3.5) Perspectives and limitations
We have presented an overview of the literature on plant trait-herbivore interactions in grasslands across the globe. Specifically, we highlight how vertebrate and invertebrate focussed studies differ in the types of traits examined, and their response to, and effect on herbivory. It is clear from the results of this review, that there is a particularly limited understanding on the impacts of invertebrates or the combined impacts of invertebrates and vertebrates on plant species in grasslands. Given the importance of grasslands for diversity, carbon cycling and human food production this is concerning. Considering that insect populations are declining across the globe (Eggleton, 2020), this highlights a particularly important area of future research to appropriately predict changes to ecosystem functioning.
Another important finding from this review is the significant bias towards domestic livestock and researching their response to, and effect on plant traits. Due to the dominant use of grassy ecosystems across the globe for livestock production, livestock potentially have a proportionally greater effect on plant traits and consequential ecosystem functioning than wild native vertebrate herbivores. Native herbivores usually exist within these pastoral landscapes however, and understanding how these species interact with domestic livestock to affect plant traits will provide a more wholistic understanding of herbivore impacts on grassland species. Further, due to differences in evolutionary history, plants likely respond differently to native compared to non-native herbivores, and this may be reflected in their trait response (Diaz et al., 2007).
As with most functional trait research, results from this review are biased toward the dominant plant species within grassland communities. This bias may be challenging to avoid, as rare species likely do not offer adequate replication, but fundamentally limits our understanding of plant-herbivore interactions. For instance, dominant plants within grazed grassland communities are likely to be selected for their traits which confer grazing tolerance or avoidance. In contrast, species which occur at low frequencies (one definition of ‘rare’) may do so because they are highly palatability and have been selectively grazed. Alternatively, they may be inherently ‘rare’, due to low fecundity and dispersal. Subordinate and rare species can have significant influence on ecosystem functioning (Baer et al., 2002; De Vries et al., 2011; Grime, 1998) and understanding their interactions with herbivory is an important area of future research.
Although there were some similarities between vertebrate and invertebrate focussed studies in terms of the plant traits examined, there is still considerable disparity, which reduced the capacity for comparisons. Differences in the potential scale, intensity and specificity of herbivory inflicted by these two groups is important to understand the differences in the types of plant traits being chosen to be researched. In saying this there is considerable capacity to expand plant trait-herbivore research to increase comparability of results between vertebrate and invertebrate herbivores. For instance, investigating the relationship between VOCs and vertebrate herbivory or examining how invertebrates influence plant morphology using longer-term experiments would be an interesting area for future research. Overall, we suggest there is also scope to shortlist the types of traits being examined to improve comparability of results, but still remaining adaptable to change as more is learned.
Despite considerable effort to ensure all relevant studies were included in this systematic review, we recognise the search terms used may not have been sufficient to find species-focussed studies that did not specify that their focal species occurred in grasslands. For instance, for some laboratory or glasshouse studies the fact that their focal plant species naturally occurs in grasslands may have been irrelevant to their study question. Overall, we found perhaps surprisingly few studies which adhered to the criteria of this systematic review. Many potentially relevant studies were excluded because they examined how plant traits responded to various intensities of herbivory but failed to include a no herbivory control. Other key reasons for exclusion were the use of categorical traits and the use of cropping species.
Overall, there is exciting opportunity for future research to explore interactions between vertebrate and invertebrate herbivores and grassland plant traits. This will add to a growing body of literature aimed at better understanding the complex relationships between plants and herbivores. As the primary pathway for nutrient cycling and movement across trophic levels, these relationships are crucial for the functioning of agricultural and ecological systems alike. By improving our understanding of these relationships, we contribute to the development of theories around plant herbivore interactions and in doing so help to guide the development of better research questions and the building of knowledge.