Ecological Archives M085-021-A4
Stephen J. Tulowiecki and Chris P. S. Larsen. 2015. Native American impact on past forest composition inferred from species distribution models, Chautauqua County, New York. Ecological Monographs 85:557–581. http://dx.doi.org/10.1890/14-2259.1
Appendix D. Native American variables (NAVs) selected by each species distribution modeling (SDM) algorithms, and variable importance in boosted regression tree (BRT) models (including NAVs).
Table D1. Native American variables (NAVs) selected by each of the four species distribution model (SDM) algorithms, for each species. An “x” indicates that a given SDM algorithm selected a given NAV for the final model.
Accessibility to a Historic Iroquoian village site |
Accessibility to a Late Woodland Iroquoian village site |
Accessibility to an Iroquoian trail |
||||||||||
Common name |
BRT |
GAM |
GLM |
MARS |
BRT |
GAM |
GLM |
MARS |
BRT |
GAM |
GLM |
MARS |
Alder |
x |
x |
x |
x |
x |
x |
x |
x |
x |
x |
x |
|
Ash, black |
x |
x |
x |
x |
x |
x |
x |
x |
||||
Ash, white |
x |
x |
x |
x |
x |
x |
x |
x |
x |
x |
||
Basswood |
x |
x |
x |
x |
x |
x |
x |
x |
x |
x |
x |
|
Beech |
x |
x |
x |
x |
x |
x |
x |
x |
x |
x |
x |
|
Birch, yellow |
x |
x |
x |
x |
x |
x |
x |
x |
x |
x |
x |
x |
Cherry, black |
x |
x |
x |
x |
x |
x |
x |
x |
x |
x |
x |
x |
Chestnut |
x |
x |
x |
x |
x |
x |
x |
x |
x |
x |
x |
x |
Cucumber magnolia |
x |
x |
x |
x |
x |
x |
x |
x |
x |
x |
x |
x |
Elm |
x |
x |
x |
x |
x |
x |
x |
x |
x |
x |
x |
x |
Fir, balsam |
x |
x |
x |
x |
x |
x |
x |
x |
x |
x |
||
Hemlock |
x |
x |
x |
x |
x |
x |
x |
x |
x |
x |
x |
|
Hickory |
x |
x |
x |
x |
x |
x |
x |
x |
x |
x |
x |
x |
Ironwood |
x |
x |
x |
x |
x |
x |
x |
x |
x |
x |
x |
|
Maple, red |
x |
x |
x |
x |
x |
x |
x |
x |
x |
x |
x |
x |
Maple, sugar |
x |
x |
x |
x |
x |
x |
x |
x |
x |
x |
||
Oak, black |
x |
x |
x |
x |
x |
x |
x |
x |
x |
x |
x |
x |
Oak, white |
x |
x |
x |
x |
x |
x |
x |
x |
x |
x |
x |
x |
Sycamore |
x |
x |
x |
x |
x |
x |
x |
x |
x |
x |
x |
|
Walnut, black |
x |
x |
x |
x |
x |
x |
x |
x |
x |
x |
x |
x |
Walnut, white (butternut) |
x |
x |
x |
x |
x |
x |
x |
x |
x |
x |
x |
|
White pine |
x |
x |
x |
x |
x |
x |
x |
x |
x |
x |
x |
x |
Whitewood |
x |
x |
x |
x |
x |
x |
x |
x |
x |
x |
x |
x |
Table D2. The relative importance of each predictor variable (including Native American variables [NAVs]), for boosted regression tree (BRT) models of tree species distributions. Predictor variable columns are sorted from highest (left) to lowest (right) mean relative importance, as averaged across all BRT models.
