Tim Vandeelen

Wolf recovery in Wisconsin and its impact on the growth of whitetailed deer populations T. R. Van Deelen, E. A Berkley, and A. P. Wydeven Department of Wildlife Ecology University of Wisconsin-Madison Wisconsin DNR

Photo : Wisconsin DNR

The players

15-18 deer/wolf/year (Fuller 1995) Photos : Wisconsin DNR

Hypothesized impact of wolf recovery in Wisconsin • Fewer deer, reduced herbivory on browse-sensitive plants • Fewer deer, reduced deer hunting opportunity • More depredations on livestock, pets, and hunting dogs

Photo : Wisconsin DNR

Anti-predator specialization of white-tailed deer • Concentration/dilution behavior • Rapid fawn growth • Speed (40 - 50 mph) • Hiding • Home range fidelity • Flagging • Vigilance • High reproductive rate Photo : Wisconsin DNR

Hunter effort at Sandhill Wildlife Area 60

Hunter-days/deer killed

55 50 45

Y = 1/(0.01809 + 0.01126X) r^2 = 0.86

40 35 30 25 20 15 10 5 0 -10

0

10

20

30

40

50

Deer/mi^2 Data from: Creed, W. A. 2001. The total removal hunt. Pp. 53-66 in J.F. Kubisiak et al. Sandhill whitetails: providing new perspective for deer management. Wisconsin DNR.

Photo : Wisconsin DNR

COMPLETELY COMPENSATED

Population growth rate (births - deaths)

PARTIALLY COMPENSATED

ADDITIVE

Mortality from wolf predation

How many wolves are there? (Source data on wolves)

Population Monitoring About 46% of packs and 16% individuals are monitored by radiotelemetry each year 25 collared in 2007 75 wolves on the air in 2007

Snow track surveys are used to count non-radiocollared packs 7000 To 8000 miles tracked

OBSERVATION REPORTS BY OTHER AGENCIES & THE PUBLIC Supplement track surveys & radio-tracking

Annual Population Meeting To Estimate State Wolf Population

State of Wisconsin Department of Natural Resources Box 7921, Madison, WI 53707-7921

Carnivore Track Survey

Form 1700-052 (R 7/03) Page 1 of 2

Volunteers are an essent ial part of the Wisconsin Carnivore Tracking Program ’s success. We appreciate your hard work and dedic ation. Thank you for participating. Notice: Use this voluntary form to monitor and report carnivore tracking activities. Inform ation reported to the Department will be used for research and management purposes. Personally identifiable information is not int ended to be used for oth er purposes. Wisconsin’s Open Records law requires the Department to provide this information upon request [ss. 19.31 - 19.69, Wis. Stats.].

Sarah R. Boles

Observers

Sheet __ 1_____ of ___2____

Survey Information Survey Block

38

Weather and Track Information Counties (List all counties tracked)

Snow Depth

Ashland

Pack Name or General Area

28" total

Temperature (at start of survey)

Torch River

10°

28

Survey Date

2-1-02

Township

42

Start Time

0800

Range

N

4

End Time

2"

48 hrs.

Begin Survey (Also indicate on accompanying map) Section

New Snow Depth on Road

2" fresh

Time of Last Snowfall (In hours if less than 48, else days)

Cloud Cover

x 0%

25%

50%

75%

100%

E/W Past Weather Hi:

1545

Canids: C =Coyote D =Dog F=Fox W =Wolf Felids: BC = Bobcat CT = Cat L = Lynx PM = Puma Roads and Direction of Travel Mileage Canida

30

Track Conditions

Low:

5 Poor

Precipitation (Last 24 Hours)

x OK

0

Good

Excellent

Mustelids: B =Badger FI=Fisher O=Otter S=Skunk Other: BR = Bear BV = Beaver P = Porcupine R = Raccoon Mustelids Felids Other Notes and Comments

