Forest cover is important across multiple scales for bird communities in vacant lots

Urban vacant lots can vary considerably in their vegetation structure, from dense, shrubby habitats to wooded remnant fragments that may provide habitat for a variety of birds. By identifying which features promote diverse bird communities, we can determine at which scale management practices should focus and the necessary habitat structure and composition features. We surveyed 150 vacant lots throughout Baltimore, Maryland for their bird communities, lot vegetation, and landscape-level forest cover. An ordination of the bird community indicated a response to a gradient of canopy cover and canopy height at the vacant lot. We also found that forest cover within 100 m of the vacant lot was the most important predictor of abundance for five bird species of interest. Species richness was spatially autocorrelated among sites, indicating that bird communities may also be driven by species’ dispersal and environmental gradients across the city. Overall, bird communities are responding to habitat features across multiple scales, from the vacant lot vegetation, to landscape-level forest cover, to city-wide dynamics. Thus, we recommend management practices to focus on increasing city-wide forest cover in order to increase species richness, yet with awareness regarding where the lot occurs within the city.     

Diversity,abundance, and species composition of ants in urban green spaces

Urbanization threatens biodiversity, yet the number and scope of studies on urban arthropod biodiversity are relatively limited. We sampled ant communities in three urban habitats (forest remnants, community gardens, vacant lots) in Detroit and Toledo, USA, to compare species richness, abundance, and species composition. We measured 24 site characteristics to examine relationships between richness and composition and habitat patch size, vegetation, and urban features. Ant richness was higher in forests (26) than in gardens (14) and intermediate in vacant lots (20). Ant richness in gardens and vacant lots negatively correlated with abundance of an exotic ant species (Tetramorium caespitum); thus this ant may affect native ant richness in urban habitats. Ant composition differed with habitat type, and abundance was lowest in forests. Site characteristics varied with habitat type: forests were larger, had more woody plants, higher woody plant richness, more branches, and leaf litter whereas lots and gardens had more concrete and buildings. Vacant lots had taller herbaceous vegetation, and gardens had higher forb richness, density, and more bare ground. Differences in vegetation did not correlate with ant richness, but several vegetation factors (e.g. patch size, number and size of trees, leaf litter, and amount of concrete and buildings) correlated with differences in ant species composition. Additional factors relating to soil, nests, or microclimatic factors may also be important for urban ant communities. Implications for biodiversity conservation in urban ecosystems are discussed.     

Urban croaking: diversity and distribution of anurans in a neotropical city

Anurans are excellent site-specific bioindicators. Although urbanization imposes diverse threats to wildlife species, little is known about its effect on tropical anurans. In this study we assessed the response of anurans to different conditions within an urban area (i.e. highly developed intra-urban residential areas, highly developed peri-urban residential areas, intra-urban abandoned lots, waterway). Our results show that: (1) 36 % of the regional anuran species pool was found within the boundaries of the studied city, (2) species richness peaked in highly developed peri-urban areas, (3) anuran abundance scores were highest in abandoned lots, and (4) highest similarity was found between the anuran community from highly developed peri-urban areas and abandoned lots. Although this study was carried out in a single city, and we only recorded four species, our results suggest that the urban semi-permeability process occurs for anuran communities, plus the fact that abandoned lots seem to offer better breeding habitat for anurans than a polluted waterway.     

