- Email: [email protected]
Beetle community response to residual forest patch size in managed boreal forest landscapes: Feeding habits matter
Forest Ecology and Management 368 (2016) 63–70
Contents lists available at ScienceDirect
Forest Ecology and Management journal homepage: www.elsevier.com/locate/foreco
Beetle community response to residual forest patch size in managed boreal forest landscapes: Feeding habits matter Mathieu Bouchard a,⇑, Christian Hébert b a b
Ministère de la Forêt, de la Faune et des Parcs du Québec, Direction de la Recherche Forestière, 2700 rue Einstein, Québec (QC) G1P 3W8, Canada Natural Resources Canada, Canadian Forest Service, Laurentian Forestry Centre, 1055 du PEPS, PO Box 10380, Québec (QC) G1V 4C7, Canada
a r t i c l e
i n f o
Article history: Received 24 November 2015 Received in revised form 23 February 2016 Accepted 24 February 2016 Available online 9 March 2016 Keywords: Insect diversity Forest fragmentation Boreal forests Beetles Feeding habits Clearcutting
a b s t r a c t Forest fragmentation by management activities has been implicated in the decline of forest biodiversity. Even though boreal ecosystems are generally deemed quite resilient to disturbance effects, high contemporary levels of disturbance might push forest-interior species toward decline or extinction. In this study, we examined beetle communities in forest patches of different sizes, including clearcuts, residual postharvest patches from 0.03 to 50 ha in size, and large mature forest tracts (>1000 ha). Overall, community structure follows a gradient between clearcuts and large mature forest tracts, even if patch size effects were more difficult to detect among patches >2.5 ha. Beetles were most abundant in clearcuts, and species richness was highest in small tree groups (0.03–0.05 ha). The effects of fragmentation were strongly conditioned by beetle feeding habits. Predators and xylophagous beetles were mostly associated with clearcuts or smaller patches (i.e., small tree groups or large tree groups [0.3–0.5 ha]), whereas fungivorous beetles were associated with forest-interior habitats. Although many forest-interior species were still present in relatively small patches 1–5 years after harvesting, negative effects of habitat fragmentation on these species might increase in the long-term. Ó 2016 Elsevier B.V. All rights reserved.
1. Introduction Habitat fragmentation has contributed to the decline and extinction of many plant and animal species around the world (Hanski and Gaggiotti, 2004; Ewers and Didham, 2006). However, the impacts of habitat fragmentation on species persistence at the landscape level probably vary among ecosystems and regions. For example, species living in the boreal forest can be expected to possess adaptations, such as good dispersal capabilities and tolerance for suboptimal habitats, that will allow them to persist in a context of frequent natural disturbances (Thompson et al., 2009). In Canada, contemporary levels of disturbance in boreal ecosystems are probably higher than during the preindustrial period. In particular, the widespread use of clearcutting, compounded by the ongoing effect of natural disturbances, has decreased the landscape-level abundance of mature forests below their natural range of variation (Cyr et al., 2009; Bouchard and Pothier, 2011). Consequently, the capacity of some species to maintain viable populations in isolated patches of mature forests could be exceeded. Species that are rare, have poor dispersal capabilities, or that are ⇑ Corresponding author. E-mail addresses: [email protected] (M. Bouchard), christian. [email protected] (C. Hébert). http://dx.doi.org/10.1016/j.foreco.2016.02.029 0378-1127/Ó 2016 Elsevier B.V. All rights reserved.
strongly dependent on particular microhabitats or environmental conditions (i.e., with a narrow niche breadth) could be more sensitive to habitat rarefaction or fragmentation (Henle et al., 2004; Ewers and Didham, 2006). Thus, even if Canadian boreal species are generally considered resilient to disturbances, some species with particular life histories could still be vulnerable to habitat rarefaction and fragmentation. Beetles are very well represented in forest ecosystems worldwide and have diversified ecological roles. By contrast with the larger animal species that are often examined with respect to fragmentation effects in boreal regions, such as woodland caribou (Festa-Bianchet et al., 2011) or birds (Schmiegelow and Mönkkönen, 2002), insects have smaller home ranges, but also very diversified life histories. This makes them useful indicators of small-scale processes (Fahrig et al., 2015). Some studies have found that beetles respond dramatically to forest fragmentation, particularly in temperate or tropical ecosystems (Didham et al., 1998; Davies et al., 2000; Driscoll and Weir, 2005; Barbaro and Van Halder, 2009). In boreal conditions, however, responses are more difficult to circumscribe (Gandhi et al., 2001; Webb et al., 2008; Saint-Germain et al., 2013). When these effects are present, few indications are provided regarding the threshold habitat size under which species persistence is threatened, or whether this threshold varies from one species to another (Lee et al., 2015).
64
M. Bouchard, C. Hébert / Forest Ecology and Management 368 (2016) 63–70
In this study, we used beetles to examine species and community responses to different sizes of residual forest patches in recently harvested boreal forests. Our hypotheses are (1) that the size of mature forest habitats is an important determinant of beetle community structure, and (2) that species response to residual forest patch size varies depending on life-history traits. We tested these hypotheses by sampling a gradient of patch sizes in landscapes recently affected by clearcut harvestings. 2. Methods 2.1. Study area The study area covers a region of about 5000 km2 in the eastern part of the Canadian boreal forest (Fig. 1), in which the natural fire cycle is ca 250 years. The forest matrix is overwhelmingly dominated by two tree species: black spruce (Picea mariana [Mill]. B.S. P.) and balsam fir (Abies balsamea [L.] Mill.) (Bouchard et al., 2008; Bélisle et al., 2011). The region has been affected by forest harvestings since the 1970s, but large tracts of mature forests of natural origin are still present throughout the study area. In 2012 and 2013, we sampled 5 landscapes in which aggregated clearcuts had been performed recently (2007–2011, Fig. 1). Each of these landscapes represents an area of 30–60 km2 dominated by clearcuts, inside which patches of residual forests were retained, varying in size from a few m2 to ca 50 ha. The guidelines used to delineate harvest areas on public lands in the province of Quebec specify that forest composition and characteristics of residual forest patches must be representative of the overall forest characteristics in the surrounding area before harvesting. 2.2. Sampling In each landscape, we sampled 6 types of habitats defined by the size of residual forest patches inside the matrix: clearcuts (0 ha, no residual forest patch), small tree groups (0.03–0.05 ha), large tree groups (0.3–0.5 ha), small blocks (2.5–4 ha), large blocks (50 ha) and large mature forest tracts (>1000 ha). The large mature forest tracts were located outside, but adjacent to, the aggregated clearcuts. One sample plot was established in each habitat, for a total of 30 plots (5 landscapes 6 habitats or residual forest patch sizes). Plots located in clearcuts were established more than 50 m from the nearest forest edge. For small tree groups, large tree groups and small blocks, we selected forest patches that were more or less circular in shape, and placed the plot at their center. For
large blocks and large mature forest tracts, the plots were established at least 200 m from the nearest clearcut. Each sample plot contained 2 flight intercept traps (FIT) and 3 pitfall traps. The FITs consisted in 4 panels (15 40 cm each, 2 made of Plexiglas and 2 of mosquito net) mounted in a cross pattern along a 10-cm diameter black ABS cylinder, with a funnel located below the cylinder leading to a collecting vial. The 2 FITs were suspended, their base about 1 m above ground level, and placed 10 m apart in each sample plot. The pitfalls were MultipherÒ traps (12.5 cm in diameter) placed flush to the ground and 10 m apart in each sample plot. For the small tree groups, which were too small to contain all the traps, we paired 2 plots located approximately 50 m apart, placing 1 FIT and 1 pitfall in the first plot, and 1 FIT and 2 pitfalls in the second. The traps were operated from May 28 to August 17 in 2012, and from May 27 to August 26 in 2013. Propylene glycol was used as a killing and preserving liquid, and the traps were emptied at 2- to 3-week intervals. All captured beetles were identified at the species level, except for the genera Epurea (Nitidulidae), Cryptophagus (Cryptophagidae) and specimens belonging to the family Ptiliidae, as well as some Agathidium (Leiodidae) and Corticaria (Latridiidae) females. We estimated basal area of living trees and snags (diameter at breast height >9.1 cm) using a prism (factor k = 2) and tallied the selected trees by species. We measured coarse woody debris (>8 cm diameter) present on the ground with the line intercept method (Van Wagner, 1968) after establishing 3 line transects (10 m each) at 120° intervals around the center of each plot. Along each transect, all coarse woody debris intersected were tallied. For strongly decayed pieces of coarse woody debris, we measured diameter in 2 perpendicular directions and calculated the geometric mean. The proportion of ground covered by Sphagnum mosses, which is a good indicator of habitat suitability for some beetles in boreal forests (Paquin, 2008), was also estimated in 4 microplots (25 m2 each) placed 5 m from the center of the plot in each cardinal direction. Certain preharvest forest characteristics can persist in postdisturbance communities. For example, some residual mature-forest beetle populations could subsist for a few years after logging. In this study, we verified if preharvest forest composition had a residual effect on beetle communities in clearcuts by calculating preharvest basal area in all clearcut stands, from an inventory of black spruce and balsam fir stumps within a 10-m radius circular plot. Together, black spruce and balsam fir represented >95% of overall plot basal area. The proportion of balsam fir was used to express variations in dominant tree species composition.
Fig. 1. Map illustrating the location of the 5 studied landscapes (left), and an example of plot spatial distribution in one landscape (right).
