SEARCH

SEARCH BY CITATION

Keywords:

  • agro-ecosystems;
  • grasslands;
  • grazing;
  • ground-gleaning;
  • Mediterranean;
  • Myotis myotis;
  • prey abundance;
  • prey availability

Summary

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

1. The management of the habitat of insectivorous species often targets increasing the availability of arthropod prey. However, this may be hindered by lack of knowledge of the mechanisms that determine food availability for insectivores. Prey abundance is often used as a surrogate for availability, but this may be incorrect wherever habitat factors limit access to prey. Ground vegetation clutter is likely to be such a limiting factor for ground foraging insectivorous bats, and we investigated this possibility using Myotis myotis as a model.

2. We performed captivity experiments to determine how ground vegetation density affects foraging. Bats were provided with crickets hidden in sparse, medium and dense grass cover. In addition, we used bat radio tracking data, prey abundance measurements, and geographic information system (GIS) modelling to determine if clutter influences how bats select foraging areas in a Mediterranean region.

3. The experiments demonstrated that ground vegetation clutter greatly reduces access to prey, affecting both capture success and time to capture. Bats detected prey in the dense vegetation, but did not attempt to capture them or did so only after a prolonged delay. Their attempts often failed, because the bats landed over prey with open wings, presumably to increase the catching surface, and the dense vegetation prevented them from reaching the ground.

4. In the study area, cover types with the densest ground vegetation harboured the most prey, but clutter made access to prey by M. myotis difficult. Corroborating this, the GIS models showed that bats avoided foraging in habitats with high prey abundance but in which availability was decreased by dense ground vegetation.

5. Ungrazed grasslands reach vegetation densities that limit access to prey by ground foraging bats, as observed in the study area. However, grazing by cattle reduced clutter to levels equivalent to the sparse treatment in our captive experiments, in which bats captured prey easily.

6.Synthesis and applications. Conservation of ground foraging bats may require the management of their feeding grounds, to increase or maintain prey availability, particularly near important colonies. While dense ground vegetation may promote prey density, it dramatically reduces access to prey and usage by bats; hence, this should be considered by managers. Moderate grazing can be used to reduce ground vegetation cover to levels that permit foraging by bats.


Introduction

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

Over the last century, the progressive conversion of natural areas for agricultural purposes has made many wild species wholly or partially dependent upon farmland (Devereux et al. 2004). There is growing evidence that changes in the management of farmland are causing strong declines in many of these species (Vickery et al. 2001; Wickramasinghe et al. 2003). These changes may include not only agricultural intensification and land use changes but also land abandonment (Russo 2007).

Grasslands are among the most widespread habitat types in the world (Silva et al. 2008), covering more than one-third of European agricultural land (FAO 2008). The management of grasslands has changed over recent decades, and farmers aim to increase herbage production through, for example, the application of fertilizers or reseeding (Haugland & Froud-Williams 1999). In Europe, despite an overall decline in grassland area, partly due to land abandonment, grassland productivity has increased in recent decades (Smit, Metzger & Ewert 2008).

It is generally accepted that one of the major consequences of these changes, is a reduction in food abundance for wild species that feed on ground dwelling arthropods. However, predators may also be affected if the changes limit access to prey, even if prey remain abundant (Devereux et al. 2006). For example, if management enhances sward density, this may increase the physical obstruction (clutter) that limits the accessibility of arthropods within it (Atkinson et al. 2005). Under such circumstances a discrepancy between prey abundance and prey availability may occur, as sites with high prey abundance may have vegetation that hinders their capture.

Bats are one of the most diverse groups of mammals, and a large number of species feed on arthropods gleaned from ground vegetation or even directly from the ground (Schnitzler, Moss & Denzinger 2003). Substrate gleaning is a hunting strategy that occurs across a variety of bat families worldwide (Faure & Barclay 1992; Hubner & Wiegrebe 2003). It has been stated that some ground foraging bats avoid areas with dense ground vegetation (Arlettaz 1996). However, there are no studies that directly evaluate how ground clutter affects access to prey by bats, and few studies for other groups of predators of ground dwelling arthropods (but see Devereux et al. 2006).

Myotis myotis (Borkhausen, 1797) is a ground gleaning bat that usually captures ground dwelling arthropods across a diversity of open habitats (Audet 1990; Arlettaz 1996). Although often associated with forested habitats, it is now recognized that, like many other ground gleaners, it forages primarily on farmland (Audet 1990; Arlettaz 1999; Zahn, Haselbach & Guttinger 2005). These bats locate their prey – mostly beetles, crickets and spiders (Pereira et al. 2002; Zahn et al. 2007) – on the ground, by passively listening to prey-generated rustling sounds (Arlettaz, Jones & Racey 2001; Siemers & Guttinger 2006; Russo, Jones & Arlettaz 2007). Myotis myotis is a vulnerable species in many countries (e.g. Cabral et al. 2005; Bulgarini et al. 2007; Palomo 2008), and the European Habitats Directive requires that national authorities classify, and manage, areas to protect this bat, in order to improve its conservation status. As for other bats, this requires not only the protection of roosts but also the correct management of foraging areas (Hutson et al. 2008).