Increased differentiation between individuals,but no genetic isolation from adjacent rural individuals in an urban red squirrel population

Whether or not individuals in cities and adjacent rural areas form separate populations depends on the origin of urbanized individuals. The rural individuals may have simply gradually invaded the city. Alternatively, the urbanization depended on arrival of individuals with specific adaptations for city life. A population within the city could also remain separated from adjacent rural populations if there are physical or behavioural dispersal barriers lowering the possibility of genetic exchange between individuals. All this is linked to dispersal behaviour of the species as gene flow is determined by distances moved during dispersal. In this study, we used microsatellite markers to evaluate whether or not Eurasian red squirrel (Sciurus vulgaris) individuals in an urban environment are genetically separated from their sub-urban/rural counterparts. Using radio-telemetry, we also studied natal dispersal distances of urban red squirrels to evaluate the potential for gene flow within the city. We observed no indication that urban squirrels had been genetically isolated from rural squirrels. In addition, increased amount of built area did not explain differentiation between urban individuals, further indicating that there were no barriers restricting movement. Still, the genetic differentiation within the city was increased and natal dispersal distances relatively short, perhaps indicating lack of need to move long distances in the urban environment. We conclude that urban and adjacent rural squirrels were part of the same larger population with gene flow in and out from the city. Yet, urbanization was related to increased genetic differentiation, indicating different spatial genetic structure in urban and rural populations.     

Evaluating the distribution of invasive woody vegetation around riparian corridors in relation to land use

Invasive species have been causing important and irreversible impacts to native species and communities of ecosystems. They distort ecosystem functions by degrading forest lands, wetlands, and agricultural habitats and replace the native vegetation and reduce biodiversity, forest productivity, and suitable wildlife habitat. To address disturbances caused by invasive species occurrence, further information is needed regarding the occurrence, extent, and dispersal of invasive species and how land use may increase the spread of these species. The objective of this study was to find the frequency and dominance of three invasive species common to riparian areas of east Alabama: Ligustrum sinense (Chinese privet), Elaeagnus pungens (silverthorn), and Triadica sebifera (Chinese tallow tree). Surveys of these species in riparian forests in and around Auburn, Alabama were conducted to show the relative extent of these shrubs and their relation to urban land use. It was expected to see the highest levels of invasive species in the city center with decreasing levels radiating outward into rural areas. Another objective was how urban land use may affect the presence-absence and prevalence of these non-native plant species within study sites. The results showed that around the city center and suburban lands, cover of both Chinese privet and silverthorn tended to increase. In contrast, Chinese tallow tree density percent cover showed an opposite trend with landscapes close to city center often having slightly less cover. This study shows that urban land use may be an important association with distribution of invasive plant species.     

Functional connectivity in urban landscapes promoted by <Emphasis Type="Italic">Ramphastos toco</Emphasis> (Toco Toucan) and its implications for policy making

The scarcity of green areas in urban landscapes hinders connectivity among sites reducing the flux of organisms and seed dispersal. Ramphastos toco is an effective plant disperser in tropical landscapes, playing an important role in conserving plant connectivity. In this study we combined two methods of landscape connectivity analysis, in a way not yet explored, to assess the potential contribution of Ramphastos toco to enhancing connectivity among urban vegetation patches. We used spatial modeling techniques to evaluate least cost routes, or those that facilitate bird movement through green urban areas, in Belo Horizonte, a city in Southeast Brazil. We also assessed the relative importance of forest patches for conserving both bird and seed dispersal fluxes using the integral index of connectivity. The resulting least cost route of greater accessibility for the species included in its course an important forest patch under environmental licensing for the construction of a residential condominium. The number of green urban areas covered by forest, of highest habitat quality for the species, summed to 155 patches. Of this total only 5.2% were relevant for regional connectivity, while the four most important patches are targeted by the city’s expansion plans. Ramphastos toco is an effective connector for tropical green urban areas given its adaptability, wide range of movement and seed dispersal effectiveness. We emphasize the need for incorporating ecological knowledge and the prioritization of green areas into city planning, since current expansion projects jeopardize forest patches that are crucial to the functional connectivity of the urban landscape.     

