Understanding the changes in spatiotemporal patterns of two carnivores in response to different anthropogenic pressures and ecological factors in Silwood Park, Ascot, London

doi:10.21203/rs.3.rs-3963369/v1

Download PDF

Research Article

Understanding the changes in spatiotemporal patterns of two carnivores in response to different anthropogenic pressures and ecological factors in Silwood Park, Ascot, London

https://doi.org/10.21203/rs.3.rs-3963369/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Human induced habitat loss and disturbances is the driving cause of mammalian extinction. Moreover, these disturbances are also significantly affecting the spatiotemporal patterns of individual species. Two species which have been able to thrive in such human modified landscapes are the red fox (Vulpes vulpes) and the European Badger (Meles meles). Across an urban gradient both species display complex behavioral and ecological adaptations. However, there have been minimal studies on the specific impacts of anthropogenic pressures and ecological factors influence the spatiotemporal patterns of both species in semiurban landscapes. In this study, camera trapping was conducted to understand how ecological factors and anthropogenic factors influenced the spatiotemporal patterns of both species in Silwood Park, Berkshire. The study revealed that foxes were more nocturnal in grasslands and closer to roads and buildings. Surprisingly, fox relative abundance was higher with increased human activity. Meanwhile, badgers, which were completely nocturnal in this study, had higher relative abundance near buildings. Lastly, both species showed varied responses to the abundances of other recorded mammalian species. Nonetheless, these multifaceted results indicate the need for studies covering a larger urban rural gradient to understand the complex behavioral adaptations to human pressures.

Behavioral Ecology

Animal Behavior

Conservation Biology

Anthropogenic pressures, such as urbanization, is severely disrupting wildlife across the world. Other than catalysing the extinction of species, human disturbances are also inducing ecological and behavioural changes amongst local wildlife populations. One taxa particularly impacted by such pervasive human disturbances are mammals (Tucker et al., 2018). In fact, in the UK only, over 50,000 European badgers (Meles meles), which compromises around 10% of the national population, are lost to roadkill annually (Fahrig & Rytwinski, 2009). Similarly, studies have shown that human disturbances hinder the dispersal (Doherty et al., 2021), foraging and reproductive behaviour (Ciuti et al., 2012) of various mammalian species.

However, the suppression of mammalian populations by anthropogenic pressures is not ubiquitous. In fact, a global study of 468 terrestrial mammal species found a positive relation between species population density and both human population density and human footprint index (Tucker et al., 2021). This trend has been attributed to a process called species filtering, where although species diversity reduces significantly in human modified landscapes, the population density of few species increases substantially (Parsons et al., 2018).

Although anthropogenic disturbances are often associated with biodiversity loss (Fahrig & Rytwinski, 2009), they also do provide opportunities for species to exploit novel environmental conditions. Human modified landscapes have increased food availability (Tucker et al., 2018) due to direct feeding by humans, the presence of introduced species including ornamental plants, and poor waste dispersal (Chamberlain et al., 2009). Moreover, these landscapes also provide shelter in the form of infrastructure (Parsons et al., 2018). Herbivores and mesocarnivores can also benefit from the predator release effect, which is the increase in population growth of prey species in a region upon the removal of their predators (Estes et al., 2001).

However, the ability of species to adapt to and exploit these disturbed landscapes varies significantly (Tucker et al., 2018). A study of the urban adaptation of 190 mammal species found that species which exhibit a broad trophic niche and high diet diversity are most adaptable to human modified landscapes (Santini et al., 2018). Meanwhile species which are rare or have highly specialized adaptations are not able to adapt to human modified landscapes. As a result, these species are vulnerable to disturbances associated with human modified

landscapes, such as habitat fragmentation, roadkill, human wildlife conflict and the spread of invasive species ns (Davies et al., 2004). Consequently, human adapted species face reduced resource competition and can thrive at higher abundances (Ruscoe et al., 2011). Similarly, species can also benefit from the human shield effect. Specifically, larger carnivores tend to spatially avoid areas with high human disturbance, hence providing refuge for prey species (Muhly et al., 2011). For example, a study in the Rocky Mountains in Alberta, found that the population density of predators reduced significantly on trails with high human footfall while the density of prey on the trails increased substantially (Muhly et al., 2011). This spatial avoidance of areas of high human activity can be attributed to the higher prevalence of human induced mortality, compared to natural predation, amongst mesocarnivores. As humans assume the role of super predators in ecosystems across the world, the landscapes of fear of predators are being significantly altered (Smith et al., 2017).

To minimise mortality risk in human modified landscapes, species often adopt shifts in temporal niches. A global analysis found that mammals have displayed an increase in nocturnal behaviour in areas with high human disturbance relative to areas with low disturbance (Gaynor et al., 2018). These shift in temporal patterns have severe implications for predator- prey interactions. For example, in Tanzania, human induced behavioural changes led to a significant increase in temporal overlap between predator and prey species (Mills & Harris, 2020).

Two species which are found across human modified landscapes are the red fox and Eurasian badger. Both species are highly adapted generalist carnivores, which allow for the exploitation for urban landscapes (Gil-Fernández et al., 2020; Mathews et al., 2018). In fact, in urban landscapes, both species have displayed increased activity in built up environments (Lovell et al., 2022; Bateman & Fleming, 2012).

Furthermore, both species display nuanced behavioural responses to anthropogenic pressures. For example, while urban foxes have displayed habituation to human activity, rural populations are timid and avoid human activity (Morton et al., 2023). Similarly, foxes are more nocturnal in areas with high hunting pressure from humans (Servin et al., 1991). Such distinct patterns are also prevalent amongst badger populations. Urban badgers have reached higher population densities compared to rural populations (Piza-Roca et al., 2018). Contrastingly, in rural landscapes, badgers avoid habitats associated to human settlements

(Lara-Romero et al., 2012). Moreover, the habitat usage of badgers in urban landscapes is ambiguous with numerous studies observing highly varied patterns (Cresswell and Harris, 1988; Davison et al., 2009; Kauhala and Auttila,2010). Furthermore, despite adaptation to urban landscapes, badgers still show spatial avoidance of humans (Lovell et al., 2022).

