Table 2 Output of the generalized linear mixed models to analyze wolf habitat selection in Scandinavia, using the scores of the PC1 as response variable (variation in PC1 was mostly explained by Elevation).

From: Wolf habitat selection when sympatric or allopatric with brown bears in Scandinavia

Type of location

Type of Home Range

Model rank

Model Coefficients

df

logLik

AICc

Delta AICc

Model weight

All locations

KERN

1

Seasons

6

−36,7

87,6

0

0,60

 

Estimate

Lower 95% CI

Upper 95% CI

t value

     

(Intercept)

0,36291

0,04687

0,67895

2,297

     

Allopatric Winter

−0,38762

−0,7811

0,00586

−1,97

     

Sympatric Summer

0,05266

−0,50846

0,61378

0,188

     

Sympatric Winter

−0,9522

−1,4583

−0,4461

−3,763

     

2

Seasons + Moose

7

−35,8

88,8

1,15

0,34

 

Estimate

Lower 95% CI

Upper 95% CI

t value

     

(Intercept)

0,43673

−0,15909

1,03255

1,466

     

Allopatric Winter

−0,38642

−0,78402

0,01118

−1,944

     

Sympatric Summer

0,02202

−0,58134

0,62538

0,073

     

Sympatric Winter

−0,98664

−1,54896

−0,42432

−3,509

     

Moose

−0,25475

−2,05037

1,54087

−0,284

     

3

Null Model

3

−43,4

93,3

5,68

0,04

4

Moose

4

−42,5

93,9

6,33

0,03

MCP

1

Seasons

6

−34,3

82,8

0

0,53

 

Estimate

Lower 95% CI

Upper 95% CI

t value

     

(Intercept)

−0,2327

−0,5319

0,0665

−1,556

     

Allopatric Winter

0,226

−0,141

0,593

1,232

     

Sympatric Summer

−0,2629

−0,7983

0,2725

−0,982

     

Sympatric Winter

0,7697

0,2837

1,2557

3,168

     

2

Seasons + Moose

7

−33,1

83,4

0,57

0,4

 

Estimate

Lower 95% CI

Upper 95% CI

t value

     

(Intercept)

−0,4383

−0,9983

0,1217

−1,565

     

Allopatric Winter

0,2164

−0,1532

0,586

1,171

     

Sympatric Summer

−0,1832

−0,7498

0,3834

−0,647

     

Sympatric Winter

0,8615

0,3321

1,3909

3,255

     

Moose

0,7463

−0,9443

2,4369

0,883

     

3

Null Model

3

−40,5

87,5

4,71

0,05

4

Moose

4

−39,8

88,6

13,38

0,001

Only moving locations

KERN

1

Seasons

6

−31,5

77,2

0

0,64

 

Estimate

Lower 95% CI

Upper 95% CI

t value

     

(Intercept)

0,387227

0,102661

0,671793

−2,722

     

Allopatric Winter

0,357956

0,027826

0,688086

2,169

     

Sympatric Summer

0,001058

−0,523912

0,526028

0,004

     

Sympatric Winter

0,954532

0,46789

1,441174

3,923

     

2

Seasons + Moose

7

−30,7

78,5

1,24

0,35

 

Estimate

Lower 95% CI

Upper 95% CI

t value

     

(Intercept)

−0,4504

−1,01538

0,11458

−1,594

     

Allopatric Winter

0,35728

0,02378

0,69078

2,143

     

Sympatric Summer

0,02766

−0,54028

0,5956

0,097

     

Sympatric Winter

0,98447

0,44175

1,52719

3,628

     

Moose

0,21745

−1,51675

1,95165

0,251

     

3

Null Model

3

−39,8

86,2

8,98

0,007

4

Moose

4

−38,9

86,8

9,58

0,005

MCP

1

Seasons

6

−30,1

74,5

0

0,52

 

Estimate

Lower 95% CI

Upper 95% CI

t value

     

(Intercept)

0,05595

−0,21069

0,32259

0,42

     

Allopatric Winter

−0,09552

−0,4332

0,24216

−0,566

     

SympatricSummer

0,5935

0,12438

1,06262

2,53

     

Sympatric Winter

−0,44264

−0,86248

−0,0228

−2,109

     

2

Seasons + Moose

7

−29

75,1

0,61

0,38

 

Estimate

Lower limit

Upper limit

t value

     

(Intercept)

0,25888

−0,22414

0,7419

1,072

     

Allopatric Winter

−0,08504

−0,42502

0,25494

−0,5

     

Sympatric Summer

0,51485

0,02347

1,00623

2,096

     

Sympatric Winter

−0,53346

−0,98548

−0,08144

−2,36

     

Moose

−0,7433

−2,18704

0,70044

−1,03

     

3

Null Model

3

−36,01

78,8

4,31

0,06

   

Moose

4

−35,22

79,5

5

0,04

  1. Habitat selection was analyzed for wolf territories sympatric or allopatric with brown bears, taking into account seasonality (winter vs spring-summer seasons), moose density, and wolf territory id (random factor). We tested models with two types of wolf GPS locations (using only moving locations in one set of models, and all locations in another set), and two proxies of habitat availability, i.e., building models with MCP and kernel methods (see Methods for further details).