Urban noise and its predictability moderate perceived risk associated with roads in grey squirrels
Data files
Sep 27, 2025 version files 12.45 KB
-
GUDdiffOutliersremoved.csv
4.68 KB
-
MeanGUD.csv
4.08 KB
-
README.md
3.69 KB
Abstract
Human activity and disturbance are thought to be perceived as a source of risk analogous to predation risk in wildlife. As such, can alter behaviour and habitat use of foraging animals. Through increased exposure to human disturbance, urban wildlife may face an increase in perceived risk during food acquisition. Urban habitats also include novel resources that could result in urban wildlife having to face distinct trade-offs associated with foraging and patch use. To examine how a successful invasive mammal, the eastern grey squirrel, balances risk and safety under human disturbance, we measured giving-up densities (GUD) at artificial food patches placed at sites subject to varying levels of urbanisation to investigate how features associated with human disturbance might influence feeding decisions. We found differences in GUDs between ‘safe’ and risky’ patches were reduced closer to roads under noisy conditions, suggesting that the risk of predation is perceived by squirrels as reduced when disturbance from human activities is highest. There was also a significant effect of the variability in noise levels on patterns of patch exploitation associated with roads, with the larger GUD differences between safe and risky patches found further from roads exacerbated under more variable noise levels, suggesting the consistency of human disturbances also moderated squirrel risk perception while foraging close to roads. Our results suggest that human activities can have doubled-edged impacts on the urban landscape of fear through offering reduced risk from predators whilst increasing foraging costs via noise disturbance.
Dataset DOI: 10.5061/dryad.d7wm37qdm
Description of the data and file structure
In this paper, we measured grey squirrel giving up densities (GUD) at sites subject to varying levels of urbanisation and human disturbance, to investigate how these features might influence feeding decisions.
MeanGUD.csv - File description
| Variable name | Description |
|---|---|
| PairID | ID of food patch – consists of two GUD foraging trays, one safe and one risky. |
| Site | Site name |
| PC1 | Site level urbanisation index based on 1st principal component score. |
| GUDMean | Average (mean) GUD, for the patch after 5 days foraging |
| RiskLevel | Risk category based on proximity to tree cover. Safe (cover/close to trees); Risky (open habitat) |
| SquirrelDensity | Site level squirrel density estimate based on individuals per acre. |
GUDdiffOutliersremoved.csv – file description
| Variable Name | Description |
|---|---|
| Site | Site name |
| PairID | ID number for each pair of patches |
| SafeGUD | Mean GUD for patch placed in cover/safe location |
| RiskyGUD | Mean GUD for patch placed at open/risky location |
| GUDDiff | Delta GUD: the difference between safe-risky patches |
| Road | Distance from Road |
| Build | Distance from Buildings |
| Footpath | Distance from Footpaths |
| population | Squirrel population density |
| VarDB | Variance in noise (dB) |
| MeanDB | Average (mean) noise (dB) |
| VarVat | Variance category high/low |
| DBCat | Noise level category high/low |
| PC1 | Urbanisation index based on PC1 |
Additional notes
Relative population density (N) was calculated from the total area of the site covered (A), total number of squirrels counted (Z), the number of 15 min observation sessions (n) and the average radial distance to squirrels counted (r)
N = A * Z / (n * π * r²)
(see Flyger, 1959; Parker and Nilon, 2008)
Flyger, V.F. (1959) ‘A Comparison of Methods for Estimating Squirrel Populations’, The Journal of Wildlife Management, 23(2), pp. 220–223
Parker, T.S. and Nilon, C.H. (2008) 'Gray squirrel density, habitat suitability, and behavior in urban parks’, Urban Ecosystems, 11(3), pp. 243–255.
Access information
Other publicly accessible locations of the data:
