SEARCH

SEARCH BY CITATION

Keywords:

  • bats;
  • Chiroptera;
  • conservation;
  • forest management;
  • LiDAR;
  • forest structure;
  • three-dimensional structure

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information

1. Forest management determines to a large degree the three-dimensional arrangement of the vegetation in production forest systems and hence has an essential influence on habitat quality for wildlife. We investigated the effects of forest structure on occurrence, activity and species composition of European bats, an ecologically important group of vertebrates known to be affected by the physical clutter of vegetation.

2. Species composition and activity of bats were assessed with acoustic monitoring on 50 one-hectare experimental plots in a biosphere reserve in Germany. Three-dimensional forest structure was assessed by Light Detection and Ranging (LiDAR), and a set of 20 mathematically derived and fine-grained structural parameters with a minimum of collinearity was used for a quantitative description of the vegetation structure.

3. Occurrence and activity of bats were positively associated with the structural parameters canopy height, standard deviation of the canopy surface roughness and edge fraction, indicating older forest stands with patches of different vegetation heights. In addition, species composition in differently managed forest stands was significantly influenced by the relative proportion of structural parameters. Species of one functional group, sharing similar adaptations in wing morphology and foraging strategy, showed similar associations with three-dimensional structural parameters. In addition, we found species-specific structural parameter associations explaining the occurrence and activity levels of individual species in differently managed production forest types.

4.Synthesis and applications. High-resolution LiDAR data are an important tool to assess structural habitat suitability for bat species. Our data revealed that bat occurrence and activity increases with structural heterogeneity in managed forest stands. Given, that bats provide an essential ecosystem service through top-down control of herbivorous insects, increasing stand structural heterogeneity through management practices (e.g. selective harvesting) is a very effective strategy to assure vital ecosystem functioning in production forest systems.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information

Natural forest cover has been largely replaced by fragments of production forest systems (FAO 2011) in the cultural landscapes of Central Europe. Within such forest systems, the composition of tree species combined with different silvicultural practices leads to considerable differences in the three-dimensional arrangement and complexity of the vegetation layers (Lindenmayer, Margules & Botkin 2000; Parker et al. 2004; McElhinny et al. 2005). Understanding the effects of vegetation structure in differently managed forests on patterns of species richness and relative abundance of animals is essential to develop sustainable conservation and management strategies within modified forest landscapes (Chazdon et al. 2009; Gardner et al. 2009).

Differences in the heterogeneity of forest canopy structures are among the major drivers of diversity of understorey plants (Getzin, Wiegand & Schöning 2012), and species richness and abundance of many animal taxa (MacArthur & MacArthur 1961; Bradbury et al. 2005), including invertebrates (Müller & Brandl 2009) and vertebrates (Clawges et al. 2008; Müller et al. 2009). The structural composition of forest stands largely determines habitat quality for animals (Bradbury et al. 2005; Vierling et al. 2008) because it influences the availability and accessibility of resources such as roost sites and food (Hayes & Loeb 2007), exposure to predators (Baxter et al. 2006) and microclimatic conditions (Chen et al. 1999). Forest structure also profoundly affects animal movement (Caras & Korine 2009), which is especially important for flying animals such as birds and bats that must navigate and forage within the three-dimensional arrangements of forests and their canopies. Both birds and bats provide essential ecosystem services by controlling many leaf-eating insects in tropical (Kalka, Smith & Kalko 2008; Van Bael et al. 2008; Williams-Guillén, Perfecto & & Vandermeer 2008) and temperate (Lowman & Wittman 1996; Cleveland et al. 2006; Böhm, Wells & Kalko 2011; Boyles et al. 2011) forest and agricultural systems, and their persistence is thus of imminent interest to sustainable landscape management (e.g. Kunz et al. 2011).

While the response of birds to the three-dimensional structure of differently managed forests stands has been well studied (e.g. Bradbury et al. 2005; Müller, Stadler & Brandl 2010), knowledge of the response of bats to forest structure is based mostly on categorical variables (e.g. forest type, age classes) or estimates of forest structure (e.g. percentage canopy cover) and assessed by ground surveys (e.g. Crampton & Barclay 1998; Kalcounis-Rüppell et al. 1999; Patriquin & Barclay 2003; Davy, Russo & Fenton 2007; Russo et al. 2010). However, very likely bats are affected by differences in the three-dimensional arrangement of differently managed forests owing to the physical clutter caused by the vegetation (Neuweiler 1989; Schnitzler & Kalko 2001), which poses a variety of morphological and sensorial challenges.

Bats foraging close to or within the vegetation (edge or cluttered space) need to be highly manoeuvrable (low wing loading and aspect ratio, Norberg & Rayner 1987), to avoid collision with the nearby vegetation. Moreover, bats foraging in edge space or cluttered habitat face the perceptual challenges of backward masking, as echoes from the surrounding vegetation may overlap in part (aerial insectivores that hawk for insects in the air) or fully (gleaning bats that take food from substrate) with the echoes of potential targets (Schnitzler & Kalko 2001). In contrast, for species foraging, further away from vegetation in rather open forest stands or above the canopy, high flight speeds at the cost of manoeuvrability (Norberg & Rayner 1987) are of benefit for a quick pursuit of insects after prey detection (Kingston et al. 2003), while the main echolocation requirement is long-range detection of small prey items in wide open space (Neuweiler 1989).