M. Bouchard, C. Hébert / Forest Ecology and Management 368 (2016) 63–70
Preharvest and postharvest proportions of balsam fir were thus considered as distinct variables in the analyses. 2.3. Statistical analyses In the analyses, the size of residual forest patches was converted to an ordinal vector ranging from clearcuts (residual forest patch size = 1) to large mature forest tracts (residual forest patch size = 6). Beetles were also classified according to their feeding habits and microsite preferences on the basis of the literature and expert opinions (Table 1). We examined the effect of residual forest patch size on beetle abundance and species richness using mixed model analyses. The abundance in each plot of individuals from each group served as an aggregated trait value (Shipley et al., 2006). The analyses were performed with the glmer function of the lme4 package (Bates et al., 2015) in R, using landscape as a random factor. Because we used count data, a Poisson error distribution was used in the models. For each statistical model, we evaluated 95% confidence intervals with bootstraps analyses, considering the explanatory variables for which the interval did not include 0 as significant. We applied the same procedure to examine the effect of residual forest patch size on species richness for each group. We used redundancy analysis (RDA) to determine the importance of residual forest patch size as a driver of the overall beetle
65
community structure, relative to the other environmental factors. Beetle data for the 2 trap types and 2 sample years were pooled for each plot, then Hellinger-transformed to reduce the importance of rare species (Borcard et al., 2011). Snags and coarse woody debris lying on the ground were modeled as distinct variables because these microhabitats are generally colonized by different species (Jonsell and Weslien, 2003). Plot elevation was used as a proxy for climatic variability within the study area. All environmental variables were standardized before analysis (Borcard et al., 2011). Redundancy analysis was performed with the rda function in the vegan package (Oksanen et al., 2013) in R version 3.02 (R Development Core Team, 2013). Relationships between individual species and residual forest patch size were tested using the multipatt function of the Indicspecies package (De Caceres and Jansen, 2015) in R. In this approach, an index that combines a species’ frequency and abundance in a target group of samples relative to its overall frequency and abundance is compared to a randomized distribution. Species that are more frequent and abundant (i.e., concentrated) in a particular group of samples (e.g., residual forest patches of a particular size) than expected by chance were considered to be indicator species for that environment. In this analysis, residual forest patch size was thus considered as a categorical predictive variable. We further restricted the analysis to beetle–patch size relationships that could be interpreted from an ecological viewpoint, using the
Table 1 Categories of life-history traits used for individual beetle species. Trait
Category
Description
Microsite preference
Litter Dead tree Living plant
Found on dead foliage or highly decomposed wood Found on dead trees standing (snags) or lying on the ground (CWD) Found on living trees or understory plants
Feeding habits
Phytophage Xylophage Predator Fungivore Detritivore
Includes insects that feed on living plant tissues such as foliage, roots and shoots during the larval stage Feed primarily on xylem and phloem Feed primarily on grubs, gastropods, collembolla, aphids, etc. Feed on carpophores, mycelium, spores or yeasts Feed on dead organic matter, except wood: leaves, dung, carcasses, dead insects, etc. Some insects feeding on carcasses were also classified as detritivores.
Fig. 2. Abundance and richness of beetle groups according to residual forest patch size. R1 = clearcuts, R2 = small tree groups (0.03–0.05 ha), R3 = large tree groups (0.3–0.5 ha), R4 = small blocks (2.5–5 ha), R5 = large blocks (50 ha), R6 = large mature forest tracts (>1000 ha). Presence of a filled arrow beside the absolute abundance or absolute species richness for a beetle species group indicates a significant response to residual forest patch size (see Table 2 for complete results).
66
M. Bouchard, C. Hébert / Forest Ecology and Management 368 (2016) 63–70
restcomb argument in the multipatt function (De Caceres and Jansen, 2015). To assess the significance of each species, we performed a permutation test (n = 4999). Only species with more than 10 individuals were considered in these analyses. For each species, sample plots where catches represented less than 1% of the overall species abundance were attributed an abundance of 0, in order to decrease the importance of vagrant individuals (Dufrêne and Legendre, 1997). Species showing significant associations with large forest tracts, patches or clearcuts were designated as forestinterior, patch and open-habitat species, respectively.
not only in large mature forest tracts, but also in smaller forest patches (Table 3). Among this group, the most sensitive to residual forest patch size was Zenodosus sanguineus, which was found only in large blocks and large mature forest tracts. Patch species were specifically associated with intermediate-sized residual forest patches (small tree groups, large tree groups, and small blocks). Finally, open-habitat were mostly confined to clearcuts and small tree groups, with one species, Syneta pilosa, being also found in large tree groups (Table 3). Among the combinations that were tested with Indicspecies which could have a biological interpretation, only one category contained no indicator species: patches from 0 to 4 ha (R1 + R2 + R3 + R4, Table 3).
3. Results A total of 8274 beetles were caught, belonging to 420 taxa. Overall, beetles were more abundant in clearcuts than in any other size of residual forest patch (Fig. 2). The size of residual forest patches significantly affected the abundance and richness of the different groups of beetles (Table 2). In larger patches, fungivores were more abundant, but abundance and species richness were lower for xylophages, detritivores and predators (Table 2). As a result, fungivores represented a much larger proportion of the overall beetle community in larger residual forest patches (>0.3 ha) than in smaller patches and clearcuts (<0.05 ha) (Fig. 2). Clearcuts and small tree groups were mostly dominated by predators, and to a smaller extent, by xylophages and phytophages. Regarding microsite preferences, residual forest patch size had a negative effect on the abundance of beetles associated with litter and dead trees, and a positive effect on the abundance of beetles associated with living plants, but a negative effect on species richness for this same group (Table 2). The RDA indicates that the selected environmental factors explained 40% of the overall community variation, including 22% along the first axis and 7% along the second axis (Fig. 3). Axis 1 (pseudo-F = 8.4, p < 0.001) and axis 2 (pseudo-F = 2.5, p = 0.002) are both significant, according to permutation tests (Borcard et al., 2011). The most important environmental factors identified in the RDA were residual forest patch size along the first axis, and elevation along the second axis (Fig. 3). Along the second axis, variability is much higher for the larger patches (>2 ha) than for the smaller ones (Fig. 3). The indicator species analysis showed that 40 species were significantly associated with certain residual forest patch sizes (Table 3). These species can be categorized into 3 groups: forestinterior species (11 species), patch species (6 species), and openhabitat species (23 species). Forest-interior species were present
4. Discussion 4.1. Feeding habits and habitat preferences as determinants of beetle response to residual forest patch size Our hypotheses that beetle community structure is influenced by patch size, and that species traits could be an important determinant to understand beetle response to residual forest patch size, were both supported by the results. Among the 5 trophic groups, fungivores were the only beetles that were collectively more abundant in larger patches, especially those larger than 2.5 ha. Fungivores represented 65% of the beetles captured in large tracts of mature forest, but only 33 and 18% in small tree groups and in clearcuts, respectively. Closed-crown forests are typically cooler and less exposed to droughts, and are probably more favorable to the presence of fungi (Durall et al., 1999) and to some groups of Coleoptera associated with them (Sippola and Renvall, 1999; Jacobs and Work, 2012; Lee et al., 2015). Moreover, although most fungivores can probably feed on or colonize different types of fungi, some are more closely associated with fungi primarily found on living or dead standing trees (Jonsell and Weslien, 2003), a substrate that is absent from clearcuts. For example, the most abundant fungivorous species captured in this study, Glischrochilus sanguinolentus, is apparently associated with mycelium growing on sap found on the trunk of injured (but live) standing trees (Jelínek et al., 2010); the abundance of this species increases progressively from clearcuts to patches >50 ha (Supplementary Material). Xylophagous beetles showed an opposite response to that of fungivorous species, with increasing abundance when size of residual patches decreases. Previous studies have also reported such a contrast between these two groups (Boulanger et al., 2010; Lee
Table 2 Summary of the GLM analyses of beetle response to residual forest patch size, according to their feeding habits and microsite preferences. Trait
Preferences
Response variable
Feeding habits
Xylophage Xylophage Fungivore Fungivore Predator Predator Phytophage Phytophage Detritivore Detritivore
Abundance Richness Abundance Richness Abundance Richness Abundance Richness Abundance Richness
Estimate 0.227 0.068 0.127 0.012 0.586 0.046 0.184 0.208 0.206 0.088
0.020 0.033 0.010 0.020 0.012 0.025 0.097 0.130 0.026 0.052
SE
11.3 2.1 13.4 0.6 48.7 1.9 1.9 1.6 7.9 1.7
( 0.27, 0.19) ( 0.13, 0) (0.11, 0.15) ( 0.03, 0.05) ( 0.61, 0.56) ( 0.09, 0) ( 0.38, 0) ( 0.47, 0.04) ( 0.26, 0.16) ( 0.19, 0.01)
Microsite preferences
Litter Litter Dead tree Dead tree Living plant Living plant
Abundance Richness Abundance Richness Abundance Richness
0.446 0.027 0.031 0.007 0.149 0.086
0.010 0.021 0.010 0.020 0.015 0.035
44.8 1.3 3.0 0.4 10.3 2.4
( 0.47, 0.43) ( 0.07, 0.01) ( 0.05, 0.01) ( 0.05, 0.03) (0.12, 0.18) ( 0.16, 0.02)
Note: Residual forest patch size was considered as a numerical vector ranging from 1 (clearcuts) to 6 (large mature forest tracts).
z-value
CI (95%)
M. Bouchard, C. Hébert / Forest Ecology and Management 368 (2016) 63–70
67
Fig. 3. Ordination of beetle communities by redundancy analysis (RDA). Environmental factors (arrows) and sites (symbols) are shown, but individual species are not shown. PreFir = Preharvest proportion of balsam fir, PostFir = Postharvest proportion of balsam fir, CWD = Coarse woody debris volume. Position of centroids is indicated with a larger symbol for each patch size category.
et al., 2015), which could be explained in part by different responses to environmental cues. Xylophagous taxa such as Curculionidae and Cerambycidae tend to be attracted to recently killed trees, particularly when the latter emit volatiles and are sunexposed. Despite the scarcity of large dead trees in clearcuts, these areas remain covered by logging residues and freshly cut stumps, which are likely to attract some species and might represent a suitable breeding substrate (Hedgren, 2007). Predator’s response followed that of xylophagous beetles. However, the greater abundance of predators in very small patches and clearcuts contrasts with the findings of Davies et al. (2000), who observed a lower abundance of predators in small patches than in larger ones. Their study took place in tropical forest ecosystems fragmented by agricultural lands, which is very different from the present study. In boreal environments, the greater abundance of predators such as ground beetles (Niemelä et al., 1993; Heliölä et al., 2001) or spiders (Buddle et al., 2000; Larrivée et al., 2005) in clearcut environments was well documented. This could be explained by the warmer sun-exposed environment, which promotes a greater biological activity in the duff layer (Chauvat et al., 2003; Malmström et al., 2009; Siira-Pietikäinen and Haimi, 2009), and results in greater abundance of decomposing microorganisms that serve as preys. Penetration and diffusion of clearcut species inside forest patches also probably explains the gradual increase in predator abundance as one moves along the gradient from large forest tracts to clearcuts. Overall, knowledge of life-history strategies was useful to understand and explain changes in community structure. Still, in the case of beetles, trait-based approaches should be considered with caution. Many species, even the most abundant ones, are still poorly described in North-America. For example, Pseudanostirus triundulatus (Coleoptera: Elateridae), the most abundant species overall, was considered in this study to be primarily litter-dwelling, based on assumed feeding habits of the larval stages and field observations. However, this species was classified differently in previous studies (Boulanger et al., 2010), indicating lingering uncertainties with respect to its biology and, by extension, to the biology of elaterids in general (Thomas et al., 2009; Saint-Germain et al., 2013). The fact that ecosystem-level studies are very dependent on high-quality, species-level descriptive work is particularly obvious for taxonomic groups such as beetles, which are extremely diversified in terms of both species richness and life-history strategies.
