Keep your forests tight with pedunculate oak

In this latest Author Blog, Lionel R. Hertzog and Michael P. Perring discuss their recent paper: ‘Overstorey composition shapes across-trophic level community relationships in deciduous forest regardless of fragmentation context.’


Community ecology is complex; some even say that it is a mess (Lawton, 1999). After all, everything apparently depends on everything else in diffuse ways. One way out of this labyrinthine complexity is to study emergent properties. Vellend (2010) proposes such a framework, by defining high-level processes governing community patterns. Yet, this particular framework does not apply to multi-trophic interactions.

Multi-trophic interactions sustain the delivery of multiple ecosystem services and are vital for efficient energy and nutrient transfer (Barnes et al., 2018). Understanding emergent properties from multiple trophic levels may assist management and awareness of biodiversity dynamics. In our paper, we provide a novel emergent metric of a multi-trophic system. We call this emergent property: “community tightness”. Inspired by previous work on pairwise and multiple co-inertia (Vleminckx et al., 2019; Wasof et al., 2019), and classic work on diversity responses in fragmented landscapes (Middleton & Merriam, 1983), we derived community tightness so as to be able to relate the strength of correlation across communities at multiple trophic levels to different putative drivers (Fig. 1).

Figure 1: The visualisation of community tightness. Taxonomic communities across trophic levels can exhibit a high (top) or low (bottom) degree of co-ordinated change across plots (bubble size indicates abundance of species in a taxonomic group; one set of bubbles per taxonomic group per level of co-ordinated change is shown for clarity). When this covariation (across multiple species within and across groups) is visualised in two-dimensional space after multiple co-inertia analyses, tight communities show a low summed distance (i.e. high congruence) from the plot synthetic score while loose communities show a high summed distance (i.e. low congruence).

In many ecosystems, including European forests (Fig. 2), sustainable management aims at preserving or restoring biodiversity at multiple trophic levels. Yet, previous works have revealed that different management strategies can have different impacts on multiple trophic levels (Leidinger et al., 2020; Schall et al., 2018). Quantifying community tightness may be one way to identify and communicate to managers best management practices, such as tree composition, for multi-trophic biodiversity. External drivers acting at larger spatial scales might also affect community tightness. For instance, the process of landscape fragmentation is expected to have great impacts on higher, more specialist, and more mobile trophic levels. Landscape fragmentation could therefore disrupt multi-trophic interactions, and decrease overall community tightness.

Figure 2: One of the studied plots in Merelbeke forest. Photo credit and copyright: Irene van Schrojenstein Lantman

To explore the impact of changing tree composition and landscape fragmentation on multi-trophic community tightness, we gathered data on 9 community groups spanning multiple trophic levels: understorey vegetation (primary producers); woodlice and millipedes (decomposers); leaf miners and gall formers (herbivores); harvestmen, carabids, spiders, bats and birds (predators) (Fig. 3). We measured species abundances across groups in 53 plots situated in 19 mature forest fragments scattered through a typical forest-urban-agricultural landscape to the south of Ghent, Belgium. Plots varied in tree species composition with all possible combinations of pedunculate oak, red oak and common beech, and were exposed to different degrees of fragmentation: from plots situated close to forest edges in small and isolated forest fragments to plots far from forest edges situated in large and well-connected forest fragments.

Figure 3: Four different taxa sampled in the plots; (A) the remainder of a leaf miner mine, (B) a Plecotus auritus individual, (C) a Parus major individual and (D) a Ligidium hypnorum individual. Photo credits and copyright: (A) and (C) Irene van Schrojenstein Lantman, (B) Rene Janssen and (D) Gert Arijs.

We found that all communities, except bats, showed significant pairwise relationships, i.e. correlations with other communities. Multi-trophic community tightness was strongly affected by tree composition: forest plots with pedunculate oak either in monoculture or in mixtures showed higher tightness than forest plots without this species. Contrary to our expectations, we did not find any disrupting effect of landscape fragmentation on the tightness metric, despite the fact that some individual community group compositions were related to fragmentation.

In our paper, we reported multi-trophic community tightness was related to tree species composition in the studied system. More correlated community matrices could suggest more efficient energy transfer across trophic levels, and ultimately improved ecosystem functioning. From a more practical perspective, tighter relations between communities could imply that conservation actions acting on one trophic level may reverberate through multiple communities and improve overall biodiversity. Such postulates require further experimental, observational and theoretical exploration.

Lionel Hertzog and Michael Perring Ghent University, Belgium.


You can read the full paper online: Overstorey composition shapes across-trophic level community relationships in deciduous forest regardless of fragmentation context

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s