Mechanisms of Host Tree Selection by the Pine Processionary Moth

La partie est intégrée dans le Chapter 6: Insect – Tree Interactions in Thaumetopoea pityocampaInternational audienceThe pine processionary moth is, by far, the most important insect defoliator of pine forests in Southern Europe and North Africa, both in terms of its temporal occurrence, geographic range and socioeconomic impact. Monitoring and pest management actions are therefore required on a regular basis, to ensure the detection, evaluation and mitigation of potential risks to forest and public health. However, we still lack some of the basic knowledge required for relevant analyses of the risk posed by the pine processionary moth. Pest risk is defined as a combination of three components: (1) pest occurrence, which depends on the spatiotemporal dynamics of populations; (2) plant vulnerability to the pest, resulting in a certain amount of damage; and (3) the socioeconomic impact of damage, depending on the potential value of the plants damaged (Jactel et al. 2012). The population dynamics of the processionary moth has been extensively studied, in particular within the context of climate change (see Battisti et al. 2014, Chap. 2, this volume). Several studies have recently addressed the question of tree and forest vulnerability to pine processionary attacks but a comprehensive review of evidence was missing. This is the first objective of this chapter. In particular we were interested in a better understanding of the ecological mechanisms responsible for the host tree selection, at both the species and individual tree levels. In a second part we show that pine susceptibility to the pine processionary moth could be reduced by improving forest diversity at different spatial scales. In the last part of this chapter we provide quantitative estimate of the growth losses caused by defoliations of the pine processionary moth. Altogether this information paves the way for quantitative risk analyses on pine processionary moth infestations based on forest growth models