For complete description of study design and model simulations, please see Turner et
al. (2021). Briefly, we simulated forest dynamics in five study landscapes in the
Greater Yellowstone Ecosystem (GYE) that were chose to represent dominant forest types
and environmental gradients of the Northern Rockies. Collectively, they encompass nearly
300,000 ha, of which 279,488 ha are potentially stockable with trees. We used iLand
(Seidl et al. 2012, 2019), an individual-based forest model that we previously adapted
for the GYE (Braziunas et al. 2018, 2021; Hansen et al. 2018, 2020; Turner et al. 2019).
We represented environmental heterogeneity within landscapes at 1-ha resolution and
assumed homogenous soils within each grid cell. Growth, mortality and competition among
trees > 4 m in height are simulated at the level of individual trees as a function of
daily radiation, canopy light interception, temperature, soil water, atmospheric CO2
concentration, and nutrients. These drivers affect canopy carbon uptake, which is
modified by species-specific tolerances for temperature extremes, drought stress,
shading, and nutrient availability. iLand also simulates tree regeneration (spatial
resolution: 2-m × 2-m cells) based on seed production, dispersal and environmental
controls on seedling establishment and sapling growth. Postfire tree regeneration
depends on species reproductive traits, age of trees that burned (which determines the
size of the canopy seed bank for serotinous species), distance to live seed sources, and
soil moisture in subsequent growing seasons (Hansen et al. 2018). Seedlings and saplings
are modeled as height cohorts in 2-m × 2-m cells until they reach a height of 4 m. Full
documