ASSESSING THE EFFECTS OF A HARVESTING BAN ON THE DYNAMICS OF WOLVES IN ALGONQUIN PARK, ONTARIO AN UPDATE Brent Patterson, Ken Mills, Karen Loveless and Dennis Murray Ontario Ministry of Natural Resources (BRP) & Trent University (BRP, KM, KL, DM), Peterborough, ON March 2006 Introduction: Algonquin Park is the largest protected area for wolves in Ontario. However, despite being protected within the Park s boundaries, many animals from the eastern half of Algonquin were shot, snared or hit by cars while following migratory deer out of the Park in winter during the 1980s and 90s. This killing (losses of up to 30% of collared wolves per winter; Theberge and Theberge 2004) was cause for concern, and in November 2001 Ontario s Ministry of Natural Resources (MNR) announced a moratorium on all wolf hunting and trapping in the 39 townships surrounding Algonquin Park. Here we describe the preliminary results from the scientific research underway to measure the effects of this harvest ban on wolf population dynamics. We start by summarizing recent trends in wolf density relative to estimates obtained during the 1980s and 90s. We then look specifically at recent trends in pack sizes, recruitment, survival and dispersal rates as these are the processes that contribute directly to changes in population size within a given area. Space use and excursions outside of the Park We placed radio-collars (or radio ear-tags or implant transmitters for some pups) on 190 different wolves, including 77 pups, between August 2002 and February 2006. During winters 2003-06 we estimated wolf densities separately in the eastern and western portions of the Park based on differences in wolf behavior as well as forest cover and topography. For example, in winter 2003, 5 of 6 packs monitored in eastern Algonquin made repeated forays to deer yards outside the Park. In contrast, only one of 10 packs within our census area in western Algonquin made excursions outside of their territory (and the Park) during the same winter. We monitored 16 packs in and around Algonquin Park during winter 2003, and approximately 25 during winters 2004-06 (e.g. Fig. 1). As the study progressed we noted fewer packs in eastern Algonquin making fewer trips to deer wintering areas outside of the Park. For example, in 2005 only 2 of 11 packs living primarily inside the eastern half of Algonquin made multiple migratory trips outside the park whereas an additional two packs were observed making one brief migratory trip each. Following the ban, we observed the formation of new wolf packs in the deer wintering areas outside of Algonquin areas where relatively high exploitation apparently precluded the formation of stable packs prior to the ban (see Pisapio 1999). Coupled with evidence of increased aggression between these new resident wolves in the deer wintering areas outside of the Park and the winter emigrants from the Park (see discussion of wolf mortality below), we initially concluded that these new packs were making it more difficult for wolves from Algonquin to leave the Park and forage freely in these areas. However, in winter 2006 we once again witnessed a high portion of the wolf packs from eastern Algonquin making migratory trips outside of the Park to feed on deer despite the continued presence of resident wolf packs throughout the deer wintering areas adjacent to the Park (see Fig. 2 for an example). We now have wolves equipped with high-tech GPS collars in migratory packs inside the Park and resident packs in the deer wintering areas outside of the Park, and hope to shed more light on the dynamics of this apparent space-sharing by these wolves. The 2006 moose aerial inventory indicated a rather dramatic decrease in moose densities in eastern Algonquin.
2 10 0 10 20 30 40 Kilometers Fig. 1. Territories of collared wolves in Algonquin Park 2004-05. Individual territories are denoted as gray polygons. The purple lines surrounding the Park indicate the boundaries of the 39 townships where all hunting and trapping of wolves has been banned since December 2001. 8 0 8 16 Kilomete Fig. 2. Migratory movements of the Radiant Lake Pack, eastern Algonquin, early winter 2006. The blue lines indicate the movements of the pack as indicated by a GPS collar fit to wolf W080, the breeding male from this pack. Along the movement path, each location (red dot) is separated by roughly 90 minutes. Round Lake, the namesake of the deer wintering area that houses many of the deer from Eastern Algonquin Park each winter, is conspicuous in the southeast corner.