Structural parameters describing the three-dimensional arrangements of the vegetation thus are likely to be highly important in explaining the differences of bat occurrence, activity and species composition, within differently composed and managed forests. However, such quantitative and fine-grained parameters are impossible to assess using ground-based forest inventories (Lefsky et al. 2002; Vierling et al. 2008) and only recently became available through airborne Light Detection and Ranging (LiDAR, for more details please refer to Vierling et al. 2008). The application of LiDAR allows a detailed description of the three-dimensional structure of forests at unprecedented quality and at high spatial resolution (Parker, Harding & Berger 2004) using a multitude of mathematically derived structural parameters and their relative proportions (J. Nieschulze, R. Zimmermann, A. Börner, E. D. Schulze, unpublished data; McElhinny et al. 2005).

Here, we present data linking three-dimensional structural parameters of differently managed forest stands (beech, oak, pine and mixed) assessed by airborne LiDAR to bats. We expected bat occurrence and activity to increase with structural parameters, indicating higher heterogeneity in forest stands as this in turn positively influences roost availability (Barclay & Kurta 2007) as well as prey diversity and abundance (Haddad et al. 2009). We further expected the composition of bat assemblages to be profoundly influenced by differences in the three-dimensional structure of forests because individual bat species are known to differ in their tolerance to the physical clutter of the vegetation (Neuweiler 1989; Schnitzler & Kalko 2001). Finally, we hypothesized that bats sharing similar adaptations in wing morphology and foraging strategy should show similar associations with structural forest parameters.

Material and methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information

Study area

Our study was located in the biosphere reserve Schorfheide-Chorin (1300 km2) in the state of Brandenburg (north-eastern Germany), a young glacial landscape with many wetlands, characterized by agricultural land use and low human population density (23 km−2). Mean annual temperature ranges at 8·0–8·5 °C and annual precipitation ranges between 520 and 600 mm (for more details please refer to Fischer et al. 2010). The Schorfheide-Chorin is one of the three ‘Biodiversity Exploratories’, established in 2006 for long-term and functional biodiversity research in Germany (Fischer et al. 2010; http://www.biodiversity-exploratories.de). The area includes 50 permanently marked 1-ha (100*100 m) experimental forest plots, all located within continuous forest areas (see Fig. S1 in Supporting Information). The 50 experimental plots encompass replicates of seven forest types representing differently managed forest stands: eight young pine (Pinus sylvestris L.) forests, seven old-growth pine forests, seven pine beech (Fagus sylvatica L.) mixed forests, seven old-growth-managed age-class pure beech forests, seven old-growth beech forest with beech thicket, seven old-growth-unmanaged beech forests and seven oak (Quercus rubor L.) forests.

Acoustic Monitoring of Bats

We conducted standardized acoustic surveys of bats between June and September 2009. Acoustic recordings began immediately after local sunset (Eberswalde-Finow, Germany, 52°49″N and 13°31″E) and continued until 1:00 a.m. to account for the main activity peak of the bats during the first half of the night (Rydell, Entwistle & Racey 1996). Three to four experimental plots in different forest types were sampled each night, and each experimental plot was visited twice during the study period with an interval of 5 weeks.

Acoustic recordings were conducted along the edges of each 100*100 m experimental plot using a combination of point-stop and transect monitoring. Point-stops were located at each corner of the experimental plot. Transects (100 m) were walked slowly and in a direct line between the point-stops. Survey time at point-stops and during transect walks was 6 min each. This resulted in a total survey time of 48 min and 400-m transects per experimental plot.

Sound recordings were made in real time (sample rate: 384 kHz, 16 bit) with a Petterson-D1000x bat detector (Pettersson Electronic AG, Uppsala, Sweden). Individual recordings were triggered manually by listening through headphones to the output of the heterodyne system while continuously scanning the frequency range between 20 and 80 kHz. The pre-trigger time of the detector was set at 10 s and the post-trigger time to 50 s resulting in standardized file length of one minute.

Acoustic Data Analysis

We used Avisoft SAS Lab Pro, version 5.0.24 (R. Specht, Avisoft Bioacoustics, Berlin, Germany) for sound analysis. Spectrograms were generated with a Hamming window (1024 FFT, 96% overlap). We evaluated the number of bat passes as a measure for bat activity. A bat pass was defined as a minimum of two consecutive echolocation calls (Fenton 2004). Successive passes within the one-minute recording files were discriminated if the time interval between calls was larger than three times the regular pulse interval of the respective species (Estrada-Villegas, Meyer & Kalko 2010). We also assessed feeding activity of bats by counting terminal phases (call sequences emitted at a high repetition rate prior to prey capture attempts; Schnitzler & Kalko 2001). Bat activity and feeding activity were positively correlated (Pearson r = 0·83, < 0·001), indicating that experimental plots with higher bat activity were also better foraging habitats for bats. We thus considered bat activity as an indicator of the intensity of habitat use. In addition, we calculated an estimate of the relative abundance (hereafter: occurrence) of bats and individual bat species per experimental plot by considering the presence-only data during each of the 6-minute transect sections (point-stops and 100-m transects; Duchamp & Swihart 2008; Estrada-Villegas, Meyer & Kalko 2010; Jung & Kalko 2011).

Species Identification

Sound sequences were manually identified to species level or sonotype following a custom-made identification key and data from literature (e.g. Denzinger et al. 2001; Russo & Jones 2002; Obrist, Boesch & Flückiger 2004). On the basis of echolocation call characteristics, particularly call structure, start-, end- and peak frequency, seven species could be identified with high certainty: Nyctalus noctula, Barbastella barbastellus, Pipistrellus nathusii, Pipistrellus pipistrellus and Pipistrellus pygmaeus, Myotis myotis and (please refer to Dietz, von Helversen & Nill 2007 for naming authorities of species). Echolocation calls of Nyctalus leisleri could only be assigned to species level if the bats regularly alternated search call frequencies between 21 and 23 kHz (Weid & Helversen 1987). Echolocation sequences with calls ranging between 24 and 29 kHz and irregular call frequency alternations were classified as sonotype Nyctaloid potentially including N. leisleri, Eptesicus serotinus, Eptesicus nilsonii and Vespertilio murinus. All four species are known to occur in the region of the Schorfheide-Chorin (Teubner et al. 2008). We further could not discriminate between Plecotus auritus and Plecotus austriacus because of the similarity in call structure and frequencies. As both are known to occur in the region (Blohm 1999; Spitzenberger et al. 2006), we combined them into the sonotypes Plecotus sp. Finally, as echolocation calls of Myotis bechsteinii, Myotis nattereri, Myotis mystacinus, Myotis brandtii, Myotis dasycneme and Myotis daubentonii are very similar with extensive overlap in time-frequency course, we grouped them into the sonotype Myotis sp. Although M. dasycneme (considered to be very rare in Brandenburg; Teubner et al. 2008) and M. daubentonii typically forage above water, we included both as potentially occurring species in forest habitats, as small water bodies are interspersed throughout the whole region.

LiDAR Data

Data acquisition for airborne laser scanning was conducted in September 2009, when deciduous trees still carried leaves. All 50 experimental plots were overflown by helicopter using a full-wave Riegl LMS-Q 560 scanner (Riegl Laser Measurement Systems, GmbH, Horb, Austria). Flight altitude was 400 m above ground, and a conducted beam divergence of 0·5 mrad resulted in a footprint diameter between 20 and 30 cm. The scanner operated at a pulse repetition rate of 240 kHz with an averaged point density of 54·74 m−2. Horizontal sampling accuracy of the LiDAR data resulted in 50 cm and vertical sampling accuracy in 15 cm.

On the basis of normalized raw data and a digital canopy height model (calculated model of the canopy surface structure based on the normalized height values of LiDAR data), we calculated 85 structural parameters for the one ha (100*100 m) forest experimental plots. We aimed to derive fine-grained structural parameters which might differ in their importance for individual bat species and included spatial derivates with different horizontal diameter thresholds (1–3 m2) for forest structures such as gaps, edges and ground hits. We also calculated different measures of canopy height, gap depth and understorey vegetation along the vertical axis. As variation of structural parameters at the stand level are particularly important for the description of habitat heterogeneity (McElhinny et al. 2005), we included the standard deviation of bole zone, outer canopy height, euphotic zone, contrast, entropy and roughness of the canopy surface into our parameter set (see also Table 1).

Table 1.   The core set of 20 structural parameters which revealed the highest importance for bat occurrence and activity and their ecological interpretation in forests. Most structural parameters represent proportions of the respective parameter fraction per experimental plot
ParameterEcological interpretation of structural parameters
RoughnessThe inhomogeneity of the outer canopy surface
Roughness SDThe variation of the inhomogeneity of the outer canopy surface
Ground AreaThe relative amount of ground (>1 m2) that is directly reached by light
Deep GapThe relative amount of ground (>3 m2) with a proportion (>1 m2) that is directly reached by light
Hits below 5 mThe relative amount of understorey and ground vegetation up to 5 m above the ground
Young standsThe relative amount of small trees, tall scrubs and shrubs in proportion to all vegetation
ShrubThe relative amount of small trees, tall scrubs and shrubs in proportion to bare ground and herbs
North-facing crown areaThe relative amount of outer canopy surface that has a north aspect and a comparatively steep slope (>2 m2)
Crown isleThe relative amount of outer canopy surface (>1 m2) that projects 2/3 over the 0·99 quantile of the outer canopy height
EdgeThe relative amount of an area (>2 m2) which differs 5 m in height at 1-m horizontal distance
Bole zone (m)The space between the herbal layer and the leafy canopy where no LiDAR hits were recorded. Values of deep gaps (>1 m2) are omitted.
Bole zone SD (m)The variation of the bole zone
Euphotic zone (m)The amount of leafy canopy space that harbours the upper 65% of the LiDAR echoes omitting values of deep gaps (>1 m2)
RegenerationThe relative amount of secured regeneration from 0 to 3·5 m
Canopy height (m)The height of the outer canopy surface corrected for topography
Canopy height SD (m)The variation of the outer canopy height
EntropyThe local vertical variation that is high for equal distributed echoes and low for clustered echoes; it is a parameter of texture. Values of deep gaps more than 3 m2 are omitted.
Entropy SDThe variation of the local vertical variation
Contrast (m²)The local vertical variation that is high for widely scattered echoes, low for dense layers of echoes, taking absolute height difference into account; it is a parameter of texture. Values of deep gaps more than 2 m2 are omitted
Contrast SD (m²)The variation of the contrast

Statistical Data Analysis

To investigate the effect of structural parameters for bat occurrence, activity and occurrence of multiple species in different forest stands, we selected a subset of the 85 structural parameters with high explanatory power for further statistical tests. We used random forest regression (package: randomForest, Breiman 2001; with 100 000 randomizations) for parameter selection as it is the preferential method in systems with strong interactions among variables (Cutler et al. 2007). Parameter selection was based on species-specific structural parameter associations to ensure that important structural parameters for rare species were not lost during the parameter selection process. On the basis of random forest regression with 85 mathematically derived structural parameters, we obtained a subset of parameters that best predicted occurrence and activity of individual bat species and the three sonotypes (Myotis sp., Nyctaloid and Plecotus sp.). We then combined the species-specific structural parameter sets into a common set of structural parameters for the whole bat assemblage and ranked them according to their importance for all species combined. We subsequently excluded highly correlated structural parameters (|r| > 0·7) such as spatial derivates of one parameter from the parameter set by keeping the higher ranked structural variable and finally obtained a core set of 20 structural parameters with high explanatory power for the whole bat assemblage (Table 1). The subset of 20 structural parameters still allowed to discriminate between the seven forest types in the Schorfheide-Chorin (discriminant function analysis (DFA): eigenvalue, 36·45, Wilks λ = 0·001, chi-square of successive roots = 446, d.f. = 120, > 0·0001, see Fig. S2).

