MULE DEER, WHITE-TAILED DEER, AND WOLVES IN JASPER NATIONAL PARK, ALBERTA: 35 YEARS OF SIGHTINGS, 1981-2016.
On sympatric range there is no inherent conflict between the 2 species, but in the long run White-tailed Deer appear to dominate (Soper 1964). For instance, in southern Manitoba, prior to European settlement, the original deer population was composed almost solely of Mule Deer. The ratio has since reversed, with a preponderance of White-tailed Deer (Soper 1964). White-tailed Deer were only rarely seen in the southwest of Alberta during the 1950s, until they gradually spread northward. Webb (1959) termed White-tailed Deer 'the deer of the future' in Alberta.
Geist (1998) attributed the range extensions of White-tailed Deer in North America to its resistance to parasites and diseases that limit its potential competitors in the northern climate. Wishart (1980) reported on hybridization of White-tailed Deer and Mule Deer in Alberta, although it is unlikely that Mule Deer are disappearing as a result of one-way hybridization with White-tailed Deer, because the courtship behavior of the 2 species is incompatible, and hybrids are very rare where predators are common and there is no hunting (Geist 1998).
White-tailed Deer began to appear in the Rocky Mountains of west-central Alberta in the 1940s, and the earliest record for Jasper National Park (JNP) is 1943 (Holroyd and VanTighem 1983). The JNP population increased during the 1950s and 1960s, particularly in semi-open habitats of the lower Athabasca River valley (Soper 1970). At higher elevations along the upper Snake Indian River, a tributary of the Athabasca, a 20-y mammal survey that began in 1965 recorded its 1st White-tailed Deer in 1980 (Dekker 1985).
Following effective protection from hunting after the park's establishment in 1907, Mule Deer became common in the valleys and seasonally at higher elevations. McTaggart-Cowan (1947) reported the densest concentrations of Mule Deer (35-60/1.6 [km.sup.2]) at Devona in the lower Athabasca River valley. Since then, ungulate surveys conducted in that district from 1981 to 2006 have shown a significant decline in Mule Deer and a simultaneous increase of White-tailed Deer (Dekker 2001, 2008). In this paper, we update previous studies and add the results of the period 2007 to 2016.
On the basis of annual observations collected over the entire 35-y span from 1981 to 2016, we document how the number of deer sighted at a study site in JNP has changed over time. Gray Wolves (Canis lupus) are common in the study area (Dekker 1997, 1998). Here we present data on the size of a territorial wolf pack sighted in the study area over 35 y, and we examine the question of whether differential predation pressure from wolves on the 2 species of deer might have been a factor in their diverging population trends. In addition, we compare photographs taken in 1985 and 2015 to show how the habitat of the study area changed during this period.
Jasper National Park (JNP) is 10,880 [km.sup.2] in size and situated in west-central Alberta, Canada, between 52[degrees]29' and 52[degrees]08' north latitude. Elevations range from 3747 m in the west to 990 m in the east. The 3 main ecoregions are alpine, subalpine, and montane (Gadd 1986). Of these, the montane is smallest in area but contains the richest diversity of flora and fauna and provides critical wintering range for the park's 7 species of ungulates and their predators. For a detailed description of the park's habitats and wildlife inventory, see Soper (1970) and Holroyd and VanTighem (1983).
The study area lies in the Devona district in the lower valley of the Athabasca River, which flows to and across the park's east boundary with the province of Alberta. Varying in height and aspect, the western slope of the valley rises 200 to 300 m above the montane bottomlands, which are 3 to 4 km wide and characterized by a mosaic of coniferous forest and semi-open meadows. The climate of the region is sub-arctic, with short summers and cold winters that are modified by periodic westerly winds bringing relatively warm weather systems from the Pacific coast. Snow cover is intermittent from October to May.