Common name |
Accessibility to a |
Mean January |
Mean temperature, |
Total precipitation, |
Slope angle |
Accessibility to a |
Accessibility to a trail |
Soil pH |
AET/PET, July |
Soil drainage class |
Compound |
Total solar radiation, |
Soil percent sand |
Soil percent clay |
Alder |
4.3% |
5.8% |
- |
3.0% |
11.7% |
10.0% |
8.0% |
8.8% |
4.4% |
14.7% |
15.2% |
8.8% |
5.5% |
- |
Ash, black |
3.3% |
3.2% |
6.2% |
3.4% |
17.5% |
3.9% |
6.6% |
9.8% |
4.1% |
24.4% |
9.1% |
3.5% |
5.0% |
- |
Ash, white |
12.3% |
9.7% |
12.4% |
8.3% |
7.7% |
8.0% |
10.5% |
4.3% |
8.0% |
3.3% |
5.5% |
6.7% |
3.2% |
- |
Basswood |
20.7% |
- |
14.1% |
15.9% |
22.9% |
11.8% |
- |
- |
- |
14.6% |
- |
- |
- |
- |
Beech |
14.0% |
5.6% |
4.9% |
6.9% |
14.9% |
4.4% |
6.7% |
5.7% |
4.8% |
13.0% |
6.0% |
4.0% |
5.6% |
3.6% |
Birch, yellow |
17.3% |
9.6% |
8.8% |
8.4% |
8.9% |
7.3% |
8.9% |
5.9% |
8.3% |
5.4% |
5.8% |
5.4% |
- |
- |
Cherry, black |
14.7% |
8.6% |
10.7% |
10.3% |
5.3% |
8.7% |
10.3% |
5.3% |
6.6% |
3.2% |
3.7% |
6.7% |
2.6% |
3.3% |
Chestnut |
24.3% |
16.8% |
8.4% |
4.2% |
6.1% |
6.6% |
5.7% |
7.5% |
5.0% |
- |
4.8% |
5.8% |
- |
4.8% |
Cucumber magnolia |
15.8% |
16.9% |
15.4% |
8.8% |
8.1% |
8.3% |
10.8% |
5.7% |
10.2% |
- |
- |
- |
- |
- |
Elm |
10.3% |
5.4% |
12.8% |
8.4% |
5.4% |
8.1% |
13.7% |
6.8% |
8.9% |
3.5% |
6.2% |
6.5% |
3.9% |
- |
Fir, balsam |
5.6% |
3.8% |
7.8% |
7.6% |
11.5% |
7.8% |
6.4% |
10.1% |
7.3% |
7.3% |
3.3% |
3.9% |
7.7% |
9.8% |
Hemlock |
8.2% |
6.4% |
9.4% |
10.6% |
8.7% |
10.1% |
7.7% |
7.5% |
5.5% |
6.3% |
6.8% |
5.2% |
4.1% |
3.6% |
Hickory |
13.9% |
19.0% |
5.7% |
8.9% |
4.9% |
11.8% |
7.7% |
2.3% |
7.1% |
- |
3.7% |
5.0% |
6.0% |
4.1% |
Ironwood |
19.2% |
12.3% |
5.6% |
20.6% |
2.8% |
5.7% |
17.8% |
2.9% |
3.5% |
- |
2.3% |
4.4% |
2.9% |
- |
Maple, red |
16.5% |
3.7% |
7.5% |
16.5% |
13.3% |
4.0% |
6.5% |
9.2% |
2.9% |
3.9% |
3.9% |
4.4% |
6.0% |
1.7% |
Maple, sugar |
17.2% |
8.1% |
8.2% |
8.3% |
16.2% |
7.2% |
- |
7.7% |
- |
13.1% |
- |
6.5% |
7.4% |
- |
Oak, black |
19.7% |
15.7% |
7.9% |
7.0% |
5.4% |
7.3% |
9.5% |
6.3% |
5.7% |
- |
5.8% |
6.5% |
- |
3.2% |
Oak, white |
28.2% |
13.7% |
13.9% |
5.7% |
3.9% |
3.7% |
8.0% |
3.7% |
6.4% |
- |
3.6% |
6.8% |
1.4% |
1.1% |
Sycamore |
14.7% |
11.3% |
8.9% |
12.3% |
6.8% |
4.9% |
8.2% |
6.4% |
- |
4.1% |
11.9% |
6.9% |
3.6% |
- |
Walnut, black |
8.9% |
17.8% |
7.0% |
4.8% |
2.6% |
17.5% |
6.0% |
4.2% |
7.9% |
2.1% |
8.7% |
5.7% |
- |
6.9% |
Walnut, white (butternut) |
11.2% |
11.2% |
7.0% |
6.9% |
5.4% |
7.3% |
7.9% |
5.0% |
3.6% |
4.6% |
13.6% |
4.9% |
5.6% |
5.8% |
White pine |
13.4% |
7.0% |
10.0% |
6.1% |
4.8% |
16.9% |
7.9% |
7.1% |
12.0% |
3.1% |
- |
5.8% |
2.7% |
3.2% |
Whitewood |
12.3% |
23.5% |
11.8% |
10.0% |
3.5% |
6.5% |
4.7% |
3.9% |
8.1% |
- |
5.6% |
5.2% |
- |
4.9% |