Skip to main content

Table 3 Analysis of summer biomass (i.e. harvested around peak biomass) in periods 1 and 2 (see Fig. 2). See Table 2 for explanation of symbols

From: No consistent effect of plant species richness on resistance to simulated climate change for above- or below-ground processes in managed grasslands

 

df

Total

Grasses

Herbsa

Legumesb

Roots

SS

F

SS

F

SS

F

SS

F

SS

F

Period 1

 Explo

2

28.7

90.6***

39.8

67.6***

18.0

2.69†

91.7

9.44***

4.14

15.7***

 LUT

2

1.46

4.60*

0.544

0.924n.s.

33.3

4.97**

79.1

8.13***

1.06

4.02*

 Treatment

1

7.33

46.3***

7.38

25.0***

20.2

6.02*

23.6

4.85*

0.0260

0.198n.s.

 Treatment:Explo

2

2.69

8.51***

2.31

3.92*

    

0.742

2.82†

 LUT:Explo

4

  

4.20

3.57*

  

85.2

4.38**

  

 Residuals

 

12.8

(df = 81)

22.7

(df = 77)

278

(df = 83)

384

(df = 79)

10.78

(df = 82)

Period 2

 Explo

2

5.16

10.4***

 

3.98*

5.24

5.58**

27.85

4.29*

4.93

4.91**

 LUT

2

0.179

0.360n.s.

0.797

0.846n.s.

2.00

2.14n.s.

6.25

0.962n.s.

0.366

0.365n.s.

 Treatment

1

0.575

2.31n.s.

0.020

0.0422n.s.

1.06

2.25n.s.

8.58

2.65n.s.

4.88

9.73**

 Treatment:Explo

2

  

2.90

3.08†

      

 LUT:Explo

4

    

5.51

2.93*

  

8.19

4.08**

 Residuals

 

20.4

(df = 82)

37.7

(df = 80)

36.6

(df = 78)

178

(df = 55)

39.12

(df = 78)

  1. Biomass data were log-transformed to achieve homogeneity of variances. Models were manually simplified until all model terms were (marginally) significant. Design effects were always kept in the model. LUT and Explo refer to land-use type and exploratory region, respectively
  2. †Indicates a P value between 0.05 and 0.1
  3. aA constant value of 0.1 was added to all response values in 2008
  4. bResponse values of 2008 were square-root, rather than log-transformed; for 2009, a constant value of 0.1 was added