FIGURE LEGENDS
FIGURE 1. Map of Chiricahua National Monument, Cochise County, Arizona, USA. Vegetation plot locations are indicated with black dots.
FIGURE 2. Differences for woody plant species between pre-fire and post-fire sample periods in Chiricahua National Monument for (a) gamma diversity, showing rarefaction accumulation curve with estimate and 95% confidence internals, (b) mean (with 95% ci) for alpha diversity (species richness, species evenness, and inverse Simpson index, and (c) mean (with 95% ci) for beta diversity (total, species turnover, and nestedness, using Sorensen dissimilarity). Statistical significance: *P <0.01, **P <0.01, ***P <0.001.
FIGURE 3. Changes for woody plant species in Chiricahua National Monument from the pre-fire to the post-fire sample periods for 138 plots across three vegetation types for (a) gamma diversity (from rarefaction with 95% ci), (b) mean (with 95% ci) alpha diversity metrics (species richness, species evenness, and the inverse Simpson index), (c) mean (with 95% ci) beta diversity expressed as all plot pairwise dissimilarities (Sorensen) for pre-fire and post-fire species presence, and (d) beta diversity expressed as pre-fire/post-fire dissimilarity for each plot independent of other plots. Asterisks (*P <0.05, **P <0.01, ***P <0.001) show pre-fire/post-fire differences from paired t-tests. For (d), P -value is for different among vegetation types, and bars with different letters are significantly different from each other (P <0.05).
FIGURE 4. Changes for woody plant species in Chiricahua National Monument from the pre-fire to the post-fire sample periods for 138 plots across levels of fire severity for (a) gamma diversity (from rarefaction with 95% ci), (b) mean (with 95% ci) for alpha diversity metrics (species richness, species evenness, and the inverse Simpson index), (c) mean (with 95% ci) for beta diversity expressed as all plot pairwise dissimilarities (Sorensen) for pre-fire and post-fire species presence, and (d) beta diversity expressed as pre-fire/post-fire dissimilarity for each plot independent of other plots. Asterisks (*P <0.05, **P <0.01, ***P <0.001) show pre-fire/post-fire differences from paired t-tests. For (d), P -values are for comparisons of plots that burned vs. those that did not burn, and bars with different letters are significantly different from each other (P <0.05).
FIGURE 5. Relationship of alpha diversity metrics (species richness, species evenness, and inverse Simpson index) with (a) fire severity (dNBR), (b) elevation, (c) topographic relative moisture index (TRMI: xeric to mesic), and (D) terrain ruggedness (TRI: less to more rugged). Values for each plot for pre-fire and post-fire are shown for all graphs. Best fit lines (with 95% ci) are shown only for relationships that are P <0.05, all of which were significant for second order polynomial regression. Full statistical model results shown in Table 1.
FIGURE 6. (a) Sorenson dissimilarity and (b) the difference in the number of species for each plot from before to after the fire in relation to dNBR. Best fit lines (with 95% ci) are shown only for relationships that are P <0.10, either linear or second order polynomial.
FIGURE 7. Relationship for woody plant species in Chiricahua National Monument of total beta diversity with fire severity (dNBR), elevation, topographic relative moisture index (TRMI: xeric to mesic), and terrain ruggedness (TRI: less to more rugged). (a) all-plot pairwise dissimilarities for pre-fire and post-fire species presence and (b) pre-fire/post-fire dissimilarity for each plot independent of other plots. Best-fit lines (linear or polynomial depending on final model) are shown with 95% ci for relationships that wereP <0.10 found with adonis tests for (a) and with linear models for (b). Full statistical model results shown in Table 2.