Urbanization impacts on land snail community composition

Urbanization has tremendous impacts on most native species. This is especially true in land snails, which are surprisingly understudied organisms. Due to their low mobility and dispersal potential, land snails are valuable indicators of ecosystem disturbance. For this study, land snails were collected in 54 city parks along an urban gradient to understand impacts of urbanization on snail communities. Sampled parks include small extensively landscaped downtown parks, neighborhood and community parks, district parks, and large nature parks, each with variable vegetation, soil characteristics, disturbance regimes, and human activities. Sampling recovered 12,153 individual snails, representing 20 families, 43 genera, and 95 species. Seven new Tennessee state and 87 new county occurrences were recorded. Five non-native and one extra-limital non-native species were found, four of which are new Tennessee state records. Results show that urbanization greatly alters land snail community structure. Nature and district parks have significantly greater species richness, species diversity and species evenness than community, neighborhood, and downtown parks. Degradation of parks, distance from the park to the commercial city center and percent of coarse woody debris accounted for most of the variation between park types. Non-metric multidimensional scaling and pairwise Jaccard indices indicate that downtown snail communities are more similar whereas snail communities in nature parks are more distinct. This suggests that urbanization promotes homogenization among land snail communities. We also show that this homogenization is thus far driven mainly by synanthropic, broadly adapted native species rather than non-native snail species.     

Vascular plants along an urban-rural gradient in the city of Tampere,Finland

The aim of this study was to analyse the spatial distribution of vascular plants along a 21-kilometre rural–urban–rural transect in the city of Tampere, Finland. The study emphasised the distribution of native and non-native species, both in absolute numbers and proportionally. The observed differences are explained by the share of forest land, the number of detached houses, distance from the city centre, and human population. Non-natives showed the highest values in suburban areas. Still, the difference in number of non-natives between suburban and central areas was quite small. In the city of Tampere, there are not continuous large areas devoid of vegetation. The number of native species remained high until the urban core and natives dominated in the rural-type areas of the city. However, there was not a great difference in the number of native species between rural and suburban areas. In the suburban areas, the detached houses and block-of-flats have little effect on the general vegetation. Proportionally, the share of natives decreases in line with the urban traits of the city. Urbanisation therefore affects native species in Finland. Overall, the characteristic features of a Finnish city, such as dispersed urban structure, small population, late urbanisation, abundant natural vegetation (forest) and the qualities of Finnish forests, guarantee the continuing diversity of urban vascular plants.     

Soil health as a predictor of lettuce productivity and quality: A case study of urban vacant lots

Urban agriculture offers a framework for local self-reliance and resilience in cities. However, there is a concern over the capacity of urban soil to provide sustainable and safe food production. We tested the effectiveness of several soil health indicators to predict food crop productivity and quality in vacant lots in a disadvantaged neighborhood in the city of Cleveland, Ohio. We defined soil health as a state of composite well being in terms of biological, chemical, and physical properties of the soil as they relate to crop productivity. Twelve city-owned vacant lots, three close to each of the four city schools, were selected for soil properties and plant growth analyses. Soil samples were analyzed for pH, moisture content (θv), soil texture, soil organic matter (SOM), active carbon (AC), ammonium (NH₄-N), nitrate (NO₃-N), microbial biomass N (MBN), and nematode community parameters including total (TNN), bacteria-feeding (BFN), fungal-feeding (FFN), and plant-parasitic (PPN) nematodes, number of nematode genera (NNG), and nematode food web enrichment index (EI) and structure index (SI). Lettuce was planted in the selected vacant lots and its growth was documented through measures of dry biomass, numbers of leaves/plant, and complementary subjective appearance scores related to physiological status. All measured parameters varied considerably among vacant lots except soil pH. Principal components analysis revealed that among the primary soil physical, chemical, and biological parameters, soil clay, NO₃-N, MBN, SOM, AC, TNN, BFN, FFN, and PPN contributed most to the variance of the entire dataset. There were also several positive correlations among these key soil health predictor variables: AC was positively correlated with clay, SOM, MBN, TNN, BFN, FFN and PPN, and TNN was positively correlated with AC, SOM, MBN, BFN, FFN and PPN. Of the identified primary soil health indicators, only clay, SOM, and MBN positively correlated with lettuce dry biomass, which was also positively correlated with a secondary soil health indicator, the nematode food web EI. Lettuce leaf necrosis was negatively correlated with clay, AC, SOM, MBN, TNN, FFN, and PPN, and the proportion of withered leaves was negatively correlated only with SOM. It is concluded that AC, PPN, TNN, SOM, MBN, clay, and nematode food web EI can serve as important soil health indicators that have potential for predicting crop productivity and quality in urban soils. It is also concluded that lettuce can serve as an important indicator of soil health with respect to crop productivity and quality in vacant lots.     

