Fig. 3: Fire vulnerability and the amplification of fire size on biogeophysical changes after fire, stratified by forest type.
From: Forest fire size amplifies postfire land surface warming
The cascade of biogeophysical drivers of postfire summer (June–August) surface radiometric warming is shown in the middle of the figure. a,e, Fire vulnerability is defined as postfire change in LAI (a) or surface radiometric temperature (ΔΤ) (e) averaged over all fire patches. For all the forest types, the mean value of ΔLAI (or ΔT) is significantly lower (or greater) than zero at α = 0.05 (one-tailed t-test), with negligible s.e.m. Tukey’s honest significant difference test for multiple comparisons shows significant differences in fire vulnerability among forest types (P < 0.05). b–d,f, Multiple linear regression models incorporating log10[fire size], forest type and their interactive effects were fitted to derive the amplification effect of fire size for different forest types (that is, the β value: the slope between postfire biogeophysical changes and log10[fire size]), using the dependent variables of ΔLAI (βΔLAI; b) and ΔET (βΔET; c), and changes in surface albedo (βΔα; d) and surface warming (βΔT; f), all for the summer period one year after fire. The asterisks show significant regressions (P < 0.05); error bars show the standard errors. All the β values show significant differences (P < 0.05, Student’s t-test) among the forest types. Figure developed using Python open-source tools.