Good management of foraging areas for bats that depend on ground dwelling arthropods requires not only the maintenance of abundant prey but also the habitat conditions that allow for good access to those prey by bats. To achieve this, it is necessary to understand the relationship between ground vegetation clutter and prey accessibility. Consequently, our main goal was to evaluate how the density of ground vegetation constrains the foraging ability of bats that capture prey from the ground, using M. myotis as a study model. Moreover, we evaluated whether these potential constraints created a discrepancy between the abundance of prey and their actual availability to bats, within the foraging range of a large colony of M. myotis in a Mediterranean region. Our ultimate aim was to integrate our results to provide recommendations for best management practices.

Materials and methods

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

Study site

The study colony of M. myotis is located in Alentejo, southeastern Portugal (38°02′N, 7°17′E). This is a dry region, characterized by a Mediterranean-Continental climate with strong rainfall and temperature annual fluctuations. Alentejo is mostly flat, with gentle slopes (200–500 m asl) and poor soils (Pinto-Correia & Vos 2004). The dominant land use is a silvo-pastoral system called montado, which usually consists of vast grasslands with a tree cover of variable density of either holm oak Quercus rotundifolia or cork oak Quercus suber. The livestock used in this system is primarily cattle, but sheep or Iberian black pigs are also common. Other important habitats in the region include olive tree groves Olea europaea, several cereal crops (largely wheat and barley) and fodder. Due to increasing demands for biodiesel, sunflower crops Helianthus annuus are becoming an increasingly common feature on the Alentejo landscape, as well.

Comparing access to prey within different ground vegetation densities

We analysed the foraging behaviour of eight male M. myotis captured in the study area, each kept in captivity for <1 week, under a permit from the Instituto da Conservação da Natureza e da Biodiversidade. Experiments were conducted in a nylon cloth tent that was 6 m long, 3 m wide and 3 m high, during May and June 2008, always between 21:00 and 03:30 h. During the day, bats were kept inside a roosting cage, following the recommendations made by Lollar & Schmidt-French (2002).

To determine how difficult it is for bats to capture prey in areas with different degrees of ground vegetation clutter, we exposed them to prey at three different levels of vegetation cover: sparse (14–22%), medium (55–64%) and dense (93–100%). Vegetation density was measured using a point interception method (Brathen & Hagberg 2004), as described below. These three levels of cover were made available inside the flight tent, in 70 × 100 cm plastic trays. To set up the trays, we dug patches with the desired cover levels out of unimproved grasslands. We maintained the trays with the desired vegetation density by changing their content regularly and watering the plants. Each experiment consisted of a set of three trials, in which one tray with a live and active cricket (Acheta domestica; 0·4 g mean weight) was made available. The mobility of the crickets was restricted by tying them to an object with thin thread. In each of the trials, we used a tray with a different vegetation cover. A trial would start when the bat was placed on the perch, and lasted until it captured the cricket, or until 30 min had passed. The perch was a 20 × 40 cm wooden board covered with small mesh plastic netting, mounted on a 1·8-m high tripod, and placed in the tent 3 m away from the tray. Water was available in the tent ad libitum. We performed no more than four experiments with each bat per night.

Bat foraging behaviour was monitored remotely using four IR CCTV-cameras (690IRBW CCTV-camera, Sony Super-HAD; Sony, Tokyo, Japan), which together covered the entire tent. Illumination was provided by custom-mounted arrays of IR LEDs. The cameras were connected to a PC through a 4-channel CCTV-capture card (Geovision GV-650; Geovision, Taipei, Taiwan) that enabled the simultaneous display and digital recording of signals from all the cameras. Both the PC and the operator were outside the tent. Using the recordings, we measured the time needed for the bat to capture the prey in each trial, divided into three phases:

Time to flight: from the start of a trial to the moment the bat flew from the perch. During this period, the bat presumably becomes aware of the presence of a prey in the tray, and decides if it is worth attempting to capture it.

Time to land: from the time of flight from the perch to the moment the bat landed in the tray. During this period, the bat presumably gathers the information it needs to determine how to land on the prey.

Time to capture: from the first time the bat lands in the tray until it captures the prey. This period presumably is proportional to the difficulty of the last phase of the attack.

In addition, we also measured capture success rate, which is the proportion of trials in which each bat manages to capture its prey.

Estimating prey abundance in different habitats

We evaluated the abundance of prey within the main habitats of the study area using pitfall traps (Ø 9 cm, 11 cm deep, filled with water, biodegradable detergent and salt). Each habitat was sampled at eight locations, and at each of these we used a row of five traps connected by 40 cm long barriers. The traps were open for 10 consecutive nights in May 2007. We preserved all captured arthropods in 70% ethanol and identified them to order, except in the case of the Orthoptera, which were identified to family. We only considered taxa known to be regularly eaten by M. myotis, and excluded all animals with a body length <10 mm (Pereira et al. 2002).

Estimating vegetation ground cover

We also estimated the density of the vegetation ground cover at each of the pitfall trap sampling stations. This was accomplished by measuring vegetation density at 15 random sites located within 12 m of the traps, using a point interception method (Brathen & Hagberg 2004). At each site, a frame with 11 metal pins, 10 cm apart, was placed perpendicular to the soil. Cover was estimated using the percentage of pins that hit vegetation. Mean height of the vegetation was measured with a ruler.

