Elsevier Editorial System(tm) for Biological Conservation Manuscript Draft Manuscript Number: BIOC-D-09-00365R2 Title: Protection from harvesting restores the natural social structure of eastern wolf packs Article Type: Full Length Article Keywords: human-caused mortality; kinship; social structure; protection; social behaviour; eastern wolves; social restoration Corresponding Author: Ms. Linda Y. Rutledge, Corresponding Author's Institution: Trent University First Author: Linda Y Rutledge Order of Authors: Linda Y Rutledge; Brent R Patterson; Kenneth J Mills; Karen M Loveless; Dennis L Murray; Bradley N White Abstract: Legal and illegal killing of animals near park borders can significantly increase the threat of extirpation for populations living within ecological reserves, especially for wide-ranging large carnivores that regularly travel into unprotected areas. While the consequences of human-caused mortality near protected areas generally focus on numerical responses, little attention has been given to impacts on social dynamics. For wolves, pack structure typically constitutes an unrelated breeding pair, their offspring, and close relatives, but intense harvest may increase adoption of unrelated individuals into packs. Concerns that high human-caused mortality outside Algonquin Park, Canada threatened the persistence of eastern wolves, led to implementation of a harvest ban in surrounding townships. We combined ecological and genetic data to show that reducing anthropogenic causes of mortality can restore the natural social structure of kin-based groups despite the absence of a marked change in density. Since implementation of the harvest ban, human-caused mortality has decreased (P = 0.000006) but been largely offset by natural mortality, such that wolf density has remained relatively constant at approximately 3 wolves/100 km2. However, the number of wolf packs with unrelated adopted animals has decreased from 80% to 6% (P = 0.00003). Despite the high kinship within packs, incestuous matings were rare. Our results indicate that even in a relatively large protected area, human harvesting outside park boundaries can affect evolutionarily important social patterns within protected areas. This research demonstrates the need for conservation policy to consider effects of harvesting beyond influences on population size.

Cover Letter

Dr. Richard B. Primack Editor-in-Chief Biological Conservation Biology Department Boston University 5 Cummington Street Boston, MA 02215, USA Phone: 1-617-353-2454 E-mail: [email protected]

Linda Rutledge Corresponding Author Natural Resources DNA Profiling & Forensic Centre Trent University DNA Building 2140 East Bank Drive Peterborough, ON, Canada K9J 7B8 Phone: (705) 755-2258 Fax: (705) 748-1132 E-mail: [email protected]

Wednesday, October 14, 2009 Dear Dr. Primack, Thank you for the recent comments on the resubmission of our manuscript “Protection from harvesting restores the natural social structure of eastern wolf packs” (MS Ref No BIOC-D-09-00365R1). We have addressed the concern of reviewer #3 by changing the wording of lines 299-300 from: “we consider the social restoration of pack structure to be a positive response to the harvest ban because it indicates a naturally-functioning ecosystem; a primary goal of Ontario Parks, the agency responsible for management of provincial parks” to read “we consider the social restoration of pack structure to be a positive response to the harvest ban because it represents an important element of a naturally-functioning ecosystem, the maintenance of which is a primary goal for Ontario Parks, the agency responsible for management of provincial parks; this social component may stimulate natural regulation at other trophic levels”. We hope you find this revision suitable. We look forward to hearing from you. Sincerely, Linda Rutledge Corresponding Author

*Revision Notes

Revision notes for manuscript “Protection from harvesting restores the natural social structure of eastern wolf packs” (MS Ref No BIOC-D-09-00365R1). Reviewer #3: We have addressed the concern of reviewer #3 by changing lines 299-300 from: “we consider the social restoration of pack structure to be a positive response to the harvest ban because it indicates a naturally-functioning ecosystem; a primary goal of Ontario Parks, the agency responsible for management of provincial parks” to: “we consider the social restoration of pack structure to be a positive response to the harvest ban because it represents an important element of a naturallyfunctioning ecosystem, the maintenance of which is a primary goal for Ontario Parks, the agency responsible for management of provincial parks; this social component may stimulate natural regulation at other trophic levels”.

1

*Manuscript

1

Protection from harvesting restores the natural social structure of eastern

2

wolf packs

3

Linda Y. Rutledge*a ([email protected]), Brent R. Pattersonb ([email protected]),

4

Kenneth J. Millsa,c ([email protected]), Karen M. Lovelessa

5

([email protected]), Dennis L. Murrayd ([email protected]) and Bradley N.

6

Whitee ([email protected])

7

* Corresponding Author: Trent University, DNA Building, 2140 East Bank Drive,

8

Peterborough, Ontario, Canada K9J 7B8, E-mail: [email protected]; Phone: 001-705-755-

9

2258; Fax: 705-748-1132

10

a

11

Bank Drive, Peterborough, Ontario, Canada K9J 7B8

12

b

13

University, DNA Building, 2140 East Bank Drive, Peterborough, Ontario, Canada K9J 7B8

14

c

15

USA 82941

16

d

17

Peterborough, Ontario, Canada K9J 7B8

18

e

19

2140 East Bank Drive , Peterborough, Ontario, Canada, K9J 7B8

20

Running Title: Protection restores kinship in wolf packs

21

Keywords: human-caused mortality, kinship, social structure, protection, social behaviour,

22

eastern wolves, social restoration

Environmental and Life Sciences Program, Trent University, DNA Building, 2140 East

Ontario Ministry of Natural Resources, Wildlife Research and Development Section, Trent

Present Address: Wyoming Game and Fish Department, PO Box 850, Pinedale, Wyoming,

Biology Department, Trent University, DNA Building, 2140 East Bank Drive,

Natural Resources DNA Profiling and Forensic Centre, Trent University, DNA Building,

23

1

24

Abstract

25

Legal and illegal killing of animals near park borders can significantly increase the threat of

26

extirpation for populations living within ecological reserves, especially for wide-ranging

27

large carnivores that regularly travel into unprotected areas. While the consequences of

28

human-caused mortality near protected areas generally focus on numerical responses, little

29

attention has been given to impacts on social dynamics. For wolves, pack structure typically

30

constitutes an unrelated breeding pair, their offspring, and close relatives, but intense harvest

31

may increase adoption of unrelated individuals into packs. Concerns that high human-caused

32

mortality outside Algonquin Park, Canada threatened the persistence of eastern wolves, led to

33

implementation of a harvest ban in surrounding townships. We combined ecological and

34

genetic data to show that reducing anthropogenic causes of mortality can restore the natural

35

social structure of kin-based groups despite the absence of a marked change in density. Since

36

implementation of the harvest ban, human-caused mortality has decreased (P = 0.000006) but

37

been largely offset by natural mortality, such that wolf density has remained relatively

38

constant at approximately 3 wolves/100 km2. However, the number of wolf packs with

39

unrelated adopted animals has decreased from 80% to 6% (P = 0.00003). Despite the high

40

kinship within packs, incestuous matings were rare. Our results indicate that even in a

41

relatively large protected area, human harvesting outside park boundaries can affect

42

evolutionarily important social patterns within protected areas. This research demonstrates

43

the need for conservation policy to consider effects of harvesting beyond influences on

44

population size.