To evaluate the relationship between occurrence, activity and occurrence of multiple bat species and three-dimensional structural forest parameters, we analysed a set of Poisson generalized linear candidate models (Pinheiro & Bates 2000, GLMM, package: lme4). Three-dimensional structural parameters were fitted as fixed factors, and experimental plots were included as random factors. The best approximating models (see Table S1) were selected and compared with the full models (see Table S2), using an information-theoretic approach based on Akaike’s Information Criterion adjusted for small samples (AICc; Burnham & Anderson 2002, packages: AICcmodavg).

We further performed a canonical correspondence analysis (CCA, vegan 1.17-3, Oksanen et al. 2008) with 1000 permutations based on square root transformed data to investigate how the combination of 20 structural forest parameters affects individual species of the bat assemblages in differently managed forest stands. All statistical tests were performed in R version 2·11·1 (R Development Core Team 2011, Vienna, Austria).

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information

Association Between Bat Occurrence and Activity with Forest Structure

We obtained 2351 recordings with a total of 11422 passes from seven species and three sonotypes of bats over 57 nights in all seven forest types of the Schorfheide-Chorin (see Fig. S3). General bat occurrence, activity and occurrence of multiple species were significantly associated with structural parameters, most of which were linked to higher structural heterogeneity of forests (see Table 2 for statistical results). Structural parameters positively associated with higher bat occurrence, activity and occurrence of multiple species were the standard deviation of the canopy roughness (roughness SD) and edge fraction (edge) (Table 2a–c). In addition, bat activity significantly increased with the structural parameter canopy height (canopy height) (Table 2b). In combination, these three structural parameters represent old-growth forests with patches of different vegetation heights including smaller clearings which promote the structural parameter edge (graphical representation of structural parameters: Fig. 1). Occurrence, activity and occurrence of multiple species were, however, negatively associated with the structural parameters canopy roughness (roughness) and standard deviation of the canopy height (canopy height SD) (Table 2a–c). Both parameters are indicative for high vertical variation of LiDAR hits as it is typical for logged or very open forests with a few remaining canopy trees.

Table 2.   Statistical results of the Poisson distributed generalized linear mixed effect models (GLMM). Presented are the best approximating models using the information-theoretic approach based on Akaike’s Information Criterion adjusted for small samples (AICc). Listed are structural parameters and their respective effect on (a) bat occurrence, (b) bat activity and (c) occurrence of multiple species in differently managed forests of the Schorfheide-Chorin
GLMM modelStructural parametersEstimateErrorZ value> (|Z|)
  1. NS, non significant. *< 0·05; **< 0·01; ****< 0·001.

(a)
Occurrence (poisson) AIC: 102·9 Deviance: 86·91Intercept3·031·192·53**
Canopy roughness−1·090·39−2·78**
Canopy roughness SD0·550·291·93*
Crown Isle−0·340·86−0·40NS
Edge7·402·852·60**
Canopy height SD−0·080·07−1·17NS
Entropy0·100·180·55NS
(b)
Activity (quasi-poisson) AICc: 2637·63Intercept10·783·972·71**
Canopy roughness−3·061·08−2·82**
Canopy roughness SD1·980·892·22*
Ground Area5·705·121·11NS
Crown isle−4·062·78−1·46NS
Edge16·287·612·14*
Canopy height0·120·033·57***
Canopy height SD0·370·24−1·58NS
Entropy0·730·53−1·38NS
(c)
Occurrence of multiple species (poisson) AICc: 383·48Intercept7·982·053·90***
Canopy roughness−2·210·57−3·87***
Canopy roughness SD1·540·463·32***
Ground area2·282·640·86NS
Crown isle−3·411·42−2·41**
Edge11·544·012·88**
Canopy height0·0530·0173·04**
Canopy height SD−0·330·12−2·67**
Entropy−0·240·27−0·90NS
image

Figure 1.  Vertical section through a 10-m-deep beech forest for a graphical representation of structural parameters derived from LiDAR hits depicted as red dots. We included the five significant structural parameters from GLMM modelling: canopy height (a); edge (b); canopy height SD (c); roughness SD (d); and roughness (e). Please refer to Table 2 for the ecological interpretations of structural parameters.