Preliminary results indicate that although wolves in Algonquin do kill moose, a large portion of adult moose preyed upon exhibit indications of physical handicaps that would likely compromise condition (i.e. malformed jaws, arthritic joints, etc.). Irrespective of decline in moose numbers, the harvested moose population in eastern Algonquin likely has a younger age structure and fewer infirmed/debilitated (i.e. vulnerable) moose for wolves to prey on. Although a decreased availability of vulnerable moose in eastern Algonquin may be partially responsible for the increase in excursions out of the Park by wolves in 2006, the tendency for some wolf packs in Algonquin Park to remain on their territories and exploit moose during winter while other packs leave the Park to forage on migratory deer may also reflect phenotypic (physiological) or genetic differences among wolves within Algonquin Park. Better understanding of predator-prey dynamics and the apparent reliance of Algonquin wolves on migratory deer (i.e. are there any conditions under which Algonquin wolves can exist without deer and without migrating) has profound implications for understanding the future of this wolf-prey system. Population density Our density estimates for both eastern (~2.9 wolves/ 100 km 2 ) and western Algonquin (2.3-2.5 wolves/ 100 km 2 ) have remained relatively stable during our study, and are comparable to those estimated by Forbes and Theberge (1995) during winters 1988-92 but higher than estimated during 1993-99 (Fig. 2; 1.6-2.5/ 100 km 2 (Theberge and Theberge 2004)). Similar to our findings, Forbes and Wolves/ 100 km 2 4 3.5 3 2.5 2 1.5 1 0.5 0 Western Algonquin Eastern Algonquin 1988-89 1989-90 1990-91 1991-92 1992-93 1993-94 1994-95 1995-96 1996-97 1997-98 1998-99 Year Ban initiated late Dec 2001 2001-02 2002-03 2003-04 2004-05 Fig. 3. Changes in wolf density in Algonquin Park, 1989-2005. Data from 1989-99 from Theberge and Theberge (2004). 3 Theberge (1995) reported higher wolf densities in eastern Algonquin, and suggested that this was due to greater abundance of deer and beaver. Yearling & adult survival: Our first collared wolf entered the study on Aug 8 th 2002. Twenty-nine radiocollared yearling or adult wolves died through to the end of February 2006, and the deaths of 20 of those animals were attributable to natural causes (includes falling through ice, strife among packs, malnutrition, mange, wounds inflicted by deer or moose). Of the 9 killed by people, 5 were hit by vehicles (3 in the Park, 2 on highways outside of Algonquin), 2 were shot (1 in Bruton Township during the deer hunt, 1 near Barry s Bay during the moose hunt) and 2 were killed in snares (both outside of Algonquin during winter). We consider survival and dispersal on the basis of the wolf-year, which runs from May 1 st through April 30 th each year. Through the end of April 2003, only 2 collared wolves were confirmed dead. Depending on the fate of an adult female that went missing in March 2003, annual survival for yearling and adult wolves during the first year of our study was 91-94%. During our second and third wolf-years estimated annual survival for yearling and adult wolves was 79-84%, depending on the assumptions around missing wolves (as discussed above for 2003). Despite the increase in natural mortality after the first year of the ban, survival rates
4 of yearling and adult wolves during our study have been relatively high for free-ranging wolves (see Fuller et al. 2003 for context). During 1988-99 survival of yearling and adult wolves in eastern Algonquin (there is insufficient data from western Algonquin) averaged ~67% (Theberge and Theberge 2004) although small sample sizes resulted in high annual variation in this estimate. Another important consideration is that, at least for wolves in eastern Algonquin, there has been a major shift in the primary causes of death for wolves. Approximately 2/3 of known mortalities of wolves during the Theberges study died of human related causes, whereas through our first 3 wolf-years, almost ¾ of documented wolf mortalities were due to natural causes. Although at that the time of writing this in March 2006 it is too early to comment on final estimates of wolf demographic parameters for 2005-06, trends seem similar to recent years. Changes in pack size Pack sizes change from one year to the next through recruitment of new pups, mortality or dispersal of existing members, and immigration of new wolves from other areas. Some researchers have suggested that changes in pack size provide a good indicator of changes in overall wolf density. Given the observed high rates of adult survival, we might have expected a noticeable increase in pack sizes Median Pack size 7 6 5 4 3 2 1 0 1988-89 1990-91 Western Algonquin Eastern Algonquin 1992-93 1994-95 1996-97 Year 1998-99 2001-02 2003-04 Ban initiated late Dec 2001 Fig. 4. Changes in wolf pack size in Algonquin Park, 1989-2005. Data from 1989-99 from Theberge and Theberge (2004). and, by extension, density since the harvest ban was implemented. During early winters 2002-05 median pack size was 4.5, 4.5, 5, 4.75 in western Algonquin, and 4.5, 4.5, and 5 (2003-05) in eastern Algonquin, respectively. Overall, pack sizes observed during 2002-05 appear similar to those observed during 1990-96, but larger than observed in eastern Algonquin during the late 1990s (Fig. 3). Pup survival, recruitment & dispersal of all ages of wolves Whereas some packs seemed to successfully recruit 2-4 pups into their ranks in winters 2003-06, others declined in size from one winter to the next despite relatively high adult survival and confirmed presence of pups with the packs during the previous summer. For example, four of the eight packs monitored in late summer and fall 2002 that we know produced pups lacked pups by winter 2003. Gaining further understanding of the seemingly low pup recruitment (whether due to poor survival, or high and early dispersal) was a high priority for our research program, and was the focus of Ken Mill s MSc thesis. Analyses of 77 wolf pups captured from 2002 2005 indicate that pup survival during the summer and fall is relatively high (~75%) for litters born within the Park. The recruitment rate of pups at the end of November (when they were ~7 months old) was only 57 pups/100 older wolves. Through the end of November a similar proportion of pups dispersed from their packs as had died by that point in the year. The combined effects of mortality and dispersal left an average of 2.2 pups with each pack by the start of winter.
5 Although most wolves will eventually disperse from their natal packs provided they live long enough, it is rare for wolves younger than 8 months old to disperse (most disperse at 22-24 months; Fuller et al. 2003). Yet we documented pups dispersing as young as 3.5 months old, the earliest record for any wolf population. The fates of these dispersing pups and the mechanisms behind this early dispersal are generally unknown and require more research. Of the 113 yearling and older wolves radio-collared since August 2002, 17 were solitary wolves based on no observed affiliations with other wolf packs or individuals, and thus were excluded from our calculations of dispersal rates. During the first year of monitoring a minimum of 3, and maximum of 4, of the 40 pack-living yearling and adult wolves dispersed from the packs they were in when we began monitoring. During 2003-04, 12-14 yearling or older wolves dispersed from the packs they started the year in (fates of 2 were uncertain). Based on these numbers, actual dispersal rates during the first 2 years of our study were 10-12 and 24-28% respectively. During 2004-05, between 12-15 yearling and adult wolves dispersed resulting in an estimated dispersal rate of 22-27%. Forbes and Theberge (1995) did not report annual or age-specific dispersal rates, but did indicate that between 1987-92 only 7 (12%) of the 57 collared wolves they monitored dispersed, all during the winter period. Only 2 of those 7 wolves were adults. Calculations of dispersal rates by known-aged wolves in Theberge and Theberge (2004: page 81-82) are in error and not useful to this discussion (i.e. the rate presented as an annual rate actually covers an approximate 3-yr period and is not comparable to an annual rate). Overall, recent dispersal rates of yearling and adult wolves are relatively high, but similar rates have been observed in other areas (e.g. Gese and Mech 1991). The radio-implants we used to monitor pups began failing after November each year, so our data on pup survival and dispersal during winter is based on small sample sizes. However, it seems safe to assume that pup dispersal continued after our radio-implants failed, thus the overall recruitment rate of pups into the adult population (assuming that recruitment officially occurs on the pups first birthday) is lower than the 57 pups/100 wolves we documented by the end of November each year. Nonetheless, our data suggests that the observed stability of pack sizes occurs because annual pup recruitment is approximately equal to the annual loss of yearling and adult wolves from the population through the combination of mortality and dispersal (~45%). General conclusions and discussion We have documented an increase in annual survival rates of yearling and adult wolves in Algonquin Park following a ban on all hunting and trapping of wolves in the 39 townships surrounding the Park. However, increased survival has not resulted in a detectable increase in either pack size or overall population density. This appears to be due to high rates of dispersal by both juvenile and adult pack members. That relatively high survival is apparently being offset by high dispersal with little overall change in wolf density suggests wolf densities may presently be self-regulated at a level suitable for the present abundance of prey (moose, deer, beaver) available to wolves in the Park. Having said this, we stress again the relatively short-term nature of our data. Further monitoring is necessary to determine whether the above observations remain consistent over time or whether, for some unknown reason, substantial population increase of wolves in Algonquin occurs several years after protection from human exploitation was initiated. The idea that the wolf population in Algonquin (as a whole) was at risk of extinction because of human-caused mortality outside the Park (Theberge and Theberge 2004) assumed that wolves from western Algonquin were leaving the Park during winter, and being killed, at a similar rate as observed in the eastern half of the Park during the 1990s. Preliminary findings suggest that most packs in western Algonquin remain within their territories, and the Park, year round. This suggests that wolves in the west side of the Park may never have been subject to the same level
6 of human caused mortality as east side wolves. Findings of considerable immigration into the Park based on genetic studies (Grewal et al. 2004) and the common emigration of collared wolves from the Park (this study) suggest that it is inappropriate to consider wolves in the Park (moreover a particular section of the Park) as a discrete biological population. Although excess harvest in some years may have severely depressed wolf numbers in eastern Algonquin it seems unlikely that complete extirpation of wolves in this area was ever a possibility. Moreover, immigration from western Algonquin, and surrounding areas, would facilitate re-colonization of vacant territories within a few years, as noted in Theberge and Theberge (2004; see also Potvin et al. 1992; Hayes and Harestad 2000). Overall then, our preliminary conclusion is that although the harvest ban does not seem necessary for wolf persistence in Algonquin, the marked shift in dominant mortality sources for wolves (from human-caused to natural), and apparent natural regulation of wolf numbers presently occurring, indicates that the ban has played a positive role in promoting a naturally functioning wolf-prey system within the Park. PLAN FOR 2006-07 1) Wolf demography Although we have reported relatively stable demographic parameters for wolves in Algonquin, our conclusions regarding natural regulation of wolf densities will not likely be accepted by the broader scientific community with <5 years of consistent and solid data. Although originally intended as a 5-year research project, a labour disruption in year 1 (2001-02) precluded any radiocollaring of wolves until year 2. To achieve 5 years of solid data we will need to continue monitoring basic demographic characteristics of wolves in Algonquin at least through March 31 2007. 