Survey Methods 1981-2016
We accessed the Devona district on foot in a series of visits during which the senior author (DD) counted all deer by visual observation. The timing and frequency of visits was irregular and took place throughout the calendar year, but the data for this survey were collected between October of 1 year and March of the following consecutive year. We assigned each winter to the year in which it began; for example, the winter of 2012 to 2013 was assigned to the year 2012. During this 6-mo period, animal viewing opportunities were enhanced because deciduous trees and shrubs were bare, while the ground tended to be covered with snow, and the river was frozen. Arriving in early afternoon of the first day and leaving again at approximately the same time on the last day, a 3-d visit represented 2 d of observation. A visit with 3 overnight stays was counted as 3 observation days. The number of observation days over 35 y was 722, with a mean of 20 d/y and a range of 2 to 39 d/y. The low of 2 days occurred in 2007 to 2008, when the Devona district was closed to public access due to oil pipeline construction.
For approximately 1 h each morning and again in the evening, respectively, just after first light and before dusk, the senior author (DD) manned an 80-m high hillside vantage point overlooking the Devona flats and the braided Snake Indian River. Deer observed, either with the unaided eye or binoculars, were recorded in a field diary and the Devona cabin logbook. Obvious same-day duplications were not counted. Deer spotted during daytime walks or snowshoe trips along standard routes in the circa 8-[km.sup.2] study area were added to the day's score. Mule Deer and White-tailed Deer could be readily distinguished by their dissimilar rump and tail (Banfield 1974).
Gray Wolves (Canis lupus) are common in JNP (Dekker and others 1995). Sightings of wolves in the study area were recorded annually, and each year the maximum pack size was noted. Sightings of Coyotes (Canis latrans), Red Foxes (Vulpes vulpes), and Cougars (Puma concolor) were also recorded.
To evaluate whether long-term habitat change might have influenced viewing conditions in the study area, we took repeat photographs from the hillside observation point.
To compare trends in deer abundance in the Devona district with matched trends in deer killed on JNP roadways over the same time period, we obtained data from JNP staff on deer killed by collisions with vehicles in the park.
We used 2 types of models to look at temporal trends in numbers of sightings, species composition, and wolf pack size. The 1st was a simple overall trend that was estimated using a Generalized Linear Model (GLM; glm function in R; Crawley 2012) to estimate trends in deer sightings over time. The model had the total number of sightings for each year as the response variable, with a Poisson error distribution suitable for count data, and included the log-transformed number of observation days as an offset to account for variable survey effort over time (Crawley 2012). No offset was used in the model for wolf pack size. The explanatory variable was Year from 1981 to 2016, with the year values reset to a sequence where 1981 had a value of 1. The 2nd model was a Generalized Additive Model (GAM; package mgcv in R; Wood 2006), again with total sightings per year as the response variable, and Poisson distribution and using log-transformed number of observation days as an offset. This 2nd model fits a smoothed curve that allowed us to evaluate the fluctuations that occurred over time. We also used these 2 types of models to examine how species composition of deer and the wolf pack size at Devona changed over time (with no offset). Using the same Year explanatory variables, we modelled the proportion of Mule Deer in the total sightings as a binomial model using both a GLM and GAM approach.
We recorded 429 deer over 35 successive winters; 274 (64%) were White-tailed Deer, and 155 (36%) were Mule Deer (Fig. 1). Sightings of all deer, either Mule or White-tailed Deer, increased from an average of 0.51 sightings/d in the first 5 winters to 0.74 sightings/d in the last 5 winters. However, over this same time span, Mule Deer declined from 0.42 sightings/d to 0.01 sightings/d. The decline estimated from the GLM was statistically significant ([[beta].sub.yr] = -0.03, SE = 0.01, t = -3.41, P < 0.01). In contrast, sightings of White-tailed Deer increased significantly from 0.08 sightings/d in 1981 to 1985 to 0.73 sightings/d in 2012 to 2016 (([[beta].sub.yr] = 0.10, SE = 0.01, t = 15.4, P < 0.001). When considered together, total sightings of deer more than doubled over time ([[beta].sub.yr] = 0.05, SE = 0.005, t = 11.3, P < 0.001), demonstrating that the decline in Mule Deer was offset by the increase in White-tailed Deer.