Download figure to PowerPoint

Influence of Structural Forest Parameters on the Composition of Bat Assemblages

As expected, differences in three-dimensional structural parameters of forest stands significantly affected the composition of bat assemblages (CCA: total inertia: 0·64, trace 0·33, F = 1·5 < 0·01, with 1000 unrestricted permutations; Fig. 2). Bats sharing similar adaptations in wing morphology and foraging style revealed similar associations with structural forest parameters. Canopy height and extent of ground LiDAR hits (ground area) were the most prominent factors separating edge space foragers such as P. pygmaeus and P. nathusii, which mainly occurred in old-growth beech forest with closed canopy from open space foragers (N. noctula, N. leisleri, Nyctaloid) that foraged in rather open old-growth pine forests or above young pine forest stands along the first canonical axis (eigenvalue: 0·133, test of significance for the first canonical axis: F = 7·59, < 0·001). In addition, structural parameters describing vertical stratification in forest stands were the most important structural parameters separating species along the second canonical axis (eigenvalue 0·056). Whereas B. barbastellus and P. pipistrellus hunted in forest stands with an open canopy (increased euphotic zone, canopy roughness) and low levels of understorey vegetation (increased contrast), gleaning species such as Myotis sp. mainly foraged in forests with high levels of understorey vegetation (increased entropy, shrub) or in the case of M. myotis and Plecotus sp. in young pine forests (young stands).

image

Figure 2.  Triplot illustrating the results of a canonical correspondence analysis of species activity (triangles) and three-dimensional structural parameters of forest stands (depicted as vectors). Experimental forest plots are plotted as filled circles, and colours represent different forest types. Vector length indicates the explanatory importance of structural parameters for bat species and experimental forest plot distribution within the multidimensional space.

Download figure to PowerPoint

In addition, individual species within functional groups revealed fine differences in the importance of three-dimensional structural parameters. This is especially obvious in the three morphologically similar Pipistrelle bats, which revealed species-specific preferences for different forest structures. The activity of P. pygmaeus and P. nathusii was highest in forest stands characterized by low stratification, which is typically found in old-growth forest of one age class and indicated by the structural parameters canopy height and bole zone. In contrast, we recorded higher activity of P. pipistrellus in forests with vertical variation of vegetation layers (bole zone SD) and open canopy (increased euphotic zone) typical of forests consisting of different age classes.

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information

In Europe, most forests (97%) are strongly influenced by humans and managed for timber production (FAO 2011). The future of biodiversity and associated processes in such production forest systems thus strongly depends on ecologically responsible forest management practices (Gardner et al. 2009).

Habitat quality for wildlife depends to a large degree on the three-dimensional arrangement of the vegetation layers (Lindenmayer, Margules & Botkin 2000; McElhinny et al. 2005). We investigated whether and how the three-dimensional structure of differently composed and managed forest stands, described at a very fine resolution, affects occurrence, activity and species composition of bats. Detailed knowledge of species-specific (micro)habitat associations allows better predictions of the way in which a species is likely to respond to structural changes in its environment (Lindenmayer, Margules & Botkin 2000; Racey & Entwistle 2003).

As expected, our results suggest that the occurrence, activity and abundance of multiple bat species are mainly associated with parameters, indicating higher structural heterogeneity in differently composed and managed forests. Here, structural heterogeneity refers to the spatial arrangement and complexity of the vegetation and is composed of a suite of mathematically derived parameters based on high-resolution LiDAR data. Structural parameters positively associated with higher bat occurrence, activity and occurrence of multiple species included roughness SD, edge fraction and canopy height. In combination (McElhinny et al. 2005), these parameters represent old-growth forests with patches of different vegetation heights including smaller clearings and forest aisles. Forest management strategies that promote a combination of these structural parameters in production forest stands through, for example the plantation of mixed tree species and single tree or group selection harvest, will foster bat occurrence, activity and diversity. In such managed forest systems, bats benefit from increased structural heterogeneity because it increases the availability of microhabitats for foraging and roosting (Barclay & Kurta 2007). Furthermore, increased heterogeneity of the canopy structure is known to promote insect abundance and diversity (Haddad et al. 2009; Müller & Brandl 2009) and thus increases prey availability for bats.

Our results suggest a negative association with the structural parameters canopy roughness and the standard deviation of the canopy height, which are both indicative of high vertical variation of LiDAR hits. In combination, these two parameters are characteristic of very open forests with a few remaining canopy trees such as old-growth pine forest stands, which were mainly planted in large monocultures during the last century in the Schorfheide-Chorin region. Both parameters are also indicative of intensively logged forests with just a few remaining canopy trees.

General Trends and Species-Specific Responses

Our results clearly show that differences in the three-dimensional structure of the vegetation influence the composition of bat species assemblages. As expected, bat species with similar wing morphology and foraging style revealed similar habitat associations in response to the physical clutter of the vegetation (Neuweiler 1989; Schnitzler & Kalko 2001). While open space foragers characterized by high flight speed and low manoeuvrability (Norberg & Rayner 1987) were positively associated with three-dimensional parameters reflecting rather open forest stands (ground area, gap; for example, pine old-growth forests, oak forest), edge space foragers and gleaning species, which are more manoeuvrable, foraged in forests with closed canopy and high levels of understorey vegetation (e.g. old-growth beech forest, beech forest with thicket). These results confirmed our hypothesis that species composition in differently managed forest stands is in part determined by morphological and foraging behavioural traits of bats.

However, in addition to this, rather general trend we also found species-specific habitat associations at the microhabitat scale underlining that morphology alone does not necessarily explain ecological differences. This was especially obvious in the genus Pipistrellus which comprises the formerly cryptic species complex of P. pipistrellus and P. pygmaeus (Barlow & Jones 1997; Mayer & von Helversen 2001). While P. pygmaeus was mainly associated with parameters indicating structural stability (canopy height, bole zone), P. pipistrellus was mainly associated with the standard deviation of bole zone (bole zone SD) indicating vertical variation. Both species are nearly identical in wing morphology and foraging behaviour, but they are known to differ in their use of forest habitats (e.g. Davidson-Watts, Walls & Jones 2006; Sattler et al. 2007). Our data supports previous work and might explain differences in the use of forest habitats with species-specific requirements for certain three-dimensional structural forest attributes.