To evaluate the relationship between habitat type, prey abundance and ground cover, we analysed a set of Poisson generalized linear candidate models (GLM) candidate models, with prey abundance as the response variable and ground cover and habitat type as predictors (see Table S1, Supporting information). Habitat type was included as a dummy variable, using cereal as an indicator. A best approximating model was selected, using an information-theoretic approach based on Akaike’s Information Criterion adjusted for small samples (AICc; Burnham & Anderson 2002) with the package Multi-model inference (MuMIn) in the r environment (v. 2.10.1; The R Foundation for Statistical Computing, Vienna, Austria). Whenever overdispersion was detected, we corrected the standard errors using a quasi-GLM model (Zuur et al. 2009).

Grazing is a potential tool by which to reduce the density of herbaceous vegetation. However, in certain circumstances, it reduces the height of the vegetation, leaving its density unchanged (Kruess & Tscharntke 2002). We used the same point intercept method to assess whether extensive cattle grazing reduces ground vegetation cover in our study region. Eight sites, in which a fence separated a recently grazed parcel from a similar but ungrazed area, were selected randomly. At each of these sites, the ground vegetation cover was measured for 10 pairs of points (one on each side of the fence, 15 m from the fence).

Habitat suitability index models

To determine whether difficulties posed by dense ground vegetation and access to prey influence how bats use different types of habitat, we constructed two Habitat Suitability Index (HSI) scenarios, following a deductive approach (Ottaviani, Lasinio & Boitani 2004). Habitat suitability modelling is a tool for predicting the quality or suitability of a land use for a given species. The resulting index ranges from 0 (unsuitable habitat) to 1 (optimal habitat) and indicates the probability that the species will forage or occur where that land use occurs by combining the interactions of key environmental variables (Juntii & Rumble 2006).

One of the models ranked suitability according to prey abundance alone, and the second used both prey abundance and accessibility, as measured in our experiments. The two scenarios were evaluated using data from radiotracking of 20 M. myotis in the study area.

For the first HSI, we simply standardized prey abundance in each main habitat by dividing its absolute value by the average prey abundance across all the studied habitats. In the second HSI, we included not only this standardized prey abundance estimate but also measurements of the difficulty of capturing prey at different levels of ground cover, as measured in the laboratory experiments. In particular, we included measurements of success rate and time needed to capture prey. Using GIS tools, we then generated maps revealing spatial variations in the values of HSI throughout the study area, for each of the two scenarios.

Foraging bats were fitted with small radio tags (BD-2A 0·66 g; Holohil Systems, Carp, ON, Canada) and tracked by triangulation from fixed telemetry stations positioned in strategic high points, and one mobile station installed on a car. Each fixed telemetry station consisted of an 8-m high metal tower equipped with two parallel 6-element Yagi antennas and a precision null combiner (Telonics Tac-5; Telonics, Mesa, AZ, USA). We followed 18 male and four female adult bats, throughout the full night, in the spring of the years 1997–2000, before the harvest of cereals and fodder. Spatio-temporal dependence of locations was avoided using the method described by Boyce et al. (2003). Areas far from the roost may be avoided by bats, and this could confound the effect of the factors that we were studying (prey abundance and ground cover). To minimize this risk, we only used locations within 10 km of the roost. Point locations that obviously corresponded to commuting animals, because they were flying very fast, were excluded. In total, 657 locations were used.

Habitat selectivity was calculated using Ivlev’s selectivity index (Krebs 1989), a method that compares relative availability of an item in the environment (p) with the relative use an animal makes of that same item (r): Ei = (r− pi)/(r+ pi). The association between habitat use by bats and the HSI models was tested using a logistic regression (Hosmer & Lemeshow 2000). The locations where we found foraging bats with radio-tracking were compared with a set of 657 random points located in the main foraging region of the studied colony, which is limited by the minimum convex polygon encompassing the bat foraging locations used in the analysis. Random points that were <500 m from a foraging bat location were excluded.

Results

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

Access to prey within different ground vegetation densities

The effect that vegetation clutter exerts on the capacity of bats to capture prey was evident in two phases of the attack sequence –time to land and time to capture– as well as on the rate of capture success (Fig. 1).

image

Figure 1.  The time needed by bats to capture prey in each ground-cover treatment: (a) time to flight; (b) time to land; (c) time to capture and (d) capture success rate. Horizontal lines represent medians, boxes the first and third quartiles, and whiskers the range. Differences tested with Kruskal–Wallis were statistically significant in (b–d).

Download figure to PowerPoint

When bats were exposed to crickets in sparse or medium vegetation cover trays, they left the perch about 45 s after the onset of the experiment (Fig. 1a). However, the time to flight increased several fold when the prey were in a dense cover tray. Once airborne, and in sparse vegetation, bats landed on or close to the crickets in about 86 s (Fig. 1b). This time grew to 144 s under medium cover conditions and rose markedly to 10 min in dense vegetation. The same pattern was observed in the time to capture, which was just 62 s in sparse vegetation, but 12 min in dense vegetation (Fig. 1c).

Capture success rate was drastically different across the three vegetation densities: over the 30 min that each trial lasted, virtually all prey were captured when bats were exposed to sparse or medium density vegetation, but only 40% were captured during the dense vegetation trials (Fig. 1d).

The effects of vegetation cover on prey abundance

Parameter estimates for the best approximating model relating prey abundance, ground cover and habitat type are given in Table 1 (see Table S1, Supporting information, for all candidate models and model selection). The density of ground cover in the main habitats in the study region was positively associated with prey abundance (Fig. 2; < 0·0001, Table 1). Ground cover and prey abundance varied significantly among habitat types (Table 1), in spite of the substantial variations in ground cover within some of the habitat types, visible in Fig. 2.

