Thermal stress gradient causes increasingly negative effects towards the range limit of an invasive mussel.

Environmental filtering (EF), the abiotic exclusion of species, can have first order, direct effects with cascading consequences for population dynamics, especially at range edges where abiotic conditions are suboptimal. Abiotic stress gradients associated with EF may also drive indirect second order effects, including exacerbating the effects of competitors, disease, and parasites on marginal populations because of suboptimal physiological performance. We predicted a cascade of first and second order EF-associated effects on marginal populations of the invasive mussel Mytilus galloprovincialis, plus a third order effect of EF of increased epibiont load due to second order shell degradation by endoliths. Mussel populations on rocky shores were surveyed across 850 km of the south-southeast coast of South Africa, from the species' warm-edge range limit to sites in the centre of their distribution, to quantify second order (endolithic shell degradation) and third order (number of barnacle epibionts) EF-associated effects as a function of along-shore distance from the range edge. Inshore temperature data were interpolated from the literature. Using in situ temperature logger data, we calculated the effective shore level for several sites by determining the duration of immersion and emersion. Summer and winter inshore water temperatures were linked to distance from the mussel's warm range edge (our proxy for an EF-associated stress gradient), suggesting that seasonality in temperature contributes to first order effects. The gradient in thermal stress clearly affected densities, but its influence on mussel size, shell degradation, and epibiosis was weaker. Relationships among mussel size, shell degradation, and epibiosis were more robust. Larger, older mussels had more degraded shells and more epibionts, with endolithic damage facilitating epibiosis. EF associated with a gradient in thermal stress directly limits the distribution, abundance, and size structure of mussel populations, with important indirect second and third order effects of parasitic disease and epibiont load, respectively.