Effects of tree diversity on insect herbivory

Xinliang Shao·Ke Cheng·Yuhua Kong·Qin Zhang·Xitian Yang

Abstract Tree diversity has long been considered a key driver of insect herbivory in forest ecosystems. However, studies have given contradictory results: increased tree diversity can have positive, negative or neutral effects on insect herbivory. Since many issues can complicate the treeherbivore interactions, the descriptor ‘tree diversity’ per se actually has only limited explanatory power for insect herbivory. Particularly, in addition to the direct bottom-up effects on insect herbivores, tree diversity may have stronger indirect top-down effects via natural enemies of insect herbivores. However, most research has addressed only direct bottom-up impacts. In fact, insect herbivory is the result of complex interactions (food webs) among all the species in a community. Although it is hard to integrate all functionally important species and dynamic information into food webs, the effects of tree diversity on insect herbivory are highly dependent on the dominant species at different trophic levels. It is important and practical to simultaneously consider the characteristics of main trees, herbivores, and natural enemies when attempting to predict the overall effects of forest tree diversity on insect herbivory. In order to ensure comparability between studies, it is necessary to classify them according to the descriptors of insect herbivory and tree diversity, and to compare results within each category. These measures might enhance our understanding of the mechanisms by which tree diversity drives insect herbivory and, in turn, help to develop sustainable pest management strategies for forests.

Keywords Enemy diversity·Food webs·Sustainable pest management·Tree diversity·Tree-herbivore interactions

Introduction

Insect herbivory has profound effects on ecosystem processes and services by influencing nutrient dynamics (Belovsky and Slade 2000; Frost and Hunter 2004; le Mellec et al. 2011; Metcalfe et al. 2016), tree growth, survival and reproduction (Hochwender et al. 2003; Zvereva et al. 2012), and plant community composition (Frost and Hunter 2008) as well as demographics and succession of forests (Barbosa et al. 2005; Karlsen et al. 2013). Tree diversity has long been regarded as a key driver of insect herbivory in forests and has been used as an important tool for sustainable pest management in forest plantations (Castagneyrol et al. 2019; Plath et al. 2012). However, studies of the relationship between insect herbivory and tree diversity have produced conflicting results. For example, increased tree diversity may reduce (Jactel et al. 2006; Jactel and Brockerhoff 2007; Kaitaniemi et al. 2007; Castagneyrol et al. 2014) or increase (Vehvil?inen et al. 2006; Schuldt et al. 2010, 2015; Haase et al. 2015; Brezzi et al. 2017) insect herbivory, or have no effect at all (Rosado-Sanchez et al. 2018). Predicting the effects of tree diversity on insect herbivory remains a challenge.

In fact, insect herbivory can be directly or indirectly influenced by many factors, including tree species (Sobek et al. 2009), traits of insect herbivores and their natural enemies (Vehvil?inen et al. 2007; Sobek et al. 2009; Plath et al. 2012; Castagneyrol et al. 2014; Abdala-Roberts et al. 2015), and spatial location within patches (Umapathy et al. 2016; Castagneyrol et al. 2019). To better understand the underlying mechanisms of tree-herbivore interactions and predict the effects of tree diversity on insect herbivory, we outline several aspects that confound these effects, briefly discuss the role of food webs, and offer suggestions for future studies.

Several issues complicate tree-herbivore interactions

Plant phylogenetic distance and genetic diversity

Since closely related host plants are more likely to have similar functional traits which herbivores must recognize to exploit their hosts (Gómez et al. 2010; Wiens et al. 2010), these plants often share common herbivores (?degaard et al. 2005; Weiblen et al. 2006). Thus, a mixed stand of closely related tree species might lead to higher risk of herbivore damage on focal plants (Pearse and Hipp 2009; Ness et al. 2011; Gilbert et al. 2012; Harvey et al. 2012; Parker et al. 2012). In addition, plant genetic diversity can also influence insect herbivory, although the direction and magnitude of such relationships are context-dependent (Tack and Roslin 2011; Barton et al. 2015). Thus, in order to better understand tree diversity effects on insect herbivory, these two factors should be taken into consideration (Castagneyrol et al. 2014).