4.2. Additional drivers of beetle community response to residual forest patch size Even if part of the community-level response could be explained by life-history strategies, there is considerable variability among beetles that share similar traits. For example, forestinterior species include many fungivorous beetles, but also species from other feeding groups (Table 3). This suggests that beetle response to residual forest patch size is also explained by lifehistory traits that were not considered directly in this study. Species tolerance for suboptimal conditions (i.e., niche breadth) could be an important factor to explain their response to habitat fragmentation (Swihart et al., 2006). For example, a narrow niche breadth could mean that a xylophagous species can use a piece of coarse woody debris if it is located in a shaded microenvironment (inside a forest patch), but not if the same piece is located in a clearcut or a forest edge (Gibb et al., 2006; Morin et al., 2015). Similarly, some species generally associated with forest-interior conditions could have behavioral, life cycle or ecophysiological adaptations that allow them to persist for long periods in suboptimal conditions (Niemelä et al., 1993). Populations of insect species narrowly associated with mature forest environments are less likely to be in direct contact with neighboring populations when surrounded by suboptimal habitats, and are thus more prone to stochastic extinction events (Ewers and Didham, 2006; Laurance et al., 2011; Slade et al., 2013). In the context of this study, fragmentation effects are thus expected to be more important for forest-interior species that are intolerant of clearcut environments. Differences in dispersal capability between species is another important mechanism that could explain the variation in fragmentation effects. Even if they cannot breed in suboptimal habitats, many species could be able to disperse through it. In fact, in the case of flying insects, recent clearcuts could facilitate dispersal, as these areas are devoid of dense understory layers that impede their flight. Individuals immigrating from the surrounding forests and dispersing though the matrix of clearcuts could thus colonize patches where the species is absent, or replenish local populations. Still, even if species with good dispersal capabilities are expected to be more likely to maintain viable populations in fragmented landscapes (Thomas et al., 2001; Tscharntke et al., 2002), dispersal remains a complex mechanism that can interact with other
68
M. Bouchard, C. Hébert / Forest Ecology and Management 368 (2016) 63–70
Table 3 Results of the indicator species analysis which allowed the categorization species as ‘‘clearcut”, ‘‘patch” and ‘‘forest-interior” specialists. Species
Family
Microsite
Feeding
Patch size
Indicspecies statistics
R1
R2
R3
R4
R5
R6
Ind.
Stat.
p
Open-habitat species
Pediacus fuscus Agriotes limosus Ampedus quebecensis Paraphotistus nitidulus Pseudanostirus triundulatus Corticarina cavicollis Mordellaria borealis Sphaeriestes virescens Heterothops fusculus Mycetoporus.consors Nudobius cephalus Acmaeops p.proteus Rhagium inquisitor Scolytus piceae Ampedus pullus Sericus incongruus Ellychnia corrusca Xylita laevigata Colopterus truncatus Gyrophaena meduxnekeagensis Pselaphus bellax Syntomium grahami Syneta pilosa
Cucujidae Elateridae Elateridae Elateridae Elateridae Latridiidae Mordellidae Salpingidae Staphylinidae Staphylinidae Staphylinidae Cerambycidae Cerambycidae Curculionidae Elateridae Elateridae Lampyridae Melandryidae Nitidulidae Staphylinidae Staphylinidae Staphylinidae Chrysomelidae
Dead tree Living plant Dead tree Litter Litter Dead tree Dead tree Dead tree Litter Litter Litter Dead tree Dead tree Living plant Dead tree Living plant Dead tree Dead tree Living plant Dead tree Litter Litter Living plant
Predator Xylophage Predator Predator Predator Fungivore Fungivore Predator Predator Fungivore Predator Xylophage Xylophage Xylophage Predator Xylophage Predator Xylophage Fungivore Fungivore Predator Detritivore Phytophage
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
1 1 1 1 1 1 1 1 1 1 1 3 3 3 3 3 3 3 3 3 3 3 6
0.816 0.749 0.735 0.739 0.913 0.726 0.749 0.739 0.600 0.655 0.655 0.632 0.707 0.721 0.687 0.807 0.822 0.961 0.632 0.707 0.938 0.717 0.759
0.002 0.006 0.007 0.009 0.000 0.019 0.040 0.012 0.045 0.017 0.029 0.033 0.007 0.012 0.019 0.004 0.003 0.000 0.038 0.005 0.000 0.014 0.036
Patch species
Trechus crassiscapus Pityokteines sparsus Abstrulia canadensis Trypodendron lineatum Pseudanostirus hoppingi Cyphon obscurus
Carabidae Curculionidae Tetratomidae Curculionidae Elateridae Scirtidae
Litter Living plant Dead tree Living plant Litter Litter
Predator Xylophage Fungivore Xylophage Predator Detritivore
0 0 0 0 0 0
1 1 1 1 1 1
0 1 1 1 1 1
0 0 0 1 1 1
0 0 0 0 0 0
0 0 0 0 0 0
2 4 4 7 7 7
0.707 0.707 0.748 0.733 0.786 0.810
0.014 0.007 0.005 0.023 0.005 0.006
Forest-interior species
Pterostichus punctatissimus Cryptophagus spp. Corticaria n.linearis Enicmus tenuicornis Dictyopterus aurora Emmesa connectens Epuraea spp. Glischrochilus s.sanguinolentus Henoticus serratus Olophrum rotundicolle Zenodosus sanauineus
Carabidae Cryptophagidae Latridiidae Latridiidae Lycidae Melandryidae Nitidulidae Nitidulidae Cryptophagidae Staphylinidae Cleridae
Litter Dead tree Dead tree Dead tree Dead tree Dead tree Dead tree Living plant Dead tree Litter Dead tree
Predator Fungivore Fungivore Fungivore Fungivore Xylophage Fungivore Fungivore Fungivore Unknown Predator
0 0 0 0 0 0 0 0 0 0 0
1 1 1 1 1 1 1 1 0 0 0
1 1 1 1 1 1 1 1 1 0 0
1 1 1 1 1 1 1 1 1 1 0
1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1
10 10 10 10 10 10 10 10 9 8 5
0.938 0.908 0.903 0.931 0.830 0.828 0.923 0.917 0.815 0.776 0.701
0.004 0.005 0.009 0.002 0.036 0.035 0.005 0.002 0.005 0.017 0.019
Note: Residual forest patch sizes were the following: R1 = clearcuts (0 ha), R2 = small tree groups (0.03–0.05 ha), R3 = large tree groups (0.3–0.5 ha), R4 = small blocks (2.5– 4 ha), R5 = large blocks (50 ha), R6 = large mature forest tracts (>1000 ha). The following treatment combinations were tested with Indicspecies: ‘‘R1”, ‘‘R2”, ‘‘R1 + R2”, ‘‘R1 + R2 + R3”, ‘‘R1 + R2 + R3 + R4”, ‘‘R2 + R3”, ‘‘R2 + R3 + R4”, ‘‘R2 + R3 + R4 + R5 + R6”, ‘‘R3 + R4 + R5 + R6”, ‘‘R4 + R5 + R6”, ‘‘R5 + R6”.
life-history traits and make species distribution patterns difficult to predict (Bonte et al., 2012). Finally, competitive or trophic interactions could also explain some fragmentation effects (Fagan et al., 1999). For example, an increased diffusion of open-habitat predators in smaller forest patches (Fig. 2) could put additional pressure on some forestinterior species that would serve as prey. Habitat fragmentation could also modify competitive interactions, as the influx of openhabitat species that share similar traits (e.g. feeding habits) could negatively affect other species with poor competitive abilities. In the present study, we do not have sufficient information to separate the potential effects of the above-mentioned processes. Gaining a better understanding of these processes will probably be crucial in the future in order to design sustainable forest practices. Three lines of action could help in this respect. First, longterm monitoring and replication across regions would be very helpful to better assess population trends and further circumscribe potentially sensitive taxa. Second, better descriptions of sensitive species could help assess their niche breadth (trait variance) and potential sensitivity to environmental change. Third, manipulative studies with critical habitats such as coarse woody debris (Gibb et al., 2006) could help isolate the effects of the different processes
involved in fragmentation effects (niche breadth, dispersal, interspecific relationships). 5. Conclusions Our study confirms that a few years after harvesting, the negative effects of forest fragmentation on forest-interior species are mostly limited to smaller patches (0.03–0.5 ha). Beetle communities found in patches of at least 2.5 ha appeared similar to those found in larger patches. At the community level, we propose that the dominant pattern in small patches comes from an influx of open-habitat species, mostly predators, and the depletion of certain forest-interior species, mostly fungivores. This pattern seems similar to the progressive loss of forest-interior beetles and gain of open-habitat species that was observed in previous studies along gradients of partial cut intensities (Stenbacka et al., 2010; Légaré et al., 2011). The basic trait-based approach used in this study was helpful, and confirms that subdividing the vast, relatively undifferentiated group of Coleoptera according to feeding habits and microsite preferences helps achieve a better understanding of the main processes that drive fragmentation effects (Driscoll and Weir, 2005).
