Habitat fragmentation and amount drive within‐species variation in dispersal kernels and limit transferability across landscapes

Dispersal is a key ecological trait that ensures connectivity, gene flow, and range dynamics, yet empirical information about how dispersal distances vary within species remains scarce. Many studies, investigating for instance connectivity conservation, use typical species‐specific dispersal kernels without accounting for the fact that these kernels actually emerge from an interaction between species traits and the landscape context. Here, we aim to analyse the influence of habitat amount and fragmentation, two key elements of landscape context, on dispersal kernels. To test this, we use an individual‐based modelling platform to simulate movement trajectories of four types of animal species, from insects to mammals, across artificial landscapes varying in habitat amount and fragmentation. From these simulated movements we derived dispersal kernels using five probability density functions and related dispersal distances to landscape context. Log‐normal (fat‐tailed) kernels consistently best described dispersal across species, but both median and long‐distance dispersal varied strongly with habitat configuration. Habitat fragmentation strongly shaped emerging dispersal kernels across species as it directly influenced movement trajectories, while habitat amount had a lesser and more ambiguous effect on dispersal distances. Our results highlight that dispersal kernels for a given species depend strongly on habitat fragmentation and amount, and therefore cannot be assumed constant across landscapes, limiting the transferability of empirically derived species‐level dispersal kernels to new landscape contexts. Improving predictions of dispersal under global change will require explicit consideration of intraspecific variation in dispersal driven by landscape context in both empirical and model‐based studies, and further investigation of additional context‐dependent drivers of dispersal variation.