=Paper=
{{Paper
|id=Vol-2323/SKI-Canada-2019-7-6-4
|storemode=property
|title=Spatial Methods for Understanding Human-Wildlife Interactions
|pdfUrl=https://ceur-ws.org/Vol-2323/SKI-Canada-2019-7-6-4.pdf
|volume=Vol-2323
|authors=Jed Long
}}
==Spatial Methods for Understanding Human-Wildlife Interactions==
https://ceur-ws.org/Vol-2323/SKI-Canada-2019-7-6-4.pdf
Spatial Knowledge and Information Canada, 2019, 7(6), 4
Spatial Methods for Understanding
Human-Wildlife Interactions
JED L ONG
Department of Geography
Western University
jed.long@uwo.ca
Braunisch, 2017). While the public health
ABSTRACT benefits of increasing participation in
outdoor recreation activities are clear
Interactions between humans and wildlife (Godbey, 2009), the long-term and spatial
are a growing concern associated with effects on local wildlife are much more
increased human presence in wildlife difficult to quantify.
habitats. Collecting reliable geographical Collecting robust data on human wildlife
data on human-wildlife interactions poses a interactions is challenging for a variety of
significant challenge owing to the cryptic reasons. First, these interactions are often
nature of wildlife and the fleeting timing of fleeting, and may not always be realized by
such interactions. In this presentation I will the human actor. Second, they may be
demonstrate a citizen science approach for associated with a distance decay effect, i.e.,
studying human-wildlife interactions, and interactions are stronger the closer the two
how it links with more traditional spatial individuals involved are. Finally, human-
ecology methods. GPS tracking is used to wildlife interactions are generally rare
collect fine-scale spatial-temporal data on events, occurring often in more remote
the locations of people along a hiking trail. areas. Thus, innovative methods are
At the same time, hikers were asked to required to collect reliable data on such
complete a wildlife viewing survey that was interactions. Citizen science offers an
linked to the GPS data based on the time opportunity for studying the impacts of
attribute. Specifically, I will demonstrate human outdoor activity on local wildlife
new tools for mapping human-wildlife (Forrester et al. 2017).
interactions and studying the environmental Here I demonstrate a study aimed at
context within which these interactions collecting, analyzing, and mapping human-
occur. wildlife interactions. I explore the types of
data that can be generated for studying
human-wildlife interaction in a citizen
1. Introduction science context. The aim of the presentation
Human activity in remote and natural areas is to demonstrate new spatial tools for
is increasing (Balmford et al. 2009). Many studying these interactions and how these
outdoor recreation activities are directly can be used to understand unique spatial
related to the presence of wildlife (e.g., events.
hunting, wildlife photography) or may be a
secondary motivation (e.g., hiking). 2. Methods and Data
However, human presence within wildlife The study took place in Glen Lyon in the
habitat can disturb wildlife, for example, Perthshire region of Scotland (Figure 1). The
causing increased vigilance (Manor & Saltz, site includes a popular 17.5 km hiking trail
2003), altering movement behaviour which includes summits to four prominent
(Marantz et al., 2016), or shifting habitat munros (defined as peaks above 3000 ft;
selection patterns in both space and time Carn Gorm, Meall Garbh, Carn Mairg, Creag
(Coppes, Burghardt, Hagen, Suchant, & Mhor). Elevation in the area ranges from
2 Human-wildlife interactions
210 m at the trailhead to a maximum of hill-walking (Figure 2). The survey was
1042 m (3419 ft; Carn Mairg). The trail is designed to be simple and easy to fill-out.
situated on an estate, which also runs The cards were then transferred to a digital
several outdoor recreation activities spreadsheet by a team member.
including red deer (Cervus elaphus)
stalking, fishing excursions, and has
domestic livestock (i.e., sheep) roaming free
throughout.
Figure 2: Example of wildlife viewing survey returned by
participants, used to map human-wildlife interactions.
