New to town: Home range size, habitat use and behavioral adaptations by urban hares

European hares ( Lepus europaeus ) have recently been shown to colonize urban areas in different parts of Europe. This appears to be a novel phenomenon, and little is known about the space use and behavioral adaptations of hares living in urban areas. Here, we describe the first findings concerning home range sizes from GPS-collared hares (n = 3) in Aarhus city, Denmark, which we compared to home range sizes of hares in a nearby agricultural area (n = 24). Home ranges by urban hares were ca. 3-fold smaller compared to farmland (mean ± SD: 13.7 ± 3.0 versus 38.8 ± 23.4 ha). These differences might be driven by resource availability or movement barriers (mainly roads). Moreover, we show that the flight initiation distance of urban hares was considerable shorter compared to hares in farmland (mean ± SD: 22 ± 17 versus 70 ± 45 m). This indicates that urban hares have adapted to human presence by increasing their tolerance towards human disturbance. Combined, our results provide a first glimpse into the movement ecology and behavioral adaptations of hares living in urban areas, which will provide the basis for further research. New to town: Home range size, habitat use and behavioral adaptations by urban hares

European hares ( Lepus europaeus ) have recently been shown to colonize urban areas in different parts of Europe. This appears to be a novel phenomenon, and little is known about the space use and behavioral adaptations of hares living in urban areas. Here, we describe the first findings concerning home range sizes from GPS-collared hares (n = 3) in Aarhus city, Denmark, which we compared to home range sizes of hares in a nearby agricultural area (n = 24). Home ranges by urban hares were ca. 3-fold smaller compared to farmland (mean ± SD: 13.7 ± 3.0 versus 38.8 ± 23.4 ha). These differences might be driven by resource availability or movement barriers (mainly roads). Moreover, we show that the flight initiation distance of urban hares was considerable shorter compared to hares in farmland (mean ± SD: 22 ± 17 versus 70 ± 45 m ). This indicates that urban hares have adapted to human presence by increasing their tolerance towards human disturbance. Combined, our results provide a first glimpse into the movement ecology and behavioral adaptations of hares living in urban areas, which will provide the basis for further research.

AKDE, flight initiation distance, Lepus europaeus, movement ecology, synurbanization, urban ecology.

Urban expansion brings wildlife into closer contact with humans, which can cause a range of issues, such as wildlife being perceived as nuisance, damaging vehicles and buildings, threatening human safety, and increasing the risk of disease transmission (Dobson et al. 2020; Santiago-Alarcon & MacGregor-Fors 2020). However, animals that are able to adapt to urban environments may benefit from increased resource availability and reduced predation risk, which can affect body condition and reproductive output (Seress et al. 2020). This may lead to higher wildlife population densities in urban compared to rural areas (Fingland et al. 2022). Using citizen observations and transect counts, it was previously shown that European hares ( Lepus europaeus ) are colonizing urban areas (Mayer & Sunde 2020; Bach, Escoubet & Mayer 2023; Krivopalova et al. 2024; Pagh et al. 2025). This is a novel phenomenon, as hares are typically associated with open and farmland areas (Vaughan et al. 2003). Hares have been declining in Europe since the 1960s, driven by agricultural intensification (Smith, Jennings & Harris 2005). As such, habitat deterioration in rural areas might have contributed to the initial colonization of urban areas, which potentially provide similar habitat quality compared to agricultural areas (Mayer & Sunde 2020). However, little is known about the space use and behavioral adaptations of hares living in urban areas (Bach, Escoubet & Mayer 2023; Krivopalova et al. 2024). Quantifying these adaptations will provide a broader understanding of the ecological and evolutionary processes allowing urban colonization by mammals. Although citizen observations and non-invasive monitoring methods can provide valuable information, they are usually limited in that they do not provide fine-scale information regarding animal movement and behavior. To that end, individual-based studies using detailed GPS data can be useful (Dunagan et al. 2019; Spelt et al. 2019). Here, as an initial step to better understand the space use by urban hares, we describe the home range sizes and habitat use of GPS-collared hares within urban areas compared to that of farmland hares. Moreover, as suitability of urban habitats for wildlife is linked to disturbance tolerance of humans, we compared flight initiation distances of hares in urban and farmland areas to investigate behavioral adaptations by urban hares.

Study areas and hare captures Our urban study area was in Aarhus, Denmark’s second largest city (273.000 inhabitants), mainly located in and around the main university campus (Fig. 1). This area consists of large university buildings surrounded by lawns, interspersed by single trees, bushes, hedgerows, streamlets, and small lakes. Further, the area is intersected by smaller footpaths and roads and is bordered by main roads on all sides. Human activity is generally high but varies between weekdays and weekends, and between semester periods and semester holidays. Our farmland study area was located in Syddjurs municipality (Fig. 1), ca. 15 km straight-line distance from the urban study area, and mostly consisted of arable fields (Mayer et al. 2018). In Aarhus, we captured three hares using a ca. 30 m long net, by using 2 to 3 people to drive hares into the net (Fig. 1). Once captured, hares were fitted with a GPS collar (LiteTrack 40 RF GPS PINPOINT, Lotek) that recorded one GPS position per hour from 18:00 to 8:00 and two-hourly GPS positions from 8:00 to 18:00 when hares are generally active and inactive, respectively. For details concerning hares capture and collaring in the farmland area, see Mayer et al. (2018). Data preparation and analysis of GPS data We excluded GPS positions with 5 (Lewis et al. 2007), leading to the exclusion of 6,718 inaccurate or unsuccessful position attempts, corresponding to a fix success rate of 47% in the urban study area. Moreover, we excluded the first 4 days of GPS data to account for potential capture effects (Mayer, Haugaard & Sunde 2021). This resulted in 5,875 urban GPS positions that were used for further analysis. From our farmland study area, we previously collected GPS data from 24 individuals (Mayer, Haugaard & Sunde 2021), which were subsampled to a one-hourly fix rate to be comparable with the urban data (and using the same data cleaning criteria as above). To quantify home range and movement differences between urban and rural hares, we calculated total (based on all GPS positions) and monthly 95% (total range area) and 50% (core range area) autocorrelated kernel density estimates (AKDE) (Fleming et al. 2015), and the straight line distance between consecutive hourly GPS positions during nighttime (from 18:00 to 7:59; hares generally active). For the estimation of monthly AKDEs, we only included months that were represented with > 180 fixes per month (> 2 weeks of data) to ensure adequate temporal coverage. To analyze (i) total and (ii) monthly home range sizes, and (iii) distance moved (response variable in separate analyses), we used linear mixed-effects models with a Gamma distribution and log-link of the R package ‘lme4’ (Bates et al. 2015), including area (urban versus farmland), sex, and month (monthly home range analysis only) as fixed effects and hare ID as random intercept. In addition, we describe habitat use and investigated habitat selection by the three urban hares in relation to buildings, roads, and paths separately for daytime (from 8:00 to 17:59; hares generally inactive) and nighttime. We obtained data of buildings, roads (all roads for car traffic), and paths (footpaths, bicycle paths, etc.) from OpenStreetMap (https://download.geofabrik.de/europe/). To describe available habitat, we created 5 random points for each hare GPS location, located within the hare’s 95% AKDE, which were assigned the same date and time as the corresponding hare location. We then calculated the closest distance to buildings, roads, and paths for all hare and random locations. To analyze habitat selection in relation to the distance from the closest building, road, and path (predictor variables), we used generalized additive mixed-effects models (GAMM) of the R package ‘mgcv’ (Wood 2017) with a binomial distribution and a REML approach to compare used (= 1) versus random l (= 0) locations as response variable, separately for daytime and nighttime. Hare ID was included as random effect. Hare approaches In addition, we conducted approaches of hares (that were not GPS-collared) in urban versus rural areas, measuring their flight initiation distance (FID) to quantify if they show behavioral adaptations towards living in urban areas. Data were collected in Aarhus and surrounding farmland areas (on fields with low vegetation comparable to lawns in urban areas) during April and May 2018 and 2020. Individual hares and groups (2 to 4 individuals) were approached by a single person in a straight line and at a constant walking pace (4-5 km/hour) during early mornings and late evenings (i.e., when hares were active). When hare groups were approached, one individual was assigned as the study subject before the approach. We conducted 40 approaches in urban areas and 20 in farmland. Using a handheld GPS (Garmin GPS64 or a smartphone), the person recorded their own location (1) from where the approach started (only recorded in 18 cases), (2) when the hare became aware of the approaching person (head lifted up, looking towards the person), (3) when the hare ran away, and (4) when reaching the initial position of the hare. From these locations, we then estimated the alert distance, defined as the distance from the approaching person to when the hare became alert, and the FID, defined as distance of the approaching person to the hare when it started to escape. We then analyzed if hare alert distance and FID (response variable in separate analyses) differs in urban versus farmland areas (predictor variable), using generalized linear models of the R package ‘glmmTMB’ (Magnusson et al. 2017), with a negative binomial error distribution to account for overdispersion of the data. For all analyses, we validated the model by performing dispersion and deviation tests, using the R package ‘DHARMa’ (Hartig 2021). Parameters that included zero within their 95% CI were considered uninformative (Arnold 2010). All statistical analyses were carried out in R 4.3.3 (R Core Team 2024).

and discussion Home range size and habitat use The AKDEs based on all GPS positions, our estimate for hare home range size, were almost 3-fold larger in farmland compared to urban areas (Table 1; estimate ± SE: 0.71 ± 0.31, p = 0.029). Similarly, monthly 95% and 50% AKDEs were significantly larger in farmland compared to urban areas (Fig. 2a, Table 1; Table 2). We could not address sex differences in hare home range size within the urban area, due to the small sample size. During nighttime, when hares are usually active, the straight-line distance between consecutive hourly GPS positions was on average 61 m for the urban hares compared to 95 m by hares in farmland, though this difference was uninformative (estimate ± SE: -0.17 ± 0.19, p = 0.387). Urban hares remained on average (± SD) 23 ± 18 m (range: 0 to 99 m) from buildings independent of the time of day. When compared to random positions, hares selected for proximity to buildings during nighttime and avoided close distance to buildings during the day (Fig. 2b). Conversely, hares remained closer to roads and footpaths during the day compared to nighttime (9 ± 8 versus 17 ± 11 m), a pattern also reflected in the habitat selection analysis (Fig. 2c-d). Generally, large roads appeared to constitute barriers that were rarely crossed by the hares (Fig. 1). Combined, our preliminary findings indicate that hares in urban areas generally move less compared to hares in farmland. We suggest two potential explanations, that are not mutually exclusive, for these differences. First, smaller range areas and reduced movement may result from barrier effects, mainly caused by roads, as well as the extreme fragmentation of the urban habitat, consisting of ‘islands’ of suitable habitat, such as parks and other areas with lawns (Mayer & Sunde 2020), surrounded by low-quality areas (dense urban fabric). For example, red squirrels ( Sciurus vulgaris ) rarely cross roads with high traffic volume during routine movements (Fey, Hämäläinen & Selonen 2016). Second, food availability may be higher and more stable in urban areas compared to farmland, as grass and herbs on lawns area available year-round. This may reduce the hares’ need to exploit larger areas to meet their energetic needs. Similarly, home ranges of northern raccoons ( Procyon lotor ) and Virginia opossums ( Didelphis virginiana ) were smaller in areas that included greater proportions of residential neighborhoods and commercial areas that contained high food resource availability, and they used these areas more than would be expected from their availability (Crandall et al. 2024). Additionally, home range differences between urban areas and farmland might have, at least in part, been driven by male hares generally having larger home ranges than females (Mayer et al. 2019), as we only had a single male collared in the urban area (and could have resulted from stochasticity caused by the small urban sample size). We will need more data from GPS-tagged hares in urban areas to answer if a reduction in home range size is driven by barrier effects or resource availability and stability, or both. Hare approaches Hares in urban areas had shorter alert distances and FIDs compared to hares in farmland (alert distance (mean ± SD): 29 ± 20 versus 87 ± 63 m; FID: 22 ± 17 versus 70 ± 45 m; Fig. 3, Table 3). We note that an important shortcoming here is that we did not account for the starting distance, which was not consistently reported, and was previously shown to influence FID in birds (Weston et al. 2012). Nevertheless, the ca. 3-fold shorter alert distances and EDs of urban hares suggest that they show tolerance towards human disturbance, either resulting from phenotypic plasticity in their behavioral response or from selection for more tolerant individuals. This is in line with findings from a recent study investigating hare FID in Prague and surrounding rural areas in Czechia, which also found that hares escaped earlier in farmland habitats than in urban habitats (Krivopalova et al. 2024), and with studies from other mammal species showing a reduced flight initiation distance in urban areas (Ritzel & Gallo 2020).