The deer trends occurred with wide annual fluctuations for both species (Fig. 1). The GAM for deer sightings showed a significant nonlinear trend over time for Mule Deer (equivalent degrees of freedom (edf) = 5.376, [chi square] = 75.38, P < 0.001), and exhibited a decline from initial high values in the early 1980s, followed by an increase in the early 2000s, and a subsequent decrease to near absence in 2005 to 2016. Similarly, sightings of White-tailed Deer showed significant fluctuations over time (edf = 8.446, [chi square] = 260.8, P < 0.001), and were rare from 1981 to the late 1990s, until sightings increased sharply to a peak in 2009, followed by a decline to 2016.
Species composition (Fig. 1c) also changed over time, such that the proportion of Mule Deer in sightings ranged from 0.60 to 1.00 in the 1980s, and declined to 0.00 to 0.15 in the 2010s, a significant linear decline according to the GLM ([[beta].sub.yr] = -0.14, SE = 0.01, t = -10.4, P < 0.001). This decline showed significant fluctuations (edf = 8.763, [chi square] = 112.1, P < 0.001). During a period in the 1990s when both deer species were rare at Devona, only approximately 20% of deer seen were Mule Deer.
The numeric size of the local pack of wolves observed in the study area varied annually from 2 to 12 animals, with a median value of 7 wolves (Fig. 1d). There was no trend in wolf pack size over time, according to the GLM ([[beta].sub.yr] = -0.004, SE = 0.006, t = -0.61, P = 0.54), and no significant fluctuations over time according to the GAM (edf = 2.811, [chi square] = 3.3, P = 0.49). Wolves are known to kill Coyotes, and sightings of Coyotes at Devona declined from 0.2/d in 1981 to 2001 to 0.03/d in 2001 to 2006 (Dekker 2001, 2008). In 2007 to 2016 the rate of Coyote sightings dropped to 0.02/d.
Repeat photographs taken in 1984 and 2015 from the observation hill showed that formerly open meadows and grassy slopes in the study area became increasingly studded with seedlings of White Spruce (Picea glauca) (Fig. 2).
Over the course of 35 winters from 1981 to 2016, White-tailed Deer increased in abundance in the Devona study area, and Mule Deer declined over the same time period. By 2005, Mule Deer were scarce, and nearly all the deer seen at the study site were White-tailed Deer. The ultimate cause of these opposing trends in abundance is not known. One proximate factor is Gray Wolf predation.
Before the proliferation of White-tailed Deer, Mule Deer were the main prey of wolves in JNP (Carbyn 1975), and Mule Deer remains were identified in 28% of Coyote scats collected in the Devona district during the mid-1970s (Bowen 1982). The anti-predator strategies of the 2 species of deer are different. Fleeing from predators, White-tailed Deer tend to stay on level ground and they are disadvantaged on rough terrain, whereas the stotting gait of Mule Deer helps to overcome obstacles in hilly terrain and facilitates ascending steep slopes where predators may lag behind (Geist 1998). In hybrid deer, the ability to escape pursuing dogs was found to be critically reduced (Geist and Francis 1990; Geist 1998).
In view of the difference in anti-predator behavior between Mule Deer and White-tailed Deer (Geist 1998), and the continued presence of wolves in JNP, the decline in Mule Deer suggests that they may be under heavier predation pressure than White-tailed Deer. Mule Deer are gregarious and occur in herds during winter, which may make them more vulnerable to wolves. White-tailed Deer tend to be solitary (Banfield 1974), although deep snow conditions and brushy terrain may force them to band together. White-tailed Deer can coexist with wolves and have been able to extend their range farther and farther north in the face of wolf predation (Nelson and Mech 1981; Heffelfinger 2011). White-tailed Deer are extending their range into Yukon (Naughton 2012), and this continental trend is mirrored at the Devona study site.
In some winters, depending on a number of variables, the territorial wolf pack observed at Devona spent more time in the study area than in other winters, which can be expected to have had an effect on local deer numbers. A temporary absence of wolves may have led to an increase in deer; conversely, persistent wolf presence may cause deer to abandon the area. In 2000, the movements of the local herd of Elk (Cervus elaphus), which included a radio-collared calf, were correlated with wolf presence at Devona. Chased by wolves, the Elk fled >3 km, crossing a railway, a river, and a highway. They returned to Devona when the wolf pack was not present (Dekker and Slatter 2009).
