A. G. Popa-Lisseanu




Дата канвертавання24.04.2016
Памер396.78 Kb.
Giant noctule bats face conflicting constraints between roosting and foraging in a fragmented and heterogeneous landscape

A. G. Popa-Lisseanu1, F. Bontadina2,3 & C. Iba´ n˜ ez1


1 Don˜ ana Biological Station, CSIC, Seville, Spain

2 Institute of Ecology and Evolution, Division of Conservation Biology, University of Bern, Switzerland

3 SWILD – Urban Ecology & Wildlife Research, Zurich, Switzerland

Keywords

deforestation; fragmentation; giant noctule; habitat selection; home range; Iberia; Nyctalus lasiopterus; radiotracking.


Correspondence

Ana G. Popa-Lisseanu, Don˜ ana Biological

Station, CSIC, Seville, Spain. Tel: +34 954

23 23 40; Fax: +34 954 62 11 25

Email: anapopa@ebd.csic.es

Abstract
The tree-dwelling giant noctule Nyctalus lasiopterus, a partially carnivorous aerial- hawking bat, is one of the least known European bats, and more information is needed to evaluate its conservation status. Using radiotracking, we obtained the first data on spatial requirements and habitat preferences for the species in an area critically affected by deforestation. Two breeding populations separated by 60 km, one roosting in a city and the other in a nature reserve encircling a vast coastal marshland, showed marked differences in the size of their home ranges, but both used selectively the marshlands for foraging. Urban bats remained in the city for foraging during pregnancy in spring, but increased largely their home ranges towards the marshland during lactation in early summer. The nature reserve, with few roosting opportunities, was only a marginal roosting site for a population which switched between roosts located up to 90 km apart. Giant noctules had to travel long distances (up to 130 km recorded) to meet both foraging and roosting requirements within the fragmented landscape, exhibiting among the largest home ranges ever reported in Microchiroptera. To promote the protection of this threatened species, a network of artificial roosting options should be provided in good foraging habitats until long-term forest restoration is achieved.




Introduction
Landscape fragmentation, forest clearance and urbaniza- tion are important threats to wildlife as they eliminate or reduce suitable foraging or roosting habitats (e.g. Riley et al., 2003; Scott et al., 2006). Understanding the effect of landscape modification on wildlife is a major issue in conservation biology. However, it requires previous knowl- edge of habitat use by species and how it varies with land- scape structure, information that is often missing.

The giant noctule Nyctalus lasiopterus is included in the IUCN Red List of Threatened Species as ‘data deficient’ (IUCN, 2006), and is classified as ‘vulnerable’ in Spain (Ministerio de Medio Ambiente, 2006). It is a large obligate tree-dwelling species that forms fission–fusion societies spread across multiple trees (Popa-Lisseanu et al., 2008). The only Western-European breeding populations of N. lasiopterus occur in the southern half of the Iberian Penin- sula (Ibanez, Guillen & Bogdanowicz, 2004), which is one of the most prominent examples of human-induced, large-scale forest destruction in the world (Arribas et al., 2003).

In south-western Spain, intense deforestation across a vast area confines this large tree-dwelling species to urban gardens. Preliminary radiotracking studies conducted in

1992 and 2000 showed that some giant noctules that roosted in urban gardens traveled at least 25 km from their roosts to forage over a vast coastal marshland belonging to the Donana complex (C. Ibanez, unpubl. data). Donana is one of the most important and largest remaining wetlands in Europe (RAMSAR site information service, http://www. wetlands.org/RSDB). It is especially known as a stop-over site for migratory birds, including waders, waterfowl and songbirds, the latter constituting large part of the diet of giant noctules during spring and autumn migration (Popa- Lisseanu et al., 2007). Most giant noctules found in the Iberian Peninsula during the early 20th century were dead or exhausted individuals found on the ground in Donana (Ibanez et al., 1995). Available evidence suggested that Donana could be a crucial foraging habitat for giant noctules in south-western Spain, but where suitable roost sites could not be found.

The occupation of a bat box by a roosting group of giant noctules in the Donana National Park provided the oppor- tunity to compare the behavior and land use patterns of two populations of this rare species roosting in two extremely different environments, a park of a populated city and a protected environment in the borders of a marshland. Using radiotelemetry, we collected the first data on spatial



requirements and habitat preferences for this species. We investigated the following questions: (1) Is the marshland area of Donana an important foraging area for giant noctules in southwestern Spain, including those belonging to a distant urban population?; (2) are there differences in space use between the urban population and the one roost- ing in the protected area?; (3) does season or reproductive period influence space use?; (4) what foraging habitats are preferred?

