Search

Land-use generates multiple stress factors, and we need to understand their effects on plant–plant interactions to predict the consequences of land-use intensification. The stress–gradient hypothesis predicts that the relative strength of positive and negative interactions changes inversely under increasing environmental stress. However, the outcome of interactions also depends on stress factor’s complexity, the scale of analysis, and the role of functional traits in structuring the community. We evaluated plant–plant co-occurrences in a temperate forest, aiming to identify changes in pairwise and network metrics under increasing silvopastoral use intensity. Proportionally, positive co-occurrences were more frequent under high than low use, while negative co-occurrences were more frequent under low than high. Networks of negative co-occurrences showed higher centralization under low use, while networks of positive co-occurrences showed lower modularity and higher centralization under high use. We found a partial relationship between co-occurrences and key functional traits expected to mediate facilitation and competition processes. Our results shows that the stress-gradient hypothesis predicts changes in spatial co-occurrences even when two stress factors interact in a complex way. Networks of negative co-occurrences showed a hierarchical effect of dominant species under low use intensity. But positive co-occurrence network structure partially presented the characteristics expected if the facilitation was an important mechanism characterizing the community under high disturbance intensity. The partial relationship between functional traits and co-occurrences may indicate that other factors besides biotic interactions may be structuring the observed negative spatial associations in temperate Patagonian forests.

The effects of grazing disturbance on insect communities were examined at a montane grassland in Central Argentina, by comparing two grazed sites differing in cattle load (heavy and continuous or moderate and discontinuous) and two cattle exclusions differing in age (7 and 19 years). Two aspects of insect diversity (taxonomic and guild structure) and two levels of taxonomic resolution (family and species) were considered. Four monthly samples were taken with a suction sampler in two 1 m2 areas at each site. Collected specimens were counted, identified to family (all insects) or species (Coleoptera) level, and allocated to trophic guilds. Abundance, richness, diversity and biomass of the insect assemblages had minimum values in the most intensely grazed habitat, which also differed from the other sites in terms of insect families and Coleoptera species composition. It also showed a distinct guild structure, with fewer secondary consumers, and chewers replacing suckers as the most abundant herbivore group. According to these observations, intense grazing in montane grasslands in Central Argentina could result in taxonomic and guild changes in the associated insect communities, but such effects would not be noticeable with less intensive use. Moreover, using family taxonomic level could be as or even more appropriate than species level in order to characterize insect communities in the studied habitats under varying disturbance regimes.