Long-term records of deer killed by wolves reported to JNP staff or by contract researchers do not reflect the relative vulnerability of the 2 species of deer, because the researchers did not differentiate between Mule Deer or White-tailed Deer if the prey remains consisted of little more than hair or splintered bone (Mark Bradley, JNP biologist, pers. comm.)
Anecdotal eye-witness accounts of Wolves killing deer are rare in the published literature. During our 35-y study, we never saw the final act of killing, but we observed wolves in pursuit of ungulates on 15 occasions, and we came across the fresh remains of 5 deer assumed to have been predated by wolves. Tracks in snow showed that a single wolf had sprinted down a steep slope and intercepted a deer running in deep snow at the base of a 100 m-high hill. In Minnesota, a single wolf was reported to have chased a White-tailed Deer over varied terrain for at least 20 km (Mech and Korb 1978). Evidence of long chases of both species of deer was observed in our study area (Dekker 1997). In contrast, a large-antlered Mule Deer stood off and faced 3 wolves, while the deer kept his rump against a dense clump of evergreens until the wolves gave up the siege, which had lasted >20 min. Another large-antlered Mule Deer buck was wounded in the rump and later killed while crossing frozen Jasper Lake, although he had managed to gore one of the attacking wolves, causing it to limp away.
Cougars also prey on deer. Cougar tracks in snow were found each winter, and Cougars were occasionally sighted on open slopes frequented by Bighorn Sheep (Ovis canadensis) wintering in the Devona district (Dekker 2010). Cougars were reported to have expanded their range in Alberta from 1991 to 2010 (Knopff and others 2014). Cougar predation on Mule Deer was nearly double the rate of predation on White-tailed Deer during winters in British Columbia (Robinson and others 2002). Therefore, predation by Cougars may also account for the different numerical trends in the 2 deer species at Devona.
Hunting is not allowed within the park but is permitted along JNP's east border with the province of Alberta, and deer can be expected to move freely across park boundaries. When hunted by humans, Mule Deer behave differently than White-tailed Deer, because Mule Deer habitually pause to watch humans approach instead of bounding off at once (Geist 1998). For that reason, Mule Deer can be expected to be more vulnerable to hunters than White-tailed Deer. However, provincial hunting quotas for Mule Deer are restricted to a limited license draw system, whereas the season on White-tailed Deer is open (Alberta Guide to Hunting Regulations, Hinton Fish and Wildlife Office, game management units #438-139). Therefore, hunting pressure on the park boundaries is likely heavier on White-tailed Deer than on Mule Deer, and thus unlikely to have resulted in the decline of Mule Deer observed in the Devona district.
The opposing trends in the 2 deer species reported in this study can be compared to a list of 191 deer killed on JNP roads and highways in 1980 to 2014. Although road kill data may be somewhat compromised by the likelihood that mammalian scavengers, such as bears (Ursus spp.) and wolves, tend to drag carrion into cover, thus reducing the chance that it is found by JNP staff, the data roughly parallel our results. The traffic toll of White-tailed Deer rose from zero strikes in the first 5 y to 17 strikes in 2010 to 2014.
The above factors could have acted independently or in concert to result in the observed opposing population trends of Mule Deer and White-tailed Deer in the Devona district. Furthermore, gradual changes in habitat structure could have entered a bias in our method of recording comparative trends in the deer population on the basis of sightings. Deer inhabit semi-open terrain and avoid closed forests (Soper 1964). Tree clearing along an oil pipeline corridor transecting the Devona district may well have played a role in attracting deer, and the White-tailed Deer is believed to be better able than the Mule Deer in adapting to man-made changes in the landscape (Webb 1959; Soper 1970; Dawe and others 2014).
Forest maturation has been progressing naturally in JNP ever since the park's establishment in 1907 when wild fires were controlled (Holroyd and VanTighem 1983). Vegetation succession has been ongoing in the montane zone for over a century (Rhemtulla 1999). White Spruce have spread into the meadows and grassy slopes of the study area and have partially restricted our view from the lookout hill (Fig. 2). Notwithstanding these changing conditions, overall deer sightings per day have more than doubled from 0.48/d to 1.13/d, evidence that sufficient visibility remains to track deer at the site.