45

2

46 47

1. Introduction Conservation and management strategies, including decisions to remove species from

48

endangered lists, are largely based on estimates of population size and sustainable harvest

49

(Pyare and Berger 2003; Whitman et al. 2004; Isaac and Cowlishaw 2004; Patterson and

50

Murray 2008). There is, however, growing evidence that maintenance of family groups

51

within species that exhibit kin-based social structure can have fitness benefits associated with

52

the adaptive evolution of sociality (Pope 2000; Silk 2007; Gobush et al. 2008). Despite the

53

potential importance of kinship, the role of social groups in long-term population persistence

54

is routinely overlooked (Haber 1996). In protected areas, exploitation near park borders

55

further complicate conservation efforts because these edge effects can significantly increase

56

risk of extirpation, especially for carnivores that have large home ranges (Woodroffe and

57

Ginsberg 1998).

58

In the absence of strong harvest pressure, wolf packs (Canis lupus, C. lycaon and

59

their hybrids) are typically kin-based (Mech and Boitani 2003). Although some variability in

60

this model has been reported (Meier et al. 1995; Forbes and Boyd 1997), exceptions are rare

61

in naturally-regulated populations. High mortality from hunting and trapping may, however,

62

disrupt this natural social structure by prompting the adoption of unrelated animals into wolf

63

packs (Grewal et al. 2004; Jedrzejewski et al. 2005). Thus, anthropogenic influences may

64

play an important role in the social structure of kin-based species. In fact, due to the high

65

propensity for compensatory demographic responses in wolf populations subject to

66

exploitation (e.g. Fuller et al. 2003; Adams et al. 2008), marked changes in wolf population

67

social structure, including those related to kinship within packs and/or inbreeding, may occur

68

even in the absence of numerical changes. This compensatory paradigm provides an

69

important challenge for the restoration and maintenance of not only viable, but also naturally-

70

functioning, populations where fitness is likely to be optimized when evolutionary adaptation

3

71

is driven by natural rather than artificial (i.e. human-mediated) selection pressures (Darimont

72

et al. 2009).

73

The eastern wolf (C. lycaon) is designated as a species of special concern by the

74

Committee on the Status of Endangered Wildlife in Canada (COSEWIC) under Canada’s

75

Species at Risk Act (SARA). One of the largest protected areas (7571 km2) for eastern

76

wolves is Algonquin Provincial Park (APP) in Ontario, where 200 – 300 resident wolves

77

have been influenced by hybridization with gray-eastern wolf hybrids (C. lupus x lycaon) that

78

occur north of the park, and with eastern coyotes (C. latrans var.) south and west of the park

79

(Grewal et al. 2004; Wilson et al. 2009). Between 1987 – 1999, eastern wolves in APP

80

suffered high mortality (56-66%) from hunting and trapping when they left the park to hunt

81

deer outside park boundaries (Forbes and Theberge 1996; Theberge et al. 2006). It was

82

speculated that this intense harvest was responsible for low kinship within packs (Grewal et

83

al. 2004) and that extirpation of wolves in APP was likely if human-caused mortality was not

84

curbed (Vucetich and Paquet 2000; Patterson and Murray 2008). In December 2001, due to

85

prevalent concern for the long-term viability of wolves in APP, the Government of Ontario,

86

amidst much public controversy, banned wolf harvest in townships adjacent to APP, thereby

87

increasing the protected area for park wolves by 6340 km2 (Fig. 1). The purpose of this study

88

was to determine whether wolf pack structure changed in APP following inception of the

89

harvest ban. Specifically, we used previously published data (Grewal et al. 2004) combined

90

with current field data and genetic profiles to test the hypothesis that the ban elicited

91

measurable effects on wolf pack structure. We predicted that extending protection for wolves

92

into areas previously experiencing high human-caused mortality would prompt the renewal

93

of kin-based wolf packs and initiate the restoration of a natural social structure for wolves in

94

APP.

95

2. Materials and Methods

4

96 97

2.1. Study area The 2700 km2 Continuous Study Area (CSA) surveyed consists of rolling hills on the

98

southern margins of the Canadian Shield. The area is forested with pines (Pinus strobus, P.

99

resinosa, P. banksiana), shade-intolerant hardwoods (Acer rubra, Populus tremuloides, P.

100

grandidentata, Betula papyrifera) and lowland conifers (Abies balsamea, Picea glauca, P.

101

mariana). On moister uplands, shade-tolerant hardwoods (Acer saccharum, Betula

102

alleghaniensis), along with Tsuga canadensis predominate. Lakes, rivers and ponds are

103

common. Although we monitored wolves across the entire park during 2002 – 2007, for

104

comparability we consider here population trend and cause of death data only for an area of

105

eastern Algonquin (881 – 2635 km2) that corresponded with the previously described CSA

106

(Theberge and Theberge 2004). It should be noted that packs used for pedigree analysis in

107

this study include wolves monitored outside the CSA and therefore the sample size for

108

animals included in the post-ban pedigree analysis (n=138) is higher than that used for the

109

post-ban density and proportional mortality data (n=112).

110

2.2. Wolf Population Density and Determination of Causes of Death.

111

Wolf population density within the CSA was estimated during eleven consecutive

112

years prior to the harvest ban (1989 – 1999) using territory mapping as described by

113

Theberge and Theberge (2004). Territories were defined based on 95% minimum convex

114

polygon (MCP; Mohr 1947) to exclude locations resulting from off-territory excursions

115

(Bekoff and Mech 1984; Potvin 1988). The effective sampling area varied annually but

116

averaged ~1250 km2. For each year’s population estimate, a census area was defined by a

117

concave polygon enclosing all adjacent territories within the study area. The total number of

118

wolves (including both territorial and non-territorial animals) was summed in the census area

119

(Messier 1985; Ballard et al. 1987; Fuller 1989) with density (Nt), given as wolves/100 km2,

120

estimated as the summed maximum pack sizes plus the estimated number of lone wolves in

5

121

the area, divided by the census area (Mech 1973; Fuller 1989). The number of lone wolves in

122

the area was estimated from the proportion of lone wolves among the radio-collared sample

123

in the study area each year. Confidence intervals are not included with density estimates

124

because they were unavailable for the pre-ban dataset (see Theberge and Theberge 2004).