Most studies have focused on the adaptations and behaviour of both species in urban landscapes. Meanwhile, there have been minimal studies focusing on the impacts of anthropogenic disturbances on the spatiotemporal patterns of both species in semiurban landscapes. Furthermore, most studies fail to account for how ecological factors, such as habitat and interspecies interactions, influence spatiotemporal patterns in semiurban landscapes.

This study aims to understand the impacts of roads, built-up areas, and human activity on the habitat usage and spatiotemporal of both species in a semiurban landscape. This was done through camera trap deployments across 50 locations in Imperial College London’s Silwood Park campus, based in Ascot, Berkshire. Temporal patterns at each deployment were determined by the species’ nocturnality, defined as the proportion of activity or observation occurring at night. Meanwhile spatial patterns were analysed by measuring the relative abundances of both species at each camera deployment. Relative abundance at each deployment was estimated using the number of records of a focal species per trap hour (Sollmann et al., 2013). This study aimed to detect how the nocturnality and relative abundance of both species varied with the proximity to buildings, proximity to roads, human activity, and habitat. Additionally, the study attempted to study how relative abundance and nocturnality of both species varied with that of other recorded species in Silwood Park. It was predicted that the spatiotemporal patterns of both species would be optimized to best exploit the novel environments associated with human disturbances, while minimizing the temporal exposure to such disturbances. Additionally, it was predicted that the spatiotemporal patterns of both species would be positively influenced by those of their prey species.

2.1 Study Site

The study was conducted in Silwood Park located in Sunninghill in Ascot Berkshire (51.406°-51.417°N, 0.655°-0.637°E). The study area covers 0.91km², out of which 0.73 km² covers a wide variety of natural habitats, including lowland deciduous woodlands, traditional orchard, wood pastures, grassland, parkland, and wet woodlands. The remainder of the campus includes built up area in the form of academic buildings, a science park and residential accommodation. Silwood Park is one of the 29 Biodiversity Opportunity Areas in Berkshire County and forms a critical link in the Sunninghill and Ascot green corridors. At least nine species of terrestrial and arboreal mammals have been observed at Silwood Park (Martin-Garcia et al., 2022) .

The boundary of the campus is surrounded by three diverse types of roads – London Road (Category A) in the south, Buckhurst Road (Category B) in the east and Cheapside Road (Category C) in the north and west. The buildings are scattered in the south-east corner of the campus with a few sporadic buildings. The map of Silwood Park is below in Fig. 1, with the locations of camera trap deployments.

2.2 Camera Trapping

2.2.1 Camera Trap Settings

30 Browning Trail Camera Traps (Model BTC 5FHD) were used for the study. The cameras were set to trail (picture) mode with photo quantity kept to low quality. To estimate relative abundance, the ideal settings were to have the highest number of photos per trigger with shortest interval between photos. Therefore, camera was set to multi-shot mode with fast trigger speed and a one second picture delay. The motion detection was set to 80 feet range with long range Infrared Flash Power.

2.2.2 Camera Calibration

The camera trap settings were calibrated utilizing a 1m calibration pole. The poles were initially placed vertically on the base of the ground, 1m directly in front of the camera for a few seconds to ensure a clear image. This was repeated for further 10–20 pole placements across the field of view and away from the camera with placements spaced about 1m apart and more whenever possible. Furthermore, each deployment was also calibrated using the same technique, continuing away from the camera to at least the maximum extent that any animals are likely to be captured.

2.2.3 Camera Trap Deployment

Camera traps were deployed between 1st May and 27th July 2023. Each deployment lasted a period of 20–25 days, amounting to a total effort of 26,594 hours. Cameras were deployed in 50 different predetermined locations in a systematic grid structure. This was to ensure each deployment was random with respect to animal movement, a prerequisite of measuring abundance through camera trapping.

Camera traps were placed parallelly to the ground between 30-45cm above ground level on vertical surfaces, such as poles and trees. It was ensured that there were no young plants are likely to grow or move in front of the camera after deployment. At all deployments, the cameras were placed facing either north or south. Cameras facing east or west were avoided as sunlight could affect photos at sunrise and sunset if the camera position were not shaded. Calibration was conducted once the cameras were rigidly attached to the structure.

2.3 Data Processing

2.3.1 Annotation of Camera Trap Images

Agouti software was used to annotate camera trap photographs. The software’s inbuilt artificial intelligence programs automatically arranged all captures made within 120 seconds of each other into photo sequences. The software also utilized the Western European species model (Version 4a) to automatically identify species in each image. However, the locations of the animal had to be added manually for each observation. Moreover, in images where multiple individuals were present, observations, annotated with locations, had to be added for each animal. Agouti utilized the locations of each observation to calculate angle, radius, and speed, which would then be utilized to calculate species abundances. The calibration poles were also manually calibrated using the Agouti software. Data was then exported from Agouti as a Camera Trap Data Package for further analysis on R.

2.3.2 Measuring Anthropogenic Disturbances

Human Activity was measured at each deployment by hourly human encounter rate. The total number of humans observed at each deployment was divided by the total number of camera trap hours for the respective deployment to produce the hourly human encounter rate. The proximity from each deployment to the nearest building was calculated afterwards using the “Measure Distance” function on Google Maps. The same process was used to measure proximity to nearest roads for each building. Habitat was classified as either grassland or woodland. Although these classifications were slightly simplistic as Silwood is a mosaic of habitats, including orchards, open woodland, parkland, wet woodland. None of the deployments were in the orchards. Moreover, the effect size for each specific habitat would be too small, hence the simplistic classification of habitats.

Nocturnality was measured as a proportion of observations which occurred in night hours, which was defined as the time between sunset and sunrise. Night hours were determined using daily sunrise and sunset times (Past weather in Ascot, England, United Kingdom, 2023).

The relative abundance of individual species was calculated using effort per unit effort (Sollmann et al., 2013). Both, the number of individuals of each focal species at each deployment and camera trapping effort was extracted using the “camtraptor” package on R (Oldoni et al., 2023). During regression analysis to study the impact of distinct factors on abundance, number of records were placed as response variable, while offsetting a logarithms of effort to provide species trap rate. species per effort.

2.4 Statistical Analysis

2.4.1 Nocturnality

A quasibinomial regression model was utilized to analyze the impacts of proximity to buildings, human encounter rate and proximity to roads on fox nocturnality while accounting for habitat. Non-significant terms were removed to produce a minimum adequate model. Additionally, the individual impact of habitat on fox nocturnality was analyzed utilizing linear regression. No analysis was conducted upon badger nocturnality as all observations of the species in this study occurred during night hours.

2.4.2 Abundance

One negative binomial regression model was conducted each for both species. This was to study the individual impacts of habitat, proximity to roads, human activity, and proximity to buildings on the relative abundance of both species. Following stepwise backward model selection, insignificant explanatory variables were removed to produce a minimum adequate model. Relative abundance was measured by the number of records of the focal species in each camera trap as the response variable and offsetting the log of camera trap effort (time). Negative binomial regressions were preferred over Poisson regression due to the high overdispersal in the latter. The impact of habitat on the abundance of both species was measured utilizing two separate linear regression models.

2.4.3 Spatial Overlap of Species in Silwood

Other than badgers and foxes, four other species were recorded across deployments. These species are roe deer (Capreolus capreolus), Reeve’s muntjac (Muntiacus reevesi), Eastern Gray squirrel (Sciurus carolinensis) and European Rabbit (Oryctolagus cuniculus). RAI was calculated for all six species.

A negative binomial regression was conducted to study the combined impact of the relative abundance index of the remaining five species on that of foxes. Species with insignificant correlations were removed to produce a minimum adequate model. Again, the log of effort was offset all nine models. The model was repeated for badgers.

2.4.4 Temporal Overlap of Species in Silwood

Similarly, a single quasibinomial regressions was conducted to study the individual relationships of the nocturnality of foxes with all other species, baring badgers. Like the analysis on spatial overlaps of species, habitat was not included in these models. No such study was conducted for badgers as they were completely nocturnal in this study.

3.1 Spatial

3.1.2 Badger (Meles meles)

There was a mean badger abundance of 1.22 (SD = 2.23) across the camera deployments (n = 50). A negative binomial regression revealed that badger abundance was only impacted by proximity to buildings. For every metre away from buildings, badger relative abundance reduced by a factor of 0.990 (SE = 0.18, df = 48, p < 0.01). The model had a pseudo-R-squared value of 0.14. Neither human encounter rate nor proximity to roads. Additionally linear regression identified that habitat did not influence badger abundance.

3.1.2 Fox (Vulpes vulpes)

Across camera trap deployments (n = 50), the mean abundance of the red fox was 9.78 (SD = 23.84). Negative binomial regressions found that fox abundance was only impacted by human encounter rate. Both proximity to roads and buildings did not impact fox abundance. Meanwhile, linear regression found that habitat did not impact fox abundance.

There was a significant positive relationship between fox abundance and human encounter rate. For every additional human recorded per hour, fox abundance increased by a factor of 1.15 (SE = 0.09, df = 48, p < 0.01). The model had a pseudo-R-squared value of 0.10. An insignificant correlation was observed between fox abundance and proximity to Lastly, there was no difference between fox relative abundance in woodlands and grasslands.

3.2 Temporal Patterns

Fox nocturnality displayed varied responses to human encounter rate, proximity to roads and proximity to buildings. Quasibinomial regression analysis revealed that fox nocturnality had significant relationship with both, proximity to buildings, and proximity to roads. Meanwhile, linear regression suggested that habitat influenced fox nocturnality.

Specifically, for each metre further away from buildings, fox nocturnality reduced by a factor of 0.998 (SE = 0.003, df = 40, p = 0.051). Beyond 174.53 m from roads, fox nocturnality reduced below 50%. This model had a pseudo-R-squared value of 0.083.

Meanwhile, a metre increase in distance to roads reduced fox nocturnality by a factor of 0.999 (SE = 0.002, df = 40, p = 0.058). When distance from nearest road was greater than 309.63m, fox nocturnality reduced below 50%. The model had a pseudo-R-squared value of 0.078. Additionally, fox nocturnality reduced by 25.4% in woodland compared to grassland (F-value = 3.965, df = (1,40), p-value: 0.053). Lastly, human encounter rate did not impact fox nocturnality (df = 40, P = 0.161). This model had a pseudo-R-squared value of 0.068.

3.3. Impact of Interspecific Relationships on the spatiotemporal patterns of foxes

3.3.1 Spatial Patterns

The relative abundances of foxes had varied correlations with that of the other recorded species. Significant positive correlations between the relative abundance of foxes and European Hare (SE < 0.01, df = 46, p < 0.01) and Roe Deer (SE = 0.01, df = 46, p < 0.01). For every additional hare, the relative abundance of foxes increased by a factor of 1.01. Meanwhile, for every additional roe deer, the relative abundance of foxes increased by a factor of 1.03. There was no significant correlation between the relative abundance of foxes and that of either badgers or squirrels. Lastly, a near-significant positive correlation was observed between the relative abundances of foxes and muntjacs (SE < 0.01, df = 46, p = 0.08). The model had a pseudo R-squared value of 0.31.

