Table 2 Summary of the SORD metrics.

From: Forest fires and climate-induced tree range shifts in the western US

 

Schoener’s D

Centroid distance

Difference in centroid distance (\({{{{{{\rm{CD}}}}}}}_{{{{{{\rm{B}}}}}}}-{{{{{{\rm{CD}}}}}}}_{{{{{{\rm{U}}}}}}}\))

 

Burned

Unburned

Burned

Unburned

Mean

95% CI

Chrysolepis chrysophyllaR

0.518*

0.641*

0.592

0.677*

−0.148

(−0.795, 0.176)

Picea engelmannii

0.369*

0.751*

0.508†

0.375*

0.132

(−0.246, 0.406)

Pinus albicaulis

0.320

0.768*

0.294

0.234*

0.0600

(−0.416, 0.204)

Pinus contortaˢ

0.534

0.692*

0.308

0.314*

−0.00600

(−0.356, 0.201)

Pinus ponderosa

0.498

0.680*

0.616*

0.198*

0.418

(−0.0870, 0.724)

Pseudotsuga menziesii

0.421*

0.704*

1.095*

0.592*

0.504*

(0.012, 0.964)

Quercus chrysolepisR

0.688*

0.852*

0.714*

0.282*

0.432*

(0.109, 0.724)

Quercus kelloggiiR

0.574

0.825*

0.641†

0.194*

0.447

(−0.242, 0.916)

  1. The sample sizes used to produce these metrics are found in Table 1. Values appended with * or † indicate statistically significant evidence of SORD (p < 0.05 and p < 0.1, respectively), where the null hypothesis of no SORD corresponds to Schoener’s D  = 1 and centroid distance = 0 (Chrysolepis chrysophylla: pSchoener’s D,Burned = 0.0199, pSchoener’s D,Unburned = 0.00398, pCentroid Distance,Burned = 0.200, pCentroid Distance,Unburned < 2.20e−16; Picea engelmannii: pSchoener’s D,Burned = 0.00398, pSchoener’s D,Unburned = 0.00398, pCentroid Distance,Burned = 0.0864, pCentroid Distance,Unburned < 2.20e−16; Pinus albicaulis: pSchoener’s D,Burned = 0.462, pSchoener’s D,Unburned = 0.00398, pCentroid Distance,Burned = 0.372, pCentroid Distance,Unburned = 6.66e−16; Pinus contorta: pSchoener’s D,Burned = 0.251, pSchoener’s D,Unburned = 0.00398, pCentroid Distance,Burned = 0.251, pCentroid Distance,Unburned < 2.20e−16; Pinus ponderosa: pSchoener’s D,Burned = 0.131, pSchoener’s D,Unburned = 0.00398, pCentroid Distance,Burned = 1.13e−4, pCentroid Distance,Unburned = 9.10e−10; Pseudotsuga menziesii: pSchoener’s D,Burned = 0.00398, pSchoener’s D,Unburned = 0.00398, pCentroid Distance,Burned = 6.42e−4, pCentroid Distance,Unburned < 2.20e−16; Quercus chrysolepis: pSchoener’s D,Burned = 0.00398, pSchoener’s D,Unburned = 0.00398, pCentroid Distance,Burned = 9.08e−6, pCentroid Distance,Unburned = 1.65e−7; Quercus kelloggii: pSchoener’s D,Burned = 0.725, pSchoener’s D,Unburned = 0.0159, pCentroid Distance,Burned = 0.00847, pCentroid Distance,Unburned = 0.00156). Lower values of Schoener’s D and greater Centroid Distance suggest greater SORD. CDB–CDU is the result of subtracting the Centroid Distance in unburned samples by the Centroid Distance in burned samples, where the null hypothesis is that CDB–CDU = 0 and wildfire occurrence does not impact SORD. Statistical significance for this metric was calculated using a two-sided bootstrap test. Superscripts appended to the names of species indicate adaptations that facilitate post-fire regeneration, where R denotes resprouting capabilities and ˢ denotes serotiny. See the discussion section for consideration on post-fire regeneration adaptations and observed SORD.
  2. Schoener’s D and centroid distance were tested for statistical significance using the one-sided niche equivalency test and two-sided Hotelling’s T2 test, respectively.
Back to article page