Mountain ecosystems are biodiversity hotspots that are particularly vulnerable to global change. Interactions between plants and their consumers, such as invertebrate herbivores and pathogens, are crucial for mountain ecosystem functioning as they strongly influence biodiversity, productivity, and food webs. Yet, research on these interactions along elevational gradients remains fragmented. Studies rarely investigate both direct, climatic drivers and indirect, plant-mediated drivers (e.g., plant traits) across entire herbaceous communities and multiple elevation gradients for invertebrate herbivores and pathogens simultaneously. We sampled 19 sites across six elevational gradients dominated by herbaceous plant communities. Using standardized protocols, we assessed plant diversity and community composition, herbivore and pathogen damage, plant functional traits, biomass, plant secondary compounds, soil, and climate variables. We used linear mixed-effects models to analyse how plant community, soil variables, and consumer damage change with elevation and fitted structural equation models to disentangle the direct andindirect, plant-mediated effects of elevation on invertebrate herbivory and pathogen damage.
Herbivore and pathogen damage both varied with elevation: while herbivory declined steadily as elevation increased, pathogen damage peaked at mid-elevations. Elevation altered the plant community in line with the growth-defense trade-off: communities at low elevations were characterised by fast-growing, resource-acquisitive plants, while high-elevation communities consisted of slow-growing, well-defended plants. In single predictor models, plant height and the abundance of phenylpropanoids were the most consistent plant community characteristics predicting both herbivore and pathogen damage. Structural equation modelling revealed that invertebrate herbivory was influenced mainly by indirect, plant-mediated responses to elevation, while pathogen damage was primarily linked to direct elevation effects. The mid-elevation peak in pathogen damage is potentially a result of direct abiotic constraints at high elevations and high land-use intensity at low elevations, an interpretation that future studies should corroborate by explicitly incorporating land-use effects.
Our results demonstrate that herbivore damage along elevation is strongly mediated by changes in plant traits instead of direct abiotic constraints. Thus, integrating indirect, plant-mediated pathways is crucial for predicting climate change impacts on herbivory, as predictions based on direct effects of elevation or temperature alone may be inaccurate if they overlook how plant trait shifts amplify or dampen overall responses. To accurately predict how global change will alter plant-consumer dynamics in mountain ecosystems, future research should integrate both direct and indirect drivers of consumer damage, including land-use intensity, to capture the full complexity of these interactions.