Materials and methods


Study area
We conducted field work in the provinces of Seville, Huelva and Cadiz, in south-western Spain (Fig. 1). The area has a typical Mediterranean climate, with long dry, hot summers. Most of the land in the Guadalquivir basin is used for extensive agriculture, with few fragmented natural vegeta- tion patches which are mostly shrubby. The Guadalquivir River crosses the city of Seville and the vast coastal marsh- land complex of Donana before flowing into the Atlantic. The marshland area lacks forested vegetation and adjacent landscapes have been intensively deforested during centuries of human settlement. Riparian vegetation around the river and in the floodplain is highly modified, and no mature gallery forests exist today.

Figure 1 Map of the study area showing the two giant noctule Nyctalus lasiopterus populations studied (1: Mar´ıa Luisa park in Seville; 2: bat box in the Don˜ ana National Park, Huelva) and two other giant noctule roosting sites in Southern Spain, (3: Zoological Gardens in Jerez de la Frontera, Ca´ diz; 4: Alcornocales Nature Park, Ca´ diz, with suspected roosts), the radiotracking stations (circles) from which bearings were taken and the limits of the Don˜ ana National Park and protected surroundings (black outline). The vast wooded region to the North is part of the Sierra Morena Mountain Range. The study area is limited by the Atlantic Sea in the south-west.

Capture and radiotracking


Bats captured at maternity roosts in the Marıa Luisa Park in Seville (hereafter Seville or urban population) were radio- tracked in summer 2003, autumn–winter 2003 and spring

2004. Bats from Donana National Park (NP) were radio- tracked only in summer 2003 (Table 1). Bats were captured with mist-nets as they emerged from their roosts at dusk. Each bat was equipped with a teflon-collared radiotransmit- ter (Pip Ag392, Biotrack, Dorset, UK) designed to wear off after 1–6 months. Each bat was classified based on age, sex and reproductive status. Lactating females could be identi- fied by the hairless skin around enlarged nipples, and juveniles by the cartilaginous plates in the metacarpal–pha- langeal joints (Anthony, 1988). The transmitter was also attached to the back of the neck with surgical cement (Skin- Bond, Smith and Nephew United, Largo, FL, USA), after clipping the fur in this area to prevent the transmitter from rotating around the neck. Total mass of the transmitter, including collar and glue was 1.85 g, representing o5% of body mass (Aldridge & Brigham, 1988). All radiotracked bats were adult females, as adult males are rarely present in maternity colonies. Capture and marking of the bats were approved by the Environmental Council of the Junta de Andalucıa.


Table 1 Female giant noctules Nyctalus lasiopterus radiotagged from two study areas, showing the period from date of tagging to date of last observation and number of locations obtained for each bat


Population


Bat ID


Date of tagging

Last observation

Number of locations

Don˜ ana NP

705

11/06/2003

05/08/2003

27




709

11/06/2003

01/07/2003

4




707

11/06/2003

19/06/2003

9




710

11/06/2003

22/07/2003

7

726 1/07/2003 15/07/2003 2a

Seville

739

3/07/2003

24/07/2003

31




740

3/07/2003

25/07/2003

37




742

3/07/2003

12/07/2003

7




785

21/10/2003

04/11/2003

26




792

21/10/2003

30/10/2003

7




793

21/10/2003

04/11/2003

21




782

21/04/2004

23/06/2004

66




783

21/04/2004

23/06/2004

58




795

20/04/2004

13/05/2004

8




796

20/04/2004

23/06/2004

94




797

20/04/2004

23/06/2004

112




798

20/04/2004

27/05/2004

41




799

21/04/2004

12/05/2004

7




800

21/04/2004

04/05/2004

15




801

21/04/2004

13/05/2004

16




802

23/04/2004

23/06/2004

100




803

23/04/2004

27/05/2004

70




804

23/04/2004

10/06/2004

59




805

23/04/2004

23/06/2004

83




807

23/04/2004

29/04/2004

9


aRoosting locations.



Bats were radiotracked during their nocturnal foraging activity from a total of 30 fixed stations situated on vantage points in the periphery of Seville, Donana and Jerez (Fig. 1) with handheld telemetry receivers (FT-250 RII, Yaesu Musen Co. Ltd, Sapporo, Japan; Falcon V, Wildlife Mate- rials International Inc., IL, USA) and three-element Yagi antennae (AF Antronics Inc., IL, USA). Bearings were taken each night simultaneously from two or three stations that were selected depending on the initial flight direction of the bats to maximize radiotracking success. Bearings were recorded at 10-min intervals to avoid time-dependent auto- correlation. The position of one of the (2–4) observers was always fixed, 200 m away from the roosting site in Donana in a 30 m high tower, to detect tagged bats returning to the roosting box. Detection range was very variable (from a few kilometers up to 60 km) and probably depended on terrain and flying heights of the bats.

