Perception-based foraging for competing resources: Assessing pest population dynamics at the landscape scale from heterogeneous resource distribution

•We developed a reaction-diffusion model of a pest population dynamics.•We accounted for foraging processes providing a mechanistic landscape dependence.•Processes under consideration are resource perception and energy supplies management.•Both processes exhibited significant effects on the population dynamics.•This supports the model relevance for addressing landscape-oriented multi-scale issues.

Resource distribution, through its effects on individual foraging and survival, drives population dynamics across the landscape. In an agricultural context, resource distribution is therefore a key information in assessing whether or not a pest population may invade and persist in a given environment. Addressing this issue by means of numerical exploration requires a population model with a sound dependence on the landscape. In this paper, we demonstrate that this dependence is effectively secured by a multi-scale description of the population. We derived a reaction-diffusion population model accounting for two individual-scale processes determining resource utilisation: (1) resource perception as a determinant of mobility and (2) energy supply management as a determinant of survival. In this model, the distribution of two competing resources (feeding and laying sites) affects the spatial population dynamics of a dipteran pest through a heterogeneous dispersion of the individuals and a metabolic currency. We conducted a global sensitivity analysis to evaluate the impact of both individual-scale processes on the population dynamics. This exploration demonstrated the biological relevance of the model according to field observations and theoretical expectations. Our key finding is that resource perception and energy supply management appear as significant as the demographic component regarding the resulting dynamics of the pest. Building on its acute multi-scale landscape dependence, this model may be particularly useful for investigating the putative relationships between agricultural landscape features and pest outbreaks.