Author Name: Rebecca Favier, BSc (Hons) Animal Behaviour and Welfare
Understanding how bats use the landscape is critical to conservation research and informing policy on habitat protection. Land-use change and habitat fragmentation is the largest threat that bats face due to rapid urbanisation and agricultural progress. Bats have been recognised by the UK government as bioindicators and priorities for habitat conservation, highlighting the need for data to be gathered. Collecting field data on highly mobile, rare or elusive species is difficult, labour intensive and costly to obtain. Predictive models are increasingly used within the industry to assess habitats for suitability and species distribution and richness, and have the potential to be used in assessing the extent of the impact of climate or land use change on bats.
Recent research has varied in its application of predictive habitat suitability modelling (HSM), from assessing single species to multiple species in the same areas, and by using different available variables. The different approaches result in differences in accuracy, highlighting the need for refinements in technology and application. HSM is currently unable to account for additional variables such as biotic interactions, historical disturbances, or dispersal and population dynamics, which could cause overestimations of habitat suitability or species richness. However, in cases where detailed field data is impractical or too costly to implement, HSM can be highly informative regarding which habitats are important for targeted conservation.
Bats are a large and widely distributed mammal order, with over 1,300 extant species inhabiting all continents except Antarctica (Voigt and Kingston, 2016). Bats span multiple ecological niches with there being insectivorous, frugivorous and carnivorous species across the world; however, all 18 UK species are insectivorous (Dietz and Kiefer, 2016; Bat Conservation Trust, 2015; Russo and Ancillotto, 2014). Anthropogenic changes such as agricultural progress and urbanization occur more rapidly than natural ecosystem change, rendering species with specialised habitat requirements unable to adapt at the same pace (Duchamp and Swihart, 2008; Johnson, Gates and Ford, 2008). The expansion of croplands and the resultant increase in the use of chemical pesticides causes invertebrate decline, and, consequently, insectivorous bat population decline (Williams-Guillén et al., 2016). Additionally, their complex habitat requirements, including discrete sites for roosting and foraging, mean land-use change may have more impact than is currently understood (Dietz and Kiefer, 2016).
Understanding how bats use the landscape and knowledge of habitat types required to ensure survival is critical to future conservation (Bellamy, Scott and Altringham, 2013). Obtaining data for rare or elusive, nocturnal and highly mobile species is challenging (Hamilton, Pollino and Jakeman, 2015). Collecting detailed field data on species distribution and habitat suitability is costly and labour intensive and as a result predictive models are increasingly utilised in conservation research (Tsoar et al., 2007). Habitat suitability modelling (HSM), or Species Distribution Modelling (SDM), enables habitat assessment for environmental attributes and estimations of species presence/absence, distribution or abundance (Bellamy and Altringham, 2015; Hirzel and Le Lay, 2008). Research in other taxa has demonstrated HSM has applications in ascertaining the extent of climate change or land use change impacts (de Rigo et al., 2017; Sultana et al., 2017; de Baan et al., 2015), and HSM can be employed by bat conservationists to the same purpose.
2.0 Critical Review:Shifting Landscapes and Bat Conservation
Land-use change is the most frequently mentioned threat in the IUCN bat species’ assessments, alongside human intrusion and urbanization (Figure 1) (Voigt and Kingston, 2016). The threats listed are interlinked, making land-use change a more significant threat than shown. Urbanization is a process of land-use change and is transforming the UK landscape (Russo and Ancillotto, 2014). Urbanisation and agriculture increase pollution and drive climate change through greenhouse gas emissions and increased temperatures, affecting ecosystem and species health (McCarthy, Best, and Betts, 2010; Grimm et al., 2008).
Figure 1: Frequency of threats listed in IUCN bat species’ assessments (Voigt and Kingston, 2016).
Bechstein’s (Myotis bechsteinii), lesser horseshoe (Rhinolophus hipposideros), greater horseshoe (Rhinolophus ferrumequinum), and barbastelle bats (Barbastella barbastellus), are UK species with currently decreasing populations (IUCN, 2017). Horseshoe bats are particularly vulnerable to disturbance and agricultural practice, declining by over 90% in the last 100 years (Bat Conservation Trust, 2010). The bats mentioned are listed in Annex II of the EC Habitats Directive, which highlights species requiring Special Areas of Conservation in the environment (SACs) (Council Directive 1992/43/EEC). As bats are sensitive to climate change and habitat deterioration, changes in bat populations are an indication of biodiversity health and habitat quality (Jones et al., 2009). The Department for the Environment, Food and Rural Affairs (DEFRA) in the UK listed eight UK bat species as ‘bioindicators’ in 2008 (Bat Conservation Trust, 2017). This highlighted the need for on-going bat habitat conservation, but the impact of the UK’s withdrawal from the EU (‘Brexit’) on conservation law is currently uncertain (Pieraccini, 2015). Whilst the UK previously had habitat protection systems (Sites of Special Scientific Interest), management measures were not as rigorous as those introduced by the EU with SACs, and there is concern that habitat conservation may be at risk post-Brexit (Pieraccini, 2015).
Bat conservation relies on threat mitigation to preserve essential habitat, and therefore an understanding of landscape and habitat use is essential (Bellamy, Scott and Altringham, 2013). Bats require a variety of foraging and roosting sites, and many species change roosts regularly (Dietz and Kiefer, 2016; Kalda, Kalda and Liira, 2015). Bats rely on the connectivity of habitat ‘patches’ within the larger landscape with linear features such as hedgerows, treelines, or water serving as commuting routes (Dietz and Kiefer, 2016). Connectivity is difficult to conserve and to study, but better understanding is required in the face of fragmentation of the landscape (Lindenmayer et al., 2007). Overall landscape structure is, therefore, as important a consideration as the features within each patch (Kalda, Kalda and Liira, 2015). Different bat species also have different specialisations and habitat requirements; some may forage in open spaces and others use edge habitat for hunting (Dietz and Kiefer, 2016; Denzinger, 2013). Different species also require different roosting habitats and some species are migratory, whilst others are more sedentary (Dietz and Kiefer, 2016).
2.1 Single species modelling
As the differences between bat species can be pronounced in terms of habitat requirements, it is common for research to focus on a single species in HSM studies. This approach simplifies HSM outputs as there is no need for multiple variable outputs for the discrete species, and it can also fill knowledge gaps for targeted rare species (Rebelo and Jones, 2010). Rebelo and Jones (2010) focused on the barbastelle bat, one of the rarest bats in Europe. The study modelled distribution in Portugal, where the barbastelle was known to be present, but few locations were known. HSM was able to give insights into new locations, predicting high suitability in 15 previously unknown sites for the species. The study confirmed presence in these locations with acoustic sampling (Rebelo and Jones, 2010). In other areas, however, the HSM output did not correlate with acoustic results. The HSM did not appear to include distance variables between resources, and Rebelo and Jones (2010) admitted that result accuracy could have been different at different spatial scales. The study encompassed the entirety of Portugal, and how resources are distributed across an area of this size could change the outcome of HSM. The inclusion of resource-based distance variables can substantially enhance HSM accuracy where species distribution is typically patchy (Allouche et al., 2008). The inclusion of distance variables is highly relevant to bat HSM in increasingly fragmented and patchy landscapes (Dietz and Kiefer, 2016). Distance variables can significantly change the output of habitat suitability maps (Figure 2) (Rainho and Palmeirim, 2011). Bats are highly mobile and the distance between resources is relevant to bats’ landscape use. Future studies should include distance variables in HSM for increased accuracy (Rainho and Palmeirim, 2011).
Figure 2: Maps of predicted foraging suitability of the study area for Miniopterus schreibersii (left) and Rhinolophus mehelyi (right). The top maps show suitability of habitat including distance variables, and the lower maps exclude distance variables. Suitability is shown on a colour scale ranging from 0 (dark blue = low suitability) to 1 (red = high suitability). (Rainho and Palmeirim, 2011).
Razgour, Hanmer and Jones (2011) focused on the grey long-eared bat (Plecotus austriacus), a rare bat in the UK, and used HSM at a finer scale than that employed by Rebelo and Jones (2010). Razgour, Hanmer and Jones (2011) highlighted that the grey long-eared bat may especially benefit from presence-only ecological modelling, as they are difficult to detect and identify acoustically, making absence data unavailable or unreliable (PTES, 2017; Fonderflick et al., 2015). Razgour, Hanmer and Jones (2011) found that seasonal temperature and precipitation, and land cover were the most important variables at the broad-scale. Whilst these are potentially difficult to overcome in conservation management, they also uncovered the importance of a specific habitat type. They found that grasslands were highly selected by grey long-eared bats in both the broad and fine spatial scales, but by using fine-scale model predictions, they discovered a stronger association with unimproved lowland grasslands compared with improved grasslands. This is an important difference to highlight. Unimproved grassland habitat declined by 97% between 1934 and 1984 to approximately 0.2 million hectares (DEFRA, 2008). The decline has continued, and estimates in the Habitat Statement reveal that less than 15,000ha of unimproved grassland remain in the UK (DEFRA, 2008). Studies demonstrating the value of specific habitats are of vital importance in highlighting key areas for targeted conservation.
Bellamy and Altringham (2013) highlighted that HSM can be simplistic and lack accuracy at the fine-scale, and the available resolution used by Razgour, Hanmer and Jones (2011) is a potential issue. Resolutions of even 100m2 can be considered too coarse to understand the effects of small landscape features such as hedgerows, and other methods would be required to gain a complete picture of the habitat features available (Ducci et al., 2015). The selected resolution in models is key in discriminating between important habitat patches and the surrounding areas; where resolution is insufficiently refined, critical patches for bats may go unnoticed (Ducci et al., 2015). Razgour, Hanmer and Jones (2011) utilised radio tracking data to incorporate distance variables, improving their accuracy and identifying additional core foraging zones, highlighting the importance of field data to support HSM.
2.2 Multiple species modelling
Bellamy and Altringham (2015) stated that ideally, separate single-species models should be developed for species with distinctive habitat requirements, but this can be impractical. Targeting multiple species maximises resources and negates the need for multiple analyses. Bellamy and Altringham (2015) created HSMs for seven species of bat in the Lake District, UK, using species records and targeting summer roost data. Roost site selection relies upon varied local resources, including foraging and drinking areas, night roosts, and day roosts, and the study used distance variables to build an accurate predictive picture (Bellamy and Altringham, 2015).
Habitat suitability maps were created using variables individualised to each species (Figure 3) (Bellamy and Altringham, 2015). In the maps it is possible to see how species may utilise the landscape differently, highlighting which areas are most suitable for targeted conservation, especially where a rarer species is a conservation priority. This is an effective demonstration of how HSM can be used to inform conservation strategies, but they cannot be relied upon completely. An issue with this study is that it cannot be used to reliably predict a complete year-round picture of habitat suitability, as species may have different requirements in different seasons (Bellamy and Altringham, 2015). Basing conservation strategies purely on summer roost data could be ineffective. Additionally, predictions were made for the roosting suitability of an area regardless of the presence of suitable roost structures, which could have a significant impact on bat use, as woodland bats rely upon tree species, height and condition for roost selection (Kühnert et al., 2016; Bellamy and Altringham, 2015; Ruczyński and Bogdanowicz, 2008).
Bellamy and Altringham (2015) found their models became less reliable depending on the number of available variables, highlighting that the N. noctula model included only one environmental variable. Field data would be required to strengthen the models, as was used by Razgour, Hanmer and Jones in 2011 with the utilisation of radio tracking, but this is not always practical in the field due to time-restraints and cost. The tracking of small bats is also limited by technology and the weight of transmitters, and the effect on behaviour and manoeuvrability is largely unknown (Dressler et al., 2016; Castle et al., 2015). Previously, only VHF could be used, with lightweight transmitters, but reception distance is short. Advances in GPS technology have facilitated smaller devices, however, reliability and functionality is debatable and open to development, and surgical attachment methods necessitate welfare considerations in licensing (Castle et al., 2015). As an alternative to physical tracking, HSM research on crayfish highlighted that knowledge gaps left by modelling could be filled by data collection of microhabitat features to overcome the limitations of refinement (Hamilton, Pollino and Jakeman, 2015). This could be applied to bat HSM in the future. Ultimately, maps provided by HSM can be a useful tool in habitat assessments, but should be applied with caution if supporting field data is unavailable (Bellamy and Altringham, 2015).
