Advertisement

Ecological Research

, Volume 33, Issue 1, pp 205–212 | Cite as

Tree diversity has contrasting effects on predation rates by birds and arthropods on three broadleaved, subtropical tree species

  • Bo Yang
  • Bin Li
  • Yuxuan He
  • Lipeng Zhang
  • Helge Bruelheide
  • Andreas Schuldt
Original Article

Abstract

Plant diversity is hypothesized to strengthen biological control by promoting top-down pressure of predators on herbivores. However, studies on the effects of plant diversity on actual predation rates are still scarce, particularly in forest ecosystems. We analyzed the effect of tree species richness, and the potential influence of neighbor tree density, on predation rates of arthropods and birds on artificial clay caterpillars in a large-scale forest biodiversity experiment in south-east China. Our study was focused on three broadleaved tree species that are frequently damaged by lepidopteran caterpillars. Predation rates were influenced by tree species richness on only one of the three tree species, on which arthropod predation increased and bird predation decreased with increasing tree species richness. Importantly, these relationships were mediated by neighbor tree density, being most pronounced when focal trees had fewer surrounding neighbor trees. Our findings indicate that low tree density reduced arthropod predator abundances and predation rates, but that negative effects of this reduction were compensated for in more diverse tree mixtures by a functionally more diverse predator community. In contrast, lower tree densities might have benefited insectivorous birds by making trees more accessible particularly in monocultures, which are often structurally more uniform and denser than tree mixtures. Overall, our results point to an important role of species-specific and density-dependent mechanisms in modifying the consequences of biodiversity loss on top-down effects in forest ecosystems. Future work should aim at separating the effects of different predator guilds and those of host diversity from host density.

Keywords

BEF-China biodiversity and ecosystem function density-dependence predator trophic interaction 

Introduction

Predators can play important roles in structuring ecological communities and modifying the diversity and functioning of ecosystems across trophic levels (Schmitz 2006; Haddad et al. 2009; Letourneau et al. 2009). The strength of predator top-down control is hypothesized to increase with plant diversity. This is because diverse plant communities might increase the diversity and stability of habitat niches and resources required for a high abundance and diversity of predators (the ‘enemies hypothesis’; Root 1973). However, studies testing for the effects of plant diversity on predator abundances and species richness have reported mixed results, and particularly study results of forest ecosystems have been inconclusive (e.g. Vehviläinen et al. 2008; Sobek et al. 2009; Schuldt et al. 2011; Zhang and Adams 2011). This also applies to potential predator effects inferred from herbivore abundances and herbivory (e.g. Jactel and Brockerhoff 2007; Haase et al. 2015; Schuldt et al. 2015; Zhang et al. 2017).

However, there is little direct evidence on how actual predation rates are affected by plant diversity (Muiruri et al. 2016; Barbaro et al. 2017). Measuring actual predation rates is complicated by the fact that predation events are difficult to observe under natural conditions. Attack rates are therefore often assessed using either live or artificial sentinel prey (Lövei and Ferrante 2017). The use of artificial prey, such as lepidopteran caterpillars formed of clay, has proven to be particularly efficient (e.g. Koh and Menge 2006; Richards and Coley 2007; Low et al. 2014; Sam et al. 2015; Roslin et al. 2017). Although artificial prey may only mimic a limited set of characteristics of live prey (Howe et al. 2009), it allows for standardized comparisons of relative predation rates among study plots.

Our knowledge of how plant diversity affects predation rates is particularly poor for the highly diverse subtropical and tropical forests (Leles et al. 2017). Top-down effects of predators are considered to be particularly important in these low-latitudinal ecosystems (Schemske et al. 2009; Roslin et al. 2017; but see Moles and Ollerton 2016; Schuldt et al. 2017a), where factors promoted by (the long-term stability of) benign environmental conditions have shaped highly diverse and tightly interlinked plant and animal communities (Brown 2014). A recent study using clay caterpillars in tropical forests showed that overall predation rates increased with plant species richness (Leles et al. 2017). This relationship was taxon-specific and largely driven by arthropod (mainly ant) predation, whereas bird predation was not related to plant species richness (but see Muiruri et al. 2016 for temperate forests). Moreover, predation rates in the understory were influenced by plant density (Leles et al. 2017). Variability in plant density can influence predator foraging behavior (e.g. Franzreb 1983) and environmental conditions that affect arthropod abundances (e.g. light, temperature; Richards and Coley 2007). However, a better understanding of the extent to which variability in plant density influences the effect of plant species richness on predation rates requires further study.

We experimentally tested for the effects of tree species richness and the variability of neighbor tree density (due to tree mortality) on arthropod and bird predation rates, using artificial clay caterpillars across a tree species richness gradient of a large-scale forest biodiversity experiment in south-east China. Our analysis was based on three focal tree species that experience high damage by lepidopteran caterpillars under natural conditions. Previous surveys at our study site indicated that the dominant arthropod predators are unaffected by changes in tree species richness (Zhang et al. 2017). Because birds also prey on predatory arthropods (Mooney et al. 2010), one explanation of this finding could be increased predation rates by birds with increasing tree species richness. We therefore hypothesized that: (1) predation rates increase with increasing tree species richness for predatory birds, but not for predatory arthropods. In addition, we expected that (2) the variability in neighbor tree density modifies the strength of the predation-tree species richness relationship.

Methods

Study site & experimental design

The study was conducted on one (Site B) of the two experimental sites of the BEF-China tree diversity experiment, located near Xingangshan (N 29.0870° E 117.9285°) in Jiangxi province, south-east China. The climate at the study site is subtropical, with a mean annual temperature of 16.7 °C and mean annual precipitation of ca. 1800 mm (Yang et al. 2013). The study site has a size of ca. 20 ha on sloped terrain (with elevation ranging from 105 to 190 m). In 2010, trees were planted on a total of 271 study plots. Each plot has a size of 25.9 m × 25.9 m. Drawing from a set of 26 broadleaved tree species, study plots were planted as monocultures or mixtures with up to 24 tree species (Bruelheide et al. 2014). Tree species were randomly assigned to planting positions within plots (400 trees in 20 rows and 20 columns, 1.3 m apart) and planted in equal proportions in mixtures.

For our study, we selected 27 study plots that harbored at least one of the following three tree species: Alniphyllum fortunei (Hemsl.) Makino (family Styracaceae), Castanopsis fargesii Franch.(Fagaceae), and Elaeocarpus chinensis (Gardn. et Champ.) Hook. ex Benth. (Elaeocarpaceae). These three tree species were selected because they were found in a previous arthropod survey (Zhang et al. 2017) to support high abundances of caterpillars, and because each species has been planted in three replicate monocultures. Average tree height of the three species in our experiment was 4.31 (± 1.24 SD) m, 3.19 (± 1.29 SD) m, and 4.34 (± 1.49 SD) m, respectively (Y. Li, G. von Oheimb, W. Haerdtle, unpublished data). The 27 study plots comprised nine monocultures (three of each tree species), nine mixtures of eight tree species, and nine mixtures of 16 tree species. Mixtures contained either two or all three of the focal species. We randomly selected six trees of the central 12 × 12 tree individuals per plot (162 in total) for our experiment (six conspecific trees in monocultures, and two (in plots with all three focal species) or three individuals (in plots with two focal species) of each species in mixtures). However, the overall number of tree individuals per species varied slightly (A. fortunei 41, C. fargesii 65, E. chinensis 56), because the species compositions of the 27 study plots did not allow for a completely balanced design.

At the end of July 2016, we installed five artificial clay caterpillars on each of the six trees per study plot (i.e. 30 caterpillars per plot). The caterpillars (~ 40 mm long, 3–4 mm diameter) were made of odorless, non-toxic, green clay (Fimo Professional Modelling Clay, 85 g, Leaf Green) rolled around a 100 mm long piece of wire used to secure each caterpillar to one of five randomly selected branches on each tree. After exposure for 3 weeks in the field, the clay caterpillars were collected and attack marks were assigned to potential damage by birds, arthropods, or other causes (e.g. mechanical damage by adjacent branches) following Low et al. (2014).

Statistical analysis

Data on the five caterpillars per tree were pooled to obtain an overall estimate of predation rates per individual tree. We removed nine trees from the analyses because of very high rates of non-predator, mechanical damage that might have obscured attack marks of arthropods or birds.

We used generalized linear mixed-effects models with a binomial error structure to analyze the effects of tree species richness, neighbor tree density, and tree species identity on predation rates by either arthropods or birds (with the proportion of damaged caterpillars per plot as response variables). Neighbor tree density (which varied owing to tree mortality) was calculated as the number of live tree individuals in the eight planting positions directly adjacent to the focal tree. Study plot was used as a random effect. As fixed effects, we included tree species richness, tree species identity, neighbor tree density, as well as all two-way and three-way interactions between these variables. All continuous predictors were standardized by subtracting the mean and dividing by 2 standard deviations (Gelman 2008) prior to the analyses.

We used structural equation models (SEM) to unravel potentially causal pathways determining the relationships observed among trees, birds, and arthropods. We calculated ‘piecewise’ SEMs that allow accounting for the hierarchical structure of our data by using mixed-effects models with the same structure as described above as input (Lefcheck 2016). Piecewise SEMs are based on the directional separation approach to analyze multilevel path models (Shipley 2009). Model fit was assessed with Fisher’s c statistic, where P > 0.05 and an absence of missing paths indicates adequate fit. We calculated two alternative SEMs. The first model tested whether the effects of tree species richness on arthropod predation rates were driven by birds. This is because birds not only prey on invertebrate herbivores, but on arthropod predators as well. This model therefore excluded direct pathways from tree species richness and the tree density: richness interaction to arthropod predation rates. In the alternative model, we assumed that bird and arthropod predation rates were both directly influenced by tree species richness, and that arthropod predation rates were independent of bird predation. The covariation between tree species richness and tree density was fitted as correlated errors rather than a direct path because mortality in the BEF-China experiment was not dependent on tree species richness (Yang et al. 2013). The support for the two alternative models was assessed by comparing the models’ Akaike Information Criterion values corrected for small sample sizes (AICc).

All analyses were conducted in R 3.3.1 (https://www.R-project.org) with the packages lmerTest (Kuznetsova et al. 2016) and piecewiseSEM (Lefcheck 2016).

Because we did not assess predator and herbivore abundances at the time of our experiment, we used the data of a previous arthropod survey conducted at our study site (Zhang et al. 2017) for a cursory inspection of whether predatory arthropod and lepidopteran caterpillar abundances on the three focal tree species were affected by tree species richness. We used the same mixed-effects model approach as detailed in Zhang et al. (2017), but we fitted separate models based on the data for each of our three focal tree species. Although the survey design on predator and herbivore abundances comprised 64 study plots along the tree species richness gradient, species identities were not replicated among plots of a given tree species richness level (i.e. each of our focal tree species was present in only one plot of each richness level). The results of the abundance analyses therefore need to be considered with care and can only provide indications of potential species richness effects.

Results

Mean predation rates of arthropods (27.8% ± 15.6 SD) and birds (26.9% ± 21.1 SD) were similar when averaged across all trees. However, predation rates differed among tree species. Arthropod predation rates tended to be lowest on E. chinensis (Table 1), whereas bird predation rates were highest for this species (Table 2). Overall, arthropod and bird predation rates were negatively correlated (Pearson’s r = − 0.30, P < 0.001).
Table 1

Mixed-effects model testing for effects of tree species richness, species identity, and neighbor tree density on arthropod predation rates of the three focal tree species

Predictor

Estimate

SE

Z-value

P

(Intercept)

− 1.42

0.28

− 5.1

< 0.001

A. fortunei

0.52

0.32

1.6

0.106

C. fargesii

0.39

0.31

1.3

0.201

Tree species richness

1.28

0.51

2.5

0.012

Neighbor tree density

1.33

0.65

2.0

0.042

Tree richness:A. fortunei

− 1.78

0.61

− 2.9

0.003

Tree richness:C. fargesii

− 1.14

0.59

− 1.9

0.053

Tree richness:neighbor tree density

− 2.16

1.07

− 2.0

0.043

Neighbor tree density:A. fortunei

− 1.03

0.75

− 1.4

0.168

Neighbor tree density:C. fargesii

− 1.05

0.70

− 1.5

0.133

Tree richness:neighbor tree density:A. fortunei

2.64

1.26

2.1

0.036

Tree richness:neighbor tree density:C. fargesii

2.42

1.17

2.1

0.038

The intercept represents the mean effects of Elaeocarpus chinensis

Non-significant relationships are italicized. All continuous predictors were standardized

Table 2

Mixed-effects model testing for effects of tree species richness, species identity, and tree neighbor density on bird predation rates on trees of the three focal tree species

Predictor

Estimate

SE

Z-value

P

(Intercept)

− 0.43

0.21

− 2.1

0.034

A. fortunei

− 0.75

0.42

− 2.2

0.025

C. fargesii

− 0.59

0.49

− 3.2

0.002

Tree species richness

− 0.59

0.27

− 1.8

0.075

Neighbor tree density

− 0.79

0.25

− 1.2

0.235

Tree richness:A. fortunei

0.50

0.53

1.5

0.137

Tree richness:C. fargesii

2.32

0.95

0.9

0.345

Tree richness:neighbor tree density

0.82

0.55

2.4

0.014

Neighbor tree density:A. fortunei

0.57

0.61

0.9

0.347

Neighbor tree density:C. fargesii

0.40

0.55

0.7

0.469

Tree richness:neighbor tree density:A. fortunei

− 2.25

1.16

− 1.9

0.051

Tree richness:neighbor tree density:C. fargesii

− 2.70

1.05

− 2.6

0.010

The intercept represents the mean effects of Elaeocarpus chinensis

Non-significant relationships are italicized. All continuous predictors were standardized

Tree species-specific patterns also emerged with respect to the effects of tree species richness and neighbor tree density. Species richness and neighbor density effects on predation rates were only significant for E. chinensis, but not for A. fortunei and C. fargesii (Table 1, Fig. S1 in Electronic Supplementary Materials (ESM) 1). For E. chinensis, arthropod predation rates significantly increased with increasing tree species richness and an increasing number of neighboring trees (Table 1). However, neighbor tree density influenced the species richness effect, which was predicted to be most pronounced when neighbor tree densities were low (Table 1, Fig. 1a). Neighbor tree density also influenced bird predation rates, which were predicted to decrease with increasing tree species richness at low neighbor tree density (Table 2, Fig. 1b). With increasing neighbor tree density, the effect of tree species richness was predicted to level off (Fig. 1b). We note that tree species richness and neighbor tree density were moderately, positively correlated for E. chinensis (Pearson’s r = 0.61; P < 0.001), but not for the two other focal tree species (P > 0.05). However, variance inflation factors were low (< 2) and changing the order in which neighbor density and species richness were fitted after the categorical species identity had no effect on model results, indicating that density and richness effects were not interchangeable.
Fig. 1

Relationships between tree species richness, the modifying effect of neighbor tree density, and predation rates of a) arthropods, and b) birds on trees of E. chinensis. Tree density, i.e. the number of live neighbor trees (between 4 and 8), is indicated by different colors (corresponding to the colored numbers on the regression lines). Regression lines show model predictions of richness effects for different levels of tree density. Colored circles show the observed neighbor tree densities for individual trees (with values corresponding to the numbers of the same color on the regression lines). The interactions between tree species richness and tree density were significant for Elaeocarpus chinensis at P ≤ 0.05 (see Tables 1, 2). Color figure online

These results were supported by the SEM analyses on the data on E. chinensis (Fig. 2). The SEMs further indicated that the alternative hypothesis of direct effects of tree species richness on arthropod predation rates was better supported (AICc = 46.8; Fig. 2b) than the assumption of bird-mediated richness effects (AICc = 48.9; Fig. 2a). Bird predation only had a marginally significant, negative effect (P = 0.06) on arthropod predation rates (Fig. 2a, Table S1 in ESM 1).
Fig. 2

Results of piecewise structural equation models for trees of Elaeocarpus chinensis, assuming that tree species richness effects on arthropod predation rates are either a) indirect and mediated by bird predation (AICc = 48.9, Fisher’s C = 8.87, df = 4, P = 0.064), or b) direct and independent of bird predation (AICc = 46.8, Fisher’s C = 2.73, df = 2, P = 0.255). Straight arrows indicate positive (black) or negative (grey) effects (bold: P ≤ 0.05, dashed: P > 0.05), numbers are standardized path coefficients (see ESM 1: Tables S1, S2 for full model results). Curved arrow indicates correlated errors

The cursory inspection of predatory arthropod and lepidopteran caterpillar data from a previous survey at our study site indicated that predator abundances tended to increase with increasing tree species richness on E. chinensis (t = 1.8, df = 14, P = 0.091), but not on the other two tree species (P ≥ 0.2 in both cases). Lepidopteran caterpillar abundances were unrelated to tree species richness in all cases (P ≥ 0.2).

Discussion

Against our expectations, arthropod predation increased, whereas bird predation declined with increasing tree species richness. However, tree species richness had significant effects for only one of the three focal tree species. As hypothesized, these effects were influenced by neighbor tree density, being most pronounced at low tree densities. These findings point to an important role of species-specific and density-dependent mechanisms in modifying the consequences of biodiversity loss on top-down effects in forest ecosystems.

Our study revealed significant effects of tree species richness on predation rates only for E. chinensis. Previous studies have shown that tree species identity can strongly affect the abundance and diversity of both predatory arthropods and insectivorous birds (Holmes and Robinson 1981; Vehviläinen et al. 2008; Schuldt and Scherer-Lorenzen 2014). This can be due, for instance, to differences among tree species in the abundance of overall prey or of specific prey types, or morphological characteristics that influence habitat availability or foraging efficiency of predators (Holmes and Robinson 1981; Whelan 2001; Riihimäki et al. 2006). These effects may, in turn, be modified by neighboring trees in distinct ways (Barbosa et al. 2009), as indicated in our study by the effects of tree species richness and neighbor tree density. Further study is necessary to identify the underlying causes of the species-specific effects in our experiment. Individual functional traits such as (specific) leaf area and leaf nutrients that might affect prey communities do not consistently differentiate E. chinensis from the other two tree species (Kröber et al. 2014). Nevertheless, the combined effects of several of these and unmeasured traits (especially those related to the structural complexity within trees; Robinson and Holmes 1982; Whelan 2001) of focal and neighboring trees might provide further insight. As the contrasting patterns of arthropod and bird predation at our study site indicate, such effects are likely to differ in their relative importance to different functional groups of predators.

The increase in arthropod predation rates on E. chinensis with increasing tree species richness is in line with the common expectation that plant species richness promotes the abundance and richness of predators and therefore increases predation pressure (Root 1973; Haddad et al. 2009; Leles et al. 2017). This is also supported by our finding that predator abundances tended to increase with increasing tree species richness on E. chinensis in an earlier arthropod survey at our study site. At the same time, arthropod predation rates for E. chinensis increased with neighbor tree density. A denser canopy may allow for easier movement and redistribution of cursorial predators, such as ants and hunting spiders, among trees (Powell et al. 2011). Interestingly, our models predicted that increasing tree density weakens the direct species richness effect on arthropod predation rates. Tree species richness might be less important at high neighbor tree densities because predatory arthropods can easily immigrate from neighboring trees and might encounter improved microclimatic conditions (Ozanne et al. 2000; Sperber et al. 2004), thereby increasing the mobility and predation pressure by individual predatory species that might be highly abundant in monocultures (see e.g. Koricheva et al. 2000). In contrast, reduced tree density that leads to lower abundances of arthropod predators might have stronger negative effects on predation pressure in monocultures if a higher (functional) diversity of predators in mixtures can compensate for the effects of reduced abundances (e.g. Schuldt et al. 2014). Although the SEM results indicated that bird effects on arthropod predation rates were only marginally significant, bird predation might have had an additional effect on these patterns, as bird predation rates were also strongly influenced by tree density. Our study considered a relatively low number of study plots, and the observed trend of bird effects might potentially become stronger when more plots are considered and statistical power of the analyses is increased.

Interestingly, mean herbivory rates on the three tree species (Schuldt et al. 2017b) were inversely related to bird predation rates, being lowest on E. chinensis (8% leaf damage), intermediate on A. fortunei (11%), and highest on C. fargesii (19%). Bird predation rates on E. chinensis were predicted to decrease with increasing tree species richness when neighbor tree density was low, but were relatively constant across the species richness gradient at high neighbor tree density. Birds are much more mobile than most predatory arthropods, and their foraging behavior may be influenced by environmental conditions across a wider range of spatial scales (Muiruri et al. 2016). Many bird species show preferences of foraging on specific tree species with optimal prey availability and detectability (Holmes and Robinson 1981; Robinson and Holmes 1982) and, on a coarser spatial scale, may select patches that feature a high abundance of their preferred tree species (Muiruri et al. 2016). The higher predation rates in monocultures (at lower tree densities) predicted in our study may reflect such an active selection of suitable plots by individual bird species. At a finer spatial scale within plots, the modifying role of neighbor tree density could be due to several factors. Trees with fewer neighbors could make prey more apparent or more accessible to insectivorous birds than trees completely surrounded by a dense vegetation of neighboring trees (i.e. with a less open canopy; Muiruri et al. 2016), and/or they could offer a higher prey availability, because sun-exposed trees might have lower tannin and higher nitrogen content in their leaves or a higher productivity (Richards and Coley 2007). We note that planting distances between trees (1.29 m) in our plots are rather small, so that reduced densities and canopy gaps due to missing neighbors may have large effects on the foraging efficiency of highly mobile bird species. The fact that high neighbor tree densities were predicted to cause a decline in bird predation rates particularly in monocultures might be due to a uniform and dense vegetation structure of monospecific forest stands. Mixed-species stands usually show a more heterogeneous canopy structure that may make individual trees more visible and better accessible to birds also at high tree densities (Muiruri et al. 2016). We lack data on bird communities at our study site. Such data would help to better understand the drivers of the observed effects on E. chinensis, and the lack of effect on the other two tree species (which is in line with the lack of plot-level effects reported from other forest ecosystems; Muiruri et al. 2016; Leles et al. 2017).

We note that the correlation between tree species richness and neighbor tree density does not allow us to completely separate their effects. While significant effects of both predictors and the low variance inflation factors indicate that the information the two predictors provide is not completely redundant, richness effects may to some extent reflect reduced neighbor tree densities for E. chinensis in the more diverse plots. On the other hand, such reduced neighbor tree densities are a direct consequence of the increased probability that the species-rich mixtures contained as initial neighbors of E. chinensis several of the tree species with high mortality rates throughout the experiment (i.e. leading to a negative selection effect of tree species richness). For our study, this suggests that tree species richness may affect predation rates also indirectly via tree density, and that tree density is an important modifier to consider in the study of biodiversity effects. An important role of plant density in influencing diversity effects has previously been shown for other ecosystem processes (e.g. Marquard et al. 2009; Baruffol et al. 2013), but still remains little explored in the general BEF context. Understanding the realistic contribution of underlying, and potentially complementary, mechanisms driving such BEF relationships will require further manipulative studies that vary density and species richness independently.

