Table 5. MLPE fitted model results on the effects of five landscape variables on genetic differentiation.
| Response variable | Model | Null | Elevation | Land cover | Land-elevation | Hybrid | Marginal R 2 |
|---|---|---|---|---|---|---|---|
| 15 populations | |||||||
| F ST | 3 | — | — | 0.40 ± 0.04 | — | n/a | 0.456 |
| 4 | — | — | — | 0.03 ± 0.01 | n/a | 0.365 | |
| 2 | — | -2.74 ± 1.04 | — | — | n/a | 0.258 | |
| 1 | 0.00 ± 0.01 | — | — | — | n/a | 0.001 | |
| D EST | 3 | — | — | 0.04 ± 0.03 | — | n/a | 0.130 |
| 4 | — | — | — | 0.04 ± 0.03 | n/a | 0.089 | |
| 2 | — | -1.26 ± 2.58 | — | — | n/a | 0.013 | |
| 1 | 0.00 ± 0.01 | — | — | — | n/a | 0.001 | |
| 12 populations | |||||||
| F ST | 3 | — | — | 0.04 ± 0.01 | — | — | 0.451 |
| 4 | — | — | — | 0.03 ± 0.01 | — | 0.372 | |
| 5 | — | — | — | — | 0.00 ± 0.00 | 0.031 | |
| 1 | 0.01 ± 0.01 | — | — | — | — | 0.006 | |
| 2 | — | 0.27 ± 1.68 | — | — | — | 0.003 | |
| D EST | 3 | — | — | 0.05 ± 0.05 | — | — | 0.247 |
| 4 | — | — | — | 0.05 ± 0.03 | — | 0.190 | |
| 5 | — | — | — | — | 0.00 ± 0.00 | 0.012 | |
| 1 | 0.01 ± 0.02 | — | — | — | — | 0.010 | |
| 2 | — | 0.21 ± 3.97 | — | — | — | 0.000 |
Four explanatory variables were fitted for MLPE models for 15 populations (above mid-rule), and five variables for 12 populations (below mid-rule) to test for the effects of landscape variables on two measures of genetic distance (F ST = above dashed line and D EST = below dashed line). Models were ranked based on marginal R 2. Explanatory variables that were not included in the fitted model are indicated by ‘—’ and variables that were not tested are indicated by ‘n/a’. Values are presented as regression slope estimates ± 95% confidence interval and have been converted to x10-4. Bold values indicate significance where 95% confidence intervals which do not overlap zero.