Figure 3: Habitat suitability maps made using each species’ set of variables. The white line denotes the boundary of the Lake District (Bellamy and Altringham, 2015).
2.3 Evaluation of HSM approaches and application
Predicting the suitability of a habitat in relation to landscape composition can be challenging when assessing mobile species which use dispersed resources, such as bats (Fonderflick et al., 2015). HSM typically does not account for additional variables such as biotic interactions, population dynamics, or historical disturbances (Hamilton, Pollino and Jakeman, 2015). The studies by Razgour, Hanmer and Jones (2011) and Bellamy and Altringham (2015) relied upon presence-only data in their modelling approaches. Whilst presence-only HSMs can be useful where absence data is unavailable, or faulty absence determinations are a problem, relying on presence-only data can be problematic without absence data for comparisons (Hirzel and Le Lay, 2008).
Depending on the target species, the predictive accuracy of HSM may differ; generalist species may be modelled with less accuracy than those with restricted ecological niches (Tsoar et al., 2007). Bat HSM was especially limited in predictive accuracy in a model evaluation by Tsoar et al. (2007). HSM has developed since this time and become more accurate with the addition of distance variables and refinements in technology which look set to continue in the future (Rainho and Palmeirim, 2011). A study by Stephenson et al. (2017) has successfully employed resolutions of 1 and 10m in HSM for a Hawaiian insect Nysius wekiuicola. Whilst resolutions this fine would be unsuitable for larger mammal species, some studies on bats have begun to be able to refine resolutions to 25m and this demonstrates that refinements to the technology are an on-going process (Stephenson et al., 2017, Le Roux et al., 2017).
Despite limitations, HSM can utilise longitudinal data sets without the expected costs of lengthy field data collection and identify potential reintroduction locations, locate unknown populations and target areas for management strategy development (Hamilton, Pollino and Jakeman, 2015; Bellamy, Scott and Altringham, 2013; Rebelo and Jones, 2010). They can also identify sites where proposed land-use changes, such as the building of new roads and wind farms, can be implemented with the least amount of disturbance to the landscape and its use by bats (Roscioni et al., 2014; Bellamy, Scott and Altringham, 2013; Santos et al., 2013). HSM is especially effective where it is possible to support model outputs with field data. Razgour, Hanmer and Jones (2011) combined radio tracking with modelling to compare habitat used by bats to the habitat suitability predictions, finding that the majority of tracked locations fell into the ranges identified. Radio tracking provides detailed data on movements and the effect of the landscape features on dispersal and foraging (Le Roux et al., 2017). This approach successfully demonstrated the value of HSM in conservation strategies for rare or elusive species that may be extrapolated to others, although the costs and practicality of this kind of data collection would need to be considered.
Whilst modelling can be a useful tool for informing conservation strategies, especially where time and money is scarce, it should not replace, but be supported by, additional field data. Habitat may be highlighted as suitable by HSM, but the species targeted may not truly be present for a number of reasons. HSM is currently unable to include all variables that may affect species presence and distribution, and this remains a challenge for the future of such studies. It would be ineffective to implement conservation planning to protect a species that is absent. HSM can, however, be of use where species are rare and not easily recorded, providing information on the species’ likely location. Initial accuracy can be improved by the inclusion of distance variables, and studies combined with acoustic sampling or radio tracking, where possible, can then confirm species presence and habitat use. Modelling is a developing science, and is utilised in ecological research across a multitude of taxa. Refinements may yet be made to HSM, employing additional variables such as climate and biotic interactions. In the face of increasing land-use and climate change and political uncertainty, identifying areas with HSM for conservation area programmes is a strategy conservationists can use to the benefit of bat populations in the future.
Allouche, O., Steinitz, O., Rotem, D., Rosenfeld, A. and Kadmon, R. (2008) Incorporating distance constraints into species distribution models. Journal of Applied Ecology [online]. 45 (2), pp. 599-609. [Accessed 13 November 2017].
Bat Conservation Trust (2010) Greater horseshoe, Rhinolophus ferrumequinum. Available from: http://www.bats.org.uk/pages/uk_bats.html [Accessed 31 October 2017].
Bat Conservation Trust (2015) Amazing Bats. Available from: http://www.bats.org.uk /pages/types_of_bats.html [Accessed 17 October 2017].
Bat Conservation Trust (2017) Why bats matter. Available from: http://www.bats. org.uk/pages/why_bats_matter.html [Accessed 17 October 2017].
Bellamy, C. and Altringham, J. (2015) Predicting Species Distributions Using Record Centre Data: Multi-Scale Modelling of Habitat Suitability for Bat Roosts. PLoS One [online]. 10 (6), p. e0128440. [Accessed 13 October 2017].
Bellamy, C. and Altringham, J. (2015) Habitat suitability maps made using each species’ set of variables. The white line denotes the boundary of the Lake District [Figure 3]. In: Predicting Species Distributions Using Record Centre Data: Multi-Scale Modelling of Habitat Suitability for Bat Roosts. PLoS One [online]. 10 (6), p. e0128440. [Accessed 13 October 2017].
Bellamy, C., Scott, C. and Altringham, J. (2013) Multiscale, presence-only habitat suitability models: fine-resolution maps for eight bat species. Journal of Applied Ecology [online]. 50, pp. 892-901. [Accessed 13 October 2017].
Castle, K.T., Weller, T.J., Cryan, P.M., Hein, C.D. and Schirmacher, M.R. (2015) Using sutures to attach miniature tracking tags to small bats for multimonth movement and behavioral studies. Ecology and Evolution [online]. 5 (14), pp. 2980-2989. [Accessed 22 November 2017].
Council Directive (1992/43/EEC) The conservation of natural habitats and of wild fauna and flora [online]. London: Department for Environment, Food & Rural Affairs. Available from: http://jncc.defra.gov.uk/page-1374 [Accessed 19 October 2017].
de Baan, L., Curran, M., Rondinini, C., Visconti, P., Hellweg, S. and Koellner, T. (2015) High-Resolution Assessment of Land Use Impacts on Biodiversity in Life Cycle Assessment Using Species Habitat Suitability Models. Environmental Science and Technology [online]. 49 (4), pp. 2237-2244. [Accessed 15 November 2017].
de Rigo, D., Caudullo, G., San-Miguel-Ayanz, J. and Barredo, J.I. (2017) Robust modelling of the impacts of climate change on the habitat suitability of forest tree species [online]. Luxembourg: Publication Office of the European Union. Available from: https://doi.org/10.2760/296501. [Accessed 14 November 2017].
DEFRA (2008) UK Biodiversity Action Plan Priority Habitat Descriptions: Lowland Meadows [online]. London: Department for Environment, Food & Rural Affairs. Available from: http://jncc.defra.gov.uk/page-5706 [Accessed 14 November 2017].
Denzinger, A. (2013) Bat guilds, a concept to classify the highly diverse foraging and echolocation behaviors of microchiropteran bats. Frontiers in physiology [online]. 4, pp. 164. [Accessed 15 November 2017].
Dietz, C. and Kiefer, A. (2016) Bats of Britain and Europe. United Kingdom: Bloomsbury Publishing.
Dressler, F., Ripperger, S., Hierold, M., Nowak, T., Eibel, C., Cassens, B., Mayer, F., Meyer-Wegener, K. and Kölpin, A. (2016) From Radio Telemetry to Ultra-Low Power Sensor Networks – Tracking Bats in the Wild. IEEE Communications Magazine [online]. 54 (1), pp. 129-135. [Accessed 21 November 2017].
Ducci, L., Agnelli, P., Di Febbraro, M., Frate, L., Russo, D., Loy, A., Carranza, M.L., Santini, G. and Roscioni, F. (2015) Different bat guilds perceive their habitat in different ways: a multiscale landscape approach for variable selection in species distribution modelling. Landscape Ecology [online]. 30 (10), pp. 2147-2159. [Accessed 19 October 2017].
Duchamp, J.E. and Swihart, R.K. (2008) Shifts in bat community structure related to evolved traitsand features of human-altered landscapes. Landscape Ecology [online]. 23, pp. 849-860. [Accessed 13 October 2017].
