Table 2. Multiple regression results showing the best subset of predictors for each considered model (EMIB, GDM, HD and CLIMATE) to explain the between-island variation in species richness of Macaronesian bryophytes.
F | P | R2adj | AICC | |
Total species richness ( STOT ) | ||||
EMIB (A) | 2.78 | 0.126 | 0.132 | 238.5 |
GDM (A, TT 2) | 7.93 | 0.002 | 0.565 | 230.1 |
HD (sdELEV) | 17.32 | <0.001 | 0.565 | 227.8 |
CLIMATE (TMAX, PMIN, MistL) | 15.67 | <0.001 | 0.728 | 221.2 |
Moss species richness ( SM ) | ||||
EMIB (A) | 4.85 | 0.042 | 0.222 | 285.0 |
GDM (A, TT 2) | 10.54 | <0.001 | 0.638 | 212.0 |
HD (sdELEV) | 26.13 | <0.001 | 0.606 | 208.8 |
CLIMATE (PMIN, MistL) | 17.07 | <0.001 | 0.662 | 208.0 |
Liverwort species richness ( SL ) | ||||
EMIB (A, DM) | 2.85 | 0.088 | 0.219 | 198.8 |
GDM (A, TT 2) | 3.83 | 0.032 | 0.363 | 197.5 |
HD (sdELEV) | 5.42 | 0.033 | 0.242 | 196.1 |
CLIMATE (TMAX, PMIN, MistL) | 19.28 | <0.001 | 0.768 | 178.3 |
The best subset of variables that were chosen using the lowest sample size-corrected Akaike information criterion (AICC) is shown in brackets. Adjusted R2 values and its statistical significance according to the F-test are also shown. Model acronyms and variable codes as in Table 1.