Skip to main content

Table 2 Regression models for seedling abundance (binomial negative models) and sapling/seedling ratio (linear models)

From: Does the persistence of sweet chestnut depend on cultural inputs? Regeneration, recruitment, and mortality in Quercus- and Castanea-dominated forests

Species

Model

Explained deviance (%)

R2

AIC

p

C. sativa

ln(seedling.abun) = 0.71 + 0.05*(BA.Cs)

47.8

 

64.21

< 0.001

ln(seedling.abun) = 1.45 + 0.01*(TD.Cs)

4.4

 

71.20

0.181

ln(seedling.abun) = 2.69–0.09*(%light)

19.3

 

69.15

0.076

sapling:seedling.ratio = 0.018 + 0.007*(%light)

 

37.6

 

0.045

Q. pyrenaica

ln(seedling.abun) = 1.23 + 0.15*(BA.Qp)

43.2

 

98.56

< 0.001

ln(seedling.abun) = 2.11 + 0.04*(TD.Qp)

26.9

 

101.69

0.004

ln(seedling.abun) = 1.99 + 0.13*(%light)

27.3

 

101.50

0.009

sapling:seedling.ratio = 0.0006 + 0.00007*(%light)

 

7.38

 

0.419

  1. Basal area (BA) and tree density (TD) of each species, and the percentage of understory light (% light) used as independent variables
  2. Cs C. sativa; Qp Q. pyrenaica