Table 2 Relative importance of variables introduced in the hazard and in the exposure boosted regression trees
Hazard model | Exposure model | ||
|---|---|---|---|
Variable | Relative importance (%) | Variable | Relative importance (%) |
Infections in 20 km | 23.23 | Infection in 20 km | 25.42 |
Volume of spruce | 10.71 | Length of roads in forest | 17.67 |
Distance to water course | 7.30 | Distance to water course | 7.79 |
Volume of oak | 6.33 | Proportion of forest | 7.75 |
Proportion of forest | 5.40 | Mean height of trees | 7.09 |
Proportion of coniferous | 5.33 | PC1: Accessibility | 6.78 |
Proportion of clear-cuts | 5.28 | Proportion of holiday houses | 6.18 |
Volume of birch | 5.24 | Distance to the sea | 5.49 |
Forest shape index | 5.04 | Standard deviation of height of trees | 5.19 |
Shannon diversity index | 4.83 | Proportion of clear-cuts | 5.19 |
Volume of pine | 4.47 | Proportion of broad-leaved forest | 3.74 |
Proportion of mixed forest | 4.26 | Proportion of waterbodies | 1.72 |
Forest proximity index | 3.60 | ||
Open areas in ecotones | 3.05 | ||
Proportion of broad-leaved forest | 2.20 | ||
PC1: Wildlife | 1.31 | ||
Proportion of waterbodies | 1.31 | ||
Roe deer | 1.05 | ||