GG & FR 335 N 335 FR 335 & FR 168 N 168 " " " " " FR 168 & Hwy. 77 E

.0 1.2 2.5 2.8 3.1 3.7 4.2 4.5 5.0 5.2

Hwy 77 & FR 170 S 170 " " FR 170 & FR 168 W 168 FR 168 & FR 335 S Backtrack to GG

8.1 8.3 8.5 8.8 11.1 12.2 12.5

Hwy GG S

15.0

Not Trackable

18.2

Logging Operation, Lots Deer sign

Hwy GG & FR 164 E Totals

1F

2W 1F

1W

2W

1 FI

Enter from E >24 hrs Sec. 11 T42N R4W (F 4.5"x3.75", 4"x3.8") Stride 38" TW's exit to [email protected]

1 FI 1 FI

1 BC

Lots of snowmo. traffic Crossed E W <12 hrs

Crossed S

N ~12 hrs.

465

Changes in Wisconsin Gray Wolf Population: 1980-2006

435 373 327 335

248

Number of Wolves

257

205 178

Number of Packs

148

06

05

04

66 70

02

20 00

99

35

98

97

96

28 14 18

95

94

93

92

91

19 90

89

88

87

86

85

84

83

82

81

19 80

40 45 40 34 28 31 27 25 21 19 17 15 16 18 13 12 5 5 4 5 4 4 5 5 6 7 11 12

47 57

01

83 57

108 113 115 83 94

03

99

Population growth of wolves in the Midwest since 1975 1000

3500 3000 2500

900

y = 4E-29e0.0366x R2 = 0.9895

800 700 600

2000

y = 3E-154e0.1797x R2 = 0.9279

500 400

1500

300 1000

MN

500 0 1975

1980

1985

1990

1995

2000

2005

MI, WI

200 100

2010

0 1975

1980

1985

1990

1995

2000

2005

2010

Consequences of more wolves.

Dogs Depredated (n=136) and Injured (n=49) by wolves in Wisconsin, 1974-2007 30 25 20 15 10 5 0 '74 '75 '76 '77 '78 '79 '80 '81 '82 '83 '84 '85 '86 '87 '88 '89 '90 '91 '92 '93 '94 '95 '96 '97 '98 '99 '00 '01 '02 '03 '04 '05 '06 07 # Of All Dog Breeds Killed (WI) # Of All Dog Breeds Injured (WI)

Farms with verified livestock depredations in Wisconsin, 19742007 35 30 25 20 15 10 5

Farms w ith Verified Losses

'06

'04

'02

'00

'98

'96

'94

'92

'90

'88

'86

'84

'82

'80

'78

'76

'74

0

Wolves captured for depredation purposes

40

38

35

32

500

27 400

25 20 15

15

0

300 200

10 5

17

18

Wolf population

Wolves captured

30

600

wolves captured wolf population

8 1

0

0

0

0

0

2

4

100 2

2 0

'91 '92 '93 '94 '95 '96 '97 '98 '99 '00 '01 '02 '03 '04 '05 '06 7 Year

1974-2007, 181 wolves captured, 130 wolves

Is population growth of wolves slowing down? (Indicates that they are limited by deer numbers)

Methods • Assemble data • Choose biologically relevant models – Mathematical models – Scientific literature

• Find the model that fits best • Study the behavior of the best-fitting model

David Mladenoff Department of Forest and Wildlife Ecology UW-Madison

Two steps • Select an optimal growth model for the combined Wisconsin and Michigan population • Apply the optimal model to Wisconsin counts through 2008

Goal of modeling: Estimates of… • • • •

Growth rate (λ) Carrying Capacity (K) Variances Form of Density Dependence (dd)

Use of these estimates • Components of relationship between sustainable harvest and population size • Predictors of Ne (number of wolves at equilibrium) – Depends on variance

Summary of step 1 • Simple density dependence • Beverton-Holt model • Ne = 1321 wolves (Wisconsin and Michigan) – 95% CI: 1215 – 1427 wolves

• λmax = 1.31 – 95% CI: 1.28 – 1.34

Step 2: Beverton-Holt model applied to Wisconsin data through 2008

546

537

467 435 373 327 335

248

Number of Wolves

257

204

Number of Packs

178 148 99 83

39 45 31 34 26 25 20 23 19 18 14 15 18 10 12 13 5 4 6 7 5 4 4 5 5 5 1980 81

82

83

84

85

86

87

88

89 1990 91

92

40 12 93

54 31 16 20 94

95

96

35

97

47

98

57

65 70

99 2000 01

83 94

02

03

108 113

04

05

116

06

source: Wisconsin DNR

141 143

07

08

Predicted growth of Wisconsin wolves (jacknifed variance) 800

Wolf numbers

700 600 500

Ne = 658 wolves 95% CI: 617 - 700

400 300 200 100 0 1985

1995

2005

2015

2025

2035

Form of Density Dependence Beverton-Holt model 1.8 Growth rate (Lambda)

1.6 1.4 1.2 1

Observed

0.8

Expected

0.6 0.4 0.2 0 0

100

200

300

400

500

600

Population size

Allee effect? 1.8 Growth rate (Lambda)

1.6 1.4 1.2 1

Observed

0.8 0.6 0.4 0.2 0 0

100

200

300 Population size

400

500

600

The common explanation for density-dependence is limited prey (Would require an impact to deer populations)

How is population growth of deer impacted by wolves?

Wolf packs 2006

Deer Management Units 2006

Deer Management Units with resident wolf packs since 1980 60

DMUs with wolf packs

50 40

y = 2E-100e0.1165x R2 = 0.972

30 20 10 0 1975

1980

1985

1990

1995 Year

2000

2005

2010

Does the presence of wolves impact growth of DMU-based deer populations? • Growth is the basis of sustainable harvest • DMUs as a sampling unit – – – –

WI DNR monitoring framework Homogeneous landuse Spatial scale approximates that of wolf pack Stable throughout the period of wolf recovery

Is the impact additive?

Explicitly Additive Model Structure pgrdmu,year = (ß0+α0I) + (ß1+α1I)x1 + (ß2+α2I)x2 + … + (ßi+αiI)xi Where I = 1 (wolf pack present) or 0 (wolf pack not present) xi = predictor variables ßi, αi = estimated parameters DMU modeled as a random spatial effect Regression parameter in presence of wolves = ßi+αi Regression parameter absent wolves = ßi

Vucetich and Peterson 2004

Predictor variables for ln(Nt/Nt-1) • WSI: yearly winter severity index • tHarDens: harvest density with Box-Cox transformation • PopDens: population density the previous year • Diff: PopDens - WI DNR goal density

Models • • • • •

Global (PopDens, tHarvDens, Diff, WSI) Density Dependent only (PopDens) Management only (tHarvDens, Diff) Winter only (WSI) Density Dependence and Management (PopDens, tHarvDens, Diff) • Density Dependence and Winter (PopDens, WSI) • Global Null (PopDens, tHarvDens, Diff, WSI with no wolf presence terms [αi])

Model selection, N = 1260 Rank

Model

AIC

wi

1

Global, rand(DMU)

-695

1.0

2

Global Null, rand(DMU)

-674

0

3

Global

-584

0

4

Global Null

-552

0

5

DD and Management, rand(DMU)

-488

0

6

DD and Management

-394

0

7

Management rand(DMU)

-156

0

8

Management only

-138

0

9

DD and winter

-87

0

10

DD and winter, rand(DMU)

-85

0

11

Winter only

1

0

12

DD, rand(DMU)

31

0

13

DD only

39

0

14

Winter, rand(DMU)

44

0

Growth with and without wolves Model: Global, random (DMU) 10,000 Monte Carlo simulations Mean + or - SE 0.15 16%