125

We employed the same methods described above to post-ban data to estimate wolf

126

density in an area of eastern Algonquin (881 – 2635 km2) similar to the CSA during winters

127

2003 – 2007. We radio-tagged 112 wolves within this study area between August 2002 and

128

February 2007 as described by Patterson et al. (2004). Each wolf was fit either with a VHF

129

radiocollar (Holohil Systems Ltd., Woodlawn, Ontario, Canada and Lotek Engineering Inc.,

130

Newmarket, Ontario, Canada) weighing approximately 400 g, or Lotek model 4400S or M

131

GPS collars (weighing approximately 500 and 950 g, respectively, Lotek Engineering, Inc.,

132

Newmarket, Ontario) that were scheduled to obtain fixes at approximately 90 minute

133

intervals during November – April. Additionally, young pups were manually captured from

134

their natal dens and weighed, sexed, and implanted with a VHF radio-transmitter (2 x 8 cm,

135

Advanced Telemetry Systems, Isanti, MN or Telonics, Inc., Mesa, AZ) in the peritoneal

136

cavity (Crawshaw et al. 2007). All radio transmitters contained mortality switches that

137

doubled the signal pulse rate if the transmitter remained motionless for >7 hours. Wolf

138

capture and handling procedures were approved by the Ontario Ministry of Natural

139

Resources’ animal care committee (permit nos. 02-75, 03-75, 04-75, 05-75, 06-75, 07-75).

140

We checked radio-tagged wolves for mortality signals from the ground or during

141

aerial tracking at <1-2 week intervals throughout the year, and when a mortality signal was

142

detected, we promptly visited the site on the ground. Cause of death for each wolf was

143

determined by assessing evidence at the mortality site and detailed necropsies conducted by

144

personnel from the Canadian Cooperative Wildlife Health Centre, University of Guelph.

145

2.3. DNA Extraction and Amplification.

6

146

Blood samples were collected during radio-collaring activities conducted from

147

August 2002 – January 4, 2007. DNA was extracted from 205 samples; 196 from blood on

148

FTA cards or blood clots and 9 from pulled hair, with a DNEasy Blood and Tissue Extraction

149

Kit (Qiagen, Mississauga, Canada). Of these, 138 were affiliated with packs and were

150

included in kinship analyses. Hair was cut into lengths of approximately 2 cm and placed

151

directly into 500 L 1X lysis buffer (4 M urea, 0.2 M NaCl, 0.5% n-lauroyl sarcosine, 10

152

mM CDTA (1,2-cyclohexanediamine), 0.1 M Tris-HCl, pH 8.0). Two 6 mm diameter hole

153

punches from the whole blood on FTA paper were placed in 500 L 1X lysis buffer and then

154

DNA was extracted according to manufacturer’s directions. For the blood clots, 350 – 400

155

mg was removed from the top portion of the clot to increase the chance of obtaining the

156

buffer coat layer where the majority of white blood cells remain after centrifugation. The clot

157

was fragmented with a scalpel blade, placed in 1 mL of 1X lysis buffer in a 15 mL tube, and

158

rotated at 37 °C overnight (12 – 18 hours). A 500 L subsample of lysate was removed and

159

placed in 1.5 mL Eppendorf tubes. Proteinase K (2.4 Units) was added and samples were

160

incubated at 65°C for 1 hour with pulse vortexing after 30 minutes and at the end of 1 hour.

161

Samples were then transferred to a 65 °C water bath inside a 37 °C incubator for one hour to

162

allow slow cooling to 37 °C, at which time a second aliquot of proteinase K (2.4 Units) was

163

added to each sample followed by pulse vortexing and incubation at 37 °C overnight. A 250

164

L subsample was removed and placed in new 1.5 mL Eppendorf tubes. DNA extraction

165

from the blood clots from this point on was according to manufacturers directions. All

166

samples were quantified with PicogreenTM (Molecular Probes) (Ahn et al. 1996) and

167

subsequently diluted to 2.5 ng/ L. For those samples below the threshold of 3 ng/ L, the

168

undiluted extract was used in PCR and quantified by gel fluorescence with ethidium bromide

169

(Ball et al. 2007) to ensure that all samples had between 0.5 – 5 ng of template DNA for each

170

PCR. We amplified a 343 – 347 bp fragment of the mitochondrial DNA control region 7

171

(Wilson et al. 2000) to assign maternal haplotypes, a 658 bp section of the Y-intron (Shami

172

2002) and 4 Y-microsatellites (Sundqvist et al. 2001) to track paternal inheritance, and 16

173

autosomal microsatellite loci (cxx377, cxx172, cxx123, cxx109, cxx225, cxx250, cxx200,

174

cxx204, cxx147, cxx253, cxx383, cxx410, cxx442, c2010, cph11, c2202) (Grewal et al.

175

2004) to determine individual genotypes and bi-parental inheritance. Amplified fragments

176

were size-separated and visualized on a MegaBace 1000 (GE Healthcare, Baie d’Urfé,

177

Quebec), sequences were edited in BioEdit 7.0.9 (Hall 2007) and genotypes were scored in

178

GeneMarker 7.1 (SoftGenetics, State College, PA).

179

2.4. Parentage and Kinship Analysis

180

All pre-ban data relating to kinship within packs was taken from Grewal et al. (2004).

181

A pack was defined as ≥3 individuals living concurrently within a group. Pack affiliations

182

were determined using multiple telemetry locations and ground tracking, as well as visual

183

observations made during telemetry tracking flights. Specifically, pack affiliations were

184

inferred when the animals in question were located together, within a common territory,

185

during >75% locations over a period extending > 30 days. In two cases (W113/C4361 in

186

McKaskill and W195/C4443 in Pretty; Supplementary Figures 1k, 1n), male individuals

187

unrelated to other pack members were not considered adopted because their presence was

188

confirmed only after contact was lost (due to dispersal, death, or collar failure) with the

189

breeder of the same sex; in such cases we could not rule out the possibility that the new

190

individual was replacing the “lost” animal as the breeder.

191

Samples genotyped at fewer than 8 loci were not included in the analysis (n = 4, all

192

from hair) to ensure high probability of identity, and an additional 5 samples were excluded

193

because they represented previously sampled animals. A total of 196 animals were included

194

in the parentage analysis; overall, missing data accounted for 1.4% of the dataset. The

195

autosomal microsatellite dataset was assessed for genotyping errors with MicroChecker

8

196

(VanOosterhout et al. 2004). To test the power of our dataset for individual identification, we

197

calculated the probability of identity (PID) and probability of identity for siblings (PIDsibs)

198

(Taberlet and Luikart 1999) in GenAlEx 6.1 (Peakall and Smouse 2006). In the parentage

199

analysis, females were excluded as the mother if the mitochondrial haplotype was

200

inconsistent with the putative offspring, and males were excluded as the candidate father of

201

male pups if either the Y-intron haplotype or Y-microsatellites were inconsistent with those

202

of the putative offspring. We then utilized two different methods to assign parents: 1) the

203

exclusion method, considered the “paragon” of parentage analysis (Jones and Ardren 2003)

204

but can result in false exclusions (Pompanon et al. 2005), and 2) a maximum likelihood

205

approach (95% confidence) implemented in CERVUS 3.0.3 (Kalinowski et al. 2007).