M. Bouchard, C. Hébert / Forest Ecology and Management 368 (2016) 63–70
Negative effects on mature forest species in harvested landscapes are expected to increase with time: for example, potential habitats such as residual stumps and slash in recently clearcut areas will gradually decompose and disappear. Dispersal limitations could also become an issue as the vegetation grows back in clearcut areas. Moreover, in boreal Canada, where old unmanaged stands are still present in the larger matrix, source habitats for many forest-interior species are expected to become less abundant in the future, as forest activities gradually expand toward the north (e.g. Bouchard and Pothier, 2011). This context could reduce the amount of colonizing species reaching residual forest patches, and further reduce the persistence of forest-interior insect populations at the landscape level (Kouki et al., 2012). Acknowledgements We thank S. Lauzon, C. Paquet, F. Berger and Y. Dubuc for their help with fieldwork, N. Robert and M.-C. Émond for their help managing and sorting the samples, A. Hibbert, J. Jacobs, C. Chantal, G. Pelletier and J. Klimaszewski for species identification, and D. Tousignant for editing the English on the manuscript. This study was financed by the Quebec Ministry of Forests, Fauna and Parks and by the Canadian Forest Service. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.foreco.2016.02. 029. References Barbaro, L., Van Halder, I., 2009. Linking bird, carabid beetle and butterfly lifehistory traits to habitat fragmentation in mosaic landscapes. Ecography 32, 321–333. Bates, D., Maechler, M., Bolker, B., Walker, S., 2015. lme4: Linear Mixed-Effects Models Using Eigen and S4. R Package Version 1.1-7... Bélisle, A.C., Gauthier, S., Cyr, D., Bergeron, Y., Morin, H., 2011. Fire regime and oldgrowth boreal forests in central Québec, Canada: an ecosystem management perspective. Silva Fennica 45, 889–908. Bonte, D., Van Dyck, H., Bullock, J.M., Coulon, A., Delgado, M., Gibbs, M., Lehouck, V., Matthysen, E., Mustin, K., Saastamoinen, M., 2012. Costs of dispersal. Biol. Rev. 87, 290–312. Borcard, D., Gillet, F., Legendre, P., 2011. Numerical Ecology with R. Springer Science & Business Media. Bouchard, M., Pothier, D., 2011. Long-term influence of fire and harvesting on boreal forest age structure and forest composition in eastern Québec. For. Ecol. Manage. 261, 811–820. Bouchard, M., Pothier, D., Gauthier, S., 2008. Fire return intervals and tree species succession in the North Shore region of eastern Quebec. Can. J. For. Res. 38, 1621–1633. Boulanger, Y., Sirois, L., Hébert, C., 2010. Distribution of saproxylic beetles in a recently burnt landscape of the northern boreal forest of Quebec. For. Ecol. Manage. 260, 1114–1123. Buddle, C.M., Spence, J.R., Langor, D.W., 2000. Succession of boreal forest spider assemblages following wildfire and harvesting. Ecography, 424–436. Chauvat, M., Zaitsev, A.S., Wolters, V., 2003. Successional changes of Collembola and soil microbiota during forest rotation. Oecologia 137, 269–276. Cyr, D., Gauthier, S., Bergeron, Y., Carcaillet, C., 2009. Forest management is driving the eastern North American boreal forest outside its natural range of variability. Front. Ecol. Environ. 7, 519–524. Davies, K.F., Margules, C.R., Lawrence, J.F., 2000. Which traits of species predict population declines in experimental forest fragments? Ecology 81, 1450–1461. De Caceres, M., Jansen, F., 2015. Package ’indicspecies’. . Didham, R.K., Hammond, P.M., Lawton, J.H., Eggleton, P., Stork, N.E., 1998. Beetle species responses to tropical forest fragmentation. Ecol. Monogr. 68, 295–323. Driscoll, D.A., Weir, T., 2005. Beetle responses to habitat fragmentation depend on ecological traits, habitat condition, and remnant size. Conserv. Biol. 19, 182– 194. Dufrêne, M., Legendre, P., 1997. Species assemblages and indicator species: the need for a flexible asymmetrical approach. Ecol. Monogr. 67, 345–366. Durall, D., Jones, M.D., Wright, E.F., Kroeger, P., Coates, K.D., 1999. Species richness of ectomycorrhizal fungi in cutblocks of different sizes in the Interior Cedar-
69
Hemlock forests of northwestern British Columbia: sporocarps and ectomycorrhizae. Can. J. For. Res. 29, 1322–1332. Ewers, R.M., Didham, R.K., 2006. Confounding factors in the detection of species responses to habitat fragmentation. Biol. Rev. 81, 117–142. Fagan, W.F., Cantrell, R.S., Cosner, C., 1999. How habitat edges change species interactions. Am. Nat. 153, 165–182. Fahrig, L., Girard, J., Duro, D., Pasher, J., Smith, A., Javorek, S., King, D., Lindsay, K.F., Mitchell, S., Tischendorf, L., 2015. Farmlands with smaller crop fields have higher within-field biodiversity. Agric. Ecosyst. Environ. 200, 219–234. Festa-Bianchet, M., Ray, J., Boutin, S., Côté, S., Gunn, A., 2011. Conservation of caribou (Rangifer tarandus) in Canada: an uncertain future. Can. J. Zool. 89, 419–434. Gandhi, K.J., Spence, J.R., Langor, D.W., Morgantini, L.E., 2001. Fire residuals as habitat reserves for epigaeic beetles (Coleoptera: Carabidae and Staphylinidae). Biol. Conserv. 102, 131–141. Gibb, H., Pettersson, R.B., Hjältén, J., Hilszczan´ski, J., Ball, J.P., Johansson, T., Atlegrim, O., Danell, K., 2006. Conservation-oriented forestry and early successional saproxylic beetles: responses of functional groups to manipulated dead wood substrates. Biol. Conserv. 129, 437–450. Hanski, I.A., Gaggiotti, O.E., 2004. Ecology, Genetics and Evolution of Metapopulations. Academic Press. Hedgren, P.O., 2007. Early arriving saproxylic beetles (Coleoptera) and parasitoids (Hymenoptera) in low and high stumps of Norway spruce. For. Ecol. Manage. 241, 155–161. Heliölä, J., Koivula, M., Niemelä, J., 2001. Distribution of carabid beetles (Coleoptera, Carabidae) across a boreal forest–clearcut ecotone. Conserv. Biol. 15, 370–377. Henle, K., Davies, K.F., Kleyer, M., Margules, C., Settele, J., 2004. Predictors of species sensitivity to fragmentation. Biodivers. Conserv. 13, 207–251. Jacobs, J.M., Work, T.T., 2012. Linking deadwood-associated beetles and fungi with wood decomposition rates in managed black spruce forests. Can. J. For. Res. 42, 1477–1490. Jelínek, J., Carlton, C., Cline, A., Leschen, R., 2010. Nitidulidae Latreille, 1802. In: Leschen, R., Beutel, R., Lawrence, J. (Eds.), Handbook of Zoology. De Gruyter, Berlin, pp. 390–407. Jonsell, M., Weslien, J., 2003. Felled or standing retained wood – it makes a difference for saproxylic beetles. For. Ecol. Manage. 175, 425–435. Kouki, J., Hyvärinen, E., Lappalainen, H., Martikainen, P., Similä, M., 2012. Landscape context affects the success of habitat restoration: large-scale colonization patterns of saproxylic and fire-associated species in boreal forests. Divers. Distrib. 18, 348–355. Larrivée, M., Fahrig, L., Drapeau, P., 2005. Effects of a recent wildfire and clearcuts on ground-dwelling boreal forest spider assemblages. Can. J. For. Res. 35, 2575– 2588. Laurance, W.F., Camargo, J.L., Luizão, R.C., Laurance, S.G., Pimm, S.L., Bruna, E.M., Stouffer, P.C., Williamson, G.B., Benítez-Malvido, J., Vasconcelos, H.L., 2011. The fate of Amazonian forest fragments: a 32-year investigation. Biol. Conserv. 144, 56–67. Lee, S.-I., Spence, J.R., Langor, D.W., Pinzon, J., 2015. Retention patch size and conservation of saproxylic beetles in boreal white spruce stands. For. Ecol. Manage. 358, 98–107. Légaré, J.-P., Hébert, C., Ruel, J.-C., 2011. Alternative silvicultural practices in irregular boreal forests: response of beetle assemblages. Silva Fennica 45, 937– 956. Malmström, A., Persson, T., Ahlström, K., Gongalsky, K.B., Bengtsson, J., 2009. Dynamics of soil meso-and macrofauna during a 5-year period after clear-cut burning in a boreal forest. Appl. Soil Ecol. 43, 61–74. Morin, M., Hébert, C., Berthiaume, R., Bauce, É., Brodeur, J., 2015. Short-term effect of selection cutting in boreal balsam fir forest on cerambycid and scolytid beetles. J. Appl. Entomol. 139, 553–566. Niemelä, J., Langor, D., Spence, J.R., 1993. Effects of clear-cut harvesting on boreal ground-beetle assemblages (Coleoptera: Carabidae) in Western Canada. Conserv. Biol. 7, 551–561. Oksanen, J., Blanchet, F.G., Kindt, R., Legendre, P., Minchin, P.R., O’Hara, R., Simpson, G.L., Solymos, P., Stevens, M., Wagner, H., 2013. Package ‘vegan’. R Packag ver 254, 20–28. Paquin, P., 2008. Carabid beetle (Coleoptera: Carabidae) diversity in the black spruce succession of eastern Canada. Biol. Conserv. 141, 261–275. R Development Core Team, 2013. In: R: A Language and Environment for Statistical Computing. R Foundation For Statistical Computing, Vienna, Austria. Saint-Germain, M., Drapeau, P., Hibbert, A., 2013. Saproxylic beetle tolerance to habitat fragmentation induced by salvage logging in a boreal mixed-cover burn. Insect Conserv. Divers. 6, 381–392. Schmiegelow, F.K., Mönkkönen, M., 2002. Habitat loss and fragmentation in dynamic landscapes: avian perspectives from the boreal forest. Ecol. Appl. 12, 375–389. Shipley, B., Vile, D., Garnier, É., 2006. From plant traits to plant communities: a statistical mechanistic approach to biodiversity. Science 314, 812–814. Siira-Pietikäinen, A., Haimi, J., 2009. Changes in soil fauna 10 years after forest harvestings: comparison between clear felling and green-tree retention methods. For. Ecol. Manage. 258, 332–338. Sippola, A.-L., Renvall, P., 1999. Wood-decomposing fungi and seed-tree cutting: a 40-year perspective. For. Ecol. Manage. 115, 183–201. Slade, E.M., Merckx, T., Riutta, T., Bebber, D.P., Redhead, D., Riordan, P., Macdonald, D.W., 2013. Life-history traits and landscape characteristics predict macromoth responses to forest fragmentation. Ecology 94, 1519–1530.