Table 1.   Parameters for the best GLM model relating the variation of prey abundance in the study area to ground cover and habitat type
 EstimateSEt-valueP value
  1. Habitat type was converted to a dummy variable using cereal crops as the indicator. Details of model construction and selection are summarized in the Methods section. Proportion of deviance explained by the model: 69%.

(Intercept)−0·050·41−0·120·910
Pasture1·620·493·290·001
Sunflower1·250·442·770·006
Olive1·340·452·950·004
Fodder0·400·550·730·466
Cover0·030·007·280·000
Cover: Pasture−0·020·01−3·310·001
Cover: Sunflower−0·010·01−1·430·156
Cover: Olive−0·020·01−3·100·002
Cover: Fodder−0·000·01−0·180·857
image

Figure 2.  Scatter plot with all sampling sites, illustrating the relationship between ground cover, habitat type and prey abundance (as defined in Methods). The fitted curves show the predicted prey abundance within the ground cover range observed in each habitat type (see the Methods section, and Table 1 for full GLM model and statistics).

Download figure to PowerPoint

A spatially explicit model for a Mediterranean case study area

The importance of considering both prey abundance and accessibility is illustrated via analysis of the spatial variations in habitat suitability around the studied roost, using both scenarios produced (Fig. 3a and b). In Fig. 3a, we show the results of a simple HSI that only reflects mean prey abundance. The areas that stand out as having the highest values are those with the densest ground cover, because they have the most prey. However, we now know that such habitats are not very suitable for ground foraging bats, because the prey is not easily accessible. Figure 3b reveals the results of a similar HSI, but this time reflecting both prey abundance and access to prey. The best areas have shifted to pastures, where ground cover vegetation is lower and prey abundance still relatively high.

image

Figure 3.  GIS models of habitat suitability for Myotis myotis, considering (a) prey abundance alone, and (b) the combination of prey abundance and prey accessibility (i.e. availability) in the main foraging region of the studied colony. Bat locations (dots) are significantly associated with areas of high suitability (darker tones) only in the availability scenario (Logistic regression: β = 0·29, Wald = 14·1, < 0·001).

Download figure to PowerPoint

The spatial output of both models was validated using radio-tracking locations of foraging M. myotis. The HSI model that reflects prey availability (Fig. 3b), i.e. both prey abundance and accessibility, better accommodated the use bats made of the area, with 46% of the bat locations within areas of high suitability (HSI >60%). In fact, bats seemed to prefer areas with high and medium access to prey (Fig. 4). In contrast, the HSI based on prey abundance alone, did not reflect how bats selected foraging sites, as the use of areas with high prey abundance was comparatively low (24%). In fact, bats seemed to avoid these areas (Fig. 4).

image

Figure 4.  Ivlev’s selectivity index for the three habitat suitability classes in the models fitted using prey abundance alone, and using both prey abundance and accessibility (i.e. prey availability). Zero corresponds to no selection. Selectivity increases with availability, but not with prey abundance.

Download figure to PowerPoint

These results were corroborated by those of the logistic regression, in which the estimated coefficient (β) revealed a statistically significant positive association between bat use and suitability, as indicated by the HSI that includes prey accessibility (β = 0·29, Wald = 14·1, < 0·001). In contrast, bat use was negatively associated with the HSI based on prey abundance alone (β = −0·19, Wald = 5·0, = 0·025; Fig. 3).

The effect of grazing on ground cover

The results of the comparison between grazed and ungrazed parcels prove that extensive cattle grazing effectively reduces ground cover (t = 28·1, < 0·001), and vegetation height (Kruess & Tscharntke 2002). Grazing maintained the vegetation cover below 30% at all the sites analysed. In contrast, all areas that had not been grazed for 1 year or more exhibited ground cover over 80%. Comparing these results with the levels of vegetation density used in the experiments, it is apparent that grazing reduced the ground cover from the level equivalent to the dense treatment to that equivalent to sparse.

Discussion

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

Ground vegetation cover decreases access to prey

Our results revealed that the influence of ground vegetation clutter on the accessibility of prey is so great that the time required to capture prey and the success rate varied several fold from the sparse to dense treatments. This difficulty of capturing prey in dense and even medium density vegetation could be due either to sensorial constraints or to mechanical difficulties capturing the prey.

Myotis myotis locates its prey by passively listening to the rustling noises that moving arthropods produce (Arlettaz, Jones & Racey 2001; Siemers & Guttinger 2006), and the presence of ground vegetation actually facilitates noise production by moving prey (Fuzessery et al. 1993). Goerlitz, Greif & Siemers (2008) proved this by showing that insects moving on meadows produce rustling sounds 3–4 dB higher than when moving on bare ground. Consequently, it is unlikely that the bats failed to strike or capture prey in denser vegetation because they had more difficulty detecting them. In our experiments, the prey was positioned close to the perch of the bat, so it could easily hear the prey rustling noises. In the experiments involving dense vegetation, bats moved their heads towards the noise the prey generated. This demonstrates that the bats were aware of the prey’s presence, even though they usually did not attack them, or took a long time to do so. Such behaviour suggests that bats recognize the difficulties associated with capturing prey located in dense vegetation, and often decide not to strike because of the high probability of failure. We observed that bats always attempted to land on their prey with outstretched wings, presumably to maximize their capture surface. This was very efficient on relatively open ground; but in denser vegetation it kept the bat from reaching the prey. Even when bats showed interest in prey located in medium and dense vegetation, it took them longer to land and capture the prey than in sparse vegetation. This fact suggests that bats have greater difficulty gathering the information necessary for a successful landing on cluttered ground. In addition, once they land, they have reduced mobility, and this further restricts their ability to make a successful capture.

