Biodiversity and Conservation

, Volume 18, Issue 9, pp 2361–2371

Do wooded streets provide connectivity for bats in an urban landscape?

Authors

    • Department of Biological, Laboratory of Chiroptera StudiesFederal University of Espírito Santo
    • Division of Mammals, MRC 108National Museum of Natural History, Smithsonian Institution
  • Poliana Mendes
    • Department of Biological, Laboratory of Chiroptera StudiesFederal University of Espírito Santo
  • Thiago B. Vieira
    • Department of Biological, Laboratory of Chiroptera StudiesFederal University of Espírito Santo
  • Albert D. Ditchfield
    • Department of Biological, Laboratory of Chiroptera StudiesFederal University of Espírito Santo
Original Paper

DOI: 10.1007/s10531-009-9593-7

Cite this article as:
Oprea, M., Mendes, P., Vieira, T.B. et al. Biodivers Conserv (2009) 18: 2361. doi:10.1007/s10531-009-9593-7
  • 449 Views

Abstract

The effects of urbanization on bats are poorly understood, but published data suggests it might be detrimental to them. Even though urban parks provide refuge to native biota, the nature of the urban landscape exacerbates the insularization process. In order to evaluate if wooded streets in an urban landscape provide connectivity for bats, we compared bat community structure in three different types of habitats: urban parks, wooded streets and non-wooded streets. Sampling occurred monthly from August 2006 to July 2007 in the city of Vitória, southeastern Brazil. Richness, relative abundance and diversity were higher in urban parks and lower in non-wooded streets. Jaccard’s similarity index showed that the wooded streets are more similar to non-wooded streets than to urban parks. Urbanization may benefit generalist species by providing new resources, but for specialist species critical resources may be lost and persistence endangered. There is evidence that wooded streets may provide some degree of connectivity for birds in urban landscapes, but our results suggest that this is not the case, with wooded streets being used by few individuals of a few species. Vegetation cover is important to maintain bat diversity in urban centers. Activities like landscape planning and gardening should include biodiversity data in their outputs in order to better design a landscape that improves the likelihood of persistence of bats.

Keywords

ChiropteraCorridorsHabitat fragmentationHabitat useUrbanizationUrban planning

Introduction

Among the many human activities that cause habitat loss (Czech et al. 2000; Bailie et al. 2004), urban development produces some of the highest rates of local extinctions and frequently eliminates the large majority of native species (Vale and Vale 1976; Luniak 1994). Also, urbanization is often more lasting than other types of habitat loss (McKinney 2002). Urbanization changes landscapes and species composition, and is generally considered to have negative effects on natural ecosystem processes (Czech and Krausman 1997; McKinney 2006). It turns natural continuous areas into disjunct and dispersed fragments embedded in a human-dominated landscape (Kammesheidt 2002; Tabarelli and Gascon 2005). Such habitat fragments are valuable not only as refuges, but they also provide connectivity, acting as stepping-stones for native wildlife within the anthropic matrix. They also provide an opportunity to address ecological principles such as landscape or gradient effects of urbanization (McDonnell and Pickett 1990; Fahrig 2003).

It is probably intuitive to even the most casual observer that the increasing fragmentation of natural habitat by human disturbances, like it happens in urban centers, will tend to reduce the species richness. There are, however, many variables that can affect the rate and pattern of species loss in the urban landscape, so empirical studies are crucial in measuring such impacts on wildlife (McKinney 2002). For many taxa, studies document that the lowest species diversities along the urban–rural gradient occur in the environments of the urban core (e.g., Mackin-Rogalska et al. 1988; Kowarik 1995; Denys and Schmidt 1998; McIntyre 2000). The novel conditions produced by urbanization may benefit generalist and exotic species by providing extra food, shelter and/or breeding sites (McKinney 2006). On the other hand, specialist species may be adversely affected if particular key resources are modified or removed (Fahrig 2003) and distances between suitable habitat patches become too great to allow for colonization (Hanski and Singer 2001).

In urban landscapes, wooded streets could be regarded as structurally similar to corridors because they are vegetative linear strips connecting urban parks (Fernández-Juricic 2000). Evidence suggests that avian richness and diversity tend to decrease with urbanization and to increase with a higher proportion of green areas in urban landscapes (Fernández-Juricic 2000).