The contrasting effect of tree species richness on arthropod and bird predation rates on E. chinensis, and the absence of effects on the two other focal tree species fit the observation that predator abundance and species richness at the plot level were not significantly related to tree species richness in a previous survey at our study site (Zhang et al. 2017). These findings indicate that an overall non-significant richness effect at the community level can be due to a high variability in species richness effects at the level of individual tree species. Such tree species-specific identity effects on trophic interactions may be important when it comes to understanding how non-random species loss or specific tree species compositions affect ecosystem processes (Hillebrand and Matthiessen 2009), or how the performance of individual plant species contributes to community assembly and diversity patterns under natural conditions (Schmitz 2006).Our study points at the modifying influence of tree density in this respect, and therefore highlights the context dependency of biodiversity effects. Considering such dependencies may be crucial for developing a realistic understanding of how biodiversity loss affects ecosystem functioning across trophic levels.

Notes

Acknowledgements

We thank Xuefei Yang, Chen Lin, Sabine Both, Keping Ma and all members of the BEF-China consortium that coordinated and helped with the establishment and maintenance of the experiment. We gratefully acknowledge funding by the German Research Foundation (DFG FOR 891/1, 891/2 and 891/3), the Sino-German Centre for Research Promotion in Beijing (GZ 524, 592, 698, 699, 785, 970 and 1020), and Jiangxi Provincial Department of Education (GJJ151285). BEF-China is supported by the German Centre for Integrative Biodiversity Research (iDiv) Halle-Jena-Leipzig (DFG FZT 118).

Compliance with ethical standards

Conflict of interest

The authors declare that they have no conflict of interest.

Supplementary material

11284_2017_1531_MOESM1_ESM.docx (156 kb)
Supplementary material 1 (DOCX 155 kb)