Analysis of home ranges


Locations of the bats were estimated through cross-triangu- lation between two or three positions using the software Locate II v1.82 (Nams, Truro, Canada) and entered into ArcView v3.2 (Environmental Systems Research Institute Inc., 1999). When the error was larger than 6 km (8% of all locations), these data were removed from every analysis. To compare home ranges, a minimum convex polygon (MCP) with all remaining locations was created for each population (urban and NP) per tracking period, using the Animal Movement extension for ArcView (Hooge & Eichenlaub,

1997), which was also used for computing the relationship between the number of locations and home range area by bootstrapping. Single locations were buffered by a distance of 2.5 km to account for mean location error and MCPs were calculated to encompass these buffers.

To test for seasonal differences in the size of home ranges, we conducted a paired t-test for those individuals tracked along a 2-month period in 2004. Non-significant deviation from normality was evaluated through Kolmogorov–Smir- nov. Two seasons were distinguished: from late April until May 31 (spring) and from June 1 until the signal was lost (early summer, Table II.1), coinciding with the pregnancy and lactation periods, respectively (Popa-Lisseanu et al.,

2008). Home ranges were represented by the 95% kernel use distribution (Worton, 1989) of the locations. The same smoothing parameter (h = 2.5 km) was used for all bats to allow for comparisons (Bontadina & Naef-Daenzer, 1996).

Habitat selection analysis
Habitat categories were inferred from the digital land-use map of the Andalusian System for Environmental Informa- tion (Moreira & Fernandez-Palacios, 1995). We grouped the

112 land use categories into four major habitat types: open land (agricultural and farmland, pastures and cleared areas); urban areas; woodland (including Mediterranean shrubs and other types of natural non-arboreal but ligneous vegeta-

tion); and riverine areas (water courses, canals, natural marshland and modified marshland for rice fields).

For habitat selection analysis, roosting locations were not included. As we could not distinguish between foraging and commuting flight, we will use the term ‘activity ranges’ instead of ‘foraging ranges.’ Locations were assigned in the field to one of three accuracy classes (high, medium and low) depending on confidence in the estimated location (see Bontadina, Naef-Daenzer & Schofield, 2002). The true accuracy was then determined from the 95% confidence area of each location, when the triangulation was performed from three fixed stations and the bearing error could be estimated; for locations obtained from two bearings, an averaged overall bearing error was applied (Locate II User’s Guide, Nams, Truro, Canada), obtained from the three- bearing triangulations when two of the telemetry stations were the same. This method was used to minimize differ- ences in bearing error due to topography. The calculations resulted in a mean location error of 2, 4 and 6 km for the three accuracy classes high, medium and low, respectively. Radiotracking locations were buffered by circles with radii of 2, 4 and 6 km depending on their assigned accuracy class to account for location error (Bontadina & Naef-Daenzer,

1996). The MCP for each bat encompassed all the buffered locations.

Habitat selection was investigated to reflect the animals’ choices at two different levels (Johnson, 1980; Porter & Church, 1987). On the first level, the animals selected an activity range within the available landscape, and on the second level, they exploited specific areas within their activity range:

(1) Selection of activity range was analyzed by comparing habitat composition within the overall activity range of each individual with habitat composition within the available landscape. The activity range was measured as the buffered MCP for each individual. The landscape considered avail- able was that area falling inside a circle around the roost with a radius equal to the maximum distance traveled from the roost (MRC or maximum range circle).

(2) Habitat selection within the activity range was analyzed by comparing the proportion of habitats used at buffered locations with the available habitat composition in its individual activity range. Habitat used took into account the intensity of use by measuring the proportion of each habitat type falling inside the buffer area of the individual locations. Habitat available was measured as the proportion of each habitat type within the MCP of the individual buffered locations.

Habitat selection was investigated through compositional analysis (Aebischer, Robertson & Kenward, 1993) using an Excel macro (Smith, 2004) which also accounted for rando- mization as recommended by Aebischer et al. (1993). We analyzed both datasets (urban and NP) together and sepa- rately. The first-level activity range selection (MRC vs. individual MCPs) could not be conducted for the NP dataset as we ignored where the bats roosted more than 70% of the time. For the urban dataset, spring (pregnancy) and summer (lactation) periods in 2004 were also analyzed separately.