Fonderflick, J., Azam, C., Brochier, C., Cosson, E. and Quékenborn, D. (2015) Testing the relevance of using spatial modeling to predict foraging habitat suitability around bat maternity: A case study in Mediterranean landscape. Biological Conservation Biology [online]. 192, pp. 120-129. [Accessed 24 November 2017].
Grimm, N.B., Faeth, S.H., Golubiewski, N.E., Redman, C.L., Wu, J., Bai, X. and Briggs, J.M. (2008) Global Change and the Ecology of Cities. Science [online]. 319 (5864), pp. 756-760. [Accessed 15 November 2017].
Hamilton, S.H., Pollino, C.A. and Jakeman, A.J. (2015) Habitat suitability modelling of rare species using Bayesian networks: Model evaluation under limited data. Ecological Modelling [online]. 299, pp. 64-78. [Accessed 15 November 2017].
Hirzel, A.H. and Le Lay, G. (2008) Habitat suitability modelling and niche theory. Journal of Applied Ecology [online]. 45, pp. 1372-1381. [Accessed 19 October 2017].
IUCN (2017) Plecotus austriacus. Available from: http://www.iucnredlist.org /details/17597/0 [Accessed 10 October 2017].
Johnson, J.B., Gates, J.E. and Ford, W.M. (2008). Distribution and activity of bats at local and landscape scales within a rural-urban gradient. Urban Ecosystems [online]. 11, pp. 227-242. [Accessed 10 October 2017].
Jones, G., Jacobs, D.S., Kunz, T.H., Willig, M.R. and Racey, P.A. (2009) Carpe noctem: the importance of bats as bioindicators. Endangered Species Research [online]. 8, pp. 93-115. [Accessed 10 October 2017].
Kalda, O., Kalda, R. and Liira, J. (2015) Multi-scale ecology of insectivorous bats in agricultural landscapes. Agriculture, Ecosystems and Environment [online]. 199, pp. 105-113. [Accessed 8 November 2017].
Kühnert, E., Schönbächler, C., Arlettaz, R. and Christe, P. (2016) Roost selection and switching in two forest-dwelling bats: implications for forest management. European Journal of Wildlife Research [online]. 62 (4), pp. 497-500. [Accessed 24 November 2017].
Le Roux, M., Redon, M., Archaux, F., Long, J., Vincent, S. and Luque, S. (2017) Conservation planning with spatially explicit models: a case for horseshoe bats in complex mountain landscapes. Landscape Ecology [online]. 32, pp. 1005-1021. [Accessed 24 November 2017].
Lindenmayer, D., Hobbs, R., Montague-Drake, R., Alexandra, J., Bennett, A., Burgman, M., Cale, P., Calhoun, A., Cramer, V., Cullen, P., Driscoll, D., Fahrig, L., Fischer, J., Franklin, J., Haila, Y., Hunter, M., Gibbons, P., Lake, S., Luck, G., MacGregor, C., McIntyre, S., Nally, R., Manning, A., Miller, J., Mooney, H., Noss, R., Possingham, H., Saunders, D., Schmiegelow, F., Scott, M., Simberloff, D., Sisk, T., Tabor, G., Walker, B., Wiens, J., Woinarski, J. and Zavaleta, E. (2007) A checklist for ecological management of landscapes for conservation. Ecology Letters, 11, pp.78-91.
McCarthy, M.P., Best, M.J. and Betts, R.A. (2010) Climate change in cities due to global warming and urban effects. Geophysical Research Letters [online]. 37 (9), pp. 1-5. [Accessed 8 November 2017].
Pieraccini, M. (2015) Democratic legitimacy and new commons: examples from English protected areas. International Journal of the Commons [online]. 9 (2), pp. 552-572. [Accessed 20 October 2017].
PTES (2017) Grey long-eared bat. Available from: https://ptes.org/get-informed/facts-figures/grey-long-eared-bat/ [Accessed 20 October 2017].
Rainho, A. and Palmeirim, J.M. (2011) The Importance of Distance to Resources in the Spatial Modelling of Bat Foraging Habitat. PLoS One [online]. 6 (4), pp. e19227. [Accessed 13 November 2017].