The greater difficulty bats have capturing prey on more cluttered ground may actually prevent bats from obtaining sufficient food to satisfy their energy requirements, even if prey is abundant. Using the formulas proposed by Speakman & Thomas (2003), we estimated that a 25 g M. myotis male needs to capture at least 12 crickets of the size that were used in our experiments each night. To capture this number of crickets in the three hours typically spent foraging in the study area (A. Rainho & J.M. Palmeirim, personal observation obtained by radio-tracking), bats would have to capture one cricket every 15 min. This capture rate does not appear to be feasible in dense ground cover. Indeed, even within the relatively simple environment used in our experiments, in which the prey were placed just 3 m from the perch and tied down, it took an average of 24 min for bats to capture them. Moreover, this average only includes the trials in which the bats succeeded (40%). The corresponding value for medium density ground cover was 8 min, and for sparse vegetation just 3 min, both with success rates close to 100%.

These results suggest that, at least in their usual 3-h foraging bouts, male M. myotis cannot satisfy their full energy needs if they have to forage in dense ground cover. Under these circumstances, bats could extend their foraging time, but this would also increase their energy expenditures and consequent caloric requirements substantially. Considering this and the low capture rate (only 40%), it seems unlikely that they ever would attain enough food; hence, habitats with dense ground cover should be considered unsuitable for M. myotis and for other bats with a similar ground foraging behaviour.

Vegetation clutter creates a discrepancy between the abundance and availability of prey

Prey availability frequently is assumed to be equivalent to prey abundance, when evaluating the quality of a bat foraging habitat (e.g. Bontadina et al. 2008; Dodd, Lacki & Rieske 2008). This assumption often is difficult to avoid, because it is usually very hard to estimate prey availability accurately. However, prey abundance and prey availability are clearly distinct parameters, and in some circumstances they may not even be related. Prey abundance simply refers to the frequency and distribution of prey organisms, while availability refers to the prey that a predator actually can detect and capture (Faure & Barclay 1992; Arlettaz, Jones & Racey 2001; Siemers & Guttinger 2006; Siemers & Swift 2006).

Our case study demonstrates how very different prey abundance and availability can be for ground foraging bats. We found that areas with the densest ground vegetation – cereal and fodder – harboured more prey. In fact, this tendency is probably even more marked than our results suggest, because pitfall trapping tends to underestimate the abundance of insect prey in denser vegetation (Brett 1999). Considering prey abundance alone, we would conclude that these two habitats provide the most suitable foraging sites for M. myotis. However, our experiments demonstrate that such dense vegetation limits access to prey, creating a major discrepancy between the abundance and availability of prey for ground foraging bats.

Is this discrepancy sufficiently pronounced to undermine the evaluation of foraging habitat suitability? The two Habitat Suitability Index models that we generated, using prey abundance alone, and prey abundance combined with access to prey (i.e. prey availability), resulted in very different maps for the distribution of suitable habitat in the study area. Radio-tracking data demonstrated that the latter model reflects habitat use by bats much better than the former, confirming the importance of incorporating access to prey when estimating and mapping habitat suitability for ground foraging species.

Managing habitat for ground foraging bats

Bats are, overall, a highly threatened group (Mickleburgh, Hutson & Racey 2002), and their conservation usually requires the combined protection of roosts and nearby foraging grounds, which are often located within farmed and grazed land (Hutson et al. 2008). In many countries, especially within Europe and North America, there are now agri-environmental programmes that encourage farmers to adjust their practices to the needs of wildlife, thereby creating opportunities to improve the feeding habitats of threatened bat species. Our results suggest that, in the case of ground foraging bats, this should be by promoting practices that favour good access to prey, not just high prey abundance.

In Mediterranean areas, practices that result in increased density of ground vegetation cover also seem to increase the abundance of prey; but, at the same time, they reduce access to them by ground foraging bats. In our study region, the best balance between prey abundance and accessibility was achieved in grazed unimproved pastures, both with and without tree cover.

When left ungrazed for long periods of time, these pastures may reach densities that, according to our results, are incompatible with successful ground foraging by bats. However, we also determined that extensive grazing by cattle can reduce ground cover to levels that allow good access to prey. It is, therefore, clear that grazing is a potentially important tool by which to manage the habitats of ground foraging bats. In fact, moderate stocking levels of domestic livestock may have a role similar to that once played by the wild herbivores that used to roam across most grasslands (Sherow 2007).

Grazing not only reduces the density of ground vegetation but also the height. In our laboratory experiments, to avoid confounding the effect of vegetation height with that of density, we varied density while attempting to keep height approximately constant (about 10–20 cm). If the dense vegetation is simultaneously taller, as was the case in the ungrazed grasslands within the study area, then the difficulties associated with accessing prey should be even greater than those observed in these experiments.