206

CERVUS is a robust parentage analysis software package that accounts for rare alleles,

207

genotyping errors, and null alleles by using simulations to statistically assign the most likely

208

parent among all non-excluded parents. Paternity simulations generated 100,000 offspring

209

with 100 candidate males (assuming a park population estimate of 200 animals and a 1:1 sex

210

ratio) and assuming 57% of the population was sampled (based on 57 males ≥1 year sampled

211

over a 5-year period and an average 5-year lifespan) and allowed a standard error rate of

212

0.010. We used KINSHIP 1.3.1 (Goodnight and Queller 1999) to test the hypothesis that

213

individuals within packs were more likely to be related at the half-sibling and full-sibling

214

level than unrelated based on a simulation series of 10,000 pairs generated from allele

215

frequency calculations of 124 adults (pups excluded). KINSHIP uses relatedness (r) values,

216

allele frequencies, and comparative genotypes to calculate the likelihood of the relationship

217

hypothesis being tested. Pairs that were assigned as not significant (based on a critical P-

218

value of 0.05) in the test of half-siblings were considered unrelated. Where relatedness was

219

indicated but specific kinship was unclear, we used ML-Relate (Kalinowski et al. 2006), a

220

program that accommodates null alleles and uses simulations and a maximum likelihood

9

221

approach, to test hypotheses between putative and alternative relationships, to assign the most

222

probable relationships based on 10,000 simulations.

223

When comparing mitochondrial DNA and Y-chromosome haplotypes (Fig. 3), for

224

pre-ban data (1995 – 2001) all animals sampled in 2 – 5 of 6 consecutive years were included

225

(pups were not sampled from the den); for post-ban data (2002 – 2007) (Figs. 3 and 4) pups

226

sampled from the den were excluded unless they were confirmed in the pack 1 year later.

227

3. Results

228

3.1 Wolf Population Density and Causes of Death

229

In winter 2003, approximately 14 months after initiation of the harvest ban, we estimated

230

wolf density in eastern APP at approximately 3 wolves/100 km2 (Fig. 2), suggesting an

231

average rate of increase (rt) = 0.20 between 1999 and 2003. However, no further increases in

232

density were observed between 2003 – 2007 despite a marked reduction in mortality from

233

hunting and trapping within the ban area and park during this period (Table 1; P = 0.000006).

234

This was due in part to natural causes largely replacing anthropogenic causes as the leading

235

mortality agents for wolves following inception of the harvest ban (B.R. Patterson et al.

236

unpublished data; Table 1).

237

3.2. Parentage and Kinship

238

The mean level of observed heterozygosity for APP wolves sampled post-ban was high (HO =

239

0.687) and similar to the levels of 0.694 – 0.725 reported by VonHoldt et al. (2008) for non-

240

inbred wolves in Yellowstone National Park. Grewal et al. (2004) also reported high levels of

241

heterozygosity during the pre-ban time period, although no estimates were given. No loci

242

showed a significant deviation from Hardy-Weinberg equilibrium after Bonferroni

243

correction. Probability of identity among siblings was low (PIDsibs = 1.06 x 10-6) indicating

244

that full-siblings in this group were unlikely to have the same genotype. We created

245

pedigrees for 138 individuals living in 17 packs over a 5-year period (Supplementary Figure

10

246

1a-1q). Parent-offspring relationships were identified in all 17 packs, nine of which had both

247

breeders identified. Only one pack (Cauliflower; Supplementary Figure 1e) had an animal

248

that was unrelated to the breeder. This female yearling was the only unrelated adult of the 59

249

non-breeding adults identified within all of the 17 packs over the 5-year period. Further, this

250

female dispersed in March 2003, the first winter of our study, and subsequently became a

251

breeding female in another territory along the southern edge of our study area. The overall

252

proportion of packs that had adopted unrelated animals (here defined as unrelated at the half

253

sibling level) decreased significantly post-ban (Table 1; P = 0.00003) demonstrating that

254

post-ban packs are less likely to adopt unrelated animals.

255

We found that incestuous matings were generally avoided despite high kinship within

256

packs. Only two of the 17 post-ban packs had related breeding pairs: one had a half-sibling

257

breeding pair (Beechnut; Supplementary Figure 1b) and another had a full-sibling breeding

258

pair (Louisa; Supplementary Figure 1j). There were three packs in which daughters became

259

subsequent breeders: two while the mother was still in the pack (Cauliflower and Leaf;

260

Supplementary Figures 1e, 1i) and one after the mother dispersed (LaFleur; Supplementary

261

Figure 1h). In one case the full sibling of the breeding male (both unrelated to the breeding

262

female) replaced his brother as breeder while the brother was still in the pack (Jocko;

263

Supplementary Figure 1g). In two cases (W113/C4361 in McKaskill, W195/C4443 in Pretty;

264

Supplementary Figures 1k, 1n) a male unrelated to all others in the pack was identified within

265

the pack after the death of the breeding male, and in one instance (W131/C4379 in Achray,

266

Supplementary Figure 1a) a male unrelated to the breeders was caught in the pack after

267

contact was lost with the breeding male, and in this case the new male became the breeder.

268

Within 5 packs known to occupy the same territory both prior to, and following, the

269

harvest ban, single mitochondrial DNA haplotypes in females were more common post-ban

270

(Fig. 3). Single Y haplotypes were common during both time periods but where a second Y

11

271

haplotype was documented in post-ban packs (Pretty and McKaskill) it was found in an

272

unrelated animal caught after the death of the breeding male (Fig. 3). This pattern was similar

273

in the other post-ban packs studied (Fig. 4). In Big Crow, the BB Y-haplotype was found in

274

animals caught after the death of the male with the CG Y-haplotype, and in Sunday the Y-

275

haplotypes represent two pairs of full siblings of which brothers with the BB Y-haplotype

276

were documented in the territory after the brothers with the AA Y-haplotype had dispersed or

277

died (Fig. 4). The C9 mtDNA haplotype in Cauliflower (Fig. 4) represents the one non-

278

breeding adult found that was unrelated to the female breeder in the pack.