References

  1. Barbaro L, Rusch A, Muiruri EW, Gravellier B, Thiery D, Castagneyrol B (2017) Avian pest control in vineyards is driven by interactions between bird functional diversity and landscape heterogeneity. J Appl Ecol 54:500–508CrossRefGoogle Scholar
  2. Barbosa P, Hines J, Kaplan I, Martinson H, Szczepaniec A, Szendrei Z (2009) Associational resistance and associational susceptibility: having right or wrong neighbors. Annu Rev Ecol Evol Syst 40:1–20CrossRefGoogle Scholar
  3. Baruffol M et al (2013) Biodiversity promotes tree growth during succession in subtropical forest. PLoS ONE 8:e81246CrossRefGoogle Scholar
  4. Brown JH (2014) Why are there so many species in the tropics? J Biogeo 41:8–22CrossRefGoogle Scholar
  5. Bruelheide H et al (2014) Designing forest biodiversity experiments: general considerations illustrated by a new large experiment in subtropical China. Methods Ecol Evol 5:74–89CrossRefGoogle Scholar
  6. Franzreb KE (1983) A comparison of avian foraging behavior in unlogged and logged mixed-coniferous forest. Wilson Bull 95:60–76Google Scholar
  7. Gelman A (2008) Scaling regression inputs by dividing by two standard deviations. Stat Med 27:2865–2873CrossRefPubMedGoogle Scholar
  8. Haase J et al (2015) Contrasting effects of tree diversity on young tree growth and resistance to insect herbivores across three biodiversity experiments. Oikos 124:1674–1685CrossRefGoogle Scholar
  9. Haddad NM, Crutsinger GM, Gross K, Haarstad J, Knops JMH, Tilman D (2009) Plant species loss decreases arthropod diversity and shifts trophic structure. Ecol Lett 12:1029–1039CrossRefPubMedGoogle Scholar
  10. Hillebrand H, Matthiessen B (2009) Biodiversity in a complex world: consolidation and progress in functional biodiversity research. Ecol Lett 12:1405–1419CrossRefPubMedGoogle Scholar
  11. Holmes RT, Robinson SK (1981) Tree species preferences of foraging insectivorous birds in a northern hardwoods forest. Oecologia 48:31–35CrossRefPubMedGoogle Scholar
  12. Howe A, Lövei GL, Nachman G (2009) Dummy caterpillars as a simple method to assess predation rates on invertebrates in a tropical agroecosystem. Entomol Exp Appl 131:325–329CrossRefGoogle Scholar
  13. Jactel H, Brockerhoff EG (2007) Tree diversity reduces herbivory by forest insects. Ecol Lett 10:835–848CrossRefPubMedGoogle Scholar
  14. Koh LP, Menge DNL (2006) Rapid assessment of lepidoptera predation rates in neotropical forest fragments. Biotropica 38:132–134Google Scholar
  15. Koricheva J, Mulder CPH, Schmid B, Joshi J, Huss-Danell K (2000) Numerical responses of different trophic groups of invertebrates to manipulations of plant diversity in grasslands. Oecologia 125:271–282CrossRefPubMedGoogle Scholar
  16. Kröber W, Zhang S, Ehmig M, Bruelheide H (2014) Linking xylem hydraulic conductivity and vulnerability to the leaf economics spectrum—A cross-species study of 39 evergreen and deciduous broadleaved subtropical tree species. PLoS ONE 9:e109211CrossRefPubMedPubMedCentralGoogle Scholar
  17. Kuznetsova A, Brockhoff PB, Christensen RHB (2016) lmerTest: tests for random and fixed effects for linear mixed effect models (lmer objects of lme4 package). R package version 2.0-32. Available at https://CRAN.R-project.org/package=lmerTest
  18. Lefcheck JS (2016) PiecewiseSEM: Piecewise structural equation modelling in r for ecology, evolution, and systematics. Methods Ecol Evol 7:573–579CrossRefGoogle Scholar
  19. Leles B, Xiao X, Pasion BO, Nakamura A, Tomlinson KW (2017) Does plant diversity increase top–down control of herbivorous insects in tropical forest? Oikos 126:1142–1149CrossRefGoogle Scholar
  20. Letourneau DK, Jedlicka JA, Bothwell SG, Moreno CR (2009) Effects of natural enemy biodiversity on the suppression of arthropod herbivores in terrestrial ecosystems. Annu Rev Ecol Evol Syst 40:573–592CrossRefGoogle Scholar
  21. Lövei GL, Ferrante M (2017) A review of the sentinel prey method as a way of quantifying invertebrate predation under field conditions. Insect Sci 24:528–542CrossRefPubMedGoogle Scholar
  22. Low PA, Sam K, McArthur C, Posa MRC, Hochuli DF (2014) Determining predator identity from attack marks left in model caterpillars: guidelines for best practice. Entomol Exp Appl 152:120–126CrossRefGoogle Scholar
  23. Marquard E, Weigelt A, Roscher C, Gubsch M, Lipowsky A, Schmid B (2009) Positive biodiversity–productivity relationship due to increased plant density. J Ecol 97:696–704CrossRefGoogle Scholar
  24. Moles AT, Ollerton J (2016) Is the notion that species interactions are stronger and more specialized in the tropics a zombie idea? Biotropica 48:141–145CrossRefGoogle Scholar
  25. Mooney KA, Gruner DS, Barber NA, Van Bael SA, Philpott SM, Greenberg R (2010) Interactions among predators and the cascading effects of vertebrate insectivores on arthropod communities and plants. Proc Natl Acad Sci USA 107:7335–7340CrossRefPubMedPubMedCentralGoogle Scholar
  26. Muiruri EW, Rainio K, Koricheva J (2016) Do birds see the forest for the trees? Scale-dependent effects of tree diversity on avian predation of artificial larvae. Oecologia 180:619–630CrossRefPubMedGoogle Scholar
