Skip to main content

Table 3 Results from linear and generalized mixed models (Table 2) analyzing the effects of the fixed effects; accumulated browsing index (ABI), moose density (pellet groups) and site productivity (Additional file 1) on moose browsing in the current winter for all tree species in young forest stands ≤5 m high

From: Moose–tree interactions: rebrowsing is common across tree species

Species

Response variable

Acc. browsing Ind.

ABI (dF = 3)

Moose density (dF = 1)

Productivity (dF = 2)

Pine

Probability of browsing (0/1)

χ 2 = 936.01, p < 0.001

χ 2 = 5.18, p = 0.023

χ 2 = 0.99, p = 0.609

Ln(browsed.shoots)

L = 190.46, p < 0.001

L = 17.85, p < 0.001

L = 1.62, p = 0.445

Bite diameter (mm)

L = 6.29, p = 0.098

L = 0.61, p = 0.435

L = 4.49, p = 0.106

Downy birch

Probability of browsing (0/1)

χ 2 = 431.44, p < 0.001

χ 2 = 3.99, p = 0.046

χ 2 = 0.28, p = 0.868

Ln(browsed.shoots)

L = 283.45, p < 0.001

L = 12.59, p < 0.001

L = 5.679, p = 0.058

Bite diameter (mm)

L = 8.12, p = 0.506

L = 1.63, p = 0.202

L = 8.17, p = 0.017

Silver birch

Probability of browsing (0/1)

χ 2 = 130.58, p < 0.001

χ 2 = 0.74, p = 0.391

χ 2 = 0.35, p = 0.552

Ln(browsed.shoots)

L = 83.97, p < 0.001

L = 15.48, p < 0.001

L = 1.63, p = 0.201

Bite diameter (mm)

L = 1.60, p = 0.660

L = 1.08, p = 0.299

L = 0.98, p = 0.321

Rowan

Probability of browsing (0/1)

χ 2 = 46.00, p < 0.001

χ 2 = 1.39, p = 0.239

χ 2 = 5.54, p = 0.019

Ln(browsed.shoots)

L = 22.93, p < 0.001

L = 0.79, p = 0.375

L = 0.18, p = 0.675

Bite diameter (mm)

L = 11.35, p = 0.010

L = 2.01, p = 0.156

L = 0.22, p = 0.634

Willows

Probability of browsing (0/1)

χ 2 = 97.22, p < 0.001

χ 2 = 3.69, p = 0.055

χ 2 = 0.03, p = 0.859

Ln(browsed.shoots)

L = 26.47, p < 0.001

L = 7.20, p = 0.007

L = 0.01, p = 0.942

Bite diameter (mm)

L = 4.61, p = 0.203

L = 0.47, p = 0.492

L = 0.73, p = 0.392

Aspen

Probability of browsing (0/1)

χ 2 = 14.48, p < 0.001

χ 2=2.34, p = 0.126

χ 2 = 1.44, p = 0.486

Ln(browsed.shoots)

L = 6.82, p = 0.033

L = 0.54, p = 0.461

L = 3.93, p = 0.140

Bite diameter (mm)

L = 15.67, p < 0.001

L = 4.54, p = 0.033

L = 1.04, p = 0.560

Juniper

Probability of browsing (0/1)

χ 2 = 46.36, p < 0.001

χ 2 = 0.45, p = 0.504

χ 2 = 0.77, p = 0.379

Ln(browsed.shoots)

L = 20.98, p < 0.001

L = 3.65, p = 0.056

L = 0.03, p = 0.866

Bite diameter (mm)

L = 4.69, p = 0.196

L = 2.11, p = 0.147

L = 2.96, p = 0.086

Spruce

Probability of browsing (0/1)

χ 2 = 92.74, p < 0.001

χ 2 = 1.24, p = 0.266

χ 2 = 9.74, p = 0.008

Ln(browsed.shoots)

L = 6.51, p = 0.089

L = 0.21, p = 0.645

L = 0.00, p = 0.953

Bite diameter (mm)

L = 1.93, p = 0.587

L = 1.61, p = 0.205

L = 2.13, p = 0.145

Grey aldera

Probability of browsing (0/1)

χ 2 = 4.83, p = 0.089

χ 2 = 0.46, p = 0.497

χ 2 = 4.50, p = 0.034

  1. For each fixed effect, nested models including/excluding the variable were compared in a likelihood ratio test, and the Likelihood ratio (L), dF and p value (<0,05 in italic) is presented for linear models, and a similar Chi square (χ2) test for the binomial model for browsing probability. For sample sizes and scientific names, see Table 1. ABI 0 = no previous browsing by moose. ABI 1 = previously browsed, but structure of the tree has not changed, ABI 2 = previous browsing has caused a change in tree structure, ABI 3 = strongly modified structure due to previous browsing
  2. aFor grey alder, the data on browsed trees were to scarce to analyze shoots browsed and bite diameter