Hollow deciduous trees are keystone structures hosting a high arthropod diversity but have been declining in numbers in central European managed forests for decades due to intensive forest management (Wetherbee et al. 2020). Modern forestry has, until recently, substantially changed the tree species- and age-composition of central European forest fragments over the last 200 years (Gossner et al. 2013; Seibold et al. 2015) and dramatically reduced the number of veteran deciduous trees (Lindenmayer et al. 2012; Müller et al. 2014; Sverdrup-Thygeson et al. 2017), resulting in a loss of dead wood habitats like tree hollows (Thorn et al. 2020).
Saproxylic beetles, i. e. those depending on dead wood or organisms living in dead wood for at least one part of their life cycle (Speight 1989), are important for ecosystem functioning in forest ecosystems as they promote wood decomposition through mutualistic relationships with fungi and microorganisms (Stokland et al. 2012; Ulyshen 2016), which drives element cycling and productivity (Gossner et al. 2013). However, due to the loss of dead wood habitats in central European forests, many saproxylic beetle species are endangered or have already gone extinct (Seibold et al. 2015; Thorn et al. 2020). Saproxylic beetle species that are specialized on a certain dead wood habitat such as tree hollows are exceptionally threatened, with about 75% of tree-hollow specialist species being listed in the Red List of Germany (Schmidl & Büche 2018).
To effectively protect saproxylic beetles in tree hollows not only knowledge of the beetles’ habitat requirements at local scales is needed (see Ranius et al. 2009; Koch Widerberg et al. 2012; Quinto et al. 2014; Micó et al. 2015; Micó 2018; Schauer et al. 2018b; Henneberg et al. 2021) but also a better understanding of how landscape composition at larger spatial scales influences saproxylic beetle assemblages in tree hollows (Franc et al. 2007; Müller & Gossner 2010; Ranius et al. 2015). Landscape composition is considered a key factor explaining species richness patterns at different spatial scales (Tscharntke et al. 2012; Gonthier et al. 2014). Habitat availability at landscape scale has been shown to be of high importance for species assemblages in old trees (Sverdrup-Thygeson et al. 2014). Some studies have emphasized the importance of the surrounding landscape for local saproxylic beetle species richness (Økland et al. 1996; Franc et al. 2007). Götmark et al. (2011) found that local dead wood amount was the main predictor of total species richness of saproxylic beetles. In contrast, the availability of woodland habitats at landscape scale was the main predictor of species richness of threatened species (Götmark et al. 2011). Similarly, Ranius et al. (2011) surveyed saproxylic beetles in oaks in southern Sweden and found that the occurrence of threatened species was positively affected by large-scale occurrence of oaks, indicating that they needed conservation efforts at larger spatial scales than common species. Overall, however, we still have limited knowledge of the relationship between saproxylic beetle diversity in forests and landscape factors at different spatial scales (Sverdrup-Thygeson et al. 2014), even though understanding scale dependency is crucial for the management of natural resources and conservation of biodiversity (Müller & Gossner 2010; Tscharntke et al. 2012; Micó et al. 2013; Ranius et al. 2015).
Besides the loss of dead wood habitats in central European forests, fragmentation of forest regions has been identified as a major driver of the decline of saproxylic beetle diversity (Ranius 2002; Brunet & Isacsson 2009; Lachat & Müller 2018). Since medieval times forest areas in central Europe have decreased greatly, resulting in a fragmented mosaic of unconnected forest patches of different size and distribution (Rüther & Walentowski 2008; Müller & Gossner 2010). Most saproxylic beetle species are assumed to have low host-tree preferences, meaning they inhabit tree hollows in different deciduous tree species (Milberg et al. 2014; Vogel et al. 2021). In general, tree hollows are far more common in deciduous than in coniferous trees (Siitonen 2012). Hence, for saproxylic beetles living in tree hollows, the landscape surrounding focal tree hollows is composed of patches of potentially suitable habitat (i. e. deciduous forest) and non-habitat (i. e. coniferous forest and open land).