Based on the time information provided by
Figure 1: Location of the study area in the Glen Lyon region
participants in the wildlife survey the
of Perthshire in Scotland. location of the walker at that point in time
was cross-referenced based on their GPS
We collected sample data during the tracking data. The locations where walkers
summer and autumn months of 2017 and viewed wildlife then served as the focal
2018 stratifying our sample days across point for estimating the location of the
weekends and weekdays. During sampling wildlife at that point in time. Where the
days, we asked all hikers entering the trail to participant provided an estimate of the
carry a GPS device while out on the hill. For distance and bearing of the wildlife
each group of hikers (groups defined as encounter, we used this information to map
individuals from the same party walking that location using simple geometry. Any
together) that agreed to participate we gave wildlife encounter recorded by a participant
them one small portable GPS device with a distance estimate of > 500 m was not
(GPSPro 747) to be carried by a single used in subsequent analysis. When a
member of each group. The GPS devices participant did not provide this information,
were pre-programmed to record position we simply mapped the encounter based on
continuously (i.e., one position fix every 5 the location of the participant at the time of
seconds) prior to being given to a the encounter.
participant. A drop-box was located at the Throughout both summers we deployed an
return point (near the car park) where GPS array of camera traps situated along
devices could be returned if the team transects at various points along the hiking
member was no longer present. We did not trail and at random locations throughout
collect any further information (e.g., age, the study area. The cameras use an infrared
gender) about hikers during this sensor to trigger photos and capable of
experiment. detecting animals in both day and night and
At the same time, we asked participants to across all weather patterns. We focused our
carry and fill-out a wildlife viewing survey, study on red deer, but the cameras also
which was a piece of card which we provided captured other animals – mostly sheep).
(along with a pencil). The survey required Camera trap photos were manually
participants to record the time, species of processed by a team member to codify
wildlife, and approximate distance and whether deer were present (and the
bearing at which wildlife were viewed while presence of other animals, i.e., sheep). We
Human-wildlife interactions 3
also identified various deer behaviors (e.g.,
running, head-up, head-down) from photos To test the quality of the participant
to study how different behavior of red deer wildlife encounter data we compare mapped
were associated with distance to the trail, encounters with the observations from the
and the presence of hikers. camera traps and visual observations in the
To explore contextual factors associated field. Here we restrict the analysis to only
with the hill walker data, we calculated red deer as they are the focal species for
sightlines for all encounters recorded in the which the camera traps were deployed (e.g.,
wildlife viewing survey and for every camera the height and calibration of the cameras),
trap location using GIS-based viewshed and thus other species are not commonly
analysis. observed.
Our preliminary results suggest issues with
3. Results relying on participant generated content to
We collected sample data on 35 (25 in identify human-wildlife interactions. Based
2017 and 10 in 2018) different days during on our in-situ visual observations we found
the summers of 2017 and 2018. Of the 197 that hikers routinely failed to see wildlife,
people that we approached, 185 agreed to even when they were within sight lines.
participate, a success rate of 93.9%. From We cross-referenced all hiker GPS data
the 185 participants we collected 153 wildlife with the camera trap data to identify the
surveys with useable data. In total, we were times when participants were near (within
able to successfully digitize 323 (2017: 259 100m) of the cameras. Over the course of
and 2018: 64) wildlife encounters. two summers we collected over 100 000
In order to map wildlife encounters, we camera trap images, of which approximately
needed useable measures of time, distance, 30% contained our focal species (red deer).
and bearing. Where distance or bearing was From this we estimated the error rates
missing we assumed the encounter occurred associated with the participant survey
in proximity with the hiker. Of the 323 wildlife encounter data and found that there
wildlife encounters collected, 143 (2017: was a high-error rate of encounters with red
102, 2018: 41) contained both distance and deer that were missed by hikers.
bearing estimates. An example of a single
participant’s GPS data and their mapped
wildlife encounters is provided in Figure 3. 4. Conclusion
Our study explores the feasibility of using a
citizen science approach for collecting
geographical data on human-wildlife
interactions. Specifically, we found an
extremely high level of engagement in our
preliminary study (greater than 90%
participation rate). This rate of participation
at initial glance is high, but other studies
have shown comparatively high willingness
by outdoor recreationalists to participate in
GPS tracking studies (e.g., Meijles, de
Bakker, Groote, & Barske, 2014). Given that
in many of these more remote areas there
are only small numbers of people out on the
landscape, such a participation rate is very
Figure 3: Example of a single participant with their GPS encouraging. Given the challenges
track and wildlife encounters mapped. Distance and bearing
between the individual participant and the location of the
associated with collecting human-wildlife
wildlife (i.e., sight lines) were estimated by the participant interaction data, maintaining such high
and shown as a purple line on the map.
