Data Manipulation
In order the establish if white-tailed deer were avoiding predators both temporally and spatially, I examined the number of joint occurrences between deer and their predators during the 61 day sampling period at the camera level and at two different time scales--1 hour and 6 hour intervals (Table 3). That is, at the same station, across all 300 stations, were predator and prey ever captured on camera within one hour or six hours of each other? Lower frequencies of overlap among camera stations would suggest that deer were avoiding overlapping with predators in both time and space. However, because predators often occur at lower densities than their prey, I also included those times where neither predator nor prey were captured at a camera. This accounted for the possibility that deer were not actively avoiding predators, but simply predators were not abundant enough to randomly encounter prey.
In addition, I randomized the deer photo frequencies while keeping camera station, date, hour, and predator photos constant. I did this to test the null hypothesis--if deer were actively avoiding predators, there would be significantly more overlap between randomized deer frequencies and actual predator frequencies, compared to the overlap between real deer frequences and real predator frequencies. Finally, I summarized the total number of real and random overlap between all three study sites (Tables 4-5).
In addition, I randomized the deer photo frequencies while keeping camera station, date, hour, and predator photos constant. I did this to test the null hypothesis--if deer were actively avoiding predators, there would be significantly more overlap between randomized deer frequencies and actual predator frequencies, compared to the overlap between real deer frequences and real predator frequencies. Finally, I summarized the total number of real and random overlap between all three study sites (Tables 4-5).
Camera Stations |
Date |
Hour |
Deer Freq |
Predator Freq |
Real Joint |
Random Deer |
Random Joint |
1 |
Sep. 1, 2016 |
12 |
5 |
0 |
0 |
1 |
0 |
2 |
Sep. 1, 2016 |
12 |
8 |
2 |
2 |
0 |
0 |
3 |
Sep. 1, 2016 |
12 |
0 |
4 |
0 |
8 |
4 |
4 |
Sep. 1, 2016 |
12 |
3 |
2 |
2 |
5 |
2 |
5 |
Sep. 1, 2016 |
12 |
1 |
3 |
1 |
3 |
3 |
Table 3. Sample table of how predator and prey overlap was estimated. The first five columns represent the frequencies of predator and prey photos taken for every hour at each station over the course of the sampling period (61 days). The "Real Joint" column is the overlap between deer and predators at any given station at any given time, and was estimated as the lowest common denominator between the two groups. The "Deer Freq" column was then randomized to test the null model. The "Random Joint" column was the overlap between the random deer frequencies against the real "Predator Freq" column. This process was then repeated using bins of six-hour intervals instead of one-hour intervals.
Results
Between both the randomized and real joint occurrences, I found almost no overlap between white-tailed deer and their predators across all three study sites (Tables 4-5). There was zero overlap between random and real photo frequencies at the Surmont and Nexen sites for both the one and six hour intervals. The Touchwood site also had zero overlap at the one hour interval for both real and random occurrences, and had very little real and randomized overlap (2 and 4 instances, respectively), at the six hour scale. These instances of overlap at the Touchwood site, although greater than zero, were insignificant, as I was randomizing several thousand observations, and only observed a real overlap of 2 occurrences and 4 randomized occurrences.
1 Hour |
Surmont |
Nexen |
Touchwood |
Real Overlap |
0 |
0 |
0 |
Random Overlap |
0 |
0 |
0 |
6 Hours |
Surmont |
Nexen |
Touchwood |
Real Overlap |
0 |
0 |
2 |
Random Overlap |
0 |
0 |
4 |
Tables 4-5. Real and randomized overlap of predator and white-tailed deer across all three study sites. There were zero instances of co-occurrence between predator and prey in both one-hour and six-hour intervals, except for Touchwood, in which there were two real overlap events and four randomized overlap events.
Discussion
The results of this study failed to support my hypothesis that white-tailed deer are avoiding predators at relatively small spatial and temporal scales. While there was zero overlap between the real frequences of predator and prey, which would initially suggest avoidance, the lack of overlap in the randomized null model points to predator and prey being unlikely to meet at the camera level within one or six hours of each other. While prey avoidance of predators is supported in multiple publications (Laundre et al. 2001, Gude et al. 2006, Bleicher 2017, Palmer et al. 2017), my particular camera study does not support this body of work. However, several factors that were beyond the scope of my study may be contributing to the distribution of deer and predators in the boreal forest that are of greater biological importance than the presence or absence of predators. For instance, despite the abundance of white-tailed deer in each study site, the presence of other prey species (snowshoe hares (Lepus americanus), woodland caribou (Rangifer tarandus caribou), etc.) may result in predators not actively hunting or seeking out white-tailed deer, lessening the degree of overlap that may occur between species at a single camera. Additionally, the degree of anthropogenic disturbances varied at each study site ranging from recreational activities to the presence of oil and gas facilities. Factors such as off-road vehicle use, human presence at well pads, and the creation of seismic lines may all have influenced predator-prey distributions at each of these study sites.
