COS 80-10 - Enhanced productivity versus increased fire: What will constrain carbon stocks in a warmer future in the Greater Yellowstone Ecosystem?

Wednesday, August 9, 2017: 11:10 AM
B112, Oregon Convention Center
Paul D. Henne, Geosciences and Environmental Change Science Center, US Geological Survey, Denver, CO and Todd J. Hawbaker, Geosciences and Environmental Change Science Center, U.S. Geological Survey, Denver, CO
Background/Question/Methods

Ecosystem responses to global change are complex and varied, which can result in contradictory projections of changes in ecosystem services. In subalpine forests of the American West, rising temperatures can enhance productivity, increasing carbon sequestration. However, warming also increases flammability, which can shorten fire rotations and eliminate forests. Therefore, it remains uncertain if subalpine forests will continue to sequester carbon in a warmer future. Resolving this question requires an approach that integrates the effects of climate-driven changes in productivity and wildfire across large landscapes. We used the Landis-II model to simulate dynamics among climate, vegetation, and fire in the Greater Yellowstone Ecosystem from 1985-2100 under the RCP 8.5 emissions scenario. To simulate changing fire regimes, we developed a new approach to relate aridity to area burned in the Intermountain West. We calculated aridity indices (1-AET/PET) and area burned for 13 ecoregions, for 1980-2015, then used regression trees to identify an aridity threshold for each ecoregion that distinguishes large from small fire years. Fire sizes from large and small fire years fit distinct log-normal distributions. We determined if each simulation year falls above or below the aridity threshold, then drew fire sizes from the corresponding fire-size distribution to simulate annual fire.

Results/Conclusions

The relationship between aridity and area burned is not linear. Instead, area burned increases exponentially beyond an aridity threshold. The position of the threshold is strongly correlated with mean annual aridity (r2 = 0.96). Whereas productive ecoregions, with typically moist growing seasons, have low aridity thresholds, arid ecoregions require much drier conditions to support large fires. Simulated area burned increases in our future projections, especially after 2050, when nearly all years in our climate scenario exceed the aridity threshold. However, productivity gains in a warming climate partially compensate for biomass lost to fire. Simulated net primary productivity increases until 2060 as temperatures rise, but returns to 1985 levels by 2100 as forested area declines due to more frequent fires. Species level changes help maintain productivity in sites where water becomes limiting. Drought tolerant Pseuodotsuga menziesii expands to occupy ~25% more area. Pinus contorta occupies less total area, but a greater proportion of remaining forests, and Picea engelmannii and Abies lasiocarpa significantly decline. Although fire and climate change will alter species distributions and forest structure, our results suggest that most forest types will persist, and the Greater Yellowstone Ecosystem will remain a carbon sink in a warmer future.