High stocking levels may lead to overgrazing and excessive trampling, and the resulting sparse ground cover may sustain only low densities of prey (Debano 2006). Consequently, low to medium stocking are preferred, not only to avoid prey depletion but also because such conditions favour the development of a mosaic of patches with varying densities of ground vegetation (Asteraki et al. 2004; Atkinson et al. 2005). Within such mosaics, patches with less ground cover, more accessible to bats, should alternate with denser patches, which are more productive for arthropod prey. In addition to controlling the density of ground vegetation, grazing by cattle increases the availability of dung, which usually enhances the diversity and abundance of arthropods (Sanchez Pinero & Avila 2004), like scarab beetles, which are an important resource for ground foraging bats (e.g. Kalka & Kalko 2006; Kervyn & Libois 2008; Storm & Whitaker 2008).

Bat species that prey on ground dwelling arthropods are found in almost all regions of the world (Fenton et al. 1977; Faure, Fullard & Dawson 1993; Jones et al. 2003; Ratcliffe, Fenton & Shettleworth 2006; Ma et al. 2008), and many of them are dependent upon grasslands (Fenton et al. 1977; Ma et al. 2008). The conclusion that dense ground vegetation affects access to prey is likely to apply to most regions. In our study, even cover approaching 60% substantially decreased access to prey, but this could be different for other bat species, arthropod prey and types of vegetation.

Grazing is now widely recognized to be a central issue in the conservation ecology of grasslands (Watkinson & Ormerod 2001; Martin & Possingham 2005), and our results add a new dimension to this issue, by demonstrating its potential importance in the maintenance of good prey accessibility for ground foraging bats. However, where ground vegetation is naturally sparse, grazing may actually be detrimental to the abundance of arthropod prey (Woodcock et al. 2005); and there are types of grassland in which grazing does not decrease ground cover (Todd 2006). In addition, grazing may be detrimental for bat species that forage in grasslands with high vegetation. This, for example, is the case with M. blythii, a bat that mostly gleans its prey directly from grassy vegetation (Arlettaz 1996), rather than from the ground, and that often roosts together with M. myotis. Consequently, before prescribing grazing for a specific area, it is important to consider its potential impact upon other species of conservation concern, and the maintenance of mosaics of grazing intensity will often be the precautionary management approach.

Acknowledgements

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

We are grateful to those who provided assistance during the experiments and field work, and to Sofia Lourenço, Andreas Zahn, Danilo Russo, Karen Haysom and Björn Siemers for helpful comments and suggestions relating to an earlier draft of this manuscript. Ana Rainho was supported by a PhD grant from Fundação para a Ciência e Tecnologia (SFRH/BD/23800/2005).