and future perspectives Our findings provide a first glimpse into how hares adjust to novel environments that constitute altered resources and stressors. Moving forward, it will be interesting to quantify how urban hares respond behaviorally to spatio-temporal variation in human activity, especially in the context of large events (‘Friday bars’, concerts, etc.) that happen in our urban study area. Specifically, we aim to quantify how movement, habitat selection and activity budgets by urban hares change in relation to variation in human activity, such as differences between weekdays and weekends, and between semester periods and holidays. Hares will be an optimal model species for this, because they do not use any dens or burrows and are thus more exposed to human disturbance compared to similar-sized mammals. Moreover, it will be of interest to investigate the reproductive ecology of urban hares. Combined, these results will help us to better understand adaptations and phenotypic plasticity of urban wildlife, and how this may affect population divergence. Insight into how fast animals can adjust to novel environmental conditions is crucial to predict if populations and species are able to persist in the future given current global changes, and to better understand their population dynamics.

Arnold, T.W. (2010) Uninformative parameters and model selection using Akaike’s Information Criterion. The Journal of Wildlife Management, 74, 1175-1178.Bach, H., Escoubet, H. & Mayer, M. (2023) Co-occurrence patterns and habitat selection of the mountain hare, European hare, and European rabbit in urban areas of Sweden. Mammalian Biology , 1-17.Bates, D., Mächler, M., Bolker, B. & Walker, S. (2015) Fitting Linear Mixed-Effects Models Using lme 4. J. Stat. Soft., 67, 48.Crandall, K.A., Pease, B.S., Dixon, J. & Cove, M.V. (2024) Human-derived food shrinks home ranges and alters resource selection of mammals at the urban-wild interface. Food Webs, 41, e00363.Dobson, A.P., Pimm, S.L., Hannah, L., Kaufman, L., Ahumada, J.A., Ando, A.W., Bernstein, A., Busch, J., Daszak, P. & Engelmann, J. (2020) Ecology and economics for pandemic prevention. Science, 369, 379-381.Dunagan, S.P., Karels, T.J., Moriarty, J.G., Brown, J.L. & Riley, S.P. (2019) Bobcat and rabbit habitat use in an urban landscape. Journal of Mammalogy, 100, 401-409.Fey, K., Hämäläinen, S. & Selonen, V. (2016) Roads are no barrier for dispersing red squirrels in an urban environment. Behavioral Ecology, 27, 741-747.Fingland, K., Ward, S.J., Bates, A.J. & Bremner‐Harrison, S. (2022) A systematic review into the suitability of urban refugia for the Eurasian red squirrel Sciurus vulgaris . Mammal Review, 52, 26-38.Fleming, C.H., Fagan, W.F., Mueller, T., Olson, K.A., Leimgruber, P. & Calabrese, J.M. (2015) Rigorous home range estimation with movement data: a new autocorrelated kernel density estimator. Ecology, 96, 1182-1188.Hartig, F. (2021) Package ‘DHARMa’. R package version 0.3, 3 .Krivopalova, A., Mikula, P., Cukor, J., Ševčík, R., Brynychová, K. & Šálek, M. (2024) Adaptation of farmland mammalian specialist to urban life: Escape behavior of European hare along the urban-rural gradient. Science of the Total Environment, 951, 175779.Lewis, J.S., Rachlow, J.L., Garton, E.O. & Vierling, L.A. (2007) Effects of habitat on GPS collar performance: using data screening to reduce location error. Journal of Applied Ecology, 44, 663-671.Magnusson, A., Skaug, H., Nielsen, A., Berg, C., Kristensen, K., Maechler, M., van Bentham, K., Bolker, B., Brooks, M. & Brooks, M.M. (2017) Package ‘glmmTMB’. R Package Version 0.2. 