As fragmentation of the landscape has resulted in isolated forest patches that are mostly surrounded by highly contrasting environmental matrices (Shepherd & Brantley 2005; Müller & Gossner 2010), saproxylic beetles must cover the distance between two habitat patches by dispersal. Based on the vulnerability of specialized species to habitat fragmentation (Oleksa et al. 2013; Sverdrup-Thygeson et al. 2017), one would expect the proportion of suitable habitat in the surrounding landscape to be of much larger importance for habitat specialists than generalists. Accordingly, many specialized saproxylic beetle species appear to not be able to overcome the distances between forest regions by dispersal (Ranius & Hedin 2001; Jonsson 2012; Oleksa et al. 2015). Jonsson (2000) argues that high and continuous availability of dead wood structures in ancient times may have led to low selection pressures for efficient dispersal among highly specialized saproxylic beetle species. Therefore, species with narrow habitat demands and low dispersal ability may have evolved (Jonsson 2000). Furthermore, species associated with long-lasting habitats like tree hollows are assumed to be rather limited in their dispersal ability (Kirby & Drake 1993; Nilsson & Baranowski 1997; Hedin et al. 2008; Oleksa et al. 2013). The relationship between degree of specialization and species’ dispersal ability is still subject to discussion (Martin & Fahrig 2018), but several studies on invertebrates have suggested that highly specialized species have lower dispersal abilities than generalist species (Entling et al. 2011; Carnicer et al. 2013; Dapporto & Dennis 2013; Stevens et al. 2014; Dahirel et al. 2015). Thus, many rare saproxylic beetles that specialized in temporally stable but spatially variable habitats like tree hollows might be dispersal-limited, as their ability to establish new populations far from present ones has been shown to be low (Ranius & Hedin 2001; Hedin et al. 2008).
Although we still lack species-specific details on dispersal abilities of most saproxylic beetle species (Feldhaar & Schauer 2018) certain morphological traits of beetles have been assumed to be related to dispersal ability (Gómez-Rodríguez et al. 2015). Body size of beetles has been shown to be a highly integrative trait representing metabolic rate, demographical properties but also dispersal ability as beetle species with a large body size are expected to be better dispersers (Hagge et al. 2019). Additionally, dispersal ability has been assumed to be positively associated with larger wing length, lower wing load, and wing aspect of beetle species (Hagge et al. 2021). Human land use can filter species assemblages by selecting against species with particular morphological traits, e. g. through habitat fragmentation or ecological degradation of forests (Hagge et al. 2021).
It has been proposed that instead of focusing solely on species richness, the diversity of functional traits of species within a community (functional diversity) should be studied (Hooper et al. 2005; Cadotte et al. 2011). A considerable amount of research has been done on predictors of saproxylic beetle species richness in hollow deciduous trees (Ranius 2002; Koch Widerberg et al. 2012; Quinto et al. 2014; Micó et al. 2015; Micó 2018; Schauer et al. 2018b; Henneberg et al. 2021), but predictors of functional diversity have remained largely unexplored (Wetherbee et al. 2020). It is unknown how landscape composition surrounding focal tree hollows will affect dispersal-associated morphological traits of the saproxylic beetle communities or their functional diversity, but we expected communities in tree hollows that are located in isolated forest patches or more fragmented forest regions to shift towards species with a better dispersal ability, i. e. beetles with relatively larger body size, larger wings and lower wing load.
In this study we address the following hypotheses: (I) Landscape composition surrounding the focal tree hollows will have a significant impact on saproxylic beetle species richness in the hollows. We expect tree hollows surrounded by a larger proportion of unsuitable habitat (i. e. open land or forest dominated by conifers) to harbor saproxylic beetle communities with lower diversity. (II) Landscape composition will have a stronger effect on threatened saproxylic beetle species compared to total species richness of saproxylic beetles as threatened species may have a lower dispersal ability. (III) Functional diversity of morphological traits that are associated with dispersal ability will be related to landscape composition surrounding the focal tree hollows. Beetle communities in isolated forest patches or fragmented forest regions are expected to contain a higher proportion of species with a better dispersal ability, reflected by morphological traits associated with dispersal, than communities in less isolated forest stands.