Canopy vs. understory: Does tree diversity affect bee and wasp communities and their natural enemies across forest strata?

Abstract

Bees and wasps provide important ecosystem services such as pollination and biocontrol in crop-dominated landscapes, but surprisingly little information is available on hymenopteran communities in temperate forest ecosystems. Species richness and abundance of bees and wasps can be hypothesised to increase with plant diversity, structural complexity, and availability of food and nesting resources. By experimentally exposing standardised nesting sites, we examined abundance and species richness of cavity-nesting bees (pollinators), wasps (predators) and their associated parasitoids across a tree diversity gradient in a temperate deciduous forest habitat. In addition, spatial distribution of individuals and species across forest strata (canopy vs. understory) was tested. Abundance and species richness was high for predatory wasps, but generally low for pollinators. Species-rich forest stands supported increased abundance, but not species richness, of pollinators and predatory wasps, and also increased abundance and species richness of natural enemies. In addition, the forests showed a distinct spatial stratification in that abundance of bees, wasps and parasitoids as well as parasitism rates were higher in the canopy than understory. We conclude that particularly the canopy in temperate forest stands can serve as an important habitat for predatory wasp species and natural enemies, but not bee pollinators. Enhanced tree diversity was related to increased hymenopteran abundance, which is likely to be linked to an increase in nesting and food resources in mixed forest stands.

Introduction

In tropical and temperate ecosystems alike, the forest canopy is known as a major source of insect diversity (Erwin, 1982, Erwin, 1988, Stork, 1988, Novotny and Basset, 2005, Dial et al., 2006), and various authors have revealed differences in insect diversity and community composition across forest strata, regardless of the studied taxon (e.g. Le Corff and Marquis, 1999, Schulze et al., 2001, Fermon et al., 2005, Roisin et al., 2006). In rare cases no distinct stratification was observed (Stork and Grimbacher, 2006). The stratification of distinct forest layers like canopy and understory results in a vertical gradient of different abiotic conditions and biotic resources including light availability, temperature, wind exposure, as well as food and nesting site availability, which impacts the diversity and distribution of insect communities due to the provision of different microhabitats (see Schulze et al., 2001, Yanoviak et al., 2003 and references therein). The relative importance of canopy and understory communities to overall diversity patterns and processes might also be linked to tree diversity, but so far data on the effect of tree diversity on vertical stratification of insect communities is lacking.

Cavity-nesting bees and wasps have been shown to be a particularly useful model system for studying abundance and diversity of hymenopteran pollinators and predators (Tscharntke et al., 1998). Cavity-nesting communities comprise guilds such as pollinators (solitary bees), predators (solitary wasps) and associated parasitoids. They can be easily assessed experimentally by introducing above-ground nesting sites to quantify biodiversity patterns and trophic interactions. Linking biodiversity of a functional group such as parasitoids or pollinators with ecosystem functioning (i.e. parasitism and pollination) is a topical issue in ecology and has recently gained major attention (e.g. Kremen, 2005, Hooper et al., 2005, Bianchi et al., 2006, Priess et al., 2007, Ebeling et al., 2008, Höhn et al., 2008, Klein et al., 2008, Winfree et al., 2008).

Several studies have dealt with the abundance and diversity of cavity-nesting bees and wasps in temperate agricultural (e.g., Gathmann et al., 1994, Kruess and Tscharntke, 2002, Sheffield et al., 2008, Steffan-Dewenter and Schiele, 2008) or tropical agroforestry habitats (e.g. Klein et al., 2006, Tylianakis et al., 2006a, Tylianakis et al., 2006b). In contrast, data on cavity-nesting hymenopteran communities in temperate forest habitats are lacking (but see Taki et al., 2008a). This is remarkable, especially since the proximity to forest habitats has recently been shown to increase local abundance and diversity of cavity-nesting wasps (Holzschuh et al., 2009) and also bee pollinators (Taki et al., 2007) in agricultural habitats. Forests are hypothesised to provide resources such as deadwood nesting sites, which are otherwise not present in primarily cleared or simple landscapes. Hence, forest habitats are believed to house source populations of hymenopteran predators, which might spill over into adjacent agricultural habitats, eventually enhancing biocontrol (Tscharntke et al., 2005, Holzschuh et al., 2009).

Taki et al., 2008a, Taki et al., 2008b provide valuable information on community structure in the forest understory, but it remains ambiguous how cavity-nesting bees, wasps and their natural enemies are spatially distributed across forest types and strata. Until now, virtually nothing has been known about how the temperate forest canopy contributes to hymenopteran diversity.

Here, we present the first data on cavity-nesting hymenopterans and their natural enemies in the canopy and understory of a temperate forest habitat. The study was conducted in the Hainich National Park, Germany's largest coherent deciduous forest. We utilised a semi-natural, mature forest neighbourhood to a priori establish a tree diversity gradient ranging from simple beech to mixed forest stands comprising various deciduous tree species. This allowed us to analyse differences in stratification, and to directly relate our results to tree diversity of the sampled forest stands. Increased plant diversity has been linked to increases in species richness of various arthropod orders (Siemann et al., 1998) including cavity-nesting bees and wasps (Tscharntke et al., 1998), and might also influence trophic interactions such as parasitism. We hypothesise that abundance, species richness and parasitism rates of cavity-nesting bee and wasp communities increase with increased tree diversity due to enhanced resource availability (e.g. nesting resources, accessory food for parasitoids), and differ between forest layers (canopy vs. understory) due to microhabitat stratification.

Specifically, we address the following questions: (1) how abundant and species-rich are bee and wasp communities (hereafter referred to as hosts) and their associated parasitoids, and how are they distributed across the tree diversity gradient? (2) How are hosts and parasitoids spatially distributed across forest strata? (3) Are parasitism rates related to the tree diversity gradient? (4) Do parasitism rates differ between forest strata?