Intraspecific herbivory variability, rather than species’ turnover effects, determined latitudinal variation in community-wide herbivory
The evidence found here for greater herbivory at lower latitudes is in agreement with the LHH, and concurs with studies of other systems including salt marsh plants (Pennings & Silliman, 2005; Penningset al. , 2009) and terrestrial plants (Anstett et al. , 2014; Baskett & Schemske, 2018). However, to the best of our knowledge, our study is the first to show a decrease in community-wide herbivory with latitude. Several studies have conducted multispecies analyses by sampling diverse sets of representative species in two or more locations (e.g. , Pennings et al. , 2009; Zvereva et al. , 2020); however, this is no substitute for the direct evaluation of community-wide herbivory in most instances (Zvereva et al. , 2020). When considering single species, a decline in herbivory at higher latitudes might also result from a reduction in abundance; abundance is a well-studied species-level feature positively affecting herbivory (the so-called “resource concentration effect”; Underwood et al. , 2014; Kim & Underwood, 2015). Therefore, earlier studies, which did not account for the abundance of individual plant species, likely yielded biased estimates of community-wide herbivory and, consequently, of the overall effects of background herbivory on ecosystem-level processes (Anstett et al. , 2016; Zvereva et al. , 2020; Moreiraet al. , 2021).
Latitudinal variation in community-wide herbivory is expected to result from two additive components, intraspecific herbivory variability and species’ turnover effects. In this study, we found that the decline in community-wide herbivory with latitude mirrored similar declines in both intraspecific herbivory variability and observed herbivory levels for three common plant species. This indicates that latitudinal variation in community-wide herbivory was mainly driven by decreases in herbivory on individual plant species, rather than by shifts in community composition. This relatively weak role for species’ turnover effects is inconsistent with a recent study of boreal forests (taiga) (Zverevaet al. , 2020). Zvereva et al. (2020) found that latitudinal shifts in community composition (the proportion of deciduous plants increased with latitude) counterbalanced a simultaneous decrease in herbivory on deciduous species; the net result was a lack of latitudinal variation in community-wide herbivory (Zvereva et al. , 2020).
This inconsistency among studies may be attributed to the higher plant species richness in grasslands (c . 20–30 plant species in a 0.5 × 0.5 m2 quadrat) and correspondingly low abundance of each plant species (e.g. , 96.3% of 54 plant species had percent cover < 5%; Ma et al. , 2017); for species showing significant latitudinal gradients in herbivory, there were no corresponding latitudinal shifts in abundance. For example, the three most broadly-distributed plant species (P . macrophyllum ,S . pulchra and T . mongolicum ) showed latitudinal gradients in herbivory, but the biomass of each species did not vary with latitude. The relatively high herbivory levels observed here could affect plant fitness, and thus competitive ability (Agrawalet al. , 2012); as a result, a compensatory increase in abundance or biomass may have been lacking. Hence, we conclude that the abundance of each plant species in a community should be taken into account when studying latitudinal variation in community-wide herbivory and detangling the relative importance of intraspecific herbivory variability versus species’ turnover.