279 280

4. Discussion

281

Our results suggest that high levels of hunting and trapping of wolves outside the borders of

282

Algonquin Provinical Park prior to the harvest ban were responsible for the low kinship

283

observed within packs. Extending protection for APP wolves has, therefore, helped restore a

284

more naturally structured population consisting of family-based wolf packs, despite stable

285

wolf densities since implementation of the ban. More specifically, adoption of unrelated

286

animals into packs is almost non-existent in the current population.

287

The restoration of a family-based social structure in APP, including the pattern of

288

female recruitment from within the pack but acceptance of unrelated immigrant males,

289

presumably as potential breeders after breeder loss, is congruent with naturally-regulated

290

gray wolf populations in park preserves where wolves have legal protection such as

291

Yellowstone National Park in Wyoming (VonHoldt et al. 2008) and the Białowieża Primeval

292

Forest in Poland (Jedrzejewski et al. 2005). Also in concordance with wolf studies in Denali

293

National Park in Alaska, Superior National Forest in northeastern Minnesota, and

294

Yellowstone National Park (Smith et al. 1997; VonHoldt et al. 2008), incestuous matings in

12

295

APP were generally avoided despite high kinship within packs. Together, these results

296

indicate that the natural social fabric has been restored for wolves in Algonquin.

297

Although the long-term viability of APP wolves has been the subject of some debate

298

(see Theberge et al. 2006; Patterson and Murray 2008), we consider the social restoration of

299

pack structure to be a positive response to the harvest ban because it represents an important

300

element of a naturally-functioning ecosystem, the maintenance of which is a primary goal for

301

Ontario Parks, the agency responsible for management of provincial parks; this social

302

component may stimulate natural regulation at other trophic levels. In general, assessments of

303

population viability typically focus on numerical responses and estimates of sustainable

304

harvest (Pyare and Berger 2003; Whitman et al. 2004; Theberge et al. 2006; Isaac and

305

Cowlishaw 2004; Patterson and Murray 2008), with the impacts of human exploitation on

306

social dynamics being largely ignored, even in highly social large mammals such as lions

307

(Whitman et al. 2004) and wolves (Haber 1996). There is, however, growing evidence

308

suggesting that maintaining kin relationships in socially structured populations is

309

evolutionarily important and can have positive effects on fitness (Silk 2007). For example,

310

female red howler monkeys (Alouatta sericulus) living in kin-based groups have higher

311

reproductive success than those living in unrelated groups (Pope 2000), and female elephants

312

(Loxodonta africana) in well established family groups with old matriarchs have lower levels

313

of stress hormones and higher reproductive output than those in groups that have been

314

socially disrupted by poaching (Gobush et al. 2008). Therefore, focussing solely on

315

abundance when assessing population status may ignore other potentially important factors

316

that can contribute to long-term fitness, and hence persistence, of populations.

317

Wolves are highly intelligent animals that have evolved under a family-based social

318

framework. Although the influence of this structure on fitness is not well understood, recent

319

work suggests that maintaining the social organization of wolf packs is important for

13

320

effective resource use (i.e. knowledge of prey distribution and ability to detect, pursue and

321

subdue prey) (Sand et al. 2006; Stahler et al. 2006), pup survival (Brainerd et al. 2008;

322

Schmidt et al. 2008), and may be effective, at least in part, at precluding hybridization with

323

coyotes (C. latrans) due to the lower turnover of individuals within packs and the tendency

324

during hybridization events for genes to flow from the more common into the rarer species

325

(Grant et al. 2005). Breeder loss is particularly influential and can result in abandonment of

326

territories, dissolution of social groups, and smaller pack size (Brainerd et al. 2008). Mate

327

loss can also result in unusual behavioural responses of the surviving breeder (Smith and

328

Ferguson 2005) or incestuous pairings if mate loss occurs close to breeding season

329

(VonHoldt et al. 2008).

330

Minimizing the anthropogenic impact on social structure in populations that form

331

highly related groups is likely to improve overall fitness by allowing evolutionary processes

332

to occur in response to natural selection, not human-mediated mortality (Darimont et al.

333

2009). In this way, conservation strategies can bolster the adaptive evolutionary potential of

334

populations facing environmental fluctuations, including climate change. When compared to

335

other conservation and management approaches such as translocations and habitat

336

restoration, reducing levels of exploitation by expanding no-harvest zones to include areas

337

outside park boundaries is a relatively simple, long-term solution to promote persistence of

338

top predators that are integral to healthy ecosystems (Terborgh et al. 2001, Soulé et al. 2003,

339

Chapron et al. 2008).

340

We conclude that the harvest ban around Algonquin has restored the natural social

341

structure of wolf packs in the park. Given the fitness benefits of kin-based groups in animals

342

that have evolved complex social patterns, these results are likely relevant to other socially

343

structured animal populations that experience high human-caused mortality near park

344

borders. Our results demonstrate the need for conservation policies that look beyond numbers

14

345

to include the subtler, but potentially important, impacts on social dynamics of wildlife.

346

Future work addressing the fitness elements associated with harvesting and the adaptive

347

evolution of family groups will add significantly to our understanding of how centuries of

348

harvesting have shaped the genetic evolutionary potential of Canis and other family-based

349

species.

350

Acknowledgements

351

Thank you to all the field crew and laboratory technicians who helped collect and process

352

samples. We are grateful to K. Middel for creating the map and to T. Frasier for his

353

comments on an early version of the manuscript. This research was funded by the Ontario

354

Ministry of Natural Resources, a Natural Sciences Engineering and Research Council of

355

Canada (NSERC) operating grant awarded to B. N. W. and an NSERC doctoral scholarship

356

awarded to L.Y.R.

357

References

358

Adams, L.G., Stephenson, R.O., Dale, B.W., Ahgook, R.T., Demma, D.J., 2008.

359

Population dynamics and harvest characteristics of wolves in the Central Brooks

360

Range, Alaska. Wildlife Monogr. 170, 1-25.

361 362 363

Ahn, S.J., Costa, J., Emanuel, J.R., 1996. PicoGreen quantification of DNA: Effective evaluation of samples pre- or post- PCR. Nucleic Acids Res. 24, 2623-2325. Ball, M.C., Pither, R., Manseau, M., Clark, J., Petersen, S.D., Kingston, S., Morrill, N.,

364

Wilson, P., 2007. Characterization of target nuclear DNA from faeces reduces

365

technical issues associated with the assumptions of low-quality and quantity template.

366

Conserv. Genet. 8, 577-586.