A systematic review of the relationship between urban agriculture and biodiversity

Urban agriculture is a unique form of agriculture that can provide fresh, local produce for urban residents, and may benefit biodiversity by decreasing the need to expand agriculture into natural areas as well as enhancing biodiversity in urban areas. However, although urban agriculture is also often cited as promoting biodiversity in urban areas, the extent of empirical evidence for such claims has not been studied. Here we systematically review the relationship between urban agriculture and biodiversity in the scientific literature. We strictly define urban agriculture as areas in cities that grow produce specifically for human consumption. We examined 148 papers from 2000 to 2017, of which only 24 studies fit our definition of urban agriculture, and of those, only 18 both involved urban agriculture and measured biodiversity. Of the studies that did measure biodiversity, some showed increases in diversity compared to urban vacant lots, but other showed no difference. Moreover, these studies were mostly focused on plants and invertebrates and were conducted almost exclusively in North America. In order to use the generalization that urban agriculture will have a positive influence on urban biodiversity, more studies will need to be conducted across a wider geographic range worldwide (particularly in developing countries in the tropics) and on a greater diversity of species and taxa (e.g., herpetiles, birds and small mammals). Such studies will likely increase in conservation importance as urban expansion and agricultural demands increase globally.     

Human land use as a driver of plant community composition in wetlands of the Chicago metropolitan region

Anthropogenic alteration of the landscape has facilitated plant community change and non-native species invasion in urban areas. We used species occurrence data from over 2000 wetlands within the Chicago metropolitan region to classify urban wetlands into community types and examined non-native species composition across community types. Non-native species were widespread across the region, occurring in over 99% of wetlands. On average, 35% of the plant species in individual wetlands were non-native. A single non-native species, Phalaris arundinacea, was present in 74% of wetlands. Six wetland community types were identified (wet meadows, marshes, forested wetlands, farmed wetlands/mudflats, roadside marshes, and an undetermined wetland type), with each having aggressively spreading non-natives amongst the most common plant species. We conducted canonical correspondence analysis to evaluate the contribution of surrounding land cover, roads, and location of wetlands to plant community composition in these wetlands, and found that similar changes to the landscape have resulted in similar combinations of native and non-native species. Differences in species composition reflected spatial gradients in land use from urban to rural areas across the region, as well as proximity to major roads. Anthropogenic drivers have resulted in profound and pervasive changes to wetland communities across the region, creating novel habitats and ultimately novel community types.     

Does demolition improve biodiversity? Linking urban green space and socioeconomic characteristics to avian richness in a shrinking city

Anthropogenic change, specifically urbanization, has lasting impacts on biodiversity and ecosystem services. Past work has linked green space variables and socioeconomic characteristics with biodiversity in urban areas. Shrinking cities present an interesting case for studying biodiversity because many of these places have implemented demolition policies to remove vacant buildings from the landscape, which returns previously urbanized land back into a “natural” state. The objective of this study is to investigate the multifaceted relationship among urban green space, socioeconomic characteristics, demolition activities, and avian species richness in the shrinking city of Buffalo, NY. We first establish a baseline understanding of how social and ecological factors are related to avian species richness through generalized linear models (GLM) and then incorporate a variable characterizing demolition to understand how the process of demolition may be impacting species richness relative to other factors. Second, since our analysis relies on “citizen science” data from eBird, we aim to better understand the factors driving eBird participation in Buffalo through zero-inflated negative binomial (ZINB) regression models to test how social and ecological characteristics and demolition affect checklists submission across the city. Our findings suggested that demolition activities do not increase species richness in Buffalo, indicating these sites may not be returning to a natural state that is usable by birds. We also found that areas that have undergone considerable demolition may be a potential deterrent to eBird participation, and eBird users may create bias by sampling areas with greater green space connectivity, and thus higher richness.     