The effects of urbanization on bats, particularly on a landscape scale, are poorly understood (Gehrt and Chelsvig 2003). Previous research suggests it may be detrimental to bats (Geggie and Fenton 1985; Kurta and Teramino 1992; Rydell et al. 1994; Walsh et al. 1995; Walsh and Harris 1996; Vaughan et al. 1997; Gaisler et al. 1998; Ávila-Flores and Fenton 2005), resulting in dramatic declines in diversity and abundance (Pierson 1998). Nevertheless, bats do occur in cities (Geggie and Fenton 1985; Brigham et al. 1989; Esbérard 2004; Ávila-Flores and Fenton 2005), and some artificial substitutes of the habitat requirements of bats (e.g., roost sites and sources of water) are present in urban areas (Everette et al. 2001). In addition, the effects of urbanization on bats may be context-specific (Fenton 1997; Gehrt and Chelsvig 2004); they may depend on the condition of the habitat and the pattern of development.

The Atlantic Forest is one of the world’s 34 biodiversity hotspots (Mittermeier et al. 2005) and one of the most threatened ecosystems on the planet (Myers et al. 2000; Mittermeier et al. 2005). Habitat fragmentation and loss have severely altered most of the Atlantic Forest, and only about 8% of its original area is left (Mittermeier et al. 2005). The Atlantic Forest houses approximately 70% of Brazil’s 169 million people, with a human population density estimated at 74.5 million people/km2 (Galindo-Leal and Câmara 2003). About 80% of Brazil’s gross domestic product is generated in the Atlantic Forest, and it shelters Brazil’s largest industrial and silvicultural centers (Galindo-Leal and Câmara 2003). In southeastern Brazil, the state of Espírito Santo is the region where bat distribution, biogeography and ecology is less studied and known. Due to the rapid pace of urbanization in the state, it is of paramount importance to gather biological knowledge about its native bats in order to create and implement conservation and management plans. Our objective is to evaluate if wooded streets are used by bats, therefore providing movement pathways and better connectivity within an urban area in southeastern Brazil.

Materials and methods

Study area

The city of Vitória (20°19′19′′S, 40°20′17′′W) is the capital of the state of Espírito Santo, in southeastern Brazil. The city is comprised of an island with 81 km2 and adjacent continental areas, and human population is estimated at 1.8 million. Vitória’s foundation dates from 1551, but until the early 1900s urbanization was insignificant. Urbanization accelerated in the 40s with the destruction of several mangroves, beaches and restingas in coastal areas and deforestation of Atlantic Forest inland. In the 60s urbanization expanded from the island of Vitória to continental adjacent areas, including the construction of an airport and a harbor, besides bridges connecting the island to the continent. In the 80s atmospheric pollution increased with the establishment of two factories (Companhia Vale do Rio Doce and Siderúrgica de Tubarão).

Despite its high rates of urbanization, Vitória is an area listed as a high priority site for biodiversity conservation in the state of Espírito Santo (IPEMA 2005). It has eight municipal parks, where forest remnants can provide refuge to wildlife, including bats. We chose three to be sampled in this study: (1) Pedra da Cebola Municipal Park (20°16′39′′S, 40°17′45′′W), with an area of 100,005 m2 was originally a transition zone between the restinga (coastal shrubland) ecosystem and the Atlantic Forest (Oprea et al. 2007); (2) Horto de Maruípe Municipal Park (20°17′44′′S, 40°18′47′′W), with an area of 60,000 m² covered by native Atlantic Forest vegetation and many introduced exotic species; and (3) Fazendinha Municipal Park (20°14′30′′S, 40°16′23′′W) with 22,853.78 m², also a transition zone between the restinga (coastal shrubland) ecosystem and the Atlantic Forest, with some exotic species (Fig. 1). Three wooded (20°18′12′′S, 40°17′29′′W; 20°18′21′′S, 40°18′36′′W; and 20°14′35′′S, 40°16′25′′W) and non-wooded streets (20°18′26′′S, 40°18′45′′W; 20°14′54′′S, 40°16′17′′W; and 20°18′51′′S, 40°17′36′′W) also were sampled. Wooded streets are characterized by the presence of trees, both native and exotic species (Fig. 1). The non-wooded streets are habitats under high anthropic influence, basically characterized by buildings and with no trees (Fig. 1).
https://static-content.springer.com/image/art%3A10.1007%2Fs10531-009-9593-7/MediaObjects/10531_2009_9593_Fig1_HTML.gif
Fig. 1