2) Predator-prey relations As discussed previously, better understanding of predator-prey dynamics and the apparent reliance of Algonquin wolves on migratory deer has profound implications for understanding the future of this wolf prey system. This is the focus of Karen Loveless s MSc thesis. Specifically, Karen is examining pack-specific patterns of prey selection in relation to availability of deer and moose within each territory, snow conditions, pack size, and wolf body size. Since 2003 we have continued to refine the use of GPS telemetry as a means of providing accurate information on prey use and consumption rates. In reviewing the issues and data collected to date (prey selection, killing rates of prey, etc.) we feel that Karen should continue field work through the end of winter 2007 to ensure this issue is adequately addressed. We would then need to continue her support through data analysis and write-up, which should conclude in December 2007. 3) Genetics Much has been done on the genetics of wolves in and around Algonquin, and the idea that Algonquin wolves belong a distinct species, C. lycaon, that extends from southern Quebec at least through to Pukaskwa National Park is gaining acceptance (Wilson et al. 2000; 2003, Grewal et al. 2004, Kyle et al. 2006). Nonetheless, several important genetic issues remain unresolved, including: Extent and Mechanisms of Hybridization - The eastern wolf, C lycaon, can hybridize with both gray wolves (C. lupus) and coyotes (C. latrans), and genetic work done on wolves from Algonquin indicates the presence of some genetic material from both coyotes (widespread) and
gray wolves (scarce). To further explore the extent, and implications of, this hybridization we propose to: 7 I. Compare major histocompatibility complex (MHC) allelic diversity among adjacent canid populations (eastern wolf, gray wolf, coyote) to help resolve where the eastern wolf fits into the canid genome. II. Determine if there is a decreased genetic presence of coyotes in the Algonquin population since the trapping/hunting ban was established. This will be done by comparing the prevalence of coyote-specific alleles among wolves sampled in Algonquin during 1962-65, 1987-99, and 2002-present. III. Explore the mechanisms of hybridization by: i. Determining the temporal stability and heritability of wolf territories within Algonquin from 1987-2007 (i.e. is pack turnover random or do territories stay in the family over time? Does this differ between migratory and year-round resident packs?). This is important in improving our understanding of how genetic change occurs among wolves in Algonquin. ii. Determining if hybridization occurs at the pack or individual level. Also, if hybridization occurs primarily at the individual level, we can use sex-specific genetic markers (e.g. Y chromosome and Mitochondrial DNA respectively) to determine whether hybridization occurs as a result of males of the predominant species accepting female invaders as mates, or visa versa. This work will parallel research in our Timmins wolf study area aimed at determining if the apparent range expansion of eastern wolves into former gray wolf range is occurring by similar mechanisms as the lycaon-latrans hybridization in southern Ontario. 4) Temporal and spatial patterns of disease exposure With exception of a rabies outbreak in 1990, disease seems to be of only minor significance to wolf population dynamics in Algonquin during most years (Theberge et al. 1994). Nonetheless, serological data on exposure to viral diseases such as canine distemper, canine parvovirus, and canine influenza indicates the continued high exposure of wolves to these diseases in Algonquin. The apparent lack of clinical disease despite high rates of exposure to these diseases could indicate either that most exposures are of low intensity (i.e. not sufficient to cause clinical disease) or that natural resistance is high. The major histocompatibility complex (MHC) is among the most important genetic systems for infectious disease resistance among vertebrates. Specifically, MHC molecules comprise an essential part of the adaptive immune system. In cooperation with the Canadian Cooperative Wildlife Health Centre (CCWHC) in Guelph, we will assess MHC heterozygosity in relation to disease exposure by combining available serology and genetic profiles from wolves sampled from 1987-present in Algonquin. Combined with data on other life history characteristics (e.g. morphology, pack sizes, feeding ecology, reproductive success), this will provide insight into how selection for disease resistant animals may have influenced the present genetic structure of wolves in Algonquin (relative to the other major factors thought to contribute to the present genetic structure of wolves in the Park).