70
M. Bouchard, C. Hébert / Forest Ecology and Management 368 (2016) 63–70
Stenbacka, F., Hjältén, J., Hilszczanski, J., Dynesius, M., 2010. Saproxylic and nonsaproxylic beetle assemblages in boreal spruce forests of different age and forestry intensity. Ecol. Appl. 20, 2310–2321. Swihart, R.K., Lusk, J.J., Duchamp, J.E., Rizkalla, C.E., Moore, J.E., 2006. The roles of landscape context, niche breadth, and range boundaries in predicting species responses to habitat alteration. Divers. Distrib. 12, 277–287. Thomas, J., Bourn, N., Clarke, R., Stewart, K., Simcox, D., Pearman, G., Curtis, R., Goodger, B., 2001. The quality and isolation of habitat patches both determine where butterflies persist in fragmented landscapes. Proc. Royal Soc. London B: Biol. Sci. 268, 1791–1796. Thomas, S.L., Wagner, R.G., Halteman, W.A., 2009. Influence of harvest gaps and coarse woody material on click beetles (Coleoptera: Elateridae) in Maine’s Acadian forest. Biodivers. Conserv. 18, 2405–2419.
Thompson, I., Mackey, B., McNulty, S., Mosseler, A., 2009. Forest resilience biodiversity and climate change. In: A Synthesis of the Biodiversity/ Resilience/Stability Relationship in Forest Ecosystems Secretariat of the Convention on Biological Diversity. Technical Series, Montreal, pp. 1–67. Tscharntke, T., Steffan-Dewenter, I., Kruess, A., Thies, C., 2002. Characteristics of insect populations on habitat fragments: a mini review. Ecol. Res. 17, 229– 239. Van Wagner, C., 1968. The line intersect method in forest fuel sampling. Forest Sci. 14, 20–26. Webb, A., Buddle, C.M., Drapeau, P., Saint-Germain, M., 2008. Use of remnant boreal forest habitats by saproxylic beetle assemblages in even-aged managed landscapes. Biol. Conserv. 141, 815–826.
Contents lists available at ScienceDirect
Forest Ecology and Management journal homepage: www.elsevier.com/locate/foreco
Beetle community response to residual forest patch size in managed boreal forest landscapes: Feeding habits matter Mathieu Bouchard a,⇑, Christian Hébert b a b
Ministère de la Forêt, de la Faune et des Parcs du Québec, Direction de la Recherche Forestière, 2700 rue Einstein, Québec (QC) G1P 3W8, Canada Natural Resources Canada, Canadian Forest Service, Laurentian Forestry Centre, 1055 du PEPS, PO Box 10380, Québec (QC) G1V 4C7, Canada
a r t i c l e
i n f o
Article history: Received 24 November 2015 Received in revised form 23 February 2016 Accepted 24 February 2016 Available online 9 March 2016 Keywords: Insect diversity Forest fragmentation Boreal forests Beetles Feeding habits Clearcutting
a b s t r a c t Forest fragmentation by management activities has been implicated in the decline of forest biodiversity. Even though boreal ecosystems are generally deemed quite resilient to disturbance effects, high contemporary levels of disturbance might push forest-interior species toward decline or extinction. In this study, we examined beetle communities in forest patches of different sizes, including clearcuts, residual postharvest patches from 0.03 to 50 ha in size, and large mature forest tracts (>1000 ha). Overall, community structure follows a gradient between clearcuts and large mature forest tracts, even if patch size effects were more difficult to detect among patches >2.5 ha. Beetles were most abundant in clearcuts, and species richness was highest in small tree groups (0.03–0.05 ha). The effects of fragmentation were strongly conditioned by beetle feeding habits. Predators and xylophagous beetles were mostly associated with clearcuts or smaller patches (i.e., small tree groups or large tree groups [0.3–0.5 ha]), whereas fungivorous beetles were associated with forest-interior habitats. Although many forest-interior species were still present in relatively small patches 1–5 years after harvesting, negative effects of habitat fragmentation on these species might increase in the long-term. Ó 2016 Elsevier B.V. All rights reserved.
1. Introduction Habitat fragmentation has contributed to the decline and extinction of many plant and animal species around the world (Hanski and Gaggiotti, 2004; Ewers and Didham, 2006). However, the impacts of habitat fragmentation on species persistence at the landscape level probably vary among ecosystems and regions. For example, species living in the boreal forest can be expected to possess adaptations, such as good dispersal capabilities and tolerance for suboptimal habitats, that will allow them to persist in a context of frequent natural disturbances (Thompson et al., 2009). In Canada, contemporary levels of disturbance in boreal ecosystems are probably higher than during the preindustrial period. In particular, the widespread use of clearcutting, compounded by the ongoing effect of natural disturbances, has decreased the landscape-level abundance of mature forests below their natural range of variation (Cyr et al., 2009; Bouchard and Pothier, 2011). Consequently, the capacity of some species to maintain viable populations in isolated patches of mature forests could be exceeded. Species that are rare, have poor dispersal capabilities, or that are ⇑ Corresponding author. E-mail addresses: [email protected] (M. Bouchard), christian. [email protected] (C. Hébert). http://dx.doi.org/10.1016/j.foreco.2016.02.029 0378-1127/Ó 2016 Elsevier B.V. All rights reserved.
strongly dependent on particular microhabitats or environmental conditions (i.e., with a narrow niche breadth) could be more sensitive to habitat rarefaction or fragmentation (Henle et al., 2004; Ewers and Didham, 2006). Thus, even if Canadian boreal species are generally considered resilient to disturbances, some species with particular life histories could still be vulnerable to habitat rarefaction and fragmentation. Beetles are very well represented in forest ecosystems worldwide and have diversified ecological roles. By contrast with the larger animal species that are often examined with respect to fragmentation effects in boreal regions, such as woodland caribou (Festa-Bianchet et al., 2011) or birds (Schmiegelow and Mönkkönen, 2002), insects have smaller home ranges, but also very diversified life histories. This makes them useful indicators of small-scale processes (Fahrig et al., 2015). Some studies have found that beetles respond dramatically to forest fragmentation, particularly in temperate or tropical ecosystems (Didham et al., 1998; Davies et al., 2000; Driscoll and Weir, 2005; Barbaro and Van Halder, 2009). In boreal conditions, however, responses are more difficult to circumscribe (Gandhi et al., 2001; Webb et al., 2008; Saint-Germain et al., 2013). When these effects are present, few indications are provided regarding the threshold habitat size under which species persistence is threatened, or whether this threshold varies from one species to another (Lee et al., 2015).
64
M. Bouchard, C. Hébert / Forest Ecology and Management 368 (2016) 63–70
In this study, we used beetles to examine species and community responses to different sizes of residual forest patches in recently harvested boreal forests. Our hypotheses are (1) that the size of mature forest habitats is an important determinant of beetle community structure, and (2) that species response to residual forest patch size varies depending on life-history traits. We tested these hypotheses by sampling a gradient of patch sizes in landscapes recently affected by clearcut harvestings. 2. Methods 2.1. Study area The study area covers a region of about 5000 km2 in the eastern part of the Canadian boreal forest (Fig. 1), in which the natural fire cycle is ca 250 years. The forest matrix is overwhelmingly dominated by two tree species: black spruce (Picea mariana [Mill]. B.S. P.) and balsam fir (Abies balsamea [L.] Mill.) (Bouchard et al., 2008; Bélisle et al., 2011). The region has been affected by forest harvestings since the 1970s, but large tracts of mature forests of natural origin are still present throughout the study area. In 2012 and 2013, we sampled 5 landscapes in which aggregated clearcuts had been performed recently (2007–2011, Fig. 1). Each of these landscapes represents an area of 30–60 km2 dominated by clearcuts, inside which patches of residual forests were retained, varying in size from a few m2 to ca 50 ha. The guidelines used to delineate harvest areas on public lands in the province of Quebec specify that forest composition and characteristics of residual forest patches must be representative of the overall forest characteristics in the surrounding area before harvesting. 2.2. Sampling In each landscape, we sampled 6 types of habitats defined by the size of residual forest patches inside the matrix: clearcuts (0 ha, no residual forest patch), small tree groups (0.03–0.05 ha), large tree groups (0.3–0.5 ha), small blocks (2.5–4 ha), large blocks (50 ha) and large mature forest tracts (>1000 ha). The large mature forest tracts were located outside, but adjacent to, the aggregated clearcuts. One sample plot was established in each habitat, for a total of 30 plots (5 landscapes 6 habitats or residual forest patch sizes). Plots located in clearcuts were established more than 50 m from the nearest forest edge. For small tree groups, large tree groups and small blocks, we selected forest patches that were more or less circular in shape, and placed the plot at their center. For
large blocks and large mature forest tracts, the plots were established at least 200 m from the nearest clearcut. Each sample plot contained 2 flight intercept traps (FIT) and 3 pitfall traps. The FITs consisted in 4 panels (15 40 cm each, 2 made of Plexiglas and 2 of mosquito net) mounted in a cross pattern along a 10-cm diameter black ABS cylinder, with a funnel located below the cylinder leading to a collecting vial. The 2 FITs were suspended, their base about 1 m above ground level, and placed 10 m apart in each sample plot. The pitfalls were MultipherÒ traps (12.5 cm in diameter) placed flush to the ground and 10 m apart in each sample plot. For the small tree groups, which were too small to contain all the traps, we paired 2 plots located approximately 50 m apart, placing 1 FIT and 1 pitfall in the first plot, and 1 FIT and 2 pitfalls in the second. The traps were operated from May 28 to August 17 in 2012, and from May 27 to August 26 in 2013. Propylene glycol was used as a killing and preserving liquid, and the traps were emptied at 2- to 3-week intervals. All captured beetles were identified at the species level, except for the genera Epurea (Nitidulidae), Cryptophagus (Cryptophagidae) and specimens belonging to the family Ptiliidae, as well as some Agathidium (Leiodidae) and Corticaria (Latridiidae) females. We estimated basal area of living trees and snags (diameter at breast height >9.1 cm) using a prism (factor k = 2) and tallied the selected trees by species. We measured coarse woody debris (>8 cm diameter) present on the ground with the line intercept method (Van Wagner, 1968) after establishing 3 line transects (10 m each) at 120° intervals around the center of each plot. Along each transect, all coarse woody debris intersected were tallied. For strongly decayed pieces of coarse woody debris, we measured diameter in 2 perpendicular directions and calculated the geometric mean. The proportion of ground covered by Sphagnum mosses, which is a good indicator of habitat suitability for some beetles in boreal forests (Paquin, 2008), was also estimated in 4 microplots (25 m2 each) placed 5 m from the center of the plot in each cardinal direction. Certain preharvest forest characteristics can persist in postdisturbance communities. For example, some residual mature-forest beetle populations could subsist for a few years after logging. In this study, we verified if preharvest forest composition had a residual effect on beetle communities in clearcuts by calculating preharvest basal area in all clearcut stands, from an inventory of black spruce and balsam fir stumps within a 10-m radius circular plot. Together, black spruce and balsam fir represented >95% of overall plot basal area. The proportion of balsam fir was used to express variations in dominant tree species composition.