0 .Mayer, M., Haugaard, L. & Sunde, P. (2021) Scared as a hare: effects of capture and experimental disturbance on survival and movement behavior of European hares. Wildlife Biology, 2021, wlb. 00840.Mayer, M. & Sunde, P. (2020) Colonization and habitat selection of a declining farmland species in urban areas. Urban Ecosystems, 23, 543–555.Mayer, M., Ullmann, W., Heinrich, R., Fischer, C., Blaum, N. & Sunde, P. (2019) Seasonal effects of habitat structure and weather on the habitat selection and home range size of a mammal in agricultural landscapes. Landscape Ecology, 34, 2279-2294.Mayer, M., Ullmann, W., Sunde, P., Fischer, C. & Blaum, N. (2018) Habitat selection by the European hare in arable landscapes: The importance of small‐scale habitat structure for conservation. Ecology and Evolution, 8, 11619-11633.Pagh, S., Jensen, L.L., Pertoldi, C. & Larsen, H.L. (2025) Monitoring Urban European Hares ( Lepus europaeus Pallas) with Citizen Science and a Thermal Spotter. Urban Science, 9, 27.R Core Team (2024) R Core Team (2024). R: A language and environment for statistical computing. R Found. Stat. Comput. Vienna, Austria .Ritzel, K. & Gallo, T. (2020) Behavior change in urban mammals: a systematic review. Frontiers in Ecology and Evolution, 8, 576665.Santiago-Alarcon, D. & MacGregor-Fors, I. (2020) Cities and pandemics: urban areas are ground zero for the transmission of emerging human infectious diseases. Journal of Urban Ecology, 6, juaa012.Seress, G., Sándor, K., Evans, K.L. & Liker, A. (2020) Food availability limits avian reproduction in the city: An experimental study on great tits Parus major. Journal of Animal Ecology, 89, 1570-1580.Smith, R.K., Jennings, N.V. & Harris, S. (2005) A quantitative analysis of the abundance and demography of European hares Lepus europaeus in relation to habitat type, intensity of agriculture and climate. Mammal Review, 35, 1-24.Spelt, A., Williamson, C., Shamoun-Baranes, J., Shepard, E., Rock, P. & Windsor, S. (2019) Habitat use of urban-nesting lesser black-backed gulls during the breeding season. Scientific Reports, 9, 10527.Vaughan, N., Lucas, E.A., Harris, S. & White, P.C. (2003) Habitat associations of European hares Lepus europaeus in England and Wales: implications for farmland management. Journal of Applied Ecology, 40, 163-175.Weston, M., McLeod, E.M., Blumstein, D. & Guay, P.-J. (2012) A review of flight-initiation distances and their application to managing disturbance to Australian birds. Emu, 112, 269-286.Wood, S.N. (2017) Generalized additive models: an introduction with R . CRC press. Fig. 1: Map (a) shows the locations (red quadrats) of the urban study area, Aarhus city, and the farmland study area. Map (b) shows the GPS locations of three GPS-collared hares (the sex is indicated in parenthesis) in Aarhus. The pictures show (c) the process of capturing a hare using a net, and (d) an urban hare that was approached by an observer. Pictures: (c) Carlos Ebrahim Vendrell and (d) Martin Mayer. Fig. 2: Plot (a) shows predicted (large dots) monthly 95% autocorrelated kernel density