The surprising finding reported here is that deer have increased in the study area, whereas Bighorn Sheep and the Elk population have declined significantly (Dekker 2008, 2010; Parks Canada unpubl. data).
The opposing trends of the 2 deer species observed in this study are remarkable and reflect the wide natural variability that can occur when wildlife populations are concurrently affected by multiple factors. Long-term monitoring in protected areas is needed to understand contemporary problems such as the impact of climate change on the living world, which is all the more important today because our last remaining wild places are being destroyed at an unprecedented rate (Jensen 2004; Schmidly 2005). Natural systems are dynamic and time series of observations collected in a standardized way without disturbing wildlife in a protected area provide a baseline to compare changes in the increasingly modified habitats outside and adjacent to national parks. What the future may bring is a matter of grave concern after Chronic Wasting Disease was identified in deer of both species at many locations across North America, including Alberta, and is feared to be spreading (Geist and others 2017).
We thank the Jasper Warden office for permission to use the Devona warden cabin. Brian Genereux and the late Peter DeMulder were frequent co-observers. JNP wildlife specialists Wes Bradford and Greg Slatter provided logistical support throughout the study. Park biologist Mark Bradley supplied the data on road-killed deer in JNP and was consulted in their analysis. This paper was improved by critical reviews of Gilbert Proulx, Valerius Geist, and the editors of this journal. All expenses incurred during this project were paid by the senior author under a temporary volunteer contract with Parks Canada.
BANFIELD AWF. 1974. The mammals of Canada. Toronto, ON: National Museum of Natural Sciences, and University of Toronto Press.
BOWEN WD. 1982. Home ranges and spatial organization of Coyotes in Jasper National Park, Alberta. Journal of Wildlife Management 46:201-216.
CARBYN LN. 1975. Wolf predation and behavioural interactions with Elk and other ungulates in an area of high prey diversity [dissertation]. Toronto, ON: University of Toronto.
CRAWLEY MJ. 2012. The R book. Chichester, UK: John Wiley & Sons.
DAWE KL, BAYNE EM, BOUTIN S. 2014. Influence of climate and human land use on the distribution of White-tailed Deer (Odocoileus virginianus) in the western boreal forest. Canadian Journal of Zoology 92:353-363.
DEKKER D. 1985. Wild hunters. Adventures with wolves, foxes, eagles and falcons based on 25 years of field observation. Edmonton, AB. CWD Publishers.
DEKKER D. 1997. Wolves of the Rocky Mountains. From Jasper to Yellowstone. Surrey, BC and Blaine, WA. Hancock House Publishers.
DEKKER D. 1998. Pack size and colour morphs of one wolf pack in Jasper National Park, Alberta, 1979-1998. Canadian Field-Naturalist 112:709-710.
DEKKER D. 2001. Two decades of wildlife investigations at Devona, Jasper National Park, 1981-2001. In collaboration with W. Bradford and other National Park Wardens. Jasper, AB: Parks Canada library.
DEKKER D. 2008. Wildlife investigations at Devona, Jasper National Park, for the winters of 2001-2006, with comparisons to 1981-2001. In collaboration with park wardens W Bradford and G Slatter. Jasper, AB: Parks Canada library.
DEKKER D. 2010. Declines of Bighorn Sheep Oris canadensis on deteriorating winter range in Jasper National Park, Alberta, 1981-2010. Canadian Field-Naturalist 123:157-164.
DEKKER D, BRADFORD W, GUNSON JR. 1995. Elk and wolves in Jasper National Park, Alberta, from historical times to 1992. In: Carbyn LN, Fritz SH, Seip DH, editors. Ecology and conservation of wolves in a changing world. Occasional publication 35. Edmonton, AB: Canadian Circumpolar Institute.
DEKKER D, SLATTER G. 2009. Wolf Cards lupus avoidance behaviour by American Elk Cervus elaphus in Jasper National Park, Alberta. Canadian Field-Naturalist 123:236-239.