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information
  • Arlettaz, R. (1996) Feeding behaviour and foraging strategy of free-living mouse-eared bats, Myotis myotis and Myotis blythii. Animal Behaviour, 51, 111.
  • Arlettaz, R. (1999) Habitat selection as a major resource partitioning mechanism between the two sympatric sibling bat species Myotis myotis and Myotis blythii. Journal of Animal Ecology, 68, 460471.
  • Arlettaz, R., Jones, G. & Racey, P.A. (2001) Effect of acoustic clutter on prey detection by bats. Nature, 414, 742745.
  • Asteraki, E.J., Hart, B.J., Ings, T.C. & Manley, W.J. (2004) Factors influencing the plant and invertebrate diversity of arable field margins. Agriculture, Ecosystems & Environment, 102, 219231.
  • Atkinson, P.W., Fuller, R.J., Vickery, J.A., Conway, G.J., Tallowin, J.R.B., Smith, R.E.N., Haysom, K.A., Ings, T.C., Asteraki, E.J. & Brown, V.K. (2005) Influence of agricultural management, sward structure and food resources on grassland field use by birds in lowland England. Journal of Applied Ecology, 42, 932942.
  • Audet, D. (1990) Foraging behavior and habitat use by a gleaning bat, Myotis myotis (Chiroptera, Vespertilionidae). Journal of Mammalogy, 71, 420427.
  • Bontadina, F., Schmied, S.F., Beck, A. & Arlettaz, R. (2008) Changes in prey abundance unlikely to explain the demography of a critically endangered Central European bat. Journal of Applied Ecology, 45, 641648.
  • Borkhausen, M.B. (1797) Deutsche Fauna, oder Kurzgeffaßte naturgeschichte der Thiera Deutschlands. 1th. Säugethiere und Vogel. Varrentrapp und Wenner, Frankfurt an Main.
  • Boyce, M.S., Mao, J.S., Merrill, E.H., Fortin, D., Turner, M.G., Fryxell, J. & Turchin, P. (2003) Scale and heterogeneity in habitat selection by elk in Yellowstone National Park. Ecoscience, 10, 421431.
  • Brathen, K.A. & Hagberg, O. (2004) More efficient estimation of plant biomass. Journal of Vegetation Science, 15, 653660.
  • Brett, A.M. (1999) Bias in the effect of habitat structure on pitfall traps: an experimental evaluation. Australian Journal of Ecology, 24, 228239.
  • Bulgarini, F., Calvario, E., Fraticelli, F., Petretti, F. & Sarrocco, S. (2007) Libro Rosso Degli Animali D’Italia. Vertebrati. WWF Italia, Roma.
  • Burnham, K.P. & Anderson, D.R. (2002) Model Selection and Multimodel Inference. A Practical Information-Theoretic Approach, 2nd edn. Springer-Verlag, New York.
  • Cabral, M.J.C., Almeida, J., Almeida, P.R., Dellinger, T., Ferrand de Almeida, N., Oliveira, M.E., Palmeirim, J.M., Queiroz, A.I., Rogado, L. & Santos-Reis, M. (2005) Livro Vermelho dos Vertebrados de Portugal, 1st edn. Instituto da Conservação da Natureza, Lisboa.
  • Debano, S. (2006) Effects of livestock grazing on aboveground insect communities in semi-arid grasslands of southeastern Arizona. Biodiversity and Conservation, 15, 25472564.
  • Devereux, C.L., Mckeever, C.U., Benton, T.G. & Whittingham, M.J. (2004) The effect of sward height and drainage on Common Starlings Sturnus vulgaris and Northern Lapwings Vanellus vanellus foraging in grassland habitats. Ibis, 146, 115122.
  • Devereux, C.L., Vickery, J.A., Fernandez-Juricic, E., Krebs, J.R. & Whittingham, M.J. (2006) Does sward density affect prey availability for grassland birds? Agriculture Ecosystems & Environment, 117, 5762.
  • Dodd, L.E., Lacki, M.J. & Rieske, L.K. (2008) Variation in moth occurrence and implications for foraging habitat of Ozark big-eared bats. Forest Ecology and Management, 255, 38663872.
  • FAO (2008) Grassland and Pasture – Crop systems (ed. AGPC). Available at: http://www.fao.org/ag/AGP/AGPC/doc/crops/4d.html (accessed on 19 May 2009).
  • Faure, P.A. & Barclay, R.M.R. (1992) The sensory basis of prey detection by the long-eared bat, Myotis evotis, and the consequences for prey selection. Animal Behaviour, 44, 3139.
  • Faure, P.A., Fullard, J.H. & Dawson, J.W. (1993) The gleaning attacks of the Northern Long-eared bat, Myotis septentrionalis, are relatively inaudible to moths. Journal of Experimental Biology, 178, 173189.
  • Fenton, M.B., Boyle, N.G.H., Harrison, T.M. & Oxley, D.J. (1977) Activity patterns, habitat use, and prey selection by some African insectivorous bats. Biotropica, 9, 7385.
  • Fuzessery, Z.M., Buttenhoff, P., Andrews, B. & Kennedy, J.M. (1993) Passive sound localization of prey by the Pallid Bat (Antrozous p. pallidus). Journal of Comparative Physiology a-Sensory Neural and Behavioral Physiology, 171, 767777.
  • Goerlitz, H.R., Greif, S. & Siemers, B.M. (2008) Cues for acoustic detection of prey: insect rustling sounds and the influence of walking substrate. Journal of Experimental Biology, 211, 27992806.
  • Haugland, E. & Froud-Williams, R.J. (1999) Improving grasslands: the influence of soil moisture and nitrogen fertilization on the establishment of seedlings. Journal of Applied Ecology, 36, 263270.
  • Hosmer, D.W. & Lemeshow, S. (2000) Applied Logistic Regression, 2nd edn. Wiley-Interscience, New York.
  • Hubner, M. & Wiegrebe, L. (2003) The effect of temporal structure on rustling-sound detection in the gleaning bat, Megaderma lyra. Journal of Comparative Physiology A-Neuroethology Sensory Neural and Behavioral Physiology, 189, 337346.
  • Hutson, A.M., Spitzenberger, F., Aulagnier, S., Coroiu, I., Karataş, A., Juste, J., Paunovic, M., Palmeirim, J.M. & Benda, P. (2008) Myotis myotis. 2008 IUCN Red List of Threatened Species (ed. IUCN). Available at: http://www.iucnredlist.org (accessed on 16 May 2009).
  • Jones, G., Webb, P.I., Sedgeley, J.A. & O’Donnell, C.F.J. (2003) Mysterious Mystacina: how the New Zealand short-tailed bat (Mystacina tuberculata) locates insect prey. Journal of Experimental Biology, 206, 42094216.
  • Juntii, T.M. & Rumble, M.A. (2006) Arc Habitat Suitability Index Computer Software. U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station, pp. 31.