Conclusion

  1. Top of page
  2. Summary
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information

We investigated the effects of three-dimensional forest structure on occurrence, activity and species composition of European bats in managed production forest stands. Our results revealed a positive influence of the structural parameters canopy height, standard deviation of the canopy surface roughness and edge fraction. Forest management that promotes a combination of these structural parameters by integrating selection harvest (single trees or tree groups), and reforestation with mixed tree species will encourage a higher bat occurrence and activity.

Our results suggest that bat species composition is largely shaped by differences in the three-dimensional structure of the forest. This highlights the importance of high-resolution LiDAR data as a tool to understand and to explain the ecological requirements of single species at a fine scale. Given that bats provide essential ecosystem services in forests through top-down control of herbivorous insects (e.g. Böhm, Wells & Kalko 2011; Boyles et al. 2011; Kunz et al. 2011), applying such knowledge to support bat populations is relevant to species conservation and sustainable forestry.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information

This work was funded by the DFG Priority Program 1374 ‘Infrastructure-Biodiversity-Exploratories’ (P 3141044) and the MPI in Jena. Field work permits were given by the state of Brandenburg (according to § 72 BbgNatSchG). We are very grateful for the logistic support of the management team Schorfheide-Chorin, especially U. Pommer and U. Schumacher. We thank R. Zimmermann, A. Börner & E.D. Schulze for providing the vertical transect section (Fig. 1). We cordially thank E.D. Schulze for his support and valuable comments which considerably helped to improve this manuscript. We thank M. Helbig for her commitment during field work and for acoustic data analysis. We also thank K. Wells, J. Hailer and I. Geipel for various scientific discussions.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information
  • Bael, S.A.V., Philpott, S.M., Geenberg, R., Bichier, P., Barber, N.A., Mooney, K.A. & Gruner, D.S. (2008) Birds as predators in tropical agroforestry systems. Ecology, 89, 928934.
  • Barclay, R.M.R. & Kurta, A. (2007) Ecology and Behaviour of bats roosting in tree cavities and under bark. Bats in Forests: Conservation and Management (eds M.J. Lacki, J.P. Hayes & A. Kurta), pp. 1737. John Hopkins University Press, Baltimore, MD.
  • Barlow, K.E. & Jones, G. (1997) Differences in songflight calls between two phonic types of the vespertilionid bat Pipistrellus Pipistrellus. Journal of Zoology, 241, 315324.
  • Baxter, D.J.M., Psyllakis, J.M., Gillingham, M.P. & O’Brien, E.L. (2006) Behavioural response of bats to perceived predation risk while foraging. Ethology, 112, 977983.
  • Blohm, T. (1999) Fund des grauen Langohres, Plecotus austriacus (Fischer 1829) in der Uckermark. Nyctalus, 7, 221222.
  • Böhm, S.M., Wells, K. & Kalko, E.K.V. (2011) Top-down control of herbivory by birds and bats in the canopy of temperate broad-leaved oaks (Quercus robur). PLoS ONE, 6, e17857.
  • Boyles, J.G., Cryan, P.M., McCracken, G.F. & Kunz, T.H. (2011) Economic importance of bats in agriculture. Science, 332, 4142.
  • Bradbury, R.B., Hill, R.A., Mason, D.C., Balzter, H., Wilson, J.D., Hinsley, S.A., Whittingham, M.J., Davenport, I.J. & Bellamy, P.E. (2005) Modelling relationships between birds and vegetation structure using airborne LiDAR data: a review with case studies from agricultural and woodland environments. Ibis, 147, 443452.
  • Breiman, L. (2001) Random forests. Machine Learning, 45, 532.
  • Burnham, K.P. & Anderson, D.R. (2002) Model Selection and Multimodel Inference a Practical Information-Theoretic Approach, 2 edn. Springer, New York.
  • Caras, T. & Korine, C. (2009) Effect of vegetation density on the use of trails by bats in a secondary tropical rain forest. Journal of Tropical Ecology, 25, 97101.
  • Chazdon, R.L., Harvey, C.A., Komar, O., Griffith, D.M., Ferguson, B.G., Martínez-Ramos, M., Morales, H., Nigh, R., Soto-Pinto, L., Van Breugel, M. & Philpott, S.M. (2009) Beyond reserves: a research agenda for conserving biodiversity in tropical human-modified landscapes. Biotropica, 41, 142153.
  • Chen, J., Saunders, S.C., Crow, T.R., Naiman, R.J., Brosofske, K.D., Mroz, G.D., Brookshire, B.L. & Franklin, J.F. (1999) Microclimate in forest ecosystem and landscape ecology. BioScience, 49, 288297.
  • Clawges, R., Vierling, K., Vierling, L. & Rowell, E. (2008) The use of airborne LiDAR to assess avian species diversity, density, and occurrence in a pine/aspen forest. Remote Sensing of Environment, 112, 20642073.
  • Cleveland, C.J., Betke, M., Federico, P., Frank, J.D., Hallam, T.G., Horn, J., Lopez, J.D., McCracken, G.F., Medellin, R.A., Moreno-Valdez, A., Sansone, C.G., Westbrook, J.K. & Kunz, T.H. (2006) Economic value of the pest control service provided by Brazilian Free-tailed bats in south-central Texas. Frontiers in Ecology and the Environment, 4, 238243.
