2020 ESA Annual Meeting (August 3 - 6)

COS 201 Abstract - Exploring recovery from fire in low-elevation Wyoming big sagebrush with a satellite-derived chronosequence and Bayesian indirect gradient analysis

Adam Mahood, Geography, University of Colorado-Boulder, Boulder, CO
Background/Question/Methods

Invasion by introduced plants and changing fire regimes are two of the more difficult ecological challenges we face today. In the U.S. Great Basin, vast areas of Wyoming big sagebrush are being invaded by cheatgrass, which has increased fire activity. While many studies show that Wyoming big sagebrush generally recovers very slow or not at all following invasion and fire, other studies have documented a relatively fast recovery (~ 20 years). Here we built a 32 year fire history atlas using two Landsat-derived data sources in northern Nevada of areas where the two products agreed that the fire frequency was one, and created a chronosequence of time since fire. We collected abundance data for every species at 35 plots, and then used Bayesian indirect gradient analysis by Markov chain Monte Carlo to analyze how the community composition was related to time since fire, cheatgrass cover and other environmental variables, diversity indexes and climatic variables around the time of the fire.

Results/Conclusions

The community composition gradient correlated with cheatgrass cover, biogeographic origin, diversity indexes and climatic gradients. Contrary to our expectations, higher time since fire was associated with degraded sites dominated by introduced plants, whereas more recently burned sites had more natives. This suggests that after fire, instead of recovering to the native plant-dominated state that typically exists in unburned plots, our sites were moving in the opposite direction after fire. Climatic water deficit after fire and actual evapotranspiration and minimum temperature before fire were correlated with more degraded sites, and higher minimum temperatures before fire corresponded to less degrated sites. Our results suggest that restoring these degraded landscapes may depend on favorable climatic conditions around the time of the fire.