Figure 2 Minimum home ranges of giant noctule Nyctalus lasiopterus bats radiotracked in summer 2003, defined by the minimum convex polygon encompassing all buffered locations. Upper polygon repre- sents the Seville population (n = 3 bats, 75 locations, shown as black dots); lower polygon designates the Don˜ ana population (n =4, 48 locations). The arrowed line shows the minimum trajectory of bat

705 in the night of July 14, 2003.

Results
Seasonal home ranges
Three bats from Seville and five bats from the NP were radiotracked in summer 2003 (Table 1). Bats from Seville, all lactating females, always returned to the site of capture in the urban park for day roosting. Individual home ranges overlapped to a large extent and its overall size stabilized after 60 (from the 75 available) locations at 1113 km2. Bats foraged up to 27 km from their roosts (Fig. 2).

Four of the five bats from the NP were lactating females (705, 707, 709 and 726). Only few locations (2–27) were recorded for each individual (Table 1). From the vantage point near the roost (high tower) from which bats could be contacted regularly up to 40 km distance, none of the bats was detectable most of the time, suggesting that they flew further away from the roost. Although we could confirm that some bats were still wearing their tags during at least 32,

30, five, two and two nights (for bats 705, 709, 707, 710 and

726, respectively; Table 1), only fast changing, very distant bearings could be obtained, and few locations by cross- triangulations were achieved. No activity locations were obtained for bat 726, but it was found roosting in two occasions in the Zoobotanical Gardens, in Jerez, about

42 km from the roost-site where it was captured (Fig. 1). Bat 705 also roosted at this site on at least four different days. The bats returned to the bat box in the NP only occasionally: out of 51 recording days, bat 705 day-roosted

at this site for nine non-consecutive days, bat 707 occupied this roost for 8 days, bat 710 for 6 days, but bats 709 and 726 never returned to the site. After July 3, no radiotagged bat ever returned to the bat box during 33 days of observation.

Bat 705 traveled a minimum of 130 km on the night of July 14 after it was located at dusk (time of emergence) in the Alcornocales Nature Park in Cadiz, 92 km from the Donana roosting site and 54 km from the roost in Jerez, where it was located the next day (Fig. 2). Bat 710 also foraged on two different nights in the Alcornocales Nature Park.

Home-range size of the NP bats was very large ( 2582 km2; Fig. 2). Maximum distance that radiotagged bats flew from a roost was 90 km, although only from the small number of locations that we could obtain. In boot- strap re-sampling, the area did not stabilize after the 48 locations available but still increased, suggesting in fact larger activity ranges.

The three bats tagged in Seville in autumn 2003 became inactive after 10 days, so that only between seven and 26 locations could be obtained (Table 1). The bats had a large home range extending to the south of Seville along the Guadalquivir River (Fig. 3b). Individual home ranges over- lapped to a large extent and while home-range size reached

1324 km2 it did not stabilize even after the maximum

number of locations that we recorded (54). Maximum distance traveled from the roosting site was c. 40 km, and the bats always returned to the urban park for day roosting. The home range of the 14 bats from Seville radiotracked

in spring 2004 was in marked contrast with the previous results, as the bats always stayed within a radius of 15 km from the roosting site (Fig. 3c). Individual home ranges overlapped to a large extent. Home-range size reached

704 km2 and stabilized after 50 of 105 available locations.

Seven bats were radiotracked through June, a period when precipitation diminished markedly in the study area (from

64.9 mm in May 2004 to 0 mm in June 2004; data from Donana Biological Reserve Monitoring Service), and coin- ciding with the lactation period. Individual home ranges overlapped to a large extent. Overall home range in June (early summer) stabilized after 100 of 195 recorded locations within an area of 1555 km2; maximum distance traveled from the roost was 40 km (Fig. 3d). A paired t-test confirmed that the increase in size of the bats’ seasonal

home ranges from spring to early summer was statistically significant (paired t-test; t6 =—3.132, P = 0.02). In both seasons, bats always returned to the urban park for day roosting.