As mentioned by Bleicher (2017), it is important to keep in mind the multi-dimensional influences of species' life history traits, individidual variation, competition, and community structures, on species interactions as they relate to the landscape of fear. In addition to the dimensions of time and space, any or all of these factors are most likely influencing the distributions of white-tailed deer, wolves, coyotes, and black bears in the Alberta boreal landscape. Despite the lack of temporal and spatial overlap between predator and prey in this study, these results do not negate the possibility of spatial (and therefore temporal) overlap at a larger scale, potentially intermediately between the camera and study site level. This would be of interest to pursue further: given the life history traits of white-tailed deer and a host of predators, at what scale would there be considerable overlap between these species? If these types of questions can be answered by larger-scale, multivariate studies, then the field of ecological research would be able to advance and explain more of the natural phenomena occurring throughout our terrestrial ecosystems.
As mentioned by Bleicher (2017), it is important to keep in mind the multi-dimensional influences of species' life history traits, individidual variation, competition, and community structures, on species interactions as they relate to the landscape of fear. In addition to the dimensions of time and space, any or all of these factors are most likely influencing the distributions of white-tailed deer, wolves, coyotes, and black bears in the Alberta boreal landscape. Despite the lack of temporal and spatial overlap between predator and prey in this study, these results do not negate the possibility of spatial (and therefore temporal) overlap at a larger scale, potentially intermediately between the camera and study site level. This would be of interest to pursue further: given the life history traits of white-tailed deer and a host of predators, at what scale would there be considerable overlap between these species? If these types of questions can be answered by larger-scale, multivariate studies, then the field of ecological research would be able to advance and explain more of the natural phenomena occurring throughout our terrestrial ecosystems.
Literature Cited
Bleicher, S.S. 2017. The landscape of fear conceptual framework: definition and a review of current applications and misuses. PeerJ 5:37-72.
Gude, J.A., R.A. Garrott, J.J. Borkowski, and F. King. 2006. Prey risk allocation in a grazing ecosystem. Ecological Applications 16:285-298.
Laundré, J., L. Hernández, and K. Altendorf. 2001. Wolves, elk, and bison: reestablishing the "landscape of fear" in Yellowstone National Park, USA. _____Canadian Journal of Zoology 79:1401-1409.
Laundré, J.W., L. Hernández, and W.J. Ripple. 2010. The landscape of fear: ecological implications of being afraid. Open Ecology Journal 3:1-7.
Lime, S.L., and P.A. Bednekoff. 1999. Temporal variation in danger drives antipredator behavior: the predation risk allocation hypothesis. _____American Naturalist 153:649-659.
Schmidt, K. and D.P.J Kuijper. 2015. A "death trap" in the landscape of fear. Mammal Research 60: 275-284.
Sih, A., L.B. Kats, and R.D. Moore. 1992. Effects of predatory sunfish on the density, drift, and refuge use of stream salamander larvae. Ecology 73:1418- 1430.
Gude, J.A., R.A. Garrott, J.J. Borkowski, and F. King. 2006. Prey risk allocation in a grazing ecosystem. Ecological Applications 16:285-298.
Laundré, J., L. Hernández, and K. Altendorf. 2001. Wolves, elk, and bison: reestablishing the "landscape of fear" in Yellowstone National Park, USA. _____Canadian Journal of Zoology 79:1401-1409.
Laundré, J.W., L. Hernández, and W.J. Ripple. 2010. The landscape of fear: ecological implications of being afraid. Open Ecology Journal 3:1-7.
Lime, S.L., and P.A. Bednekoff. 1999. Temporal variation in danger drives antipredator behavior: the predation risk allocation hypothesis. _____American Naturalist 153:649-659.
Schmidt, K. and D.P.J Kuijper. 2015. A "death trap" in the landscape of fear. Mammal Research 60: 275-284.
Sih, A., L.B. Kats, and R.D. Moore. 1992. Effects of predatory sunfish on the density, drift, and refuge use of stream salamander larvae. Ecology 73:1418- 1430.