3.3.2 Temporal Patterns

Fox nocturnality was not influence by the temporal patterns of any of the other recorded mammalian species. There was an insignificant positive correlation noted between fox nocturnality and that of roe deer (F-value = 0.940, df = (1,40), p = 0.338), muntjac (F-value = 0.143, df = (1,40), p = 0.708), squirrels (F-value = 0.002, df = (1,40), p = 0.962) and rabbits (F-value = 2.045, df = (1,40), p = 0.161).

3.4. Impact of Interspecific Relationships on Badger Spatial Patterns

The relative abundance of Badgers only had significant correlations with that of hares and roe deer. For every additional roe deer, the relative abundance of badgers reduced by a factor of 0.95 (SE = 0.02, df = 47, p < 0.05). Meanwhile, the relative abundance of badgers increased by a factor of 1.01 for each additional increase in the relative abundance of hares (SE < 0.01, df = 47, p < 0.01). No significant correlation was noted between the relative abundance of badgers and that of fox, squirrel and munjacts,

4.1. Spatiotemporal Patterns of Foxes

The response of fox relative abundance to the three anthropogenic pressures was multifaceted. Fox relative abundance was not influenced by proximity to buildings. This is inconsistent with previous studies which found that foxes have been able to achieve higher population densities in built up studies (Lovell et al., 2022). This phenomenon has been attributed to the ability of red foxes to exploit novel anthropogenic food sources (Gil-Fernández et al., 2020). Meanwhile the results of this study found that fox relative abundance had a strong positive correlation with the relative abundances of hare, their primary prey. This indicates that prey availability could be a more influential factor in habitat selection.

However, this study also found that foxes were more nocturnal closer to buildings. Previous studies have found a substantial reduction in diurnal activities of fox near human settlements in the Mediterranean (Díaz-Ruiz et al., 2015). Similarly, around the UK, fox display increased nocturnality in urban areas compared to rural (Lovell et al., 2022). Minimizing temporal overlap with human activity would minimize risk from anthropogenic disturbances. This would allow for the proper exploitation of the novel environments associated with human dominated landscapes associated (Lovell et al., 2022).

There was no correlation between the relative abundance of foxes and proximity to road in this study. Previous found that fox activity reduced significantly closer to roads (Fahrig & Rytwinski, 2009). However, fox nocturnality increased with proximity to roads. This is consistent with previous studies where various canid species displayed increased nocturnality closer to roads. Increased nocturnality closer to roads would minimize mortality risk due to reduced traffic (Kautz et al., 2021).

Surprisingly, the relative abundance of foxes increased in areas with high human encounter rate. Previous studied found that foxes in rural and suburban landscapes were less habituated to humans (Padovani et al., 2021). Moreover, foxes in this study did not display any temporal shifts in response to increased human activity. The high tolerance of foxes towards human activity at Silwood would need to be investigated further.

In this study, fox relative abundance did not significantly change across habitats. However, foxes were significantly more diurnal in woodlands. Similarly, introduced fox populations in coastal Australia were found to be more active in areas with higher vegetation (Kimber et al., 2020). In habitats, such as woodlands, increased vegetation reduces conspicuousness of an animal and hence risk. This would allow for not only increased activity, but also more diurnal activity. Meanwhile, in open habitats, such as grasslands, increased nocturnality is a critical adaptation to reduce conspicuousness. (Schwegmann et al., 2023).

4.2 Spatiotemporal Patterns of Badgers

Badger relative abundance was only influenced by proximity to buildings. Closer to buildings, the relative abundance of badgers increased significantly. This is consistent with previous studies in urban areas, which found increased badger activity near built up areas (Lovell et al., 2022). In fact, the ability of badgers to exploit anthropogenic food sources and shelter (Bateman & Fleming, 2012), have allowed them to achieve higher population densities than those under natural conditions (Piza-Roca et al., 2018).

Surprisingly, roads did not have a significant impact on the abundance of badgers, particularly as roadkill is the leading source of mortality amongst badgers in the United Kingdom (Tuyttens et al., 2001). Previous studies have found a decrease in badger activity in areas with higher density of paved roads (Bateman & Fleming, 2012). Moreover, studies in Netherlands (Piza-Roca et al., 2018), Italy (Balestrieri et al., 2009) and Scotland (Silva et al., 2017) have indicated that in rural landscapes, badger abundance and activity reduces in close proximity to urban infrastructure. This is indicative of the risk averse nature of rural badgers.

In this study, badger abundance was not impacted by either human activity nor habitat. This is surprising as a broad range of studies have found that human disturbance negatively affected both badger abundance (Lara-Romero et al., 2012) and badger activity (Lovell et al., 2022). Moreover, previous studies have suggested that badgers have a strong preference of woodlands in human modified landscapes (Lovell et al., 2022). Another potential cause of a lack of correlation between badger relative abundance and proximity to roads and human trap rate in this study could be due to the low sample size and small scales of effects. A similar study is required over a larger rural-urban gradient, covering a larger variety of habitats is required to understand the driver of spatiotemporal patterns of suburban badgers.

4.3. Impact of Interspecific Interactions spatiotemporal patterns

4.3.1 Fox

Eropean rabbits from a large proportion of the red fox diet in the United Kingdom (Baker et al., 2006). It is no surprise, therefore, that a significant positive correlation between the relative abundance of foxes and rabbits were observed in this study. Moreover, this pattern is consistent with previous studies which have found that fox activity increased with rabbit availability (Díaz-Ruiz et al., 2015). Surprisingly, there was no significant correlation observed between squirrels and foxes, as there was numerous anecdotal evidence of depredation of the former in this study.

There was no correlation between the relative abundance of foxes and muntjacs. This was expected because both muntjacs and roe deer are very rarely predated upon by red fox (Doncaster et al., 1990). Surprisingly, the relative abundances of roe deer and foxes were positively correlated. At most, the spatial avoidance of red fox was expected as numerous roe deer fawn were recorded in the camera trapping effort. This is because previous studies have implied that there are significant lifetime fitness benefits to rob deer inhabiting highly productive, predator-free habitats (Panzacchi et al., 2008). However, the habitats in Silwood may act as attractive sinks, where the costs of high predation risk may outweigh the potential fitness benefits of high resource habitat (Panzacchi et al., 2008). An interesting extension to this study would be to study the impacts of anthropogenic disturbances on the energetics of predator-prey interactions. However, a larger study site than Silwood would be required.

