FIG. A1. Bivariate event analysis of the temporal association of extreme climate events (C) before and during randomly generated fire events (F, n = 22). Compare mostly nonsignificant results here to significant trends in Fig. 4 based on 22 fire events from subalpine forests in western Colorado. Years of significant synchrony are highlighted with gray shading. L^CF(t) values (black lines) above the upper confidence limit (dotted lines) indicate synchrony between the two series of events (i.e., extreme climate events occur more often than expected t years before fire events), values below a lower confidence limit indicate asynchrony (i.e., extreme climate events occur less often than expected t years before fire events), and values between indicate independence. 95% confidence envelopes (dotted lines) are based on 1000 Monte-Carlo simulations. See Methods: Fire-climate analysis for further explanation.

   FIG. A2. Bivariate event analysis of the temporal association of extreme climate events (C) before and during randomly generated fire events (F, n = 22). Compare relatively nonsignificant results here to significant trends in Fig. 6 based on 22 fire events from subalpine forests in western Colorado. Years of significant synchrony are highlighted with gray shading. L^CF(t) values (black lines) above the upper confidence limit (dotted lines) indicate synchrony between the two series of events (i.e., extreme climate events occur more often than expected t years before fire events), values below a lower confidence limit indicate asynchrony (i.e., extreme climate events occur less often than expected t years before fire events), and values between indicate independence. 95% confidence envelopes (dotted lines) are based on 1000 Monte-Carlo simulations. See Methods: Fire-climate analysis for further explanation.