Habitat selection
Selection of activity range
Percentage habitat composition of individual activity ranges, compared with that of the available landscape with- in a circle (MRC) with radius of 40 km, was non-random (Seville all: n = 20, Wilks’ l = 0.0710, w2 = 52.91, d.f. = 3, Po0.001; Seville spring 2004: n = 14, Wilks’ l = 0.0019, w2 = 88.10, d.f. = 3, Po0.001; Seville summer 2004: n =7,



Figure 3 Point locations and overall home ranges of all bats from Seville radiotracked throughout the study period; (a) summer 2003 (n = 3 bats); (b) autumn 2003 (n = 3); (c) spring

2004 (n = 14); (d) summer 2004 (n =7).

Wilks’ l = 0.0194, w2 = 27.61, d.f. = 3, P = 0.015). Urban areas were on average used more often than they were available; riverine areas were used more often than they were available except for spring 2004 (Fig. 4). Open land and woodland were on average always used below availability. Woodlands were significantly avoided; urban areas and open land were significantly selected over riverine areas, except during summer of 2004 (Table 2).

Habitat selection within the activity range
The selection of habitats within the individual activity ranges of all bats, Seville bats in the spring 2004 and Seville bats in the summer 2004 were significantly different from their availability (all: n = 24, Wilks’ l = 0.3866, w2 = 24.71, d.f. = 3, Po0.001; Seville all: n = 20, Wilks’ l = 0.3978, w2 = 22.13, d.f. = 3, Po0.001; Seville spring 2004: n = 14, Wilks’ l = 0.0852, w2 = 44.32, d.f. = 3, Po0.001; Seville summer 2004: n = 7, Wilks’ l = 0.1877, w2 = 18.40, d.f. = 3, P = 0.005). Urban and riverine areas were used, on average, more than was available (Fig. 5). Open land and woodland, the first of which was the main habitat type ( 60%) within



Figure 4 Comparisons of habitat proportions within an MRC with a radius of 40 km (mean percentage area; habitat available) versus habitat proportions within the individual MCPs (mean percen- tage area+SD; habitat used) for Seville bats, all grouped together and separated by season, that is spring 2004 and summer 2004 (habitat available is the same for all groups). MCP, minimum convex polygon.

Table 2 Habitat ranking matrix derived from compositional analysis, from most preferred to least preferred, on a first level habitat selection



(selection of activity ranges within the landscape) and on a second level habitat selection (habitat selection within the activity range)


Level

Group

Habitat ranking

First level

Seville MRC (40 km)

UrbanadOpenb4RiverinebdWoodc




Seville spring 2004 MRC

UrbanadOpenbdRiverinecdWoodd




Seville summer 2004 MRC

Urbana4Riverineab4OpenbdWoodc

Second level

All (Seville and Don˜ ana)

Riverinea4UrbanadWoodb4Openb




Seville

Riverinea4UrbanadWoodb4Openb




Seville spring 2004

RiverineadUrbanbdOpenc4Woodc




Seville summer 2004

RiverineadUrbanbdWoodc4Openbc

Values marked with three signs (d) indicate significant deviation from zero at P = 0.01. Habitat categories that are not significantly different from one another are indicated with the same lower-case letter.




Figure 5 Comparisons of habitat available within the individual buf- fered MCPs (mean percentage area+SD; black line) versus habitat used defined as mean habitat composition inside the individual buffered locations (gray line). 1: complete dataset; 2: Seville; 3: Seville spring 2004; 4: Seville summer 2004.

the activity ranges of the bats, were used on average below availability. Riverine areas and urban areas were signifi- cantly selected over all other habitat types (Table 2). During spring and summer 2004, riverine/marshland was signifi- cantly selected over urban areas. Differences in selection of woodland and open land were non-significant.


Discussion
Home-range size varied seasonally and was strikingly differ- ent between the two populations studied, urban and NP. Our results support the key importance of the Donana marshland as foraging grounds for giant noctule popula- tions spread across a vast area. Contrary to our predictions, giant noctules from the NP exhibited the largest home range, almost three times that of urban bats in the same season, and 15% larger than the overall home range of urban bats throughout the whole study period. Its size, larger than 2500 km2, is among the largest reported for any microchiropteran species, and represents a conservative

estimate as bats were absent from the study area most of the time. One of the radiotagged giant noctules flew an

astounding distance of more than 130 km in a single night. So far, daily foraging movements of over 50 km have only been reported for Tadarida brasiliensis (Best & Geluso,

2003).


