Skip to main content

Table 8 Upper part: Results of the pairwise comparisons by fitting Eq. 3 (corresponding to Q3)

From: Drought can favour the growth of small in relation to tall trees in mature stands of Norway spruce and European beech

Pairwise comparisons with regression models after Eq. 1

Comparison

a 0

SE

a 1

SE

b 0

SE

b 1

SE

R 2

p (model)

N.sp. all vs. E.be. all

−0.0425

0.0339

0.1004***

0.0106

0.0022

0.0016

n.a.

n.a.

0.55

0.000

N.sp. mono vs. N.sp. mixed

−0.0668*

0.0321

0.0469

0.0454

0.0036*

0.0016

−0.0027

0.0022

0.16

0.112

E.be. mono vs. E.be. mixed

0.0101

0.0388

0.0972***

0.0122

0.0022

0.0019

n.a.

n.a.

0.65

0.000

Resulting linear models GDC = a + bMgs

Category

a

b

 

diff

      

N.sp. all

−0.0425

0.0022

 

A

diff A is due to a difference in the intercept (p = 0.000)

E.be. all

0.0579

0.0022

 

A

diff B is due to a difference in the intercept (p = 0.029)

N.sp. mono

−0.0668

0.0036

        

N.sp. mixed

−0.0199

0.0001

        

E.be. mono

0.0101

0.0022

 

B

      

E.be. mixed

0.0972

0.0022

 

B

      
  1. In some cases, the parameter b1 was left out due to model overfitting (entry n.a.). Lower part: Linear models for the different species/stand categories resulting from combining the corresponding parameter estimates from the upper part. If two categories have the same letter under ‘diff’, their linear models differ significantly due to the significances shown in the upper part of the table. Not all possible combinations of categories were tested against each other; see column ‘comparison’ for the performed tests. N.sp, Norway spruce, E.be European beech. Significant differences of the parameters between the groups are indicated by the levels of significance (*p < 0.05, **p < 0.01, ***p < 0.001)