Characteristics of herbivores

Diet breadth of insect herbivores is an important factor influencing the effect of tree diversity on insect herbivory. Meta-analysis has shown that tree diversity more effectively reduces herbivory by monophagous insect herbivores than by oligophagous herbivores, but has no overall effect on polyphagous herbivores (Castagneyrol et al. 2014). In addition, feeding mode (Wright et al. 2012) and mobility (Abdala-Roberts et al. 2015) of dominant herbivores in each community can also influence the effects of tree diversity. For example, ectophagous and endophagous herbivores may respond differently to tree diversity (Wright et al. 2012). Once the adult female finds a host plant and oviposits, the feeding stage of sedentary herbivores (e.g., caterpillars) can occur on a single host plant (Barbosa et al. 2009). In contrast, mobile herbivores (e.g., generalist leafhoppers have highly mobile adult and nymph stages) can readily disperse and choose among plant patches (Bommarco and Banks 2003).

Insect foraging behavior and preference for a given host plant can be influenced by interactions among herbivores (inter- and intraspecific) (Underwood 2010; Utsumi et al. 2011; Karban et al. 2013; Parent et al. 2014; Carrasco et al. 2015), which can further complicate the relationship between tree diversity and insect herbivory.

Enemy diversity

Plant diversity can directly influence host-finding by insect herbivores by altering the physical and chemical appearance of host plants (Finch and Collier 2000; Mauchline et al. 2005; Karban 2007; Castagneyrol et al. 2013; Moreira et al. 2016; Vidal et al. 2018). Beyond the direct bottomup effects, plant diversity can also have indirect top-down effects on insect herbivores by influencing the abundance and diversity of natural enemies (Root 1973; Russell 1989; Vidal et al. 2018).

Although the net effect of herbivore suppression in diverse enemy communities is dependent on ‘balance’ among multiple mechanisms (Ives et al. 2004; Jonsson et al. 2017), the suppression of arthropod herbivore in terrestrial ecosystems generally increases with enemy richness (Letourneau et al. 2009; Maas et al. 2016). A recent meta-analysis demonstrated that in most cases the indirect top-down forces conveyed by natural enemies are stronger than the direct bottom-up forces (Vidal et al. 2018). Thus, without taking into account the impacts of natural enemies, we cannot truly understand the effects of tree diversity on insect herbivory (Vidal et al. 2018). However, most studies involve only direct bottom-up effects, possibly explaining why these studies often yield conflicting results.

Dilution effect and some other influencing factors

The dilution (or reduction) of host-plant densities can reduce insect herbivory (Otway et al. 2005; Castagneyrol et al. 2014). However, this effect can occur in stands with only two tree species or might increase in positive association with tree diversity (Otway et al. 2005; Conner et al. 2014) (Fig. 1), which is often neglected in studies focused on ‘diversity’ alone. Insect herbivory can also be influenced by forest stratum (Castagneyrol et al. 2019), location within patches (Umapathy et al. 2016), landscape context (Castagneyrol et al. 2019; Valdés-Correcher et al. 2019), mast seeding of some tree species (e.g. oak) (Murakami and Wada 1997), climatic factors (Dyer et al. 2013; Rosenblatt and Schmitz 2014), and the mutualism between some plants/herbivores and their partners (Ferrari et al. 2004; Mooney 2007; Koricheva et al. 2009; Moen et al. 2012; Vidal et al. 2016). The individual and/or combined effects of these factors can obscure the real impact of tree diversity.

