Table 4. Relative variable contributions of the predictor variables to Bsal SDMs.
| ECA70 | ECA150 | ECAfull | ||||
| pc | pi | pc | pi | pc | pi | |
| csu 25 | 34.4 | 32.8 | 26.5 | 24.0 | 39.4 | 49.4 |
| csu 5 | 0.0 | 0.0 | 6.4 | 4.1 | 18.8 | 20.8 |
| su 10–15 | 0.0 | 0.0 | 5.9 | 0.3 | 6.5 | 1.7 |
| tr 10–15 | 56.2 | 55.8 | 32.1 | 48.0 | 4.0 | 10.9 |
| cddn | 0.3 | 0.5 | 19.5 | 7.1 | 30.1 | 15.2 |
| cwdn | 0.8 | 1.4 | 2.7 | 4.6 | 0.5 | 0.7 |
| r 10 | 8.3 | 9.5 | 7.0 | 11.8 | 0.7 | 1.3 |
| BIO70 | BIO150 | BIOfull | ||||
| pc | pi | pc | pi | pc | pi | |
| bio 8 | 21.6 | 8.4 | 9.2 | 6.0 | 8.1 | 7.0 |
| bio 10 | 0.0 | 0.0 | 0.0 | 0.0 | 2.1 | 4.1 |
| bio 11 | 35.8 | 50.2 | 37.3 | 46.3 | 26.7 | 43.8 |
| bio 15 | 3.0 | 1.1 | 12.6 | 2.6 | 21.1 | 3.3 |
| bio 16 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 |
| bio 17 | 39.6 | 40.2 | 40.9 | 45.0 | 42.0 | 41.9 |
Variable contributions are shown as average percent contribution (pc) and permutation importance (pi) (for details see https://www.cs.princeton.edu/~schapire/maxent/tutorial/tutorial.doc). For each model the three highest values are in bold.