Fig. 1. Map illustrating the location of the 5 studied landscapes (left), and an example of plot spatial distribution in one landscape (right).
M. Bouchard, C. Hébert / Forest Ecology and Management 368 (2016) 63–70
Preharvest and postharvest proportions of balsam fir were thus considered as distinct variables in the analyses. 2.3. Statistical analyses In the analyses, the size of residual forest patches was converted to an ordinal vector ranging from clearcuts (residual forest patch size = 1) to large mature forest tracts (residual forest patch size = 6). Beetles were also classified according to their feeding habits and microsite preferences on the basis of the literature and expert opinions (Table 1). We examined the effect of residual forest patch size on beetle abundance and species richness using mixed model analyses. The abundance in each plot of individuals from each group served as an aggregated trait value (Shipley et al., 2006). The analyses were performed with the glmer function of the lme4 package (Bates et al., 2015) in R, using landscape as a random factor. Because we used count data, a Poisson error distribution was used in the models. For each statistical model, we evaluated 95% confidence intervals with bootstraps analyses, considering the explanatory variables for which the interval did not include 0 as significant. We applied the same procedure to examine the effect of residual forest patch size on species richness for each group. We used redundancy analysis (RDA) to determine the importance of residual forest patch size as a driver of the overall beetle
65
community structure, relative to the other environmental factors. Beetle data for the 2 trap types and 2 sample years were pooled for each plot, then Hellinger-transformed to reduce the importance of rare species (Borcard et al., 2011). Snags and coarse woody debris lying on the ground were modeled as distinct variables because these microhabitats are generally colonized by different species (Jonsell and Weslien, 2003). Plot elevation was used as a proxy for climatic variability within the study area. All environmental variables were standardized before analysis (Borcard et al., 2011). Redundancy analysis was performed with the rda function in the vegan package (Oksanen et al., 2013) in R version 3.02 (R Development Core Team, 2013). Relationships between individual species and residual forest patch size were tested using the multipatt function of the Indicspecies package (De Caceres and Jansen, 2015) in R. In this approach, an index that combines a species’ frequency and abundance in a target group of samples relative to its overall frequency and abundance is compared to a randomized distribution. Species that are more frequent and abundant (i.e., concentrated) in a particular group of samples (e.g., residual forest patches of a particular size) than expected by chance were considered to be indicator species for that environment. In this analysis, residual forest patch size was thus considered as a categorical predictive variable. We further restricted the analysis to beetle–patch size relationships that could be interpreted from an ecological viewpoint, using the
Table 1 Categories of life-history traits used for individual beetle species. Trait
Category
Description
Microsite preference
Litter Dead tree Living plant
Found on dead foliage or highly decomposed wood Found on dead trees standing (snags) or lying on the ground (CWD) Found on living trees or understory plants
Feeding habits
Phytophage Xylophage Predator Fungivore Detritivore
Includes insects that feed on living plant tissues such as foliage, roots and shoots during the larval stage Feed primarily on xylem and phloem Feed primarily on grubs, gastropods, collembolla, aphids, etc. Feed on carpophores, mycelium, spores or yeasts Feed on dead organic matter, except wood: leaves, dung, carcasses, dead insects, etc. Some insects feeding on carcasses were also classified as detritivores.
Fig. 2. Abundance and richness of beetle groups according to residual forest patch size. R1 = clearcuts, R2 = small tree groups (0.03–0.05 ha), R3 = large tree groups (0.3–0.5 ha), R4 = small blocks (2.5–5 ha), R5 = large blocks (50 ha), R6 = large mature forest tracts (>1000 ha). Presence of a filled arrow beside the absolute abundance or absolute species richness for a beetle species group indicates a significant response to residual forest patch size (see Table 2 for complete results).
66
M. Bouchard, C. Hébert / Forest Ecology and Management 368 (2016) 63–70
restcomb argument in the multipatt function (De Caceres and Jansen, 2015). To assess the significance of each species, we performed a permutation test (n = 4999). Only species with more than 10 individuals were considered in these analyses. For each species, sample plots where catches represented less than 1% of the overall species abundance were attributed an abundance of 0, in order to decrease the importance of vagrant individuals (Dufrêne and Legendre, 1997). Species showing significant associations with large forest tracts, patches or clearcuts were designated as forestinterior, patch and open-habitat species, respectively.
not only in large mature forest tracts, but also in smaller forest patches (Table 3). Among this group, the most sensitive to residual forest patch size was Zenodosus sanguineus, which was found only in large blocks and large mature forest tracts. Patch species were specifically associated with intermediate-sized residual forest patches (small tree groups, large tree groups, and small blocks). Finally, open-habitat were mostly confined to clearcuts and small tree groups, with one species, Syneta pilosa, being also found in large tree groups (Table 3). Among the combinations that were tested with Indicspecies which could have a biological interpretation, only one category contained no indicator species: patches from 0 to 4 ha (R1 + R2 + R3 + R4, Table 3).
3. Results A total of 8274 beetles were caught, belonging to 420 taxa. Overall, beetles were more abundant in clearcuts than in any other size of residual forest patch (Fig. 2). The size of residual forest patches significantly affected the abundance and richness of the different groups of beetles (Table 2). In larger patches, fungivores were more abundant, but abundance and species richness were lower for xylophages, detritivores and predators (Table 2). As a result, fungivores represented a much larger proportion of the overall beetle community in larger residual forest patches (>0.3 ha) than in smaller patches and clearcuts (<0.05 ha) (Fig. 2). Clearcuts and small tree groups were mostly dominated by predators, and to a smaller extent, by xylophages and phytophages. Regarding microsite preferences, residual forest patch size had a negative effect on the abundance of beetles associated with litter and dead trees, and a positive effect on the abundance of beetles associated with living plants, but a negative effect on species richness for this same group (Table 2). The RDA indicates that the selected environmental factors explained 40% of the overall community variation, including 22% along the first axis and 7% along the second axis (Fig. 3). Axis 1 (pseudo-F = 8.4, p < 0.001) and axis 2 (pseudo-F = 2.5, p = 0.002) are both significant, according to permutation tests (Borcard et al., 2011). The most important environmental factors identified in the RDA were residual forest patch size along the first axis, and elevation along the second axis (Fig. 3). Along the second axis, variability is much higher for the larger patches (>2 ha) than for the smaller ones (Fig. 3). The indicator species analysis showed that 40 species were significantly associated with certain residual forest patch sizes (Table 3). These species can be categorized into 3 groups: forestinterior species (11 species), patch species (6 species), and openhabitat species (23 species). Forest-interior species were present
4. Discussion 4.1. Feeding habits and habitat preferences as determinants of beetle response to residual forest patch size Our hypotheses that beetle community structure is influenced by patch size, and that species traits could be an important determinant to understand beetle response to residual forest patch size, were both supported by the results. Among the 5 trophic groups, fungivores were the only beetles that were collectively more abundant in larger patches, especially those larger than 2.5 ha. Fungivores represented 65% of the beetles captured in large tracts of mature forest, but only 33 and 18% in small tree groups and in clearcuts, respectively. Closed-crown forests are typically cooler and less exposed to droughts, and are probably more favorable to the presence of fungi (Durall et al., 1999) and to some groups of Coleoptera associated with them (Sippola and Renvall, 1999; Jacobs and Work, 2012; Lee et al., 2015). Moreover, although most fungivores can probably feed on or colonize different types of fungi, some are more closely associated with fungi primarily found on living or dead standing trees (Jonsell and Weslien, 2003), a substrate that is absent from clearcuts. For example, the most abundant fungivorous species captured in this study, Glischrochilus sanguinolentus, is apparently associated with mycelium growing on sap found on the trunk of injured (but live) standing trees (Jelínek et al., 2010); the abundance of this species increases progressively from clearcuts to patches >50 ha (Supplementary Material). Xylophagous beetles showed an opposite response to that of fungivorous species, with increasing abundance when size of residual patches decreases. Previous studies have also reported such a contrast between these two groups (Boulanger et al., 2010; Lee
Table 2 Summary of the GLM analyses of beetle response to residual forest patch size, according to their feeding habits and microsite preferences. Trait
Preferences
Response variable
Feeding habits
Xylophage Xylophage Fungivore Fungivore Predator Predator Phytophage Phytophage Detritivore Detritivore
Abundance Richness Abundance Richness Abundance Richness Abundance Richness Abundance Richness
Estimate 0.227 0.068 0.127 0.012 0.586 0.046 0.184 0.208 0.206 0.088
0.020 0.033 0.010 0.020 0.012 0.025 0.097 0.130 0.026 0.052
SE
11.3 2.1 13.4 0.6 48.7 1.9 1.9 1.6 7.9 1.7
( 0.27, 0.19) ( 0.13, 0) (0.11, 0.15) ( 0.03, 0.05) ( 0.61, 0.56) ( 0.09, 0) ( 0.38, 0) ( 0.47, 0.04) ( 0.26, 0.16) ( 0.19, 0.01)
Microsite preferences
Litter Litter Dead tree Dead tree Living plant Living plant
Abundance Richness Abundance Richness Abundance Richness
0.446 0.027 0.031 0.007 0.149 0.086
0.010 0.021 0.010 0.020 0.015 0.035
44.8 1.3 3.0 0.4 10.3 2.4
( 0.47, 0.43) ( 0.07, 0.01) ( 0.05, 0.01) ( 0.05, 0.03) (0.12, 0.18) ( 0.16, 0.02)
Note: Residual forest patch size was considered as a numerical vector ranging from 1 (clearcuts) to 6 (large mature forest tracts).
z-value
CI (95%)
M. Bouchard, C. Hébert / Forest Ecology and Management 368 (2016) 63–70
67
Fig. 3. Ordination of beetle communities by redundancy analysis (RDA). Environmental factors (arrows) and sites (symbols) are shown, but individual species are not shown. PreFir = Preharvest proportion of balsam fir, PostFir = Postharvest proportion of balsam fir, CWD = Coarse woody debris volume. Position of centroids is indicated with a larger symbol for each patch size category.