Map showing the sampling sites within the city of Vitória: urban parks (triangle), wooded streets (squares), and non-wooded streets (circles)

Sampling sessions

In the whole project there were 12 sampling sessions, occurring monthly from August 2006 to July 2007. Each sampling session comprised 3 days of work (one in each habitat). In each site we used eight mist nets (6 nets with 9.0 × 2.5 m and 2 nets with 7.0 × 2.5 m). Due to safety reasons, nets were opened at sunset and stayed opened for 4 h. Sampling effort was calculated following Straube and Bianconi (2002). During sampling sessions, we also conducted active search for bat roosts, and used an ultrasound detector Pettersson D240 (Pettersson Elektronik, AB Upsala) on Heterodyne system at 35 Hz.

The following data was recorded for all captured bats: species, sex, age, reproductive condition (for females: pregnant, lactating, post-lactating, and inactive; for males: active or inactive), weight, and forearm length. All captured bats were individually marked with collars for identification. These collars had colored plastic rings, which were used to organize an individual color code based on a method described by Esbérard and Daemon (1999).

Community structure

In order to study how feeding guilds were structured within the bat community in Vitória, we did our classification on guild structure based on that proposed by Kalko et al. (1996): aerial insectivores, gleaning insectivores, frugivores, nectarivores, carnivores, piscivores, sanguinivores and omnivores. We took into account sampling effort and cumulative species richness to construct a collector curve, a curve demonstrating diminishing returns over time, for each area (urban parks, wooded streets and non-wooded street) and for the study as a whole. In order to assess similarity between these different habitats, we used Jaccard’s similarity index (Ludwig and Reynolds 1988). Bat diversity was calculated for each area (park, wooded street and non-wooded street) using the Shannon’s diversity index (Ludwig and Reynolds 1988).

Results

We obtained a total of 174 captures of 172 individuals, representing ten bat species in four different families (Table 1). Phyllostomidae was the most abundant family. Individuals of the family Noctilionidae were observed while foraging. Using an ultrasound detector we could detect individuals from the family Molossidae, but we could not identify which species. In each sampling area our daily effort was 680 m2 h. Our total sampling effort was 21,080 m2 h. One individual of P. lineatus and one individual of A. lituratus were recaptured in the same night and same area that they were first captured.
Table 1

Number of individuals and captures for the species recorded in the urban landscape of Vitória

Family

Subfamily

Species

Captures

Individuals

Phyllostomidae

Glossophaginae

Glossophaga soricina

n = 11

n = 11

Stenodermatinae

Artibeus lituratus

n = 115

n = 114*

 

Uroderma magnirostrum

= 1

= 1

 

Plathyrrinus lineatus

n = 25

n = 24*

Phyllostominae

Phyllostomus discolor

n = 8

n = 8

Vespertilionidae

Myotinae

Myotis nigricans

n = 6

n = 6

 

Myotis albescens

n = 1

n = 1

Molossidae

Molossinae

Molossus molossus

n = 1

n = 1

 

Nyctinomops laticaudatus

n = 5

n = 5

Emballonuridae

Emballonurinae

Peropteryx macrotis

n = 1

n = 1

Noctilionidae

 

Noctilio sp.

Visual obs.

Visual obs.

* Recaptures

Richness and abundance were higher in urban parks, followed by wooded streets, and lower in non-wooded streets (Table 2). Although the urban parks are open to the public, there are many trees, small lakes and, other resources within them. A. lituratus was the only species captured in all three habitats, but its abundance decreases along the park-wooded street-non-wooded street gradient (Table 2). This was the only species captured in non-wooded streets (Table 2). A. lituratus and P. lineatus were the only species found in all three parks and wooded streets. The great majority of the species captured were only recorded in the urban parks (Table 2).
Table 2

Richness and abundance in urban parks, wooded streets, and non-wooded streets

Species

Urban parks

Wooded streets

Non-wooded streets

Total

Glossophaga soricina

11 (11)

11 (11)

Artibeus lituratus

72 (71)

40 (40)

3 (3)

115 (114)

Uroderma magnirostrum

1 (1)

1 (1)

Plathyrrhinus lineatus

19 (18)

6 (6)

25 (24)

Phyllostomus discolor

8 (8)

8 (8)

Myotis nigricans

6 (6)

6 (6)

Myotis albescens

1 (1)

 

1 (1)

Molossus molossus

1 (1)

 

1 (1)

Nyctinomops laticaudatus

5 (5)

 

5 (5)

Peropteryx macrotis

1 (1)

 

1 (1)

Noctilio sp.