Moreover, while this study attemps to study spatiotemporal patterns within a landscape, a between landscape study also might prove useful. This will allow for a more comprehensive analysis across a more extensive disturbance gradient. This type of study would also counter for random variations in affecting spatiotemporal patterns.

Surprisingly, no correlation between the relative abundance index of badgers and fox. Previous studies have found significant spatial avoidances between the two species (Ferreiro-Arias et al., 2021). Meanwhile, previous studies have found high spatial overlap between the two species, which could be fostered by the high dietary niche partitioning between the two species (Mori & Menchetti, 2019).

Fox nocturnality was not impacted by that of other species. This is contradictory to previous studies which have also found temporal overlap between fox and prey species (Díaz-Ruiz et al., 2015). A multi-species distribution model might be useful in this scenario to understand how human disturbances impact the temporal overlap between species. Moreover, additional ecological factors, such as proximity to water sources, diurnal temperature, habitat cover, could also be included in further studies. This could prove more insightful in understanding the drivers ot spatiotemporal patterns in both of the studied species.

4.3.2. Badgers

As predicted, the relative abundance of badger was positively correlated with those of rabbits. Badgers are intraguild predators of these small mammals, such as rabbits and squirrels (Cleary et al., 2009). While these small mammals are often depredaded upon by badger, they also share a similar diet of fruits and nuts. (Hof et al., 2019). Spatial overlap of badgers with rabbit were therefore expected. Surprisingly, badger abundance was negatively correlated with the abundances of roe deer. There have been very few previous studies focusing on the spatiotemporal pattern of both species. Hence conclusions regarding this observation will be inappropriate.

Human disturbances had induced changes in spatial and temporal in foxes and badgers, respectively. However, these results, like previous studies remain ambiguous. For example, badger abundance reduced with increasing human activity in woodlands but increased in grasslands. Moreover, underlying driver behind increased abundance of badgers in built up area, as observed in this and prior studies remains unclear. Moreover, foxes did not display any spatiotemporal avoidance in areas with increased human activity, contradicting the idea of rural fox populations being timider. Meanwhile foxes did show increased nocturnality near roads and buildings, hence reducing temporal overlap with high human activity. This could potentially allow the species to properly the novel environments associated with human modified landscape.

Nonetheless, to truly comprehend the impact of anthropogenic pressures on the ecology and spatiotemporal patterns of UK’s two largest carnivores, a more elaborate study is required. Both ecological and anthropogenic pressures need to be incorporated into studies over a larger geographical range, covering an extensive urban-rural gradient.

I declare that the results presented in this study was based on the data I collected during fieldwork carried out by myself between May and July 2023. My supervisors, Professor Rob Ewers, and Dr. Marcus Rowcliffe provided me with guidance regarding sampling design for the camera trap survey. Dr. Catalina Estrada supported me with the execution of the required fieldwork. All images collected from camera trapping were annotated and exported in the appropriate format through the Agouti software, developed by the Research Institute for Nature and Forest (INBO). To estimate mammalian densities, I conducted Random Encounter Modelling, a method developed by Dr. Marcus Rowcliffe. To conduct this analysis, I utilized the “camtraptor” package on R, developed by Dr. Damiano Oldnoi, Mr. Peter Desmet and Mr. Pieter Hybrchts from INBO.

I declare that this thesis is an original work by myself and all materials which are not original are quoted and referenced properly.

Data Code and Availability Statement

All data collected and utilized in this study has been made public, alongside the R code used for analysis. This can be accessed here: https://github.com/AnishWildlife/Silwood- Mammal-s-Master-s-Thesis

Acknowledgements

I thank both my supervisors, Professor Rob Ewers (Imperial College London) and Dr. Marcus Rowcliffe, for their constant support and guidance throughout the project. I would also like to acknowledge the logistical support provided by Dr. Catalina Estrada during fieldwork. Additionally, I would like to acknowledge the valuable feedback provided by members of my lab group, especially Dr. David Orme, Dr. Jacob Cook, Dr. Olivia Daniel, Tijmen de Lorm. Lastly, I would like to acknowledge the feedback and support of my peers, including Mahika Dixit, Daniel Dunleavy, and Megan Worsely.

Baker, P. et al. (2006) ‘The potential impact of red fox vulpes vulpes predation in agricultural landscapes in lowland Britain’, Wildlife Biology, 12(1), pp. 39–50. doi:10.2981/0909- 6396(2006)12[39:tpiorf]2.0.co;2.
Balestrieri, A., Remonti, L. and Prigioni, C. (2009) ‘Exploitation of food resources by the Eurasian badger (Meles meles) at the altitudinal limit of its alpine range (NW Italy)’, Zoological Science, 26(12), pp. 821–827. doi:10.2108/zsj.26.821.
Bateman, P.W. and Fleming, P.A. (2012a) ‘Big City Life: Carnivores in urban environments’, Journal of Zoology, 287(1), pp. 1–23. doi:10.1111/j.1469- 7998.2011.00887.x.
Bonnot, N. et al. (2012) ‘Habitat use under predation risk: Hunting, roads and human dwellings influence the spatial behaviour of Roe Deer’, European Journal of Wildlife Research, 59(2), pp. 185–193. doi:10.1007/s10344-012-0665-8.
Bonnot, N.C. et al. (2017) ‘Stick or twist: Roe deer adjust their flight behaviour to the perceived trade-off between risk and reward’, Animal Behaviour, 124, pp. 35–46. doi:10.1016/j.anbehav.2016.11.031.
Brown, J.S. and Kotler, B.P. (2004) ‘Hazardous duty pay and the foraging cost of predation’, Ecology Letters, 7(10), pp. 999–1014. doi:10.1111/j.1461-0248.2004.00661.x.
Chamberlain, D.E. et al. (2009) ‘Avian productivity in urban landscapes: A review and meta- analysis’, Ibis, 151(1), pp. 1–18. doi:10.1111/j.1474-919x.2008.00899.x.
Ciuti, S. et al. (2012) ‘Effects of humans on behaviour of wildlife exceed those of natural predators in a landscape of fear’, PLoS ONE, 7(11). doi:10.1371/journal.pone.0050611.
Cleary, G.P. et al. (2009) ‘The diet of the Badger Meles Meles in the Republic of Ireland’, Mammalian Biology, 74(6), pp. 438–447. doi:10.1016/j.mambio.2009.07.003.
Clinchy, M. et al. (2016) ‘Fear of the human “Super predator” far exceeds the fear of large carnivores in a model mesocarnivore’, Behavioral Ecology, pp. 1826–1832. doi:10.1093/beheco/arw117.
Coleman, J.L. and Barclay, R.M. (2011) ‘Urbanization and the abundance and diversity of Prairie Bats’, Urban Ecosystems, 15(1), pp. 87–102. doi:10.1007/s11252-011-0181-8.
Coulon, A. et al. (2008) ‘Inferring the effects of landscape structure on roe deer (capreolus capreolus) movements using a step selection function’, Landscape Ecology, 23(5), pp. 603–614. doi:10.1007/s10980-008-9220-0.
Cresswell, W.J. and Harris, S. (1988) ‘Foraging behaviour and home-range utilization in a suburban Badger (Meles tneles) population’, Mammal Review, 18, pp. 37–49. doi:https://doi.org/10.1111/j.1365-2907.1988.tb00069.x. Data Packages_. R package version 0.20.0.
Davies, K.F., Margules, C.R. and Lawrence, J.F. (2004) ‘A synergistic effect puts rare, specialized species at greater risk of extinction’, Ecology, 85(1), pp. 265–271. doi:10.1890/03-0110.
Davison, J. et al. (2009) ‘Restricted ranging behaviour in a high-density population of urban badgers’, Journal of Zoology, 277, pp. 45–53. doi:https://doi.org/10.1111/j.1469- 7998.2008.00509.x.
Díaz‐Ruiz, F. et al. (2015) ‘Drivers of red fox (Vulpes vuples) daily activity: Prey availability, human disturbance or habitat structure?’, Journal of Zoology, 298(2), pp. 128–138. doi:10.1111/jzo.12294.
Doherty, T.S., Hays, G.C. and Driscoll, D.A. (2021) ‘Human disturbance causes widespread disruption of animal movement’, Nature Ecology & Evolution, 5(4), pp. 513–519. doi:10.1038/s41559-020-01380-1.
Doncaster, C.P., Dickman, C.R. and Macdonald, D.W. (1990) ‘Feeding ecology of red foxes (Vulpes vulpes) in the city of Oxford, England’, Journal of Mammalogy, 71(2), pp. 188–194. doi:10.2307/1382166.
Dupke, C. et al. (2016) ‘Habitat selection by a large herbivore at multiple spatial and temporal scales is primarily governed by Food Resources’, Ecography, 40(8), pp. 1014–1027. doi:10.1111/ecog.02152.
Fahrig, L. and Rytwinski, T. (2009) ‘Effects of roads on Animal Abundance: An empirical review and synthesis’, Ecology and Society, 14(1). doi:10.5751/es-02815-140121.
Feore, S. and Montgomery, W.I. (1999) ‘Habitat effects on the spatial ecology of the European badger Meles meles’, Journal of Zoology, 247(4), pp. 537–549. doi:10.1111/j.1469-7998.1999.tb01015.x.
Ferreiro‐Arias, I. et al. (2021) ‘Fine‐scale coexistence between Mediterranean mesocarnivores is mediated by spatial, temporal, and trophic resource partitioning’, Ecology and Evolution, 11(22), pp. 15520–15533. doi:10.1002/ece3.8077.
Gaynor, K.M. et al. (2018a) ‘The influence of human disturbance on Wildlife Nocturnality’, Science, 360(6394), pp. 1232–1235. doi:10.1126/science.aar7121.
Gil-Fernández, M. et al. (2020) ‘Adaptations of the red fox (Vulpes vulpes) to urban environments in Sydney, Australia’, Journal of Urban Ecology, 6(1). doi:10.1093/jue/juaa009.
Harris, S. (1981) ‘An estimation of the number of foxes (Vulpes vulpes) in the city of Bristol, and some possible factors affecting their distribution’, The Journal of Applied Ecology, 18(2), pp. 455–465. doi:10.2307/2402406.
Harris, S. (1984) ‘Ecology of urban badgers Meles Meles: Distribution in Britain and habitat selection, persecution, food and damage in the city of Bristol’, Biological Conservation, 28(4), pp. 349–375. doi:10.1016/0006-3207(84)90041-7.
Hof, Allen and Bright (2019) ‘Investigating the role of the Eurasian badger (Meles Meles) in the nationwide distribution of the Western European Hedgehog (erinaceus europaeus) in England’, Animals, 9(10), p. 759. doi:10.3390/ani9100759.
Hohnen, R. et al. (2016) ‘Occupancy of the invasive feral cat varies with Habitat complexity’, PLOS ONE, 11(9). doi:10.1371/journal.pone.0152520.
Huck, M., Davison, J. and Roper, T.J. (2008a) ‘Predicting European badger Meles meles sett distribution in urban environments’, Wildlife Biology, 14(2), pp. 188–198. doi:10.2981/0909-6396(2008)14[188:pebmms]2.0.co;2.
Kauhala, K. and Auttila, M. (2010) ‘Habitat preferences of the native badger and the invasive raccoon dog in southern Finland’, Acta Theriologica , 55, pp. 231–240. doi:http://dx.doi.org/10.4098/j.at.0001-7051.040.2009.
Kautz, T.M. et al. (2021) ‘Large carnivore response to human road use suggests a landscape of coexistence’, Global Ecology and Conservation, 30, pp. 1–12. doi:10.1016/j.gecco.2021.e01772.
Kelt, D.A. and Van Vuren, D.H. (2001) ‘The ecology and Macroecology of mammalian home range area’, The American Naturalist, 157(6), pp. 637–645. doi:10.1086/320621.
Kimber, O. et al. (2020) ‘The Fox and the beach: Coastal landscape topography and urbanisation predict the distribution of carnivores at the edge of the sea’, Global Ecology and Conservation, 23. doi:10.1016/j.gecco.2020.e01071.
Lara-Romero, C. et al. (2012) ‘Habitat selection by European badgers in Mediterranean semi- arid ecosystems’, Journal of Arid Environments, 76, pp. 43–48. doi:10.1016/j.jaridenv.2011.08.004.
Lovell, C. et al. (2022) ‘The effect of habitat and human disturbance on the spatiotemporal activity of two urban carnivores: The results of an intensive camera trap study’, Ecology and Evolution, 12. doi:https://doi.org/10.1002%2Fece3.8746.
Martin-Garcia, S. et al. (2022) ‘Comparing relative abundance models from different indices, a study case on the red fox’, Ecological Indicators, 137, p. 108778. doi:10.1016/j.ecolind.2022.108778.
McKinney, M.L. (2006) ‘Urbanization as a major cause of biotic homogenization’, Biological Conservation, 127(3), pp. 247–260. doi:10.1016/j.biocon.2005.09.005.