4 Human-wildlife interactions
participation rates will be advantageous to timing of human-wildlife interactions. We
future work in this area. employed a citizen science approach to
While the data we have collected appears collect data on wildlife encounters along a
to be of relatively good quality upon initial popular hill-walking route in the Glen Lyon
inspection. However, in 2018 we situated a region of the Scottish Uplands. Specifically,
team member at a viewpoint within the site we used voluntary GPS tracking of hikers
and found that hikers routinely did not and a paper-based wildlife viewing survey to
identify deer that were within viewing map the locations of wildlife encounters
range. Other problematic aspects of citizen along a hiking trail.
science studies however need further study,
for example, what might be more important Acknowledgements
is the variability between participants in Funding and contributions to this work have
their capability to observe (or report) been provided by The James Hutton
wildlife sightings (Moyer-Horner, Smith, & Institute, The Carnegie Trust, The British
Belt, 2012), rather than overall measures of Deer Society, The Association of Deer
error. Management Groups, and the North
The approach we have taken here is highly Chesthill Estate.
labor intensive (i.e., it requires a study
member be present to pass out GPS devices
and the survey). Future work will explore References
how to upscale data collection of human
Balmford A, Beresford J, Green J, Naidoo R,
wildlife interactions using mobile-phone
Walpole M, Manica A. (2009). A Global
based apps. Other studies have
Perspective on Trends in Nature-Based
demonstrated how mobile-phone apps can
Tourism. PLoS Biol. 7: e1000144.
be used effectively in collecting similar types
Coppes, J., Burghardt, F., Hagen, R.,
data in ecological field studies (Teacher,
Suchant, R., & Braunisch, V. (2017).
Griffiths, Hodgson, & Inger, 2013) and we
Human recreation affects spatio-
will look to draw on these studies in our
temporal habitat use patterns in red
developments.
deer (Cervus elaphus). PLoS ONE,
It is generally unknown at what distance
12(5), 1–19.
human presence influences different wildlife
doi:10.1371/journal.pone.0175134
species. Previous studies have explored this
Forrester, T., Baker, M., Costello, R., Kays,
in different contexts, for example previous
R., Parsons, A., McShea, W. (2017).
research has found that rocky mountain elk
Creating advocates for mamal
(Cervus elaphus L.) respond at large
conservation through citizen science.
distances (i.e., up to 2000 m) to all-terrain
Biological Conservation, 2018: 98-105.
vehicles (Preisler, Ager, & Wisdom, 2006).
Godbey, G. (2009). Outdoor recreation,
Along the trail in our site wildlife (especially
health, and wellness: Understanding
red deer) may be encountered (i.e., sighted)
and enhancing the relationship.
at similarly long distances (e.g., using
Recreation, (May), 1–42.
binoculars, when visibility is high). For
doi:10.2139/ssrn.1408694
example, we had some wildlife encounters
Manor, R., & Saltz, D. (2003). Impact of
with distances measurements of greater
Human Nuisance Disturbance on
than 1000 m. At what distance such an
Vigilance and Group Size of a Social
encounter represents a true interaction with
Ungulate. Ecological Applications,
a hiker is another question that needs to be
13(6), 1830–1834.
explored further in future research.
Marantz, S. A., Long, J. A., Webb, S. L., Gee,
In summary, collecting reliable and robust
K. L., Little, A. R., & Demarais, S.
geographical data on human-wildlife
(2016). Impacts of human hunting on
interactions is a challenge, owing to the
spatial behavior of white-tailed deer (
cryptic nature of wildlife and the fleeting
Odocoileus virginianus). Canadian
Human-wildlife interactions 5
Journal of Zoology, 94(12), 853–861. doi:10.1002/jwmg.373
doi:10.1139/cjz-2016-0125 Preisler, H. K., Ager, A. A., & Wisdom, M. J.
Meijles, E. W., de Bakker, M., Groote, P. D., (2006). Statistical methods for
& Barske, R. (2014). Analysing hiker analysing responses of wildlife to
movement patterns using GPS data: human disturbance. Journal of Applied
Implications for park management. Ecology, 43(1), 164–172.
Computers, Environment and Urban doi:10.1111/j.1365-2664.2005.01123.x
Systems, 47, 44–57. Teacher, A. G. F., Griffiths, D. J., Hodgson,
doi:10.1016/j.compenvurbsys.2013.07. D. J., & Inger, R. (2013). Smartphones
005 in ecology and evolution: A guide for
Moyer-Horner, L., Smith, M. M., & Belt, J. the app-rehensive. Ecology and
(2012). Citizen science and observer Evolution, 3(16), 5268–5278.
variability during American pika doi:10.1002/ece3.888
surveys. Journal of Wildlife
Management, 76(7), 1472–1479.