Home ranges of urban bats showed considerable seasonal variation, being minimal in spring 2004 (within 15 km from the roost) and maximal in late autumn 2003 and early summer 2004 (within 40 km from the roost). Bats enlarged their home ranges southwards during June, coinciding with the lactation period. Energetic demands are higher during lactation than during pregnancy (Anthony & Kunz, 1977; Racey & Speakman, 1987; Kurta, Kunz & Nagy, 1990), and as a response some bats spend more time foraging (Barclay, 1989). However, the opposite was found in several bat species, with the largest distances traveled by non- reproductive females (Entwistle, Racey & Speakman, 1996; O’Donnell, 2001); common noctules in England did not show differences in home-range size relative to reproductive state (Mackie & Racey 2007). Apart from reproductive condition, seasonality in climatic conditions linked to insect abundance in the study area, with negative water balance between May and October (Prenda, Lopez-Nieves & Bravo,

2001) and peak insect abundance in spring (Herrera, 1988; Sanchez-Pinero & Avila, 2004), could force bats to travel longer distances in search of insect prey during summer. High energetic demands could also result in large ranges during autumn because of the need to accumulate fat reserves before hibernation, or because of low insect abun- dance due to colder temperatures. The diet of giant noctules experiences a major shift toward nocturnally migrating passerines in autumn (Ibanez et al., 2001; Popa-Lisseanu et al., 2007). Although the density distribution of this resource is not known at regional scales, its spatial variation, including the concentrations of birds departing from stop- over sites at dusk, could have influenced the foraging move- ments of this species.

On a landscape level, urban bats avoided woodland habitats during their nocturnal flights. On a core area level, all bats used riverine and urban areas more intensively within their MCPs. The selection of urban areas might be an artifact owing to the location of roosting sites, and because we could not differentiate between commuting and foraging. Hunting around city lights, a behavior commonly



reported for several bat species living in urban areas (e.g. Kronwitter, 1988; Lee & McCracken, 2002; Haupt, Menzler

& Schmidt, 2006), has never been observed for giant noctules. Riverine areas were in all cases the most used habitat type relative to their availability. Bats followed the river when moving south (Fig. 3b and d), and the marsh- lands in Donana were used by both urban and NP popula- tions. A preference for this habitat could respond to the high insect availability of riparian environments compared with all other habitats in semi-arid ecosystems (Gregory et al.,

1991). The arid conditions during the Mediterranean sum- mer could be the cause why bats from the urban population enlarged their home ranges toward the marshes. Lactation could also motivate bats to search for more productive habitats such as marshes to meet increased energetic de- mands (Anthony & Kunz, 1977; Racey & Speakman, 1987; Kurta et al., 1990). Both woodland and open areas, the latter being the most abundant habitat type within the MCPs of the bats, were negatively selected. The avoidance of woodland as foraging habitat by giant noctules contrasts with the preference for this habitat type by common noc- tules in Great Britain (Mackie & Racey, 2007). Open areas, mostly agricultural, were used with low intensity, suggesting that agricultural landscapes do not represent optimal fora- ging grounds for N. lasiopterus in the study area.

The low fidelity shown by bats in the NP to their roosting areas compared to the urban population can be explained by the roosting dynamics and social organization of giant noctules. This species forms fission–fusion colonies dividing every day into smaller roosting groups that experience continuous re-mixing (Popa-Lisseanu et al., 2008). In fact, the population of Seville used at least 73 roost trees during the course of this study. Likewise, bats using the box in the NP constituted a ‘roosting group’ belonging to one or to several colonies whose main roosts were most likely located outside the area of Donana. This is supported by the fact that some of the radiotracked bats roosted on several occasions in a known giant noctule roost in Jerez 42 km away, and at least once in the area of the Alcornocales Nature Park (around 90 km away). Our results suggest that the larger home ranges of giant noctules from the NP compared with those from the urban are a consequence of the roost-switching behavior typical of this species, in an area where foraging is optimal but where roosts are a scarce and dispersed resource.

The availability and distribution of resources is likely to affect the activity of giant noctule bats on two fronts: (1) roosting sites; (2) adequate foraging grounds. In south- western Spain, these two resources are segregated, as (1) the area around Donana and the Guadalquivir marshlands, which were positively selected by the bats during their foraging activity, currently lacks sufficient roosting sites for a tree-dwelling bat that exhibits roost-switching; (2) the city of Seville, where a historic park offers a high density of roosts for giant noctules, is not an optimal foraging ground throughout the year. For this reason, this species must travel long distances on a daily basis to reach its foraging grounds or for moving between alternative roosting sites. This might

not be optimal, as flying involves increased energetic costs for bats (Racey & Speakman, 1987).