Vacant land conversion to community gardens: influences on generalist arthropod predators and biocontrol services in urban greenspaces

Due to economic decline and the recent rise in home foreclosure, many U.S. cities are faced with managing large acreages of vacant land. Interest in local food production on this land has the potential to dramatically reshape the composition of greenspace found within urban landscapes. This study examined how the conversion of vacant land to urban gardens and farms influenced arthropod generalist predator populations and their ability to support biocontrol services. We found that the abundance of Coccinellidae and Syrphidae, and the activity density of Carabidae, Formicidae, and Lycosidae were equivalent among vacant lot and urban garden sites. Dolichopodidae abundance and the activity density of Linyphiidae and Opiliones were reduced in urban gardens whereas Anthocoridae abundance and the activity density of Staphylinidae were greater within urban gardens relative to vacant lots. The biocontrol service supplied by generalist predators was measured using sentinel eggs (Helicoverpa zea (Noctuiidae)) and pupae (Sarcophaga bullata (Sarcophagidae) and Musca domestica (Muscidae)). We found no difference in the biocontrol of H. zea eggs or M. domestica pupae among the focal greenspaces. We found no difference in the removal of S. bullata pupae by predators in June 2010, whereas in August 2010 a greater proportion of pupae were removed in urban garden sites relative to vacant lots. Based on these findings, we discuss the potential of the urban landscape to support generalist predators and sustainable urban gardening and farming.     

Effects of urban structure on plant species richness in a large European city

Urban areas have developed under anthropogenic impact and hence feature strongly altered environmental conditions that influence today’s plant species distribution. Generally, urban areas are rich in plant species, but on smaller scales richness can vary considerably. Our study aim is to identify areas of high species richness and to assess richness distribution patterns in the city of Hamburg by analyzing a floristic mapping dataset on the scale of 1 km². Differences in plant species richness were analyzed between three urbanization zones. With multiple regression analyses, we tested effects of urban structure, habitat, and environmental conditions on the distribution of species richness measures. Total species richness per 1 km² was 274?±?60 on average and differed only slightly between the urbanization zones. It increased with habitat diversity and decreased with Ellenberg indicator values (EIV) for nutrients (multiple R²?=?0.31). Proportion of non-native species increased with mean annual temperature and decreased with EIVs for moisture (multiple R²?=?0.72), while proportion of endangered species increased with EIVs for moisture and decreased with EIVs for nutrients (multiple R²?=?0.66). Proportion of thermophilic species (multiple R²?=?0.58) could be explained by mean annual temperature. The emerging patterns probably differ from those in other cities due to the central port harboring a particular flora. Besides the expected high proportions of non-native species, high proportions of endangered species were also found in this area. Our results contribute to identifying drivers of biodiversity in cities and can thus be used to develop measures for the conservation of urban biodiversity.     