Visual obs.

Visual obs.

Total

125 (123)

46 (46)

3 (3)

 

The numbers of individuals are in brackets

The collector curve for the study as a whole showed a continuous increase, and it is possible that a greater effort would result in additional species being captured within the city’s boundaries (Fig. 2). We also constructed a collector’s curve for each habitat (urban parks, wooded streets and non-wooded streets), and it is noticeable that the curves for wooded and non-wooded streets quickly reached their asymptote and stabilized (Fig. 3). However, the curve for urban parks is still increasing and indicates that new species may be found in Vitória.
https://static-content.springer.com/image/art%3A10.1007%2Fs10531-009-9593-7/MediaObjects/10531_2009_9593_Fig2_HTML.gif
Fig. 2

Species accumulation curve for the species recorded in the urban landscape of Vitória

https://static-content.springer.com/image/art%3A10.1007%2Fs10531-009-9593-7/MediaObjects/10531_2009_9593_Fig3_HTML.gif
Fig. 3

Species accumulation curve for each habitat: urban parks (https://static-content.springer.com/image/art%3A10.1007%2Fs10531-009-9593-7/MediaObjects/10531_2009_9593_Figa_HTML.gif ), wooded streets (https://static-content.springer.com/image/art%3A10.1007%2Fs10531-009-9593-7/MediaObjects/10531_2009_9593_Figb_HTML.gif ), and non-wooded streets (https://static-content.springer.com/image/art%3A10.1007%2Fs10531-009-9593-7/MediaObjects/10531_2009_9593_Figc_HTML.gif )

We found day roosts for A. lituratus and P. lineatus in palm tree leaves near the sampling areas. During the searches for roosts we observed some Noctilionidae bats foraging near the Federal University of Espírito Santo (Universidade Federal do Espírito Santo) campus, below a bridge connecting Vitória’s island to the continent, and on a small lake in Fazendinha Municipal Park, but we were not able to capture them. These bats can be recognized by their large size, red color, agile flight, and behavior of dipping their feet and skimming the water as they fly (Emmons and Feer 1997).

Regarding the feeding guilds (Kalko et al. 1996), frugivores (A. lituratus, Uroderma magnirostrum and P. lineatus) represented the great majority of individuals captured, whereas aerial insectivores (Myotis nigricans, Myotis albescens, Molossus molossus and Peropteryx macrotis), gleaning insectivores (Nyctinomops laticaudatus), nectarivores (G. soricina) and omnivores (Phyllostomus discolor) comprised a small portion of the bat community in Vitória (Fig. 4). Noctilionids were excluded from this analysis because they were recorded only through direct observation and we do not have a quantifiable number of individuals.
https://static-content.springer.com/image/art%3A10.1007%2Fs10531-009-9593-7/MediaObjects/10531_2009_9593_Fig4_HTML.gif
Fig. 4

Proportion of bats in each feeding guild based in Kalko et al. (1996)

The urban parks showed higher bat diversity (H’ = 1.415) than wooded streets (H’ = 0.387). Since only one species was captured in non-wooded streets, Shannon–Wiener’s diversity index for this habitat equaled to zero. Jaccard’s similarity index showed that the wooded and non-wooded streets are more similar to each other than to the urban parks. The lower similarity was observed between urban parks and non-wooded streets (Table 3). These results may reflect the higher richness and abundance found in urban parks compared to smaller richness and abundances found in both wooded and non-wooded streets.
Table 3

Jaccard similarity index for each sampld area in the urban landscape of Vitória

 

Urban parks

Wooded streets

Non-wooded streets

Urban parks

  

Wooded streets

0.273

 