367 368 369

Ballard, W.B., Whitman, J.S., Gardner, C.L., 1987. Ecology of an exploited wolf population in south-central Alaska. Wildlife Monogr. 98, 1-2. Bekoff, M., Mech, L.D., 1984. Simulation analyses of space use: home range estimates,

15

370 371

variability, and sample size. Behav. Res. Methods Instrum. Comput. 16, 32-37. Brainerd, S. M., Andrén, H., Bangs, E.E., Bradley, E.H., Fontaine, J.A., Hall, W.,

372

Iliopoulos, Y., Jimenez, M.D., Jozwiak, E.A., Liberg, O., 2008. The effects of

373

breeder loss on wolves. J. Wildl. Manage. 72, 89-98.

374 375 376

Chapron, G., Andrén, H., Liberg, O., 2008. Conserving top predators in ecosystems. Science 320, 47. Crawshaw, G.J., Mills, K.J., Mosley, C., Patterson, B.R., 2007. Field implantation of

377

intraperitoneal radiotransmitters in eastern wolf (Canis lycaon) pups using inhalation

378

anesthesia with sevoflurane. J. Wildl. Dis. 43, 711-718.

379

Darimont, C.T., Carlson, S.M., Kinnison, M.T., Paquet, P.C., Reimchen, T.E., Wilmers,

380

C.C., 2009. Human predators outpace other agents of trait change in the wild. Proc.

381

Natl. Acad.Sci. U.S.A. 106, 952-954.

382 383 384 385 386

Forbes, G.J., Theberge, J.B., 1996. Cross-boundary management of Algonquin Park wolves. Conserv. Biol. 10, 1091-1097. Forbes, S.H., Boyd, D.K., 1997. Genetic structure and migration in native and reintroduced Rocky Mountain wolf populations. Conserv. Biol. 11, 1226-1234. Fuller, T.K., Mech, D.L., Cochrane, J.F., 2003. Wolf population dynamics, in: Mech, D.L.,

387

Boitaini, L. (Eds.), Wolves: Behavior, Ecology, and Conservation. University of

388

Chicago Press, Chicago, pp. 161-191.

389 390 391

Fuller, T.K. 1989. Population dynamics of wolves in north-central Minnesota. Wildlife Monogr. 105, 1-41. Gobush, K.S., Mutayoba, B.M., Wasser, S.K., 2008. Long-term impacts of poaching on

392

relatedness, stress physiology, and reproductive output of adult female African

393

elephants. Conserv. Biol. 22, 1590-1599.

16

394 395 396 397 398

Goodnight, K.F., Queller, D.C., 1999. Computer software for performing likelihood tests of pedigree relationship using genetic markers. Mol. Ecol. 8, 1231-1234. Grant, P.R., Grant, B.R., Petren, K. 2005. Hybridization in the recent past. Amer. Nat. 166, 56-67. Grewal, S.K., Wilson, P.J., Kung, T.K., Shami, K., Theberge, M.T., Theberge, J.B., White,

399

B.N., 2004. A genetic assessment of the eastern wolf (Canis lycaon) in Algonquin

400

Provincial Park. J. Mammal. 85, 625-632.

401 402

Haber, G.C. 1996. Biological, conservation, and ethical implications of exploiting and controlling wolves. Conserv. Biol. 10, 1068-1081.

403

Hall, T. 2007. BioEdit v7. http://www.mbio.ncsu.edu/BioEdit/BioEdit.html

404

Isaac, N.J.B., Cowlishaw, G., 2004. How species respond to multiple extinction threats.

405 406

Proc. R. Soc. Lond., B, Biol. Sci. 271, 1135-1141. Jędrzejewski, W., Branicki, W., Veit, C., Međugorac, I., Pilot, M., Bunevich, A.N.,

407

Jędrzejewska, B., Schmidt, K., Theuerkauf, J., Oyarma, H., Gula, R., Szymura, L.,

408

Förster, M., 2005. Genetic diversity and relatedness within packs in an intensely

409

hunted population of wolves Canis lupus. Acta Theriol. 50, 3-22.

410 411 412

Jones, A.G., Ardren, W.R., 2003. Methods of parentage analysis in natural populations. Mol. Ecol. 12, 2511-2523. Kalinowski, S.T., Wagner, A.P., Taper, M.L., 2006. ML-Relate: a computer program for

413

maximum likelihood estimation of relatedness and relationship. Mol. Ecol. Notes 6,

414

576-579.

415

Kalinowski, S.T., Taper, M.L., Marshall, T.C., 2007. Revising how the computer

416

program CERVUS accommodates genotyping error increases success in paternity

417

assignment. Mol. Ecol. 16, 1099-1106.

418

Mech, L.D., Boitani, L., 2003. Wolf Social Ecology, in: Mech, D.L., Boitaini, L. (Eds.),

17

419

Wolves: Behavior, Ecology, and Conservation. University of Chicago Press, Chicago,

420

pp. 1-34.

421

Mech, L.D. 1973. Wolf Numbers in the Superior National Forest of Minnesota. Research

422

Paper NC-97. St. Paul, MN: US Dept. of Agriculture, Forest Service, North Central

423

Forest Experiment Station.

424

Meier, R.J., Burch, J.W., Mech, D.L., Adams, L.G., 1995. Pack structure and genetic

425

relatedness among wolf packs in a naturally regulated population, in: Carbyn, L.N.,

426

Fritts, S.H., Seip, D.R. (Eds.), Ecology and Conservation of Wolves in a Changing

427

World. Canadian Circumpolar Institute, Edmonton, Alberta, Canada, pp. 293-302.

428 429 430 431

Messier, F. 1985. Social organization, spatial distribution, and population density of wolves in relation to moose density. Can. J. Zool. 65, 1068-1077. Mohr, C.O. 1947. Table of equivalent populations of North American small mammals. Am. Midl. Nat. 37, 223-249.

432

Patterson, B.R., Murray, D.L., 2008. Flawed population viability analysis can result in

433

misleading population assessment: A case study for wolves in Algonquin park,

434

Canada. Biol. Conserv. 141, 669-680.

435

Patterson, B.R., Quinn, N.W.S., Becker, E.F., Meier, D.B., 2004. Estimating wolf

436

densities in forested areas using network sampling of tracks in snow. Wildl. Soc. Bull.

437

32, 938-947.

438

Peakall, R., Smouse, P.E., 2006. GENALEX 6: genetic analysis in Excel. Population

439

genetic software for teaching and research. Mol. Ecol. Notes 6, 288-295.

440

Pompanon, F., Bonin, A., Bellemain, E., Taberlet, P., 2005. Genotyping errors: causes,

441 442

consequences and solutions. Nat. Rev. Genet. 6, 847-859. Pope, T.R. 2000. Reproductive success increases with degree of kinship in cooperative

18

443

coalitions of female red howler monkeys (Alouatta seniculus). Behav. Ecol.

444

Sociobiol. 48, 253-267.