  • Kalka, M. & Kalko, E.K.V. (2006) Gleaning bats as underestimated predators of herbivorous insects: diet of Micronycteris microtis (Phyllostomidae) in Panama. Journal of Tropical Ecology, 22, 110.
  • Kervyn, T. & Libois, R. (2008) The diet of the serotine bat – a comparison between rural and urban environments. Belgian Journal of Zoology, 138, 4149.
  • Krebs, C.J. (1989) Ecological Methodology. Harper Collins Publishers, New York.
  • Kruess, A. & Tscharntke, T. (2002) Contrasting responses of plant and insect diversity to variation in grazing intensity. Biological Conservation, 106, 293302.
  • Lollar, A. & Schmidt-French, B. (2002) Captive Care and Medical Reference for the Rehabilitation of Insectivorous Bats, 2nd edn. Bat World Publication, Mineral Wells.
  • Ma, J., Liang, B., Zhang, S. & Metzner, W. (2008) Dietary composition and echolocation call design of three sympatric insectivorous bat species from China. Ecological Research, 23, 113119.
  • Martin, T.G. & Possingham, H.P. (2005) Predicting the impact of livestock grazing on birds using foraging height data. Journal of Applied Ecology, 42, 400408.
  • Mickleburgh, S.P., Hutson, A.M. & Racey, P.A. (2002) A review of the global conservation status of bats. Oryx, 36, 1834.
  • Ottaviani, D., Lasinio, G.J. & Boitani, L. (2004) Two statistical methods to validate habitat suitability models using presence-only data. Ecological Modelling, 179, 417443.
  • Palomo, J.. (2008) Atlas y Libro Rojo de los Mamiferos Terrestres de España, 1st edn. ICONA, Madrid.
  • Pereira, M.J.R., Rebelo, H., Rainho, A. & Palmeirim, J.M. (2002) Prey selection by Myotis myotis (Vespertilionidae) in a Mediterranean region. Acta Chiropterologica, 4, 183193.
  • Pinto-Correia, T. & Vos, W. (2004) Multifunctionality in Mediterranean Landscapes – past and future. The New Dimension of the European Landscapes (ed R.Jongman), pp. 135164. Springer, Dordrecht.
  • Ratcliffe, J.M., Fenton, M.B. & Shettleworth, S.J. (2006) Behavioral flexibility positively correlated with relative brain volume in predatory bats. Brain, Behavior & Evolution, 67, 165176.
  • Russo, D. (2007) Effects of Land Abandonment on Animal Species in Europe: Conservation and Management Implications. European Commission: Office for Official Publications of the European Communities, Luxembourg.
  • Russo, D., Jones, G. & Arlettaz, R. (2007) Echolocation and passive listening by foraging mouse-eared bats Myotis myotis and M. blythii. Journal of Experimental Biology, 210, 166176.
  • Sanchez Pinero, F. & Avila, J.M. (2004) Dung-insect community composition in arid zones of south-eastern Spain. Journal of Arid Environments, 56, 303327.
  • Schnitzler, H.U., Moss, C.F. & Denzinger, A. (2003) From spatial orientation to food acquisition in echolocating bats. Trends in Ecology & Evolution, 18, 386394.
  • Sherow, J.E. (2007) The Grasslands of the United States – An Environmental History, 1st edn. ABC-CLIO, CA.
  • Siemers, B.M. & Guttinger, R. (2006) Prey conspicuousness can explain apparent prey selectivity. Current Biology, 16, R157R159.
  • Siemers, B.M. & Swift, S.M. (2006) Differences in sensory ecology contribute to resource partitioning in the bats Myotis bechsteinii and Myotis nattereri (Chiroptera: Vespertilionidae). Behavioral Ecology and Sociobiology, 59, 373380.
  • Silva, J.P., Toland, J., Jones, W., Eldridge, J., Thorpe, E. & O’Hara, E. (2008) LIFE and Europe’s Grasslands: Restoring a Forgotten Habitat. European Communities, Luxembourg.
  • Smit, H.J., Metzger, M.J. & Ewert, F. (2008) Spatial distribution of grassland productivity and land use in Europe. Agricultural Systems, 98, 208219.
  • Speakman, J.S. & Thomas, D.W. (2003) Physiological ecology and energetics of bats. Bat Ecology (eds T.H.Kunz & M.B.Fenton), pp. 430489. University of Chicago Press, Chicago.
  • Storm, J.J. & Whitaker, J.O. (2008) Prey Selection of Big Brown Bats (Eptesicus fuscus) during an emergence of 17-year Cicadas (Magicicada spp.). American Midland Naturalist, 160, 350357.
  • Todd, S.W. (2006) Gradients in vegetation cover, structure and species richness of Nama-Karoo shrublands in relation to distance from livestock watering points. Journal of Applied Ecology, 43, 293304.
  • Vickery, J.A., Tallowin, J.R., Feber, R.E., Asteraki, E.J., Atkinson, P.W., Fuller, R.J. & Brown, V.K. (2001) The management of lowland neutral grasslands in Britain: effects of agricultural practices on birds and their food resources. Journal of Applied Ecology, 38, 647664.
  • Watkinson, A.R. & Ormerod, S.J. (2001) Grasslands, grazing and biodiversity: editors’ introduction. Journal of Applied Ecology, 38, 233237.
  • Wickramasinghe, L.P., Harris, S., Jones, G. & Vaughan, N. (2003) Bat activity and species richness on organic and conventional farms: impact of agricultural intensification. Journal of Applied Ecology, 40, 984993.
  • Woodcock, B.A., Pywell, R.F., Roy, D.B., Rose, R.J. & Bell, D. (2005) Grazing management of calcareous grasslands and its implications for the conservation of beetle communities. Biological Conservation, 125, 193202.
  • Zahn, A., Haselbach, H. & Guttinger, R. (2005) Foraging activity of central European Myotis myotis in a landscape dominated by spruce monocultures. Mammalian Biology – Zeitschrift fur Saugetierkunde, 70, 265270.
  • Zahn, A., Rodrigues, L., Rainho, A. & Palmeirim, J.M. (2007) Critical times of the year for Myotis myotis, a temperate zone bat: roles of climate and food resources. Acta Chiropterologica, 9, 115125.
  • Zuur, A.F., Ieno, E.N., Walker, N.J., Saveliev, A.A. & Smith, G.M. (2009) Mixed Effects Models and Extensions in Ecology With R. Springer, New York.

Supporting Information

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

Table S1.Selection of models relating the variation of prey abundance in the study area to ground cover and habitat type, based on an information-theoretic approach

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_1820_sm_tableS1.pdf51KSupporting 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.