Rainho, A. and Palmeirim, J.M. (2011) Maps of predicted foraging suitability of the study area for Miniopterus schreibersii (left) and Rhinolophus mehelyi (right). The top maps shows suitability of the habitat including distance variables, and the lower maps excluded distance variables. Suitability is shown on a colour scale ranging from 0 (dark blue = low suitability) to 1 (red = high suitability) [figure 2]. In: The Importance of Distance to Resources in the Spatial Modelling of Bat Foraging Habitat. PLoS One [online]. 6 (4), pp. e19227. [Accessed 13 November 2017].
Razgour, O., Hanmer, J. and Jones, G. (2011) Using multi-scale modelling to predict habitat suitability for species of conservation concern: The grey long-eared bat as a case study. Biological Conservation [online]. 144 (12), pp. 2922-2930. [Accessed 19 October 2017].
Rebelo, H. and Jones, G. (2010) Ground validation of presence-only modelling with rare species: a case study on barbastelles Barbastella barbastellus (Chiroptera: Vespertilionidae). Journal of Applied Ecology [online]. 47, pp. 410-420. [Accessed 13 November 2017].
Roscioni, F., Rebelo, H., Russo, D., Carranza, M.L., Di Febbraro, M. and Loy, A. (2014) A modelling approach to infer the effects of wind farms on landscape connectivity for bats. Landscape Ecology [online]. 29 (5), pp. 891-903. [Accessed 13 November 2017].
Ruczyński, I. and Bogdanowicz, W. Summer Roost Selection by Tree-Dwelling Bats Nyctalus noctula and N. leisleri: A Multiscale Analysis. Journal of Mammalogy [online]. 89 (4), pp. 942-951. [Accessed 24 November 2017].
Russo, D. and Ancillotto, L. (2014) Sensitivity of bats to urbanization: A review. Mammalian Biology [online]. 80 (3), pp. 205-212. [Accessed 17 October 2017].
Santos, H., Rodrigues, L., Jones, G. and Rebelo, H. (2013) Using species distribution modelling to predict bat fatality risk at wind farms. Biological Conservation [online]. 157, pp. 178-186. [Accessed 15 November 2017].
Stephenson, N., Perroy, R., Eiben, J. and Klasner, F. (2017) High resolution habitat suitability modelling for an endemic restricted-range Hawaiian insect (Nysius wekiuicola, Hemiptera: Lygaeidae). Journal of Insect Conservation [online]. 21 (1), pp. 87-96. [Accessed 15 November 2017].
Sultana, S., Baumgartner, J.B., Dominiak, B.C., Royer, J.E. and Beaumont, L.J. (2017) Potential impacts of climate change on habitat suitability for the Queensland fruit fly. Scientific Reports [online]. 1 (7). [Accessed 14 November 2017].
Tsoar, A., Allouche, O., Steinitz, O., Rotem, D. and Kadmon, R. (2007) A comparative evaluation of presence-only methods for modelling species distribution. Diversity and Distributions [online]. 13, pp. 397-405. [Accessed 19 October 2017].
Voigt, C. and Kingston, T. (2016) Bats in the anthropocene. In: Voigt, C. and Kingston, T., eds. (2016) Bats in the Anthropocene: Conservation of Bats in a Changing World [online]. Springer International Publishing, pp. 1-9. [Accessed 13 October 2017].
Voigt, C. and Kingston, T. (2016) Frequency of threats listed in IUCN bat species’ assessments [Figure 1]. In: Bats in the Anthropocene: Conservation of Bats in a Changing World [online]. Springer International Publishing, p. 5. [Accessed 13 October 2017].
Williams-Guillén, K., Olimpi, E., Maas, B., Taylor, P.J., and Arlettaz, R. (2016) Bats in the anthropogenic matrix: challenges and opportunities for the conservation of chiroptera and their ecosystem services in agricultural landscapes. In: Voigt, C. and Kingston, T., eds. (2016) Bats in the Anthropocene: Conservation of Bats in a Changing World [online]. Springer International Publishing, pp. 151-186. [Accessed 13 October 2017].