Environmental variable / Native units / Description
Vegetation class / — / Refined vegetation categories derived from Existing Vegetation Layer obtained from LANDFIRE (2009): barren, desert grassland, desert shrubland, mixed conifer forest and woodland, montane grassland, montane shrubland, montane-deciduous forest, pinyon-juniper woodland, ponderosa pine and pine-oak forest, water, other.
Ponderosa pine proportion / — / Proportion of ponderosa pine pixels calculated using the LANDFIRE (2009) Existing Vegetation Layer.
Tree densitya / trees/ha / Estimate of the average number of trees per hectare > 2.54 cm in diameter.
Basal areaa / m2/ha / Estimate of the average tree basal area per hectare.
Overstory canopy covera / % / Estimate of average tree canopy cover.
Tree heighta / m / Estimate of average tree height.
Stand density indexa / — / Index of tree density based on the number of trees per unit area and diameter of trees of average basal area.
Quadratic mean diametera / in / Quadratic mean tree diameter = square root (basal area / (0. 0054 × number of trees)).
Wood volumea / m3/ha / Total cubic foot stem volume.
Canopy biomassa / dry tons/ha / Estimate of total crown biomass.
Total biomassa / dry tons/ha / Estimate of total tree biomass.
Canopy bulk densitya / kg/m3 / Dry weight of the available canopy fuel per unit of canopy volume (Scott and Reinhart 2005).
Canopy base heighta / m / Estimate of the average height from the ground to tree canopy bottom.
Dominant forest structure class / — / Categories are old-forest dense, old-forest open, intermediate, young forest, open or non-forest. Derived using mean tree height and overstory canopy cover.
Forest edge density / m/m2 / Calculated using refined forest category pixels identified in Existing Vegetation Layer obtained from LANDFIRE (2009).
Distance to edge / m / Estimate of average distance from forest edge, derived using overstory canopy cover.
NDVI / — / Estimate of normalized difference vegetation index calculated for leaf on (Sept.) and leaf-off (Oct.) periods in 2006. Derived using Landsat Thematic Mapper imagery obtained from the U.S. Geological Survey. See equation in Appendix B.
Delta NDVI / — / Estimate of change in NDVI calculated for leaf on (Sept.) and leaf-off (Oct.) periods in 2006. Derived using Landsat Thematic Mapper imagery obtained from the U.S. Geological Survey.
Elevation / m / Estimate of average elevation derived using National Elevation Dataset (NED), obtained from LANDFIRE (2007a).
Slope / deg. / Estimate of average slope, obtained from LANDFIRE (2007b).
Aspect / deg. / Cosine transformed aspect 1-cos[(aspect-30)×(π/180)]/2. Original aspect layer obtained from LANDFIRE (2007c).
Northeast / — / Sum of all cosine transformed aspect values, scaled between 0-1, where 45 degrees (northeast) was set to maximum value of 1.
Topographic position index / — / Four-category topographic position: 1=canyon bottom, 2=gentle slope, 3=steep slope, 4=ridgeline. Derived using NED and approach of Dickson and Beier (2007).
Density of primary and secondary roads / km/km2 / Primary and secondary roads determined using 2007 US Census Bureau TIGER/Line File and U.S. Forest Service road data, obtained from U.S. Census Bureau and Kaibab and Coconino National Forests, respectively.
Distance to primary roads / m / Primary roads determined using 2007 US Census Bureau TIGER/Line File and U.S. Forest Service road data, obtained from U.S Census Bureau and Kaibab and Coconino National Forests, respectively.
aAll forest structure variables were derived using predictors variables described in Appendix B. Detailed methods presented in Hampton et al. (2011).
References
Dickson BG, Beier P (2007) Quantifying the influence of topographic position on cougar (Puma concolor) movement in southern California, USA. Journal of Zoology 271:270–277.
Hampton HM, Sesnie SE, Bailey JD, Snider GB (2011) Estimating regional wood supply based on stakeholder consensus for forest restoration in northern Arizona. Journal of Forestry 109:15–26.
LANDFIRE. 2007a. Digital Elevation Model, v 1.0.0. U.S. Department of Agriculture, Forest Service; U.S. Department of the Interior, Washington D.C. (accessed May 2007).
LANDFIRE. 2007b. Slope, v 1.0.0. U.S. Department of Agriculture, Forest Service; U.S. Department of the Interior, Washington D.C. (accessed May 2007).
LANDFIRE. 2007c. Aspect, v 1.0.0. U.S. Department of Agriculture, Forest Service; U.S. Department of the Interior, Washington D.C. (accessed May 2007).
LANDFIRE. 2009. Existing Vegetation Layer, v 1.0.1. U.S. Department of Agriculture, Forest Service; U.S. Department of the Interior, Washington D.C. (accessed January 2009).
Scott JH, Reinhardt ED (2005) Stereo photo guide for estimating canopy fuel characteristics in conifer stands. USDA Forest Service, Rocky Mountain Research Station, General Technical Report RMRS-GTR-145. Fort Collins, CO, USA. 49 p.
Appendix B. Spectral and spatial predictor variables used to model forest structure in the study area. Variables were developed from 30-m resolution multi-date Landsat Thematic Mapper (TM) imagery and National Elevation Data (NED).Predictor variable / Equations and units / Ancillary information
Landsat TM
TM1 / 0.45-0.52 μm / Leaf on/Leaf offa
TM2 / 0.52-0.60 μm / Leaf on/Leaf offa
TM3 / 0.63-0.69 μm / Leaf on/Leaf offa
TM4 / 0.76-0.90 μm / Leaf on/Leaf offa
TM5 / 1.55-1.75 μm / Leaf on/Leaf offa
TM7 / 2.08-2.35 μm / Leaf on/Leaf offa
Brightness / Tasseled cap / Leaf on/Leaf offa
Greenness / Tasseled cap / Leaf on/Leaf offa
Wetness / Tasseled cap / Leaf on/Leaf offa
MNF1b / 1st Minimum noise fraction (TM bands) / Leaf on/Leaf offa
MNF2b / 2nd Minimum noise fraction(TM bands) / Leaf on/Leaf offa
NDVI / TM4 - TM3/TM4 + TM3 / Leaf on/Leaf offa
NDVIc / NDVI × [1-(TM5-TM5min)/(TM5max-TM5min)] / Leaf on/Leaf offa
NDVIc4 / NDVIc/TM4 / Leaf on/Leaf offa
NDVI ratio / (NDVIc on)/(NDVIc off) / -
NED
Elevation / m / -
Slope / degrees / -
Aspect (Cosine transform) / 1 - cos[(aspect - 30) × (π / 180)]/2 / 0.0 - 1.0
Roughness / Square root (standard deviation of elevation) / -
a Two separate predictor variables are included in forest structure models, one for a leaf on TM image (September) and one for a leaf off TM image (October).[JMR1]
b Minimum noise fraction rotation (two cascaded principal components analysis) of TM bands was used as a data and noise reduction technique (Green et al. 1988).
References
Green AA, Berman M, Switzer P, Craig MD (1988) A transformation for ordering multispectral data in terms of image quality with implication for noise removal. IEEE Transactions on Geoscience and Remote Sensing 26:65–74.
[JMR1]Add 2006?