Table 1.
Multivariate regression model explaining the effect of environmental variables on the species abundances of the last relevé of each plot using forward selection in CCA, using the countries as covariables
| A | ||||
|---|---|---|---|---|
| Variable | Compartment | F | p | Percentage explained variance |
| pH | Organic | 8.10 | .001 | 1.7 |
| Mediterr. oak | Tree | 6.79 | .001 | 1.4 |
| Temperate oak | Tree | 5.95 | .001 | 1.3 |
| Pinus sylv + nigra | Tree | 4.83 | .001 | 1.0 |
| Fagus | Tree | 4.42 | .001 | 0.9 |
| CEC | Mineral | 2.84 | .001 | 0.6 |
| N/C | Organic | 2.46 | .005 | 0.5 |
| Latitude | Climate | 2.44 | .001 | 0.4 |
| NO3 (2000) | Deposition | 2.36 | .001 | 0.5 |
| Longitude | Climate | 2.28 | .001 | 0.5 |
| Coniferous “other” | Tree | 2.23 | .003 | 0.4 |
| Deciduous “other” | Tree | 2.13 | .061 | 0.4 |
| Ca | Organic | 2.10 | .002 | 0.4 |
| Atlantic South | Climate | 1.96 | .008 | 0.4 |
| Age | Tree | 1.89 | .003 | 0.3 |
| Atlantic North | Climate | 1.85 | .005 | 0.3 |
| K | Organic | 1.87 | .011 | 0.4 |
| Boreal | Climate | 1.72 | .012 | 0.3 |
| P | Organic | 1.70 | .007 | 0.3 |
| Altitude | Climate | 1.36 | .092 | 0.3 |
| N/C | Mineral | 1.23 | .173 | 0.3 |
| Further terms not given | ||||
| Sum if p < .05 | 12.4 | |||
| B | ||
|---|---|---|
| Compartment | % Explained variance (data) | % Explained variance (fitted values) |
| Tree layer | 5.9 | 47.3 |
| Soil organic layer | 3.4 | 27.4 |
| Soil mineral layer | 0.6 | 4.8 |
| Climate | 2.0 | 16.4 |
| Deposition | 0.5 | 4.1 |
| Sum | 12.4 | 100.0 |
Eigenvalues: λ1 = 0.257, λ2 = 0.235, λ3 = 0.185, λ4 = 0.120, sum of all eigenvalues = 11.739, sum of all canonical eigenvalues = 1.458, number of plots = 477, number of species = 170. Rare species are downweighted. F = (regression mean square with this term—regression mean square without this term)/error mean square; p = probability of this, or a higher F‐value under the null hypothesis as determined on the basis of 999 bootstrap samples. The pool of environmental variables from which the terms were selected included all soil chemical variables, tree species, and climate zones; and altitude, geographical coordinates, stand age; and EMEP deposition estimates of NO3 and NH4 for both 1995 and 2000, with the constraint that their mutual absolute correlation coefficient should always be below 0.5 (see Appendix S8). A: the selection results per variable, B: a summary per compartment giving the percentage explained variance with respect to the data (left column) and with respect to the fitted values (right column).