South-western Spain, with the largest populations known worldwide of giant noctules, is a crucial geographical area for this threatened species and thus deserves attention and concerted conservation efforts. However, current park man- agement practices that cut down old and hollow trees are seriously menacing the last shelters of these bats in this urban environment (Popa-Lisseanu et al., 2008). We recom- mend the creation of a network of artificial roost sites in Donana within the limits of the NP. This would reduce commuting distances between potential tree roosts and between roosts and foraging grounds, presumably reducing fitness costs as well as the subsequent risks of mortality during long flights. As the NP is banned to human access, it would also provide safe roosting sites for the bats. This would be a temporary solution until the natural vegetation in Donana has recovered and mature forests with suitable cavities become available for the bats. Given the importance of Donana as foraging habitat for the largest known giant noctule populations, these measures could have significant implications for the conservation of this rare species.

Acknowledgments


We thank C. Ruiz, O. Mora and I. Cardiel for their participation in the field work and to J. Juste and three anonymous reviewers for substantially improving the paper. Research was funded by the Organismo Autonomo de Parques Nacionales (MMA) of Spain, project 021/2002. MEC of Spain funded A.G. Popa-Lisseanu (fellowship AP-

2002-3721).

References
Aebischer, N.J., Robertson, P.A. & Kenward, R.E. (1993).

Compositional analysis of habitat use from animal radio-tracking data. Ecology 74, 1313–1325.

Aldridge, H.D.J.N. & Brigham, R.M. (1988). Load carrying and maneuverability in an insectivorous bat: a test of the

5% ‘‘rule’’ of radio telemetry. J. Mammal. 69, 379–383. Anthony, E.L.P. (1988). Age determination in bats. In Ecologi-

cal and behavioral methods for the study of bats: 47–58. Kunz, T.H. (Ed.). Washington: Smithsonian Institution Press.

Anthony, E.L.P. & Kunz, T.H. (1977). Feeding strategies of the little brown bat, Myotis lucifugus, in southern New Hampshire. Ecology 58, 775–786.

Arribas, A., Gallardo, C., Gaertner, M.A. & Castro, M. (2003). Sensitivity of the Iberian Peninsula climate to land degradation. Clim. Dyn. 20, 477–489.

Barclay, R.M.R. (1989). The effect of reproductive condition on the foraging behavior of female hoary bats, Lasiurus cinereus. Behav. Ecol. Sociobiol. 24, 31–37.

Best, T.L. & Geluso, K.N. (2003). Summer foraging range of Mexican free-tailed bats (Tadarida brasiliensis mexicana) from Carlsbad Cavern, New Mexico. Southwest. Nat. 48, 590–596.



Bontadina, F. & Naef-Daenzer, B. (1996). Analysing spatial data of different accuracy: the case of greater horseshoe bats foraging. In Wildlife – biotelemetry. Le Maho, Y. (Ed.). Strasbourg: CNRS-CEPE http://www.rhinolophus. net/wildtele.pdf.

Bontadina, F., Naef-Daenzer, B. & Schofield, H. (2002).

Radio-tracking reveals that lesser horseshoe bats (Rhinolophus hipposideros) forage in woodland. J. Zool. (Lond.) 258, 281–290.

Entwistle, A.C., Racey, P.A. & Speakman, J.R. (1996).

Habitat exploitation by a gleaning bat, Plecotus auritus.

Philos. Trans. Roy. Soc. B 351, 921–931.

Gregory, S.V., Swanson, F.J., McKee, W.A. & Cummins, K.W. (1991). An ecosystem perspective of riparian zones. BioScience 41, 540–551.

Haupt, M., Menzler, S. & Schmidt, S. (2006). Flexibility of habitat use in Eptesicus nilssonii: does the species profit from anthropogenically altered habitats? J. Mammal. 87, 351–361.

Herrera, J. (1988). Pollination relationships in southern

Spanish Mediterranean shrublands. J. Ecol. 76, 274–287. Hooge, P.N. & Eichenlaub, B. (1997). Animal movement

extension to ArcView ver. 1.1. Anchorage, AK: Alaska

Biological Science Center, US Geological Survey.

Ibanez, C., Guillen, A. & Bogdanowicz, W. (2004). Nyctalus lasiopterus (Schreber, 1780) – Riesenabendsegler. In Handbuch der Sa¨ ugetiere Europas, Vol. 4: 695–716. Krapp, F. (Ed.). Wiesbaden: AULA-Verlag.

Ibanez, C., Guillen, A., Juste, J., Migens, E., Perez, J.L. & Ruiz, C. (1995). Quiropteros del Parque Nacional de Donana: Especies, y Tamano y Metodos de Estima de sus Poblaciones. ICONA-CSIC, Sevilla 114.

