Specific leaf area modulates but does not explain the association between abiotic stress tolerance and insect feeding...

Plant adaptations to abiotic stress and insect herbivory may share functional bases, yet large-scale evidence linking stress tolerance strategies to herbivory patterns remains limited. We tested whether abiotic stress tolerance is associated with insect herbivory patterns in Northern Hemisphere woody plants, and whether specific leaf area (SLA) and plant functional type (PFT) modulate this relationship.

We combined data on 5927 lepidopteran and hymenopteran species with 645 woody plant species from three PFTs (deciduous angiosperms, evergreen angiosperms, evergreen gymnosperms) for which information on their shade, drought, cold and waterlogging tolerance was also available. We then modelled the prevalence of three feeding guilds (chewers, borers, miners) as a function of stress tolerance, SLA, and their interaction at both species and assemblage levels.

Assemblage-level associations were substantially stronger than species-level patterns (54 % vs. 18 % of significant effects, respectively). Despite significant SLA × stress tolerance interactions, these did not provide a clear mechanistic explanation of the observed patterns. In addition, spatial autocorrelation in model residuals was substantial (mean Moran's I = 0.66), and most of the observed effects were not robust to correction for spatial dependency.

Abiotic stress tolerance and SLA interact to shape herbivore guild composition, but SLA modulates rather than mediates this relationship, defining a context within which other, unidentified factors (e.g. most specific defence traits) operate. The strong spatial structure in herbivore communities probably reflects biogeographical processes (dispersal limitation, host specificity, regional species pool assembly) that operate independently of contemporary trait distributions. Understanding plant-herbivore interactions at macroecological scales requires integrating trait-based filtering with historical and biogeographical constraints.