Non-wooded streets

0.100

0.500

Discussion

The species pool that occurs within the study area region is much larger than the 10 species recorded for the urban landscape of Vitória, with 36 species recorded in the state of Espírito Santo so far (Reis et al. 2007). There are very few sites where bat surveys were conducted within Espírito Santo, but some inventories provide data that suggests a lower richness and diversity for the urban landscape of Vitória. For example, a total of 14 species were recorded for Paulo Cesar Vinha State Park (Oprea 2006); Fonte Grande State Park has 17 species (Ditchfield, unpublished data). The numbers of bat species recorded for protected areas in Espírito Santo are higher than the richness observed within Vitória’s urban limits, and diversity is also greater in protected areas (Oprea et al. 2007). Our results are in accordance with previous studies on bat communities in urban environments that show typically lower species richness and abundance in urban habitats than those in less disturbed habitats (Geggie and Fenton 1985; Walsh et al. 1995; Walsh and Harris 1996; Vaughan et al. 1997; Gaisler et al. 1998; Ávila-Flores and Fenton 2005). However, the continuation of sampling, using complementary methodologies (i.e., ultrasound detectors) may provide new records for the urban bat community of Vitória providing a better picture of the community structure in this urban landscape (Bergallo et al. 2003). Phyllostomidae was the family with the highest number of individuals captured. This may be explained by the method used, since mist nets are more efficient to capture bats that fly in low altitudes (Pedro and Taddei 1997). We believe that the use of other methods could include more insectivorous species to the present list.

Our results agree with other research on bats (e.g., Ávila-Flores and Fenton 2005), showing that even though some species exploit urbanized sites, the presence of parks are important to maintain bat diversity. Wooded streets have already been reported as potential corridors for birds in an urban landscape, connecting urban parks immersed in an urban matrix (Fernández-Juricic 2000). Our results suggest that for bats such streets are not providing the same degree of connectivity as observed for birds, with wooded streets being much more similar to non-wooded streets than to urban parks and being used by only two species and in small numbers.

Almeida et al. (2007) monitored bat activity patterns in Vitória with an ultra-sound detector. They showed that the areas with more vegetation and water are important for feeding and general activities of bats, concluding that the vegetation may offer shelter and food. In some of our sampling sessions we used an ultrasound detector, and could detect the presence of molossid bats. Species that roost in man-made structures such as buildings and bridges also have a higher abundance in urban areas than species that are not able to explore these artificial resources that replace natural roosting sites in urban areas (Fenton 1997; Brosset et al. 1996).

In Vitória, A. lituratus was the dominant species, representing 66% of all captures. A. lituratus was caught in all sampled areas, and was the dominant species in all of them. This species has been regularly reported to occur in urban areas in Brazil, possibly due to its opportunistic behavior (Galetti and Morellato 1994; Sazima et al. 1994; Zortéa and Chiarello 1994; Bredt and Uieda 1996), suggesting it may have a certain tolerance to urbanization (Barros et al. 2006). The majority of the bat species recorded in the present study were already found to occur in urban areas in Brazil (Reis et al. 1993; Pedro et al. 1995; Bredt and Uieda 1996; Silva et al. 1996; Félix et al. 2001; Bredt et al. 2002; Esbérard 2003; Nogueira et al. 2003; Passos and Passamani 2003). Bredt and Uieda (1996) state that A. lituratus, P. lineatus, and G. soricina all find food and roosts in urban habitats. Insectivorous bats are reported to be extremely abundant in urban areas due to new roosts, open spaces to fly, and large amount of insects that are attracted by artificial light (Bredt and Uieda 1996; Silva et al. 1996). The low capture rate for this group in the present study may be related with the sampling method adopted, since insectivorous bats are capable to detect and avoid mist nets (Bergallo et al. 2003).

Our results will not only improve our knowledge on urban bats in Vitória but they provide data that may be useful in developing management strategies to maintain such diversity within the urban landscape and how to monitor it. Urban gardening and urban landscape planning should include biodiversity data in their outputs in order to better design a landscape that improves the likelihood of persistence of bats within Vitória, and better planning tree plantation in streets, using native trees instead of exotic ones and higher densities, could improve the use of wooded streets by bats.

Acknowledgments

We would like to thank our collegues Vinicius Pimenta, Márcio Almeida, Sílvia Ramira, Ricardo M. Fonseca, Rafael Z. Coutinho, and Geovana Mendes for all the help on the fieldwork. Daniel Brito for reviewing this manuscript prior to submission. Pedro L. Peloso for helping with the map. We also thank Prefeitura Municipal de Vitória and park managers for fieldwork permission. Instituto Brasileiro do Meio Ambiente e dos Recursos Naturais Renováveis for the research authorization. M. Oprea recieved a scholarship from Bat Conservation International through the Student Research Scholarship.

Copyright information

© Springer Science+Business Media B.V. 2009