  • Crampton, L.H. & Barclay, R.M. (1998) Selection of roosting and foraging habitat by bats in different-aged aspen mixedwood stands. Conservation Biology, 12, 13471358.
  • Cutler, D.R., Edwards, T.C., Beard, K.H., Cutler, A., Hess, K.T., Gibson, J. & Lawler, J.J. (2007) Random forest for classification in ecology. Ecology, 88, 27832792.
  • Davidson-Watts, I., Walls, S. & Jones, G. (2006) Differential habitat selection by Pipistrellus pipistrellus and Pipistrellus pygmaeus identifies distinct conservation needs for cryptic species of echolocating bats. Biological Conservation, 133, 118127.
  • Davy, C.M., Russo, D. & Fenton, M.B. (2007) Use of native woodlands and traditional olive groves by foraging bats on a Mediterranean island: consequences for conservation. Journal of Zoology, 273, 397405.
  • Denzinger, A., Siemers, B.M., Schaub, A. & Schnitzler, H.-U. (2001) Echolocation by the barbastelle bat, Barbastella barbastellus. Journal of Comparative Physiology A, 187, 521528.
  • Dietz, C., von Helversen, O. & Nill, D. (2007) Handbuch der Fledermäuse Europas und Nordwestafrikas, 1st edn. Kosmos Naturführer, Stuttgart.
  • Duchamp, J. & Swihart, R. (2008) Shifts in bat community structure related to evolved traits and features of human-altered landscapes. Landscape Ecology, 23, 849860.
  • Estrada-Villegas, S., Meyer, C.F.J. & Kalko, E.K.V. (2010) Effects of tropical forest fragmentation on aerial insectivorous bats in a land-bridge island system. Conservation Biology, 143, 597608.
  • FAO (2011) The state of forest resources – a regional analysis. Available from http://www.fao.org/docrep/013/i2000e/i2000e00.htm (accessed 19 August 2011).
  • Fenton, M.B. (2004) Reporting: essential information and analysis. Bat Echolocation Research: Tools, Techniques and Analysis (eds M. Brigham, E.K.V. Kalko, G. Jones, S. Parsons & H.J.G.A. Limpens), pp. 133140. Bat Conservation International, Austin, TX.
  • Fischer, M., Bossdorf, O., Gockel, S., Hänsel, F., Hemp, A., Hessenmöller, D., Korte, G., Nieschulze, J., Pfeiffer, S., Prati, D., Renner, S., Schöning, I., Schumacher, U., Wells, K., Buscot, F., Kalko, E.K.V., Linsenmair, K.E., Schulze, E.-D. & Weisser, W.W. (2010) Implementing large-scale and long-term functional biodiversity research: the Biodiversity Exploratories. Basic and Applied Ecology, 11, 473485.
  • Gardner, T.A., Barlow, J., Chazdon, R., Ewers, R.M., Harvey, C.A., Peres, C.A. & Sodhi, N.S. (2009) Prospects for tropical forest biodiversity in a human-modified world. Ecology Letters, 12, 561582.
  • Getzin, S., Wiegand, K. & Schöning, I. (2012) Assessing biodiversity in forests using very high-resolution images and unmanned aerial vehicles. Methods in Ecology and Evolution, DOI: 10.1111/j.2041-1210X.2011.00158.x.
  • Haddad, N.M., Crutsinger, G.M., Gross, K., Haarstad, J., Knops, J.M. & Tilman, D. (2009) Plant species loss decreases arthropod diversity and shifts trophic structure. Ecology Letters, 12, 10291039.
  • Hayes, J.P. & Loeb, S.C. (2007) The influence of forest management on bats in North America. Bats in Forests: Conservation and Management (eds M.J. Lackie, J.P. Hayes & A. Kurta), pp. 207236. John Hopkins University Press, Baltimore, MD.
  • Jung, K. & Kalko, E.K.V. (2011) Adaptability and vulnerability of high flying Neotropical aerial insectivorous bats to urbanization. Diversity and Distributions, 17, 262274.
  • Kalcounis-Rüppell, M.C., Hobson, K.A., Brigham, M.R. & Hecker, K.R. (1999) Bat activity in the boreal forest: importance of stand type and vertical strata. Journal of Mammalogy, 80, 673682.
  • Kalka, M.B., Smith, A.R. & Kalko, E.K.V. (2008) Bats limit arthropods and herbivory in a tropical forest. Science, 320, 71.
  • Kingston, T., Jones, G., Akbar, Z. & Kunz, T.H. (2003) Alternation of echolocation calls in 5 species of aerial-feeding insectivorous bats from Malaysia. Journal of Mammalogy, 84, 205215.
  • Kunz, T.H., Braun de Torrez, E., Bauer, D., Lobova, T. & Fleming, T.H. (2011) Ecosystem services provided by bats. Annals of the New York Academy of Sciences, 1223, 138.
  • Lefsky, M.A., Cohen, W.B., Parker, G.G. & Harding, D.J. (2002) LiDAR remote sensing for ecosystem studies. BioScience, 52, 1930.
  • Lindenmayer, D.B., Margules, C.R. & Botkin, D.B. (2000) Indicators of biodiversity for ecologically sustainable forest management. Conservation Biology, 14, 941950.
  • Lowman, M.D. & Wittman, P.K. (1996) Forest canopies: methods, hypotheses, and future directions. Annual Review of Ecology & Systematics, 27, 5581.
  • MacArthur, R.H. & MacArthur, J.W. (1961) On bird species diversity. Ecology, 42, 594598.
  • Mayer, F. & von Helversen, O. (2001) Sympatric distribution of two cryptic bat species across Europe. Biological Journal of the Linnean Society, 74, 365374.
  • McElhinny, C., Gibbons, P., Brack, C. & Bauhus, J. (2005) Forest and woodland stand structural complexity: its definition and measurement. Forest Ecology and Management, 218, 124.