  27. Ozanne CMP, Speight MR, Hambler C, Evans HF (2000) Isolated trees and forest patches: Patterns in canopy arthropod abundance and diversity in Pinus sylvestris (Scots Pine). Forest Ecol Manag 137:53–63CrossRefGoogle Scholar
  28. Powell S, Costa AN, Lopes CT, Vasconcelos HL (2011) Canopy connectivity and the availability of diverse nesting resources affect species coexistence in arboreal ants. J Anim Ecol 80:352–360CrossRefPubMedGoogle Scholar
  29. Richards LA, Coley PD (2007) Seasonal and habitat differences affect the impact of food and predation on herbivores: a comparison between gaps and understory of a tropical forest. Oikos 116:31–40CrossRefGoogle Scholar
  30. Riihimäki J, Vehviläinen H, Kaitaniemi P, Koricheva J (2006) Host tree architecture mediates the effect of predators on herbivore survival. Ecol Entomol 31:227–235CrossRefGoogle Scholar
  31. Robinson SK, Holmes RT (1982) Foraging behavior of forest birds: the relationships among search tactics, diet, and habitat structure. Ecology 63:1918–1931CrossRefGoogle Scholar
  32. Root RB (1973) Organization of a plant-arthropod association in simple and diverse habitats: the fauna of collards (Brassica oleracea). Ecol Monogr 43:95–124CrossRefGoogle Scholar
  33. Roslin T et al (2017) Higher predation risk for insect prey at low latitudes and elevations. Science 356:742–744CrossRefPubMedGoogle Scholar
  34. Sam K, Remmel T, Molleman F (2015) Material affects attack rates on dummy caterpillars in tropical forest where arthropod predators dominate: an experiment using clay and dough dummies with green colourants on various plant species. Entomol Exp Appl 157:317–324CrossRefGoogle Scholar
  35. Schemske DW, Mittelbach GG, Cornell HV, Sobel JM, Roy K (2009) Is there a latitudinal gradient in the importance of biotic interactions? Annu Rev Ecol Evol Syst 40:245–269CrossRefGoogle Scholar
  36. Schmitz OJ (2006) Predators have large effects on ecosystem properties by changing plant diversity, not plant biomass. Ecology 87:1432–1437CrossRefPubMedGoogle Scholar
  37. Schuldt A, Scherer-Lorenzen M (2014) Non-native tree species (Pseudotsuga menziesii) strongly decreases predator biomass and abundance in mixed-species plantations of a tree diversity experiment. Forest Ecol Manag 327:10–17CrossRefGoogle Scholar
  38. Schuldt A, Both S, Bruelheide H, Härdtle W, Schmid B, Zhou H, Assmann T (2011) Predator diversity and abundance provide little support for the enemies hypothesis in forests of high tree diversity. PLoS ONE 6:e22905CrossRefPubMedPubMedCentralGoogle Scholar
  39. Schuldt A, Bruelheide H, Durka W, Michalski SG, Purschke O, Assmann T (2014) Tree diversity promotes functional dissimilarity and maintains functional richness despite species loss in predator assemblages. Oecologia 174:533–543CrossRefPubMedGoogle Scholar
  40. Schuldt A et al (2015) Early positive effects of tree species richness on herbivory in a large-scale forest biodiversity experiment influence tree growth. J Ecol 103:563–571CrossRefPubMedPubMedCentralGoogle Scholar
  41. Schuldt A et al (2017a) Belowground top-down and aboveground bottom-up effects structure multitrophic community relationships in a biodiverse forest. Sci Rep 7:4222CrossRefPubMedPubMedCentralGoogle Scholar
  42. Schuldt A et al (2017b) Herbivore and pathogen effects on tree growth are additive, but mediated by tree diversity and plant traits. Ecol Evol 7:7462–7474CrossRefPubMedPubMedCentralGoogle Scholar
  43. Shipley B (2009) Confirmatory path analysis in a generalized multilevel context. Ecology 90:363–368CrossRefPubMedGoogle Scholar
  44. Sobek S, Scherber C, Steffan-Dewenter I, Tscharntke T (2009) Sapling herbivory, invertebrate herbivores and predators across a natural tree diversity gradient in Germany’s largest connected deciduous forest. Oecologia 160:279–288CrossRefPubMedPubMedCentralGoogle Scholar
  45. Sperber CF, Nakayama K, Valverde MJ, Neves FdS (2004) Tree species richness and density affect parasitoid diversity in cacao agroforestry. Basic Appl Ecol 5:241–251CrossRefGoogle Scholar
  46. Vehviläinen H, Koricheva J, Ruohomaki K (2008) Effects of stand tree species composition and diversity on abundance of predatory arthropods. Oikos 117:935–943CrossRefGoogle Scholar
  47. Whelan CJ (2001) Foliage structure influences foraging of insectivorous forest birds: an experimental study. Ecology 82:219–231CrossRefGoogle Scholar
  48. Yang X et al (2013) Establishment success in a forest biodiversity and ecosystem functioning experiment in subtropical China (BEF-China). Eur J Forest Res 132:593–606CrossRefGoogle Scholar
  49. Zhang Y, Adams J (2011) Top-down control of herbivores varies with ecosystem types. J Ecol 99:370–372Google Scholar
  50. Zhang J et al (2017) Tree diversity promotes generalist herbivore community patterns in a young subtropical forest experiment. Oecologia 183:455–467CrossRefPubMedGoogle Scholar

Copyright information

© The Ecological Society of Japan 2017

Authors and Affiliations

  1. 1.Key Laboratory of Plant Resources and Biodiversity of Jiangxi ProvinceJingdezhen UniversityJingdezhenChina
  2. 2.Institute of Biology/Geobotany and Botanical GardenMartin-Luther-University Halle-WittenbergHalleGermany
  3. 3.German Centre for Integrative Biodiversity Research (iDiv) Halle-Jena-LeipzigLeipzigGermany

Personalised recommendations