estimates (AKDE) shown separately for female (dots) and male (squares) hares, and for farmland (orange) and urban areas (blue). Raw data are shown as small symbols and 95% confidence intervals as lines. Plots (b-d) show the predicted probability of use (solid lines) by the 3 urban hares in relation to the distance from the closest (b) building, (c) road, and (d) path, separately for nighttime (red) and daytime (turquoise) positions. The 95% confidence intervals are shown as shading. Fig. 3: The predicted (large dots) alert and flight initiation distance (FID) of hares in farmland and urban areas. The 95% confidence intervals are shown as bars and raw data as small dots. Tables Table 1: Overview of the number of GPS-collared hares in urban areas and farmland, showing the home range size (in ha), estimated as 95% and 50% autocorrelated kernel density estimates (AKDE), and the number and mean estimates (95% confidence intervals) of monthly 95% and 50% range area (in ha), | Farmland | Female | 10 | 20.9 (11.4 - 32.3) | 4.5 (2.4 - 6.9) | 68 | 13.5 (9.8 - 18.6) | 3.2 (2.7 - 3.7) | | Farmland | Male | 14 | 51.7 (39.9 - 91.6) | 11.1 (8.1 - 19.2) | 78 | 36 (27.3 - 47.5) | 8.4 (7.3 - 9.5) | | Urban | Female | 2 | 12.5 (9.3 - 21.8) | 2.6 (1.7 - 4.3) | 13 | 7.2 (3.9 - 13.1) | 2.3 (1.8 - 2.8) | | Urban | Male | 1 | 16.1 | 3.5 | 6 | 19.2 (10.2 - 36.4) | 3.2 (2.1 - 4.3) | Table 2: Estimate, standard error (SE), and lower (LCI) and upper (UCI) 95% confidence interval of the analyses investigating monthly home range size by GPS-collared hares in urban areas and farmland. Informative parameters are in bold. | Intercept | 2.73 | 0.21 | 2.32 | 3.15 | | Sex male | 0.98 | 0.21 | 0.58 | 1.39 | | Month 2 | 0.21 | 0.19 | -0.16 | 0.59 | | Month 3 | 0.32 | 0.20 | -0.07 | 0.71 | | Month 4 | 0.59 | 0.30 | 0.01 | 1.17 | | Month 5 | -0.10 | 0.17 | -0.43 | 0.24 | | Month 6 | -0.55 | 0.17 | -0.87 | -0.22 | | Month 7 | -0.68 | 0.16 | -0.99 | -0.37 | | Month 8 | -0.09 | 0.16 | -0.41 | 0.22 | | Month 9 | -0.04 | 0.16 | -0.35 | 0.27 | | Month 10 | 0.02 | 0.16 | -0.29 | 0.33 | | Month 11 | 0.20 | 0.16 | -0.12 | 0.52 | | Month 12 | 0.00 | 0.16 | -0.32 | 0.32 | | Area urban | -0.63 | 0.32 | -1.25 | 0.00 | Table 3: Estimate, standard error (SE), and lower (LCI) and upper (UCI) 95% confidence interval of the analyses investigating (1) alert distance and (2) flight initiation distance by 60 hares that were approached in urban areas and farmland. | Alert distance | |||| | Intercept | 4.47 | 0.15 | 4.17 | 4.77 | | Urban area | -1.10 | 0.19 | -1.47 | -0.74 | | Flight initiation distance | |||| | Intercept | 4.25 | 0.16 | 3.93 | 4.56 | | Urban area | -1.16 | 0.19 | -1.54 | -0.77 | Information & Authors Information Version history Copyright This work is licensed under a Non Exclusive No Reuse License. Collection

Authors Metrics & Citations Metrics Article Usage 353views 216downloads Citations Download citation Martin Mayer, Rasmus Mohr Mortensen, Peter Sunde. New to town: Home range size, habitat use and behavioral adaptations by urban hares. Authorea. 28 March 2025. DOI: https://doi.org/10.22541/au.174317438.83894494/v1 DOI: https://doi.org/10.22541/au.174317438.83894494/v1 If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download. For more information or tips please see 'Downloading to a citation manager' in the Help menu.