Prédiction des zones favorables à l'installation de l'espèce exotique envahissante Ambrosia artemisiifolia L. en région wallonne

Abstract

The common ragweed (Ambrosia artemisiifolia L.) is an invasive alien species native to North America that has conquered Europe since the turn of the 20th century. Despite various observations in Belgium, the species has few presence data and is not very well-known in the Walloon region. The species causes extensive health and economic damage, as it is highly allergenic cultivated weed. Spatial modelling of areas favourable to the installation would allow better sensitization, as well as targeting of samples and management of the species. This study brings together different modeling methods used in the literature in order to better understand the issues encountered by modeling invasive species. Using the presence points and the Worldclim bioclimatic rasters of the native area, a first modeling was carried out by extrapolation over the Walloon territory. A temporal extrapolation was also performed on the model, to point the impact of climate change on the potential distribution of common ragweed. A second modeling was carried out using environmental data from ‘Lifewatch Ecotope’ and the Walloon presence data, despite their poor robustness. The model was validated with the AUC index, and the results of the two models showed favorable areas over a large part of the Walloon territory in the provinces of Hainaut, Namur, Brabant Wallon and Liège. The variables constraining the extrapolated model the most are summer temperatures and precipitation as well as the variation in precipitation. The variables constraining the independent model the most are the minimum winter temperatures, the distance to the railways and the presence of soils with average drainage. The temporal extrapolation has shown that almost all of the Walloon territory will be favorable to the colonization of the common ragweed, with the exception of the east of the province of Luxembourg and the east of the province of Liège. The independent model performed less well than the extrapolated model, mainly because of the weakness of the data set used. It is important to remain attentive to the proliferation of common ragweed in the Walloon region. While it is not yet very harmful in the region, the potential damage it can cause is enough to raise as much awareness as possible among those involved in the introduction of the species and its management. It is essential to increase the number of observation points in the Walloon region in order to better understand the distribution of the species and to better fight against it in the future.