et al., 2015), which could be explained in part by different responses to environmental cues. Xylophagous taxa such as Curculionidae and Cerambycidae tend to be attracted to recently killed trees, particularly when the latter emit volatiles and are sunexposed. Despite the scarcity of large dead trees in clearcuts, these areas remain covered by logging residues and freshly cut stumps, which are likely to attract some species and might represent a suitable breeding substrate (Hedgren, 2007). Predator’s response followed that of xylophagous beetles. However, the greater abundance of predators in very small patches and clearcuts contrasts with the findings of Davies et al. (2000), who observed a lower abundance of predators in small patches than in larger ones. Their study took place in tropical forest ecosystems fragmented by agricultural lands, which is very different from the present study. In boreal environments, the greater abundance of predators such as ground beetles (Niemelä et al., 1993; Heliölä et al., 2001) or spiders (Buddle et al., 2000; Larrivée et al., 2005) in clearcut environments was well documented. This could be explained by the warmer sun-exposed environment, which promotes a greater biological activity in the duff layer (Chauvat et al., 2003; Malmström et al., 2009; Siira-Pietikäinen and Haimi, 2009), and results in greater abundance of decomposing microorganisms that serve as preys. Penetration and diffusion of clearcut species inside forest patches also probably explains the gradual increase in predator abundance as one moves along the gradient from large forest tracts to clearcuts. Overall, knowledge of life-history strategies was useful to understand and explain changes in community structure. Still, in the case of beetles, trait-based approaches should be considered with caution. Many species, even the most abundant ones, are still poorly described in North-America. For example, Pseudanostirus triundulatus (Coleoptera: Elateridae), the most abundant species overall, was considered in this study to be primarily litter-dwelling, based on assumed feeding habits of the larval stages and field observations. However, this species was classified differently in previous studies (Boulanger et al., 2010), indicating lingering uncertainties with respect to its biology and, by extension, to the biology of elaterids in general (Thomas et al., 2009; Saint-Germain et al., 2013). The fact that ecosystem-level studies are very dependent on high-quality, species-level descriptive work is particularly obvious for taxonomic groups such as beetles, which are extremely diversified in terms of both species richness and life-history strategies.
4.2. Additional drivers of beetle community response to residual forest patch size Even if part of the community-level response could be explained by life-history strategies, there is considerable variability among beetles that share similar traits. For example, forestinterior species include many fungivorous beetles, but also species from other feeding groups (Table 3). This suggests that beetle response to residual forest patch size is also explained by lifehistory traits that were not considered directly in this study. Species tolerance for suboptimal conditions (i.e., niche breadth) could be an important factor to explain their response to habitat fragmentation (Swihart et al., 2006). For example, a narrow niche breadth could mean that a xylophagous species can use a piece of coarse woody debris if it is located in a shaded microenvironment (inside a forest patch), but not if the same piece is located in a clearcut or a forest edge (Gibb et al., 2006; Morin et al., 2015). Similarly, some species generally associated with forest-interior conditions could have behavioral, life cycle or ecophysiological adaptations that allow them to persist for long periods in suboptimal conditions (Niemelä et al., 1993). Populations of insect species narrowly associated with mature forest environments are less likely to be in direct contact with neighboring populations when surrounded by suboptimal habitats, and are thus more prone to stochastic extinction events (Ewers and Didham, 2006; Laurance et al., 2011; Slade et al., 2013). In the context of this study, fragmentation effects are thus expected to be more important for forest-interior species that are intolerant of clearcut environments. Differences in dispersal capability between species is another important mechanism that could explain the variation in fragmentation effects. Even if they cannot breed in suboptimal habitats, many species could be able to disperse through it. In fact, in the case of flying insects, recent clearcuts could facilitate dispersal, as these areas are devoid of dense understory layers that impede their flight. Individuals immigrating from the surrounding forests and dispersing though the matrix of clearcuts could thus colonize patches where the species is absent, or replenish local populations. Still, even if species with good dispersal capabilities are expected to be more likely to maintain viable populations in fragmented landscapes (Thomas et al., 2001; Tscharntke et al., 2002), dispersal remains a complex mechanism that can interact with other
68
M. Bouchard, C. Hébert / Forest Ecology and Management 368 (2016) 63–70
Table 3 Results of the indicator species analysis which allowed the categorization species as ‘‘clearcut”, ‘‘patch” and ‘‘forest-interior” specialists. Species
Family
Microsite
Feeding
Patch size
Indicspecies statistics
R1
R2
R3
R4
R5
R6
Ind.
Stat.
p
Open-habitat species
Pediacus fuscus Agriotes limosus Ampedus quebecensis Paraphotistus nitidulus Pseudanostirus triundulatus Corticarina cavicollis Mordellaria borealis Sphaeriestes virescens Heterothops fusculus Mycetoporus.consors Nudobius cephalus Acmaeops p.proteus Rhagium inquisitor Scolytus piceae Ampedus pullus Sericus incongruus Ellychnia corrusca Xylita laevigata Colopterus truncatus Gyrophaena meduxnekeagensis Pselaphus bellax Syntomium grahami Syneta pilosa
Cucujidae Elateridae Elateridae Elateridae Elateridae Latridiidae Mordellidae Salpingidae Staphylinidae Staphylinidae Staphylinidae Cerambycidae Cerambycidae Curculionidae Elateridae Elateridae Lampyridae Melandryidae Nitidulidae Staphylinidae Staphylinidae Staphylinidae Chrysomelidae
Dead tree Living plant Dead tree Litter Litter Dead tree Dead tree Dead tree Litter Litter Litter Dead tree Dead tree Living plant Dead tree Living plant Dead tree Dead tree Living plant Dead tree Litter Litter Living plant
Predator Xylophage Predator Predator Predator Fungivore Fungivore Predator Predator Fungivore Predator Xylophage Xylophage Xylophage Predator Xylophage Predator Xylophage Fungivore Fungivore Predator Detritivore Phytophage
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
1 1 1 1 1 1 1 1 1 1 1 3 3 3 3 3 3 3 3 3 3 3 6
0.816 0.749 0.735 0.739 0.913 0.726 0.749 0.739 0.600 0.655 0.655 0.632 0.707 0.721 0.687 0.807 0.822 0.961 0.632 0.707 0.938 0.717 0.759
0.002 0.006 0.007 0.009 0.000 0.019 0.040 0.012 0.045 0.017 0.029 0.033 0.007 0.012 0.019 0.004 0.003 0.000 0.038 0.005 0.000 0.014 0.036
Patch species
Trechus crassiscapus Pityokteines sparsus Abstrulia canadensis Trypodendron lineatum Pseudanostirus hoppingi Cyphon obscurus
Carabidae Curculionidae Tetratomidae Curculionidae Elateridae Scirtidae
Litter Living plant Dead tree Living plant Litter Litter
Predator Xylophage Fungivore Xylophage Predator Detritivore
0 0 0 0 0 0
1 1 1 1 1 1
0 1 1 1 1 1
0 0 0 1 1 1
0 0 0 0 0 0
0 0 0 0 0 0
2 4 4 7 7 7
0.707 0.707 0.748 0.733 0.786 0.810
0.014 0.007 0.005 0.023 0.005 0.006
Forest-interior species
Pterostichus punctatissimus Cryptophagus spp. Corticaria n.linearis Enicmus tenuicornis Dictyopterus aurora Emmesa connectens Epuraea spp. Glischrochilus s.sanguinolentus Henoticus serratus Olophrum rotundicolle Zenodosus sanauineus
Carabidae Cryptophagidae Latridiidae Latridiidae Lycidae Melandryidae Nitidulidae Nitidulidae Cryptophagidae Staphylinidae Cleridae
Litter Dead tree Dead tree Dead tree Dead tree Dead tree Dead tree Living plant Dead tree Litter Dead tree
Predator Fungivore Fungivore Fungivore Fungivore Xylophage Fungivore Fungivore Fungivore Unknown Predator
0 0 0 0 0 0 0 0 0 0 0
1 1 1 1 1 1 1 1 0 0 0
1 1 1 1 1 1 1 1 1 0 0
1 1 1 1 1 1 1 1 1 1 0
1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1
10 10 10 10 10 10 10 10 9 8 5
0.938 0.908 0.903 0.931 0.830 0.828 0.923 0.917 0.815 0.776 0.701
0.004 0.005 0.009 0.002 0.036 0.035 0.005 0.002 0.005 0.017 0.019
Note: Residual forest patch sizes were the following: R1 = clearcuts (0 ha), R2 = small tree groups (0.03–0.05 ha), R3 = large tree groups (0.3–0.5 ha), R4 = small blocks (2.5– 4 ha), R5 = large blocks (50 ha), R6 = large mature forest tracts (>1000 ha). The following treatment combinations were tested with Indicspecies: ‘‘R1”, ‘‘R2”, ‘‘R1 + R2”, ‘‘R1 + R2 + R3”, ‘‘R1 + R2 + R3 + R4”, ‘‘R2 + R3”, ‘‘R2 + R3 + R4”, ‘‘R2 + R3 + R4 + R5 + R6”, ‘‘R3 + R4 + R5 + R6”, ‘‘R4 + R5 + R6”, ‘‘R5 + R6”.
life-history traits and make species distribution patterns difficult to predict (Bonte et al., 2012). Finally, competitive or trophic interactions could also explain some fragmentation effects (Fagan et al., 1999). For example, an increased diffusion of open-habitat predators in smaller forest patches (Fig. 2) could put additional pressure on some forestinterior species that would serve as prey. Habitat fragmentation could also modify competitive interactions, as the influx of openhabitat species that share similar traits (e.g. feeding habits) could negatively affect other species with poor competitive abilities. In the present study, we do not have sufficient information to separate the potential effects of the above-mentioned processes. Gaining a better understanding of these processes will probably be crucial in the future in order to design sustainable forest practices. Three lines of action could help in this respect. First, longterm monitoring and replication across regions would be very helpful to better assess population trends and further circumscribe potentially sensitive taxa. Second, better descriptions of sensitive species could help assess their niche breadth (trait variance) and potential sensitivity to environmental change. Third, manipulative studies with critical habitats such as coarse woody debris (Gibb et al., 2006) could help isolate the effects of the different processes
involved in fragmentation effects (niche breadth, dispersal, interspecific relationships). 5. Conclusions Our study confirms that a few years after harvesting, the negative effects of forest fragmentation on forest-interior species are mostly limited to smaller patches (0.03–0.5 ha). Beetle communities found in patches of at least 2.5 ha appeared similar to those found in larger patches. At the community level, we propose that the dominant pattern in small patches comes from an influx of open-habitat species, mostly predators, and the depletion of certain forest-interior species, mostly fungivores. This pattern seems similar to the progressive loss of forest-interior beetles and gain of open-habitat species that was observed in previous studies along gradients of partial cut intensities (Stenbacka et al., 2010; Légaré et al., 2011). The basic trait-based approach used in this study was helpful, and confirms that subdividing the vast, relatively undifferentiated group of Coleoptera according to feeding habits and microsite preferences helps achieve a better understanding of the main processes that drive fragmentation effects (Driscoll and Weir, 2005).