11% 0.1

Growth rate

0.05 5%

No wolves

0 Wolves -0.05 -5%

-11% -0.1 -0.15 -16%

-22% -0.2

Sensitivity analysis Model: Global, random(DMU) 10,000 Monte Carlo Simulations across range in data set 0.5

0.5

No wolves

0.4

0.4

0.3

0.3

0.2

0.2

0.1

0.1

0 -0.1

0 10%tile

20%tile

30%tile

40%tile

50%tile

60%tile

70%tile

80%tile

90%tile

Wolves

-0.2 -0.3

-0.5

0.4

0.3

0.3

0.2

0.2

0.1

0.1

0

50%tile

60%tile

70%tile

80%tile

90%tile

0 10%tile

20%tile

30%tile

40%tile

-0.2

-0.5

40%tile

-0.5

0.4

-0.4

30%tile

tHarvDens

-0.4

0.5

-0.3

20%tile

-0.3

0.5

-0.1

10%tile

-0.2

PopDens

-0.4

-0.1

50%tile

60%tile

70%tile

80%tile

90%tile

-0.1

10%tile

20%tile

30%tile

40%tile

50%tile

-0.2

Diff

-0.3 -0.4 -0.5

WSI

60%tile

70%tile

80%tile

90%tile

Conclusions so far • Optimal model contained most structure (most variables) • Additive effect of wolves is small, especially when partitioned among predictor variables • High variances

Photo : Wisconsin DNR

Inferences about ecology of wolves and deer in Wisconsin • Growth in deer populations is influenced by deer density, harvest, winter severity, habitat, and wolves • Wolves may depress deer population growth but the magnitude of this effect is small and probably swamped by the effects of other variables – Compensation? – Swamping by a high deer population?

So what?

Federal Wolf Delisting • Wolves in WI and other portions of the Western Great Lakes DPS de-listed on March 12, 2007 • Humane Society et al. is currently legally challenging delisting

Wisconsin Conservation Congress Question 71: Wolf management. Vote: Yes, 4848 – No 772. Passed in 72 of 72 counties The question: “Do you favor the Department of Natural Resources, Wisconsin conservation Congress and the Wisconsin Legislature develop a season framework and harvest goals to maintain the wolf population with management objectives?”

Wolf Harvest by US States Population

• • • • • • •

Hunting

Trapping

AK ~7700+ ~630/yr ~950/yr ID 732 no* no MI 520 no no MN ~3000 no (5+ yr) no (5+ yr) MT 422 no* no WI 537 not currently not currently WY 359 no* no *Hunting Planned but tabled when wolves were relisted on 7/11/08

FACTORS AFFECTING WOLF POPULATION SIZE Illegal Kill

Diseases

Vehicle Collisions

WOLF WOLF POPULATION POPULATION Prey Abundance (Deer & Beaver) Depredation Control Activities

Other Wolves, (Wolf Population Density) Public Harvest?

Sustainable Yield at 350 wolves (Goal of Wolf Management Plan)

Yield

Sustained Yield Curve for Wisconsin Wolves

45 40 35 30 25 20 15 10 5 0

41 wolves

350 wolves 0

100

200

300

400

Population size

500

600

700

Sustainable Yield at 75% of Carrying Capacity

Yield

Sustained Yield Curve for Wisconsin Wolves

45 40 35 30 25 20 15 10 5 0

29 wolves

494 wolves 0

100

200

300

400

Population size

500

600

700

Discussion • Simple density dependence – Ungulate biomass is very high – Wolf impact on deer population growth is trivial – Social effect?

• Predicted Ne substantially higher than recovery goals – Harvest, hunting, trapping? – Revise recovery goals?

• Uncertain landscape relationships – What’s the quality threshold for pack occupancy?

Caveats • Density dependent behavior is a recent feature of these data • Assumes harvest is additive • Predictions will change with added observations • Wisconsin’s wolf population is not isolated

Thanks

Photo : Wisconsin DNR