Miller, J.R.B. and Schmitz, O.J. (2019) ‘Landscape of fear and human-predator coexistence: Applying spatial predator-prey interaction theory to understand and reduce carnivore- livestock conflict’, Biological Conservation, 236, pp. 464–473. doi:10.1016/j.biocon.2019.06.009.
Mills, K.L. and Harris, N.C. (2020) ‘Humans disrupt access to prey for large African carnivores’, eLife, 9. doi:10.7554/elife.60690.
Mori, E. and Menchetti, M. (2019) ‘Living with roommates in a shared den: Spatial and temporal segregation among semifossorial mammals’, Behavioural Processes, 164, pp. 48–53. doi:10.1016/j.beproc.2019.04.013.
Morton, F.B. et al. (2023) ‘Urban Foxes are bolder but not more innovative than their rural conspecifics’, Animal Behaviour, 203, pp. 101–113. doi:10.1016/j.anbehav.2023.07.003.
Muhly, T.B. et al. (2011) ‘Human activity helps prey win the predator-prey space race’, PLoS ONE, 6(3), pp. 1–8. doi:10.1371/journal.pone.0017050.
Murphy, A. et al. (2021) ‘Threading the needle: How humans influence predator–prey spatiotemporal interactions in a multiple‐predator system’, Journal of Animal Ecology, 90(10), pp. 2377–2390. doi:10.1111/1365-2656.13548.
Oldoni D, Desmet P, Huybrechts P (2023). _camtraptor: Read, Explore and Visualize Camera Trap
Padovani, R., Shi, Z. and Harris, S. (2021) ‘Are British urban foxes (Vulpes vulpes) “bold”? The importance of understanding human–wildlife interactions in urban areas’, Ecology and Evolution, 11(2), pp. 835–851. doi:https://doi.org/10.1002%2Fece3.7087.
Palencia, P. et al. (2022) ‘Random encounter model is a reliable method for estimating population density of multiple species using camera traps’, Remote Sensing in Ecology and Conservation, 8(5), pp. 670–682. doi:10.1002/rse2.269.
Panzacchi, M. et al. (2008) ‘When a generalist becomes a specialist: Patterns of red fox predation on roe deer fawns under contrasting conditions’, Canadian Journal of Zoology, 86(2), pp. 116–126. doi:10.1139/z07-120.
Panzacchi, M. et al. (2008) ‘When a generalist becomes a specialist: Patterns of red fox predation on roe deer fawns under contrasting conditions’, Canadian Journal of Zoology, 86(2), pp. 116–126. doi:10.1139/z07-120.
Parsons, A.W. et al. (2018) ‘Mammal communities are larger and more diverse in moderately developed areas’, eLife, 7. doi:10.7554/elife.38012.
Pita, R. et al. (2020) ‘Roads, forestry plantations and hedgerows affect badger occupancy in intensive Mediterranean farmland’, Agriculture, Ecosystems & Environment, 289, p. 106721. doi:10.1016/j.agee.2019.106721.
Piza‐Roca, C. et al. (2018) ‘European badger habitat requirements in the Netherlands – combining ecological niche models with neighbourhood analysis’, Wildlife Biology, 2018(1), pp. 1–11. doi:10.2981/wlb.00453.
Raymond, S. et al. (2021) ‘Temporal patterns of wildlife roadkill in the UK’, PLOS ONE, 16(10). doi:10.1371/journal.pone.0258083.
Ruscoe, W.A. et al. (2011) ‘Unexpected consequences of control: Competitive vs. Predator release in a four-species assemblage of invasive mammals’, Ecology Letters, 14(10), pp. 1035–1042. doi:10.1111/j.1461-0248.2011.01673.x.
Šálek, M., Drahníková, L. and Tkadlec, E. (2014) ‘Changes in home range sizes and population densities of carnivore species along the natural to urban habitat gradient’, Mammal Review, 45(1), pp. 1–14. doi:10.1111/mam.12027.
Santini, G. et al. (2022) ‘Population assessment without individual identification using camera-traps: A comparison of four methods’, Basic and Applied Ecology, 61, pp. 68–81. doi:10.1016/j.baae.2022.03.007.
Santini, L. et al. (2016) ‘A trait‐based approach for predicting species responses to environmental change from sparse data: How well might terrestrial mammals track climate change?’, Global Change Biology, 22(7), pp. 2415–2424. doi:10.1111/gcb.13271.
Santini, L. et al. (2018) ‘One strategy does not fit all: Determinants of urban adaptation in mammals’, Ecology Letters, 22(2), pp. 365–376. doi:10.1111/ele.13199.
Schmitz, O.J., Krivan, V. and Ovadia, O. (2004a) ‘Trophic cascades: The primacy of trait- mediated indirect interactions’, Ecology Letters, 7(2), pp. 153–163. doi:10.1111/j.1461- 0248.2003.00560.x.
Schwegmann, S. et al. (2023) ‘Forage, forest structure or landscape: What drives roe deer habitat use in a fragmented multiple-use forest ecosystem?’, Forest Ecology and Management, 532, p. 120830. doi:10.1016/j.foreco.2023.120830.
Scott, D.M. et al. (2014) ‘Changes in the distribution of Red Foxes (Vulpes vulpes) in urban areas in Great Britain: Findings and limitations of a media-driven nationwide survey’, PLoS ONE, 9(6), pp. 1–11. doi:10.1371/journal.pone.0099059.
Servin, J., Rau, J.R. and Delibes, M. (1991) ‘Activity pattern of the red fox vulpes vulpes in Doñana, SW Spain’, Acta Theriologica, 36, pp. 369–373. doi:10.4098/at.arch.91-39.
Setsaas, T.H. et al. (2007) ‘How does human exploitation affect Impala populations in protected and partially protected areas? – A case study from the Serengeti Ecosystem, Tanzania’, Biological Conservation, 136(4), pp. 563–570. doi:10.1016/j.biocon.2007.01.001.
Silva, A.P. et al. (2017) ‘Climate and anthropogenic factors determine site occupancy in Scotland’s northern-range badger population: Implications of context-dependent responses under environmental change’, Diversity and Distributions, 23(6), pp. 627– 639. doi:10.1111/ddi.12564.
Smith, J.A. et al. (2017a) ‘Fear of the human “super predator” reduces feeding time in large carnivores’, Proceedings of the Royal Society B: Biological Sciences, 284(1857), p. 20170433. doi:10.1098/rspb.2017.0433.
Smith, J.A., Wang, Y. and Wilmers, C.C. (2015) ‘Top carnivores increase their kill rates on prey as a response to human-induced fear’, Proceedings of the Royal Society B: Biological Sciences, 282(1802), p. 20142711. doi:10.1098/rspb.2014.2711.
Sollmann, R. et al. (2013) ‘Risky business or simple solution – relative abundance indices from camera-trapping’, Biological Conservation, 159, pp. 405–412. doi:10.1016/j.biocon.2012.12.025.
Stephens, P.A. et al. (2019) ‘The limits to population density in birds and mammals’, Ecology Letters, 22, pp. 654–663. doi:10.1111/ele.13227.
Støen, O.-G. et al. (2015) ‘Physiological evidence for a human-induced landscape of fear in brown bears (Ursus arctos)’, Physiology & Behavior, 152, pp. 244–248. doi:10.1016/j.physbeh.2015.09.030.
Tucker, M.A. et al. (2018) ‘Moving in the anthropocene: Global reductions in terrestrial mammalian movements’, Science, 359(6374), pp. 466–469. doi:10.1126/science.aam9712.
Tuyttens, F.A. et al. (2001) ‘Vigilance in Badgersmeles Meles: The effects of group size and human persecution’, Acta Theriologica, 46(1), pp. 79–86. doi:10.1007/bf03192419.
Werhahn, G. et al. (2019) ‘Himalayan wolf foraging ecology and the importance of wild prey’, Global Ecology and Conservation, 20. doi:10.1016/j.gecco.2019.e00780.
Wilkinson, D. and Smith, G.C. (2001) ‘A preliminary survey for changes in urban fox (Vulpes vulpes) densities in England and Wales, and implications for rabies control’, Mammal Review, 31(1), pp. 107–110. doi:10.1046/j.1365-2907.2001.00076.x.
Zini, V. et al. (2021) ‘Human and environmental associates of local species-specific abundance in a multi-species deer assemblage’, European Journal of Wildlife Research, 67(6), pp. 98–110. doi:10.1007/s10344-021-01539-6.