  • Müller, J. & Brandl, R. (2009) Assessing biodiversity by remote sensing in mountainous terrain: the potential of LiDAR to predict forest beetle assemblages. Journal of Applied Ecology, 46, 897905.
  • Müller, J., Stadler, J. & Brandl, R. (2010) Composition versus physiognomy of vegetation as predictors of bird assemblages: the role of LiDAR. Remote Sensing of Environment, 114, 490495.
  • Müller, J., Moning, C., Baessler, C., Heurich, M. & Brandl, R. (2009) Using airborne laser scanning to model potential abundance and assemblages of forest passerines. Basic and Applied Ecology, 10, 671681.
  • Neuweiler, G. (1989) Foraging ecology and audition in echolocation bats. Trends in Ecology and Evolution, 4, 160166.
  • Norberg, U.M. & Rayner, J.M.V. (1987) Ecological morphology and flight in bats (Mammalia; Chiroptera): wing adaptations, flight performance, foraging strategy and echolocation. Philosophical Transactions of the Royal Society B, 316, 335427.
  • Obrist, M., Boesch, R. & Flückiger, P. (2004) Variability in echolocation call design of 26 Swiss bat species: consequences, limits and options for automated field identification woth a synergetic pattern recognition approach. Mammalia, 68, 307322.
  • Oksanen, J., Kindt, R., Legendre, P., O'Hara, B., Simpson, G.L., Solymos, P., Stevens, M.H.H. & Wagner, H. (2008) vegan: Community Ecology Package. R package version, 1.15-0.
  • Parker, G.G., Harding, D.J. & Berger, M.L. (2004) A portable LIDAR system for rapid determination of forest canopy structure. Journal of Applied Ecology, 41, 755767.
  • Parker, G.G., Harmon, M.E., Lefsky, M.A., Chen, J., Pelt, R.V., Weis, S.B., Thomas, S.C., Winner, W.E., Shaw, D.C. & Frankling, J.F. (2004) Three-dimensional structure of an old-growth Pseudotsuga-Tsuga canopy and its implications for radiation balance, microclimate, and gas exchange. Ecosystems, 7, 440453.
  • Patriquin, K.J. & Barclay, R.T.M.R. (2003) Foraging by bats in cleared, thinned and unharvested boreal forest. Journal of Applied Ecology, 40, 646657.
  • Pinheiro, J.C. & Bates, D.M (2000) Mixed-effect models in S and S plus. Springer, London, Berlin, New York.
  • R Development Core Team (2011) R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria. http://www.R-project.org .
  • Racey, P.A. & Entwistle, A.E. (2003) Conservation ecology. Bat Ecology (eds T.H. Kunz & M.B. Fenton), pp. 680743. Chicago University Press, Chicago, IL.
  • Russo, D. & Jones, G. (2002) Identification of twenty-two bat species (Mammalia: Chiroptera) from Italy by analysis of time-expanded recordings of echolocation calls. Journal of Zoology, 258, 91103.
  • Russo, D., Cistrone, L., Garonna, A. & Jones, G. (2010) Reconsidering the importance of harvested forests for the conservation of tree-dwelling bats. Biodiversity and Conservation, 19, 25012515.
  • Rydell, J., Entwistle, A. & Racey, P.A. (1996) Timing of foraging flights of three species of bats in relation to insect activity and predation risk. Oikos, 76, 243252.
  • Sattler, T., Bontadina, F., Hirzel, A.H. & Arlettaz, R. (2007) Ecological niche modelling of two cryptic bat species calls for a reassessment of their conservation status. Journal of Applied Ecology, 44, 11881199.
  • Schnitzler, H.-U. & Kalko, E.K.V. (2001) Echolocation by insect eating bats. BioSience, 51, 557569.
  • Spitzenberger, F., Strelkov, P.P., Winkler, H. & Haring, E. (2006) A preliminary revision of the genus Plecotus (Chiroptera, Vespertilionidae) based on genetic and morphological results. Zoologica Scripta, 35, 187230.
  • Teubner, T., Teubner, J., Dolch, D. & Heise, G. (2008) Säugetierfauna des Landes Brandenburg – Teil1: Fledermäuse, Naturschutz und Landschaftspflege in Brandenburg. Landesumweltamt Brandenburg (LUA), Velten.
  • Vierling, K.T., Vierling, L.A., Gould, W.A., Martinuzzi, S. & Clawges, R.M. (2008) LiDAR: shedding new light on habitat characterization and modeling. Frontiers in Ecology and the Environment, 6, 9098.
  • Weid, R. & von Helversen, O. (1987) Ortungsrufe europäischer Fledermäuse beim Jagdflug im Freiland. Myotis, 25, 527.
  • Williams-Guillén, K., Perfecto, I. & & Vandermeer, J. (2008) Bats limit insects in a neotropical agroforestry system. Science, 320, 70.

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information

Figure S1. Map of the study area.

Figure S2. Graphical representation of the discriminant function analysis of forest types based on structural LiDAR parameters.

Figure S3. Numbers of bat species and general bat activity in the seven forest types of the Schorfheide-Chorin.

Table S1. Eighty-nine percent confidence set of models.

Table S2. Full GLMM models.

As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials may be re-organized for online delivery, but are not copy-edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.

FilenameFormatSizeDescription
JPE_2116_sm_FigS1.doc235KSupporting info item
JPE_2116_sm_FigS2.doc61KSupporting info item
JPE_2116_sm_FigS3.doc106KSupporting info item
JPE_2116_sm_TableS1.doc63KSupporting info item
JPE_2116_sm_TableS2.doc75KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.