Mammal functional diversity increases with vegetation structural complexity in two forest types

Highlights

• Functional diversity links species diversity with ecosystem function.

• We surveyed mammals and vegetation structure in southeast Australian eucalypt forest.

• Mammal functional diversity responded positively to vegetation structural complexity.

• Relationships were consistent in wet and dry forest.

• Conserving structurally complex vegetation may enhance ecosystem function.

Abstract

Enhanced understanding of relationships between functional diversity (FD) and environmental gradients is crucial given accelerating rates of land-cover change and disturbance worldwide. Functional diversity measures the abundance, range and distribution of traits in a community, and links species diversity with ecosystem function. Several studies have related bird or invertebrate FD to environmental gradients, but information on the responses of mammal FD to vegetation structural diversity at scales of management relevance is scarce. We addressed this knowledge gap by examining responses of ground-dwelling mammal FD to vegetation structural complexity in wet (high-productivity) and dry (low-productivity) eucalypt forest in the Otway Ranges, southeast Australia. In dry forest, we expected a positive relationship between FD and vegetation structural complexity because more resources should enable species with a greater diversity of traits to co-exist. We expected negative correlations in wet forest, where competitive dominance may drive a decrease in FD as structural complexity increases. Ground-dwelling mammals were surveyed using camera traps, and we used five traits to construct four FD indices (richness, evenness, divergence, and dispersion). Six vegetation structure variables were used to calculate two indices of vegetation structure, and we used linear mixed models to relate functional diversity and species richness to vegetation structural complexity (the total abundance of structural attributes) and heterogeneity (the level of contrast or patchiness in structure), in wet and dry forest. Camera traps detected ten native ground-dwelling mammal species. All FD indices were positively correlated with vegetation structural complexity, but only functional dispersion responded to structural heterogeneity. Contrary to expectations, relationships between FD and structural complexity were consistent in both forest types, and we suggest that low levels of functional niche occupancy prevented competitive dominance in wet forest. Species richness did not respond to any predictor variables, and is unlikely to be a useful surrogate of ground-dwelling mammal FD. Our results indicate that forest managers may sustain ecosystem functions performed by ground-dwelling mammals by conserving structurally complex vegetation.

Introduction

Land-cover change is occurring at rapid rates worldwide, and the consequences for ecosystem function are poorly known (Luck et al., 2013, Watson et al., 2014). Direct measurement of ecosystem function is often intractable, but life-history traits can reveal links between land-cover change, species diversity and ecosystem function (Driscoll et al., 2010, Pimm et al., 1988, Webb et al., 2010). There is ongoing interest in the implications of environmental change for functional diversity (FD), which is defined as the value, range, and density of functional traits in a community (Díaz et al., 2007, Díaz and Cabido, 2001, Mason et al., 2005). A large body of literature demonstrates that FD, rather than species diversity per se, regulates ecological processes (Chapin et al., 1998, Loreau, 2000, Villéger et al., 2008), and flexible new methods for measuring FD have fuelled its application to a range of taxonomic groups and scales (Gonzalez-Maya et al., 2016, Schirmel et al., 2012, Seymour et al., 2015).

Mammal FD has been related to land-cover types and environmental gradients in climate and productivity (Gonzalez-Maya et al., 2016, Mason-Romo et al., 2017). Gradients in climatic factors such as temperature and rainfall result in spatial variability in vegetation structure, which has often been correlated with faunal diversity (Chesson, 2000, Tews et al., 2004). In theory, correlations between structural complexity and species diversity should be positive because a high diversity of vegetation structural attributes is expected to provide an array of resources which support a greater number of species (Fischer et al., 2006, MacArthur and MacArthur, 1961, Tews et al., 2004). Similarly, positive relationships between FD and vegetation structural complexity are expected to arise from associations between individual traits and individual structural characteristics. Several studies have identified positive relationships between bird or invertebrate FD and vegetation structural diversity or surrogate variables (e.g. Schirmel et al., 2012, Ibarra and Martin, 2015, Seymour et al., 2015, Sitters et al., 2016a, Sitters et al., 2016b), but the responses of mammal FD to direct measures of vegetation structural complexity in continuous forest are largely unknown (Marcot and Aubry, 2003, Zimbres et al., 2017).

In forests, positive responses of species diversity or FD to vegetation structural diversity are expected in less productive systems because limited supplies of energy and resources prevent competitive dominance (Huston, 2004). Conversely, a few species can become dominant in productive systems where energy and resources are plentiful, and FD can decline when resource availability is high (Sitters et al., 2016b, Verschuyl et al., 2008). A study of associations between bird FD and vegetation structural diversity in wet (high-productivity) and dry (low-productivity) forest revealed marked differences in the strength and direction of the FD-structure relationship between forest types (Sitters et al., 2016b). It is plausible that vegetation structural complexity and productivity interact to influence mammal FD.

Our principal aim was to examine the responses of ground-dwelling mammal FD to vegetation structural complexity in the wet and dry eucalypt forests of the Otway Ranges, southeast Australia, where a rainfall gradient, in combination with topographic complexity and prescribed burning, creates substantial variability in vegetation structure. We used six vegetation structure variables to derive two indices; vegetation structural complexity (the total abundance of structural attributes) and heterogeneity (the level of contrast or patchiness in structure). In dry forest, we predicted a positive relationship between FD and vegetation structure indices, and in wet forest, we expected negative correlations because competitive dominance may drive a decrease in FD as structural complexity or heterogeneity increase (Huston, 2004, Schoener, 1974, Sitters et al., 2016b). For comparison, we used ground-dwelling mammal species richness as a response variable, and again anticipated positive responses to vegetation structure indices in dry forest and negative responses in wet forest.

Ground-dwelling mammals are recognised for their important roles in ecosystem functioning (Fleming et al., 2014, Roberts et al., 2015) and are likely to be affected by changing disturbance regimes, such as increased use of prescribed fire in Australia and elsewhere (Adams, 2013, Fernandes et al., 2013, Penman et al., 2007). Our results will help to guide forest management by identifying key drivers of ground-dwelling mammal FD.