445 446 447 448 449

Potvin, F. 1988. Wolf movements and population dynamics in Papineau-Labelle Reserve, Quebec. Can. J. Zool. 66, 1266-1273. Pyare, S., Berger, J., 2003. Beyond demography and delisting: ecological recovery for Yellowstone’s grizzly bears and wolves. Biol. Conserv. 113, 63-73. Sand, H., Wikenros, C., Wabakken, P., Liberg, O., 2006. Effects of hunting group size,

450

snow depth and age on the success of wolves hunting moose. Anim. Behav. 72, 781-

451

789.

452

Schmidt, K., Jędrzejewski, W., Theuerkauf, J., Kowalczyk, R., Okarma, H., Jędrzejewska,

453

B., 2008. Reproductive behaviour of wild-living wolves in Bialowieza Primeval

454

Forest (Poland). J. Ethol. 26, 69-78.

455

Shami, K. 2002. Evaluation of the change in distribution of the eastern timber wolf (Canis

456

lycaon) using the Y chromosome. MSc Thesis, Trent University, Peterborough,

457

Ontario Canada.

458 459 460 461 462 463

Silk, J.B., 2007. The adaptive value of sociality in mammalian groups. Philos. Trans. R. Soc. Lond., B, Biol. Sci. 362, 539-559. Smith, D.W., Ferguson, G., 2005. Decade of the Wolf: Returning the Wild to Yellowstone. The Lyons Press, Guilford, CT. Smith, D., Meier, T.J., Geffen, E., Mech, L.D., Burch, J.W., Adams, L.G., Wayne, R.K., 1997. Is incest common in gray wolf packs? Behav. Ecol. 8, 384-391.

464

Soulé, M.E., Estes, J.A., Berger, J., Martinez del Rio, C., 2003. Ecological effectiveness:

465

conservation goals for interactive species. Conserv. Biol. 17, 1238-1250.

466

Stahler, D.R., Smith, D.W., Guernsey, D.S., 2006. Foraging and Feeding Ecology of the

19

467

Gray Wolf (Canis lupus): Lessons from Yellowstone National Park, Wyoming, USA.

468

J. Nutr. 36, 1923S.

469

Sundqvist, A.K., Ellegren, H., Olivier, M., Vilà, C., 2001. Y chromosome haplotyping in

470

Scandinavian wolves (Canis lupus) based on microsatellite markers. Mol. Ecol. 10,

471

1959-1966.

472 473

Taberlet, P., Luikart, G., 1999. Non-invasive genetic sampling and individual identification. Biol. J. Linn. Soc. Lond. 68, 41-55.

474

Terborgh, J., Lopez, L., Nuñez V, P., Rao, J., Shahabuddin, G., Orihuela, G., Riveros, M.,

475

Ascanio, R., Adler, G.H., Lambert, T.D., Balbas, L., 2001. Ecological meltdown in

476

predator-free forest fragments. Science 294, 1923-1926.

477

Theberge, J.B., Theberge, M.T., 2004. The Wolves of Algonquin Park: A 12 year

478

Ecological Study. Department of Geography, University of Waterloo, Waterloo,

479

Ontario, Canada.

480

Theberge, J.B., Theberge, M.T., Vucetich, J.A., Paquet, P.C., 2006. Pitfalls of applying

481

adaptive management to a wolf population in Algonquin Provincial Park, Ontario.

482

Environ. Manage. 37, 451-460.

483

VanOosterhout, C., Hutchinson, W.F., Wills, D.P.M., Shipley, P., 2004. MICRO-

484

CHECKER: software for identifying and correcting genotyping errors in

485

microsatellite data. Mol. Ecol. Notes 4, 535-538.

486

VonHoldt, B.M., Stahler, D.R., Smith, D.W., Earl, D.A., Pollinger, J.P., Wayne, R.K.,

487

2008. The genealogy and genetic viability of reintroduced Yellowstone grey wolves.

488

Mol. Ecol. 17, 252-274.

489

Vucetich, J., Paquet, P., 2000. The demographic populations viability of Algonquin wolves.

490

Unpublished report prepared for the Algonquin Wolf Advisory Committee, [online]

491

URL: www.cpaws-ov.org/algonquinwolves/docs/vucetichpaquet.pdf (accessed 12

20

492 493 494 495 496 497

October 2007). Whitman, K., Starfield, A.M., Quadling, H.S., Packer, C., 2004. Sustainable trophy hunting of African lions. Nature 428, 175-178. Wilson, P.J., Grewal, S.K., Mallory, F.F., White, B.N., 2009. Genetic characterization of hybrid wolves across Ontario. J. Hered. 100, S80-S89. Wilson, P.J., Grewal, S., Lawford, I.D., Heal, J.N.M., Granacki, A.G., Pennock, D.,

498

Theberge, J.B., Theberge, M.T., Voigt, D.R., Waddell, W., Chambers, R.E., Paquet,

499

P.C., Goulet, G., Cluff, D., White, B.N., 2000. DNA profiles of the eastern Canadian

500

wolf and the red wolf provide evidence for a common evolutionary history

501

independent of the gray wolf. Can. J. Zool. 78, 2156-2166.

502 503

Woodroffe, R., Ginsberg, J.R., 1998. Edge effects and the extinction of populations inside protected areas. Science 280, 2126.

504 505

21

506 Table 1. Impact of harvest ban on cause of deatha and adoption of unrelated wolves into packsb in Algonquin Park, Ontario. Number (%) of packs Number (%) of humanNumber (%) of Number of Time period that had unrelated caused deaths* natural deaths packs with ≥ 3 animals** Pre-Ban

42 (67)

21 (33)

15

12 (80)

Post-Ban

5 (16)

26 (84)

17

1 (5.9)

a

Data are from radio-collared animals. Pre-ban mortality data are based on an 11-year sampling period (1989 – 1999) (Theberge and Theberge 2004); post-ban mortality data are based on a 5-year sampling period (2002 – 2007). b Pre-ban pack data are based on sampling between 1987 – 2001 (Grewal et al. 2004); post-ban pack data are based on sampling between 2002 – 2007. * and ** indicate significance based on a two-tailed Fisher’s exact test (P = 0.000006 and P = 0.00003, respectively).

507 508

22

509

Figure Legends

510

Figure 1. Map of study area. Pack territories are fixed kernel home ranges and those outlined

511

in black represent packs compared in Figure 3. Deer wintering areas are occupied by deer in

512

early winter or when snow cover is light and less than 30 cm in depth; deer yards are the core

513

of the deer wintering areas and is used when movement of deer is restricted due to severe

514

weather conditions when snow depth is greater than 46 cm.

515 516

Figure 2. Wolf density in Algonquin Park, Canada. Pre-ban (1989 – 1999) data are from

517

Theberge and Theberge (2004). Post-ban data was collected after a hunting and trapping ban

518

was implemented in townships surrounding the park.

