Table 4.
Analysis of variable contributions and model assessment
| ENM_1 | ENM_2 | |||
|---|---|---|---|---|
| Variable | Percent contribution | Permutation importance | Percent contribution | Permutation importance |
| Sheep10 | 23.4 | 10.5 | 22 | 8.1 |
| Rumi05 | 20.2 | 9.8 | 22.6 | 4.8 |
| Alt | 15.3 | 25.2 | 13.8 | 32.9 |
| Bio 15 | 14.7 | 8.7 | 11.8 | 6.8 |
| Bio 7 | 9.5 | 3.1 | 12.1 | 9.5 |
| Bio 3 | 7.3 | 6.0 | 6.2 | 1.9 |
| Access | 5.9 | 1.4 | 6.1 | 0.7 |
| Bio 10 | 1.3 | 6.6 | 2.4 | 5.9 |
| Pop | 0.9 | 12.5 | 1.0 | 9.6 |
| Bio 2 | 0.7 | 3.4 | 0.4 | 4.1 |
| Cattle05 | 0.5 | 2.3 | 0 | 1.0 |
| Evi mean | 0.2 | 2.8 | 0.2 | 3.6 |
| Cattle10 | 0.2 | 1.2 | 0.2 | 1.3 |
| Prec mean | 0.1 | 5.3 | 0.5 | 6.8 |
| lst mean | 0.1 | 0.6 | 0.1 | 1.8 |
| Goat10 | 0 | 0.3 | 0.5 | 0.8 |
| Mir mean | 0 | 0.5 | 0 | 0.2 |
| ENM_1 | ENM_2 | |||
| Training gain | 2.634 | 2.614 | ||
| Test gain | 2.667 | 1.252 | ||
| Training AUC | 0.976 | 0.975 | ||
| Test AUC | 0.973 | 0.936 | ||