5) Spatial patterns of disease exposure Given the potential role of domestic dogs as vectors of these viral diseases, we will analyze spatial patterns of disease exposure among wolves in relation to intensity of human use of different areas of the Park from 1987-present. 8 Literature cited Forbes, G.J., and J.B. Theberge. 1995. Influences of a migratory deer herd on wolf movements and mortality in and near Algonquin Park, Ontario. Pages 303-313 in Ecology and Conservation of Wolves in a Changing World. Canadian Circumpolar Institute, Edmonton, Alberta. Fuller, T.K., L.D. Mech, and J. F. Cochrane. 2003. Wolf population dynamics. Pp. 161-191 in Mech, L.D., and L. Boitani, eds. Wolves: Behavior, ecology, and conservation. Univ. Chicago Press, Chicago, IL. Gese, E. M., and L. D. Mech. 1991. Dispersal of wolves (Canis lupus) in northeastern Minnesota, 1969-1989. Canadian Journal of Zoology 69:2946-2955. Grewal, S.K., Wilson, P.J., Kung, T.K., Shami, K., Theberge, M.T., Theberge, J.B., B.N. White. 2004. A genetic assessment of the eastern wolf, canis lycaon, in Algonquin Provincial Park. Journal of Mammalogy 84(4): 625-632. Hayes, R. D. and A. S. Harestad. 2000. Demography of a Recovering Wolf Population in the Yukon. Canadian Journal of Zoology 78:36-48. Kyle, C.J., A.R. Johnson, B.R. Patterson, P.J. Wilson, K. Shami, S.K. Grewal, B.N. White. 2006. Genetic nature of eastern wolves: Past, present and future. Conservation Genetics, in press. Pisapio, J. M. 1999. Spatial relations between migratory Algonquin Park wolves and non-migratory resident wolves in a winter deer yard. Msc. Thesis, Univ. of Waterloo, Ontario. Potvin, F., H. Jolicoeur, and L. Breton. 1992. Evaluation of an experimental wolf reduction and its impact on deer in Papineau-Labelle, Quebec. Canadian Journal of Zoology 70:1595-1603. Theberge, J. B., G. J. Forbes, I. K. Barker, and T. Bollinger. 1994. Rabies in wolves of the Great Lakes Region. Journal of Wildlife Diseases 30:563-566. Theberge, J.B., and M. Theberge. 2004. The Wolves of Algonquin Park: a 12 ecological study. Department of geography publication series 56, University of Waterloo 168 pp. Wilson, P. J., S. Grewal, I. D. Lawford, J. N. M. Heal, A. G. Granacki, D. Pennock, J. B. Theberge, M. T. Theberge, D. R. Voigt, W. Waddell, R. E. Chambers, P. C. Paquet, G. Goulet, D. Cluff, and B. N. White. 2000. Dna Profiles of the Eastern Canadian Wolf and the Red Wolf Provide Evidence for a Common Evolutionary History Independent of the Gray Wolf. Canadian Journal of Zoology 78:2156-2166. Wilson, P. J., S. Grewal, T. McFadden, R. C. Chambers, and B. N. White. 2003. Mitochondrial DNA extracted from eastern North American wolves killed in the 1800s is not of gray wolf origin. Canadian Journal of Zoology 81:936-940.