519 520

Figure 3. Comparison of pre- and post-ban haplotypes. Mitochondrial DNA control region

521

and Y-chromosome microsatellite haplotypes found in packs occupying the same territory

522

during pre- and post-ban time periods. Maternal haplotypes are based on the mitochondrial

523

DNA control region and paternal haplotypes are based on Y-microsatellites. Where different,

524

the pre-ban pack name is included in parentheses.

525 526

Figure 4. Maternal and paternal haplotypes in post-ban packs. These packs are in addition to

527

those shown in Fig. 3. * indicates inferred haplotype based on paternity analysis. As in Fig. 3,

528

pups sampled from the den were excluded unless they were confirmed in the pack 1 year

529

later. Two packs (Flat Iron and LaFleur) are not shown because there was only 1 female and

530

1 male adult representative in the pack.

531 532

Supplementary Figures 1a-1q. Pedigrees for 17 post-ban eastern wolf packs in Algonquin

533

Provincial Park, Ontario. Pack names are shown at the top of each pedigree. Pink circles are

23

534

females, blue squares are males, individuals have a unique identifier made up of the field

535

sample tag (Wxxx) plus a database number (Cxxxx), dashed lines identify additional

536

relationships between individuals, HS = half-siblings, FS = full-siblings, UR = unrelated, ? =

537

unknown individual, check mark indicates those animals that were identified at some point

538

over the 5 year period as a non-breeding adult, triple lines around circle or square indicate an

539

individual that is, for clarity, duplicated in the pedigree, red outline identifies the one

540

individual that was an adopted adult animal that was unrelated to any other animal in the

541

pack, a dashed box outline indicates male animals that were captured after the death of the

542

breeding male or loss of contact with the breeding male. W58 is represented twice in these

543

figures because she dispersed from the Achray territory and subsequently became the breeder

544

in Flat Iron. Sample sizes (n) are indicated for each pack. However, it should be noted that all

545

animals represented in the pedigree were not necessarily alive or located together at the same

546

time because the pedigrees represent up to 5 years of monitoring.

547

24

Figure 1 Click here to download high resolution image

Figure 2 Click here to download high resolution image

Figure 3 Click here to download high resolution image

Figure 4 Click here to download high resolution image

Supplementary Figure 1a-q

n=14

Achray W89 C4337

HS

✔

?

UR

UR W131 C4379

UR

W14 C4262

W44 C4292

W58 C4306

W101 C4349

W146 C4394

W147 C4395

W148 C4396

W149 C4397

W150 C4398

W8 C4256

✔

W9 C4257

W151 C4399

Supplementary Fig. 1a

n=5

Beechnut

HS

W130 C4378

W119 C4367

✔ W143 C4391

✔ W144 C4392

W145 C4393

Supplementary Fig. 1b

n=4

Bena W21 C4269

UR

W42 C4290

✔ W22 C4270

W23 C4271

✔

Supplementary Fig. 1c

n=6

Big Crow ?

?

W75 C4323

?

W76 C4324

✔

W88 C4336

UR

UR

W191 C4439

W72 C4320

✔

W207 C4455

✔

✔

Supplementary Fig. 1d

n=12

Cauliflower ?

?

W49 C4297

✔

?

W16 C4264

W122 CPup2 C4370

CPup1 C4535

✔

✔

?

W176 C4424

W109 C4357 W15 C4263

W110 C4358

✔

W111 C4359

✔

W112 C4360

✔ W135 C4383

W136 C4384

Supplementary Fig. 1e

n=4

Flat Iron

?

W58b C4306

✔ W185 C4433

W173 C4421

W174 C4422

Supplementary Fig. 1f

Jocko

FS UR

W46 C4294

✔ W28 C4276

W43 C4291

n=14 UR

W200 C4448

W2 C4250

✔ ✔ W118 C4366 W53 C4301

W201 is same animal as W106

✔ W102 C4350

✔ W201 C4449

✔ W103 C4351

W104 C4352

✔ W105 C4353

W106 C4354

W107 C4355

W108 C4356

Supplementary Fig. 1g

LaFleur

n=4

?

W1 C4249

W3 C4251

?

W4 C4252

W116 C4364

Supplementary Fig. 1h

n=14

Leaf ?

W13 C4261

HS

UR

?

W13 C4261

✔

UR W68 C4316

W37 C4285

✔ W63 C4311

W68 C4316 W60 C4308

W93 C4341

W162 C4410 W198 C4446

✔ W163 C4411

✔

W164 C4412

W94 C4342

W69 C4317

✔

✔ W165 C4413

W166 C4414

Supplementary Fig. 1i

Louisa

n=6

FS

W7 C4255

W12 C4260

?

W35 C4283

W10 C4258

✔

W11 C4259

✔ W128 C4376

Supplementary Fig. 1j

McKaskill

n=16

W50 C4298

?

✔ W26 C4274

W40 C4288

✔ W54 (W193) C4302

✔ W95 C4343

✔ W96 C4344

W97 C4345

✔ W98 C4346

✔ W99 C4347

W100 C4348

✔ W113 C4361

W137 C4385

✔

W138 C4386

✔ W139 C4387

W140 C4388

W141 C4389

W142 C4390

Supplementary Fig. 1k

Pine UR

n=3

W45 C4293

W91 C4339

✔

W45/C4293 was previously collared and tracked by John Theberge as a Zig Zag pack member during the pre-ban time period. She was recaptured and recollared during the post-ban study.

W29 C4277

Supplementary Fig. 1l

Potter ?

n=5

?

✔ W41 C4289

W90 C4338

?

✔ W38 C4286

W190 C4438

?

✔ W205 C4453

Supplementary Fig. 1m

Pretty

n=11

?

✔

UR

W66 C4314

W65 C4313

W152 C4400

W195 C4443

✔ W153 C4401

W154 C4402

✔

?

W155 C4403

W156 C4404

W78 C4326

✔

✔ W70 C4318

W157 C4405

W64 C4312

✔

Supplementary Fig. 1n

Radiant

n=5

UR

W84 C4332

W62 C4310

✔ W81 C4329

✔ W82 C4330

W83 C4331

Supplementary Fig. 1o

Spoor

n=9

UR

W125 C4373

W124 C4372

✔ W133 C4381

✔ W126 C4374

W182 C4430

✔ W158 C4406

W159 C4407

W160 C4408

W161 C4409

Supplementary Fig. 1p

Sunday

n=7

?

?

✔

W47 C4295

W24 C4272

✔

W80 C4328

✔

FS

W180 C4428

UR

W25 C4273

UR

W113 C4361

SunPup C4534

Supplementary Fig. 1q