The authors declare no competing interests.

ThesisData.csv

Download PDF

Version 1

posted

You are reading this latest preprint version

Understanding the changes in spatiotemporal patterns of two carnivores in response to different anthropogenic pressures and ecological factors in Silwood Park, Ascot, London

Status:

Version 1

Abstract

Figures

1. Introduction

2. Method

2.1 Study Site

2.2 Camera Trapping

2.2.1 Camera Trap Settings

2.2.2 Camera Calibration

2.2.3 Camera Trap Deployment

2.3 Data Processing

2.3.1 Annotation of Camera Trap Images

2.3.2 Measuring Anthropogenic Disturbances

2.4 Statistical Analysis

2.4.1 Nocturnality

2.4.2 Abundance

2.4.3 Spatial Overlap of Species in Silwood

2.4.4 Temporal Overlap of Species in Silwood

3. Results

3.1 Spatial

3.1.2 Badger (Meles meles)

3.1.2 Fox (Vulpes vulpes)

3.2 Temporal Patterns

3.3. Impact of Interspecific Relationships on the spatiotemporal patterns of foxes

3.3.1 Spatial Patterns

3.3.2 Temporal Patterns

3.4. Impact of Interspecific Relationships on Badger Spatial Patterns

4. Discussion

4.1. Spatiotemporal Patterns of Foxes

4.2 Spatiotemporal Patterns of Badgers

4.3. Impact of Interspecific Interactions spatiotemporal patterns

4.3.1 Fox

4.3.2. Badgers

5. ​Conclusion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1

5. Conclusion