M. Bouchard, C. Hébert / Forest Ecology and Management 368 (2016) 63–70
Negative effects on mature forest species in harvested landscapes are expected to increase with time: for example, potential habitats such as residual stumps and slash in recently clearcut areas will gradually decompose and disappear. Dispersal limitations could also become an issue as the vegetation grows back in clearcut areas. Moreover, in boreal Canada, where old unmanaged stands are still present in the larger matrix, source habitats for many forest-interior species are expected to become less abundant in the future, as forest activities gradually expand toward the north (e.g. Bouchard and Pothier, 2011). This context could reduce the amount of colonizing species reaching residual forest patches, and further reduce the persistence of forest-interior insect populations at the landscape level (Kouki et al., 2012). Acknowledgements We thank S. Lauzon, C. Paquet, F. Berger and Y. Dubuc for their help with fieldwork, N. Robert and M.-C. Émond for their help managing and sorting the samples, A. Hibbert, J. Jacobs, C. Chantal, G. Pelletier and J. Klimaszewski for species identification, and D. Tousignant for editing the English on the manuscript. This study was financed by the Quebec Ministry of Forests, Fauna and Parks and by the Canadian Forest Service. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.foreco.2016.02. 029. References Barbaro, L., Van Halder, I., 2009. Linking bird, carabid beetle and butterfly lifehistory traits to habitat fragmentation in mosaic landscapes. Ecography 32, 321–333. Bates, D., Maechler, M., Bolker, B., Walker, S., 2015. lme4: Linear Mixed-Effects Models Using Eigen and S4. R Package Version 1.1-7.
69
Hemlock forests of northwestern British Columbia: sporocarps and ectomycorrhizae. Can. J. For. Res. 29, 1322–1332. Ewers, R.M., Didham, R.K., 2006. Confounding factors in the detection of species responses to habitat fragmentation. Biol. Rev. 81, 117–142. Fagan, W.F., Cantrell, R.S., Cosner, C., 1999. How habitat edges change species interactions. Am. Nat. 153, 165–182. Fahrig, L., Girard, J., Duro, D., Pasher, J., Smith, A., Javorek, S., King, D., Lindsay, K.F., Mitchell, S., Tischendorf, L., 2015. Farmlands with smaller crop fields have higher within-field biodiversity. Agric. Ecosyst. Environ. 200, 219–234. Festa-Bianchet, M., Ray, J., Boutin, S., Côté, S., Gunn, A., 2011. Conservation of caribou (Rangifer tarandus) in Canada: an uncertain future. Can. J. Zool. 89, 419–434. Gandhi, K.J., Spence, J.R., Langor, D.W., Morgantini, L.E., 2001. Fire residuals as habitat reserves for epigaeic beetles (Coleoptera: Carabidae and Staphylinidae). Biol. Conserv. 102, 131–141. Gibb, H., Pettersson, R.B., Hjältén, J., Hilszczan´ski, J., Ball, J.P., Johansson, T., Atlegrim, O., Danell, K., 2006. Conservation-oriented forestry and early successional saproxylic beetles: responses of functional groups to manipulated dead wood substrates. Biol. Conserv. 129, 437–450. Hanski, I.A., Gaggiotti, O.E., 2004. Ecology, Genetics and Evolution of Metapopulations. Academic Press. Hedgren, P.O., 2007. Early arriving saproxylic beetles (Coleoptera) and parasitoids (Hymenoptera) in low and high stumps of Norway spruce. For. Ecol. Manage. 241, 155–161. Heliölä, J., Koivula, M., Niemelä, J., 2001. Distribution of carabid beetles (Coleoptera, Carabidae) across a boreal forest–clearcut ecotone. Conserv. Biol. 15, 370–377. Henle, K., Davies, K.F., Kleyer, M., Margules, C., Settele, J., 2004. Predictors of species sensitivity to fragmentation. Biodivers. Conserv. 13, 207–251. Jacobs, J.M., Work, T.T., 2012. Linking deadwood-associated beetles and fungi with wood decomposition rates in managed black spruce forests. Can. J. For. Res. 42, 1477–1490. Jelínek, J., Carlton, C., Cline, A., Leschen, R., 2010. Nitidulidae Latreille, 1802. In: Leschen, R., Beutel, R., Lawrence, J. (Eds.), Handbook of Zoology. De Gruyter, Berlin, pp. 390–407. Jonsell, M., Weslien, J., 2003. Felled or standing retained wood – it makes a difference for saproxylic beetles. For. Ecol. Manage. 175, 425–435. Kouki, J., Hyvärinen, E., Lappalainen, H., Martikainen, P., Similä, M., 2012. Landscape context affects the success of habitat restoration: large-scale colonization patterns of saproxylic and fire-associated species in boreal forests. Divers. Distrib. 18, 348–355. Larrivée, M., Fahrig, L., Drapeau, P., 2005. Effects of a recent wildfire and clearcuts on ground-dwelling boreal forest spider assemblages. Can. J. For. Res. 35, 2575– 2588. Laurance, W.F., Camargo, J.L., Luizão, R.C., Laurance, S.G., Pimm, S.L., Bruna, E.M., Stouffer, P.C., Williamson, G.B., Benítez-Malvido, J., Vasconcelos, H.L., 2011. The fate of Amazonian forest fragments: a 32-year investigation. Biol. Conserv. 144, 56–67. Lee, S.-I., Spence, J.R., Langor, D.W., Pinzon, J., 2015. Retention patch size and conservation of saproxylic beetles in boreal white spruce stands. For. Ecol. Manage. 358, 98–107. Légaré, J.-P., Hébert, C., Ruel, J.-C., 2011. Alternative silvicultural practices in irregular boreal forests: response of beetle assemblages. Silva Fennica 45, 937– 956. Malmström, A., Persson, T., Ahlström, K., Gongalsky, K.B., Bengtsson, J., 2009. Dynamics of soil meso-and macrofauna during a 5-year period after clear-cut burning in a boreal forest. Appl. Soil Ecol. 43, 61–74. Morin, M., Hébert, C., Berthiaume, R., Bauce, É., Brodeur, J., 2015. Short-term effect of selection cutting in boreal balsam fir forest on cerambycid and scolytid beetles. J. Appl. Entomol. 139, 553–566. Niemelä, J., Langor, D., Spence, J.R., 1993. Effects of clear-cut harvesting on boreal ground-beetle assemblages (Coleoptera: Carabidae) in Western Canada. Conserv. Biol. 7, 551–561. Oksanen, J., Blanchet, F.G., Kindt, R., Legendre, P., Minchin, P.R., O’Hara, R., Simpson, G.L., Solymos, P., Stevens, M., Wagner, H., 2013. Package ‘vegan’. R Packag ver 254, 20–28. Paquin, P., 2008. Carabid beetle (Coleoptera: Carabidae) diversity in the black spruce succession of eastern Canada. Biol. Conserv. 141, 261–275. R Development Core Team, 2013. In: R: A Language and Environment for Statistical Computing. R Foundation For Statistical Computing, Vienna, Austria. Saint-Germain, M., Drapeau, P., Hibbert, A., 2013. Saproxylic beetle tolerance to habitat fragmentation induced by salvage logging in a boreal mixed-cover burn. Insect Conserv. Divers. 6, 381–392. Schmiegelow, F.K., Mönkkönen, M., 2002. Habitat loss and fragmentation in dynamic landscapes: avian perspectives from the boreal forest. Ecol. Appl. 12, 375–389. Shipley, B., Vile, D., Garnier, É., 2006. From plant traits to plant communities: a statistical mechanistic approach to biodiversity. Science 314, 812–814. Siira-Pietikäinen, A., Haimi, J., 2009. Changes in soil fauna 10 years after forest harvestings: comparison between clear felling and green-tree retention methods. For. Ecol. Manage. 258, 332–338. Sippola, A.-L., Renvall, P., 1999. Wood-decomposing fungi and seed-tree cutting: a 40-year perspective. For. Ecol. Manage. 115, 183–201. Slade, E.M., Merckx, T., Riutta, T., Bebber, D.P., Redhead, D., Riordan, P., Macdonald, D.W., 2013. Life-history traits and landscape characteristics predict macromoth responses to forest fragmentation. Ecology 94, 1519–1530.
70
M. Bouchard, C. Hébert / Forest Ecology and Management 368 (2016) 63–70
Stenbacka, F., Hjältén, J., Hilszczanski, J., Dynesius, M., 2010. Saproxylic and nonsaproxylic beetle assemblages in boreal spruce forests of different age and forestry intensity. Ecol. Appl. 20, 2310–2321. Swihart, R.K., Lusk, J.J., Duchamp, J.E., Rizkalla, C.E., Moore, J.E., 2006. The roles of landscape context, niche breadth, and range boundaries in predicting species responses to habitat alteration. Divers. Distrib. 12, 277–287. Thomas, J., Bourn, N., Clarke, R., Stewart, K., Simcox, D., Pearman, G., Curtis, R., Goodger, B., 2001. The quality and isolation of habitat patches both determine where butterflies persist in fragmented landscapes. Proc. Royal Soc. London B: Biol. Sci. 268, 1791–1796. Thomas, S.L., Wagner, R.G., Halteman, W.A., 2009. Influence of harvest gaps and coarse woody material on click beetles (Coleoptera: Elateridae) in Maine’s Acadian forest. Biodivers. Conserv. 18, 2405–2419.
Thompson, I., Mackey, B., McNulty, S., Mosseler, A., 2009. Forest resilience biodiversity and climate change. In: A Synthesis of the Biodiversity/ Resilience/Stability Relationship in Forest Ecosystems Secretariat of the Convention on Biological Diversity. Technical Series, Montreal, pp. 1–67. Tscharntke, T., Steffan-Dewenter, I., Kruess, A., Thies, C., 2002. Characteristics of insect populations on habitat fragments: a mini review. Ecol. Res. 17, 229– 239. Van Wagner, C., 1968. The line intersect method in forest fuel sampling. Forest Sci. 14, 20–26. Webb, A., Buddle, C.M., Drapeau, P., Saint-Germain, M., 2008. Use of remnant boreal forest habitats by saproxylic beetle assemblages in even-aged managed landscapes. Biol. Conserv. 141, 815–826.