Ibanez, C., Juste, J., Garcıa-Mudarra, J.L. & Agirre-Mendi, P.T. (2001). Bat predation on nocturnally migrating birds. Proc. Natl. Acad. Sci. USA 98, 9700–9702.

IUCN. (2006). Red list of threatened species. http://www. iucn.org/themes/ssc/redlist2006/redlist2006.htm

Johnson, D.H. (1980). The comparison of usage and avail- ability measurements for evaluating resource preference. Ecology 61, 65–71.

Kronwitter, F. (1988). Population structure, habitat use and activity patterns of the noctule bat, Nyctalus noctula Schreb., 1774 (Chiroptera: Vespertilionidae) revealed by radio-tracking. Myotis 26, 23–85.

Kurta, A., Kunz, T.H. & Nagy, K.A. (1990). Energetics and water flux of free-ranging big brown bats (Eptesicus fuscus) during pregnancy and lactation. J. Mammal. 71, 59–65.

Lee, Y.F. & McCracken, G. (2002). Foraging activity and food resource use of Brazilian free-tailed bats, Tadarida brasiliensis (Molossidae). Ecoscience 9, 306–313.

Mackie, I. & Racey, P. (2007). Habitat use varies with reproductive state in noctule bats (Nyctalus noctula): implications for conservation. Biol. Conserv. 140, 70–77.

Ministerio de Medio Ambiente. (2006) Catalogo Nacional de Especies Amenazadas http://www.mma.es/portal/secciones/ biodiversidad/especies_amenazadas/catalogo_especies/ acceso_catalogo.htm.

Moreira, J.M. & Fernandez-Palacios, A. (1995) Usos y coberturas vegetales del suelo en Andalucı´a. Seguimiento a trave´s de ima´ genes de sate´lite (Land-use and land-cover in Andalusia. Monitoring from satellite imagery). Junta de Andalucıa, Consejerıa de medio ambiente.

O’Donnell, C.F.J. (2001). Home range and use of space by Chalinolobus tuberculatus, a temperate rainforest bat from New Zealand. J. Zool. (Lond.) 253, 253–264.

Popa-Lisseanu, A.G., Bontadina, F., Mora, O. & Ibanez, C. (2008). Highly structured fission–fusion societies in an aerial-hawking, carnivorous bat. Anim. Behav. 75, 471–482 (Online DOI: 10.1016/j.anbehav.2007.05.011).

Popa-Lisseanu, A.G., Delgado-Huertas, A., Forero, M., Rodrıguez, A., Arlettaz, R. & Ibanez, C. (2007). Bats’ conquest of a formidable foraging niche: the myriads of nocturnally migrating songbirds. PLoS ONE 2, e205 (Online DOI: 10.1371/journal.pone.0000205).

Porter, W.F. & Church, K.E. (1987). Effects of environmental pattern on habitat preference analysis. J. Wildl. Mgmt. 51,

681–685.


Prenda, J., Lopez-Nieves, P. & Bravo, R. (2001). Conserva- tion of otter (Lutra lutra) in a Mediterranean area: the importance of habitat quality and temporal variation in water availability. Aq. Conserv.: Mar. Freshw. Ecos. 11,

343–355.


Racey, P.A. & Speakman, J.R. (1987). The energy costs of pregnancy and lactation in heterothermic bats. Symp. Zool. Soc. Lond. 57, 107–125.

Riley, S.P., Sauvajot, R.M., Fuller, T.K., York, E.C., Kamradt, D.A., Bromley, C. & Wayne, R.K. (2003). Effects of urbanization and habitat fragmentation on bobcats and coyotes in southern California. Conserv. Biol. 17, 566–576.

Sanchez-Pinero, F. & Avila, J.M. (2004). Dung-insect community composition in arid zones of south-eastern Spain. J. Arid Environ. 56, 303–327.

Scott, D.M., Brown, D., Mahood, S., Denton, B., Silburn, A.

& Rakotondraparany, F. (2006). The impacts of forest clearance on lizard, small mammal and bird communities in the arid spiny forest, southern Madagascar. Biol. Conserv. 127, 72–87.

Smith, P.G. (2004). Automated log-ratio analysis of compositional data: software suited to analysis of habitat preference from radiotracking data. Bat Res. News



45, 16.

Worton, B.J. (1989). Kernel methods for estimating the utilization distribution in home-range studies. Ecology 70,



164–168.





База данных защищена авторским правом ©shkola.of.by 2016
звярнуцца да адміністрацыі

    Галоўная старонка