GADD B. 1986. Handbook of the Canadian Rockies. Jasper, AB. Corax Press.
GEIST V. 1998. Deer of the world: Their evolution, behaviour, and ecology. Mechanicsburg, PA: Stack-pole Books.
GEIST V, FRANCIS MH. 1990. Mule Deer country. Minocqua, WI: Northword Press.
GEIST V, CLAUSEN D, CRICHTON V, ROWLEDGE D. 2017. The challenge of CWD: Insidious and dire. Living Legacy White Paper Series. Alliance for Public Wildlife. http://apwildlife.org/publications/
HEFFELFINGER JR. 2011. Taxonomy, evolutionary history, and distribution. In: Hewitt DG, editor. Biology and management of White-tailed Deer. New York, NY: CRC Press.
HOLROYD GL, VANTIGHEM KJ. 1983. Ecological (biophysical) land classification of Banff and Jasper National Parks. Volume III: The wildlife inventory. Edmonton, AB: Canadian Wildlife Service.
JENSEN MN. 2004. Climate warming shakes up species. BioScience 54:722-729.
KNOPFF KH, WEBB NF, BOYCE MS. 2014. Cougar population status and range expansion in Alberta during 1991-2010. Wildlife Society Bulletin 38:116-121.
MCTAGGART-COWAN I. 1947. The Timber Wolf in the Rocky Mountain national parks of Canada. Canadian Journal of Research 25:139-174.
MECH LD, KORB M. 1978. An unusually long pursuit of a deer by a wolf. Journal of Mammalogy 59:860-861.
NAUGHTON D. 2012. The natural history of Canadian mammals. Toronto, ON: Canadian Museum of Nature and University of Toronto Press.
NELSON ME, MECII LD. 1981. Causes of a 3-decade dearth of deer in a wolf dominated ecosystem. American Midland Naturalist 155:373-382.
RHEMTULLA JM. 1999. Eighty years of change: The montane vegetation of Jasper National Park [thesis]. Edmonton, AB: University of Alberta.
ROBINSON HS, WIELCUS RB, GWILLIAM JC. 2002. Cougar predation and population growth of sympatric Mule Deer and White-tailed Deer. Canadian Journal of Zoology 80:556-568.
SCHMIDLY DJ. 2005. What it means to be a naturalist, and the future of natural history at American universities. Journal of Mammalogy 86:449-456.
SOPER JD. 1964. The mammals of Alberta. Edmonton, AB: The Hamly Press.
SOPER JD. 1970. The mammals of Jasper National Park, Alberta. Canadian Wildlife Service Report Series 10.
WEBB R. 1959. Alberta's big game resources. Edmonton, AB: Alberta Fish and Wildlife Division, Queen's Printer.
WISHART WD. 1980. Hybrids of White-tailed and Mule Deer in Alberta. Journal of Mammalogy 61:716-720.
WOOD SN. 2006, Generalized additive models: An introduction with R. Boca Raton, FL: Chapman and Hall/CRC Press.
3819-112 A Street NW, Edmonton, AB T6J 1K4 Canada; email@example.com
MARK C DREVER
Canadian Wildlife Service, Environment and Climate Change Canada, 5421 Robertson Road, Delta, BC V4K 3N2 Canada; firstname.lastname@example.org
Submitted 9 August 2017, accepted 19 January 2018. Corresponding Editor: Chelsea Waddell.
|Printer friendly Cite/link Email Feedback|
|Author:||Dekker, Dick; Drever, Mark C.|
|Publication:||Northwestern Naturalist: A Journal of Vertebrate Biology|
|Date:||Sep 22, 2018|
|Previous Article:||THE NORTHERN EXTENT OF PACIFIC WHITE-SIDED DOLPHIN (LAGENORHYNCHUS OBLIQUIDENS) DISTRIBUTION IN THE EASTERN NORTH PACIFIC.|
|Next Article:||OTOLITH MICROCHEMISTRY OF COASTAL CUTTHROAT TROUT FROM THE MARYS AND WILLAMETTE RIVERS.|