Fig. 1 Conceptual diagram representing the two dilution effects that are often overlooked in studies about the effects of tree diversity on insect herbivory. a The dilution effect can occur when tree diversity remains constant; b The dilution effect can increase with tree diversity

Fig. 2 Simplified schematic diagram of the interactions among trees, insect herbivores and natural enemies (direct and indirect effects of tree diversity on insect herbivores). Each letter indicates a main species at a trophic level. Arrows manifest consumption. This example shows the divergent effects of tree diversity on different insect herbivores. For insect herbivore E, tree diversity (AB) could have direct suppression effects on it by influencing the accessibility of host trees and indirect suppression effects on it via natural enemies F and G. For insect herbivore C, tree diversity only has indirect suppression effects on it via natural enemy F. For insect herbivore D, tree diversity may have no suppression effects on it

Parameters of herbivory and metrics of plant diversity

Previous studies usually focused on different parameters of herbivory (e.g., herbivore damage, abundance, incidence rate, or species richness) and on different metrics of tree diversity (e.g., Shannon/Simpson diversity, species richness, or comparison between pure and mixed stands) (Doi et al. 2016). Because different descriptors can reflect different aspects of insect herbivory and of plant diversity, the heterogeneity among studies can lead to differing conclusions regarding the effects of plant diversity on insect herbivory (Doi et al. 2016).

For example, incidence rate might reflect the foraging behavior of herbivores, while herbivore damage could additionally reflect the effects of tree diversity on plant quality/palatability and herbivore performance/survival (Doi et al. 2016). While diversity indices reflect the relative frequency of host plants, species richness only indicates the number of plant species.

Herbivore damage and herbivore abundance are the two most commonly monitored parameters, however, herbivore damage is not necessarily correlated with herbivore abundance (Stoepler et al. 2012; López-Villamor et al. 2019), and these two parameters can respond differently to plant diversity (Barbosa et al. 2009).

Food webs

Insect herbivory is, in fact, the result of complex interactions (food webs) among all the species in a community (Macfadyen et al. 2009). Food webs might be a powerful tool for examining the direct and indirect effects on herbivory in complex communities of species (McCann 2007; Mishra et al. 2015). However, the occurrence of pairwise interactions in food webs depends on many characteristics of species (e.g., life history, activity time, and location of predator and prey) (Berlow et al. 2004; Gravel et al. 2016). Thus, it is hard to integrate all the functionally important species and dynamic information into the food webs to draw general conclusions (Macfadyen et al. 2009). Nevertheless, because the effects of plant diversity on insect herbivory are highly dependent on the identity (traits) of main plant species (Sobek et al. 2009; Plath et al. 2012; Abdala-Roberts et al. 2015), the types (characteristics) of dominant herbivores (Vehvil?inen et al. 2007; Castagneyrol et al. 2014; Abdala-Roberts et al. 2015) and their enemies (Jonsson et al. 2017), general patterns of insect herbivory are actually the result of interactions among the traits of dominant species of different trophic levels (Abdala-Roberts et al. 2015; Vidal et al. 2018). Whether the nature of forest trees affects the intensity of insect herbivory depends on the nature of coexisting herbivores and natural enemies (Fig. 2). Therefore, it is important and practical to simultaneously consider the traits of dominant trees as well as those of herbivores and their natural enemies when attempting to predict the effects of plant diversity on insect herbivory in forests (Ives et al. 2004; Plath et al. 2012; Abdala-Roberts et al. 2015; Jonsson et al. 2017).

Conclusions

As a traditional descriptor, plant diversity per se has only limited explanatory power for insect herbivory. Because scientists usually need to compare the results of different studies, it is necessary to classify the studies according to the descriptors of insect herbivory and tree diversity and compare results within each category to ensure comparability between studies. For predicting the effects of tree diversity on insect herbivory, the characteristics of dominant local trees, herbivores, and natural enemies should first be considered simultaneously, and then other factors (e.g., dilution effects and landscape context) should be considered according to the specific situation. Generally, tree diversity can have significant suppression effects on insect herbivory if the local main insect herbivores (e.g. herbivore E in Fig. 2) have narrow diet breadth (relative to the main tree species) and efficient natural enemies (high abundance/diversity or have strong predation ability such as some birds, see Maas et al. 2016). If the main insect herbivores (e.g., herbivore C and D in Fig. 2) can feed on all (or most of) the local main tree species and are subject to low or no predation pressure in the local tree stand, tree diversity might have no significant suppression effects on insect herbivory.

AcknowledgementsWe are thankful to the editors and reviewers for improving this manuscript. We thank Zhiming Zhang, Jianqiang Qian and all the members in our lab for their valuable comments.