Drivers of variability in water use of native and non-native urban trees in the greater Los Angeles area

Trees in urban ecosystems are valued for shade and cooling effects, reduction of CO₂ emissions and pollution, and aesthetics. However, in arid and semi-arid regions, urban trees must be maintained through supplemental irrigation, in competition with other water needs. Currently, a comprehensive understanding of the factors which influence water use of urban tree species is lacking. In order to study the drivers of whole tree water use of two common species in the Los Angeles Basin urban forest, four sites in Los Angeles and Orange County were instrumented with sap flow and meteorological sensors. These sites allowed comparisons of the water use of a native riparian (Platanus racemosa Nutt.; California sycamore) and non-native (Pinus canariensis C. Sm.; Canary Island pine) Mediterranean species, as well as the spatial variability in water use under different environmental and management conditions. We found higher rates of sapflux (J _O ) in native California sycamore as compared to non-native Canary Island pine. Within each species, we found considerable site-to-site variability in the magnitude and seasonality of J _O . For Canary Island pine, the majority of inter-site variability derived from differences in water availability: response to vapor pressure deficit was similar during a period without water limitations. In contrast, California sycamore did not appear to experience water limitation at any site; however, there was considerable spatial variability in water use, potentially linked to differences in nutrient availability. Whole tree transpiration (E) was similar for the two species when water was not limiting, but Canary Island pine was able to withstand unirrigated conditions with a very low E. These results add to the currently small pool of data on urban tree water use and ecophysiology, and contribute to establishing a more quantitative understanding of urban tree function.     

Can urban greening increase vector abundance in cities? The impact of mowing,local vegetation,and landscape composition on adult mosquito populations

Worldwide, cities are investing in greenspace to enhance urban quality of life and conserve biodiversity. Cities should ensure these investments do not unintentionally result in ecosystem disservices. Municipal management decisions regarding urban greenspaces, such as mowing frequency, could influence mosquito communities and public health. We examined how mowing, resultant vegetation characteristics, and landscape context influenced adult mosquito abundance in urban vacant lots. We sampled adult Culex and Aedes mosquitoes in a network of vacant lots within eight Cleveland, Ohio, USA neighborhoods in 2015 and 2016 using CO₂-baited light traps and grass-infused gravid traps. For each lot, we quantified vegetation characteristics, including plant diversity, bloom area, and biomass, as well as the surrounding landscape composition at radii of 60 and 1000 m. We found that periodic mowing did not significantly affect mosquito abundances. However, vacant lots with more diverse plant communities were associated with a greater light trap capture of both Culex and Aedes. Both mosquito genera declined in light trap catches with increased impervious surface at 60 m. Similarly, Culex (gravid trap) declined with the amount of built infrastructure at 1000 m. In contrast, Aedes (light trap) increased with the concentration of buildings in the landscape at 1000 m. Our findings indicate that reducing the frequency of mowing within vacant lots will not necessarily increase adult mosquito abundance. Nonetheless, mosquito surveillance and management should be considered when planning conservation-focused greenspaces, as vegetation design choices and the landscape context of a site do influence vector abundance and potentially disease risk.

     

Bird species assemblages across a rural urban gradient around Kolkata,India

Urban development leads to changes in habitat structure and resource base. Bird communities are known to respond sharply to such changes. Our result from study of bird community along urbanization gradient around Kolkata metropolitan city clearly separated the urban bird community from the rural and rural 3 habitats in terms of species diversity and foraging groups. Rural and rural 3 sites had more number of rare species and higher percentage of unique species as compared to the urban habitats. Functional group analysis showed higher abundance of granivores in urban habitats and absence of insectivore and carnivorous species that were found in the rural and rural 3 habitats. The bird species assemblage along the gradient was significantly nested where bird species recorded in urban areas were subset of the species rich rural areas. There was no difference in individual counts between urban and rural habitats, therefore nullifying the hypothesis that rural areas are more species rich because of higher population size. Bird community in the urban areas was less even as compared to the rural areas due to the dominance of omnivorous guild. Bird diversity was negatively correlated to the density of house lots.     

Ecological and biological determination of invasion success of non-native plant species in urban woodlands with special regard to short-lived monocarps

Today’s urban green space is exposed to a constant flow of newly introduced exotic plant species, resulting from multiple, intentional and inadvertent introductions. A substantial share of introduced plant species are represented by potential invaders of plant communities, which allows considering that the risk of rapid dispersion by aggressive plant species to be today’s most serious threat to the urban biodiversity. Alien plant species with a short life span (monocarps, therophytes, annuals and biennials) are the most successful invaders across anthropogenically transformed landscapes. These species can have a significant impact on the functioning of previously disturbed ecosystems. At the same time, it is still unclear which mechanisms enable short-lived monocarps to invade under specific conditions of urban forest interiors. Therefore, I have conducted a literature analysis aiming at identifying possible biological and ecological determinants of monocarpic plants’ invasive success in urban woodlands. The analysis was made at three qualitative levels taking into account characteristics of the urban environment, those of different types of urban woodlands as well as traits of non-native monocarpic species. Considering the city level, the main ecological factors are the high level of human-induced disturbance and propagule pressure on the urban-woodland interface sites. The level of invasion as well as invasibility of urban woodlands is mainly determined by the landscape context, whereas the habitat type, species composition and spatial structure play a less important role. In the forest environment, where competition for light and nutrients is the main shaping factor of the phytocoenosis, short-lived alien species must possess a certain level of competitive ability. These findings indicate that this ability may be due to such traits as high fecundity rates, effective dispersal mechanisms, early germination, high rate of growth and the overall big size of plants, presence of autonomous pollination and high shade-tolerance.     

Bat boxes in urban non-native forests: a popular practice that should be reconsidered

Impact mitigation practices are currently one of the hottest topics in conservation and regarded as priorities worldwide. Forest bat populations are known to provide important ecosystem services such as pest control and bat boxes have become one of the most popular management options for counteracting the loss of roosts. However, bat boxes tend to be employed in non-native forests near highly humanized areas where human disturbance is higher. The aim of this study was to evaluate how the surrounding landscape composition affects bat box occupancy in urban non-native forests along the Mediterranean corridor in the Northeastern Iberian Peninsula. Two hundred wooden bat boxes were monitored in young non-native forests in the period 2004–2012. The influence of land cover on occupancy rate of bat boxes was analysed at the landscape level in a 5 km buffer around bat-box stations. In total, 1659 inspections were carried out, in which a 15 % occupancy rate was detected. Bat boxes hosted three different species (Pipistrellus pygmaeus, Pipistrellus kuhlii and Nyctalus leisleri). More than 70 % of the occupancy can be explained by habitat and spatial composition. The presence of urban areas around bat boxes tends to have a negative impact on bat occupation rates; by contrast, forest coverage has a positive effect, especially for the tree-dwelling bats. These patterns could be associated with the large number of available roosts in buildings, microhabitat or phylopatry. Thus, to increase success, we suggest that landscape composition should be considered when using bat boxes for conservation.     

Ecological corridors also operate in an urban matrix: A test case with garden shrews

The fragmentation of landscapes, induced by the growing urbanisation, is a major cause of biodiversity loss. Ecological corridors are landscape elements that make up for the negative effects of habitat fragmentation in agricultural or natural landscapes. However, their effectiveness have been rarely assess in an urban context. To do so, we analysed shrews in woodlots, corridors and domestic gardens that are connected or disconnected from a corridor. Indeed, these mammals are very sensitive to fragmentation and could play an important ecological role even in small and isolated areas such as domestic gardens. Ninety-seven shrews, from three species were trapped. In our study, Sorex coronatus, Sorex minutus and Crocidura russula were not widely distributed and mainly occurred in woodlots, corridors and connected gardens. In gardens, shrew occurrences were mainly determined by landscape configuration with negative effects of the distance to the corridor and the distance to the woodlot, stronger for C.russula than Sorex species. At the local scale, garden management and vegetation showed a positive effect mainly for Sorex species. The quality of gardens seems good enough to permit the co-occurrence of several species of shrews and thus competition was not identified as a major process for determining the occurrence. We showed that, as in other ecosystems, ecological corridors in urban context can play an effective role for the distribution of organisms with low dispersal capabilities and should be develop in planning strategies.