the century. However, temperature-induced reproductive constraints and particularly the temperature-dependence of virus amplification limit areas of possible transmission. We illustrate that assessing the risk of a spread of diseases requires the combination of projections for vector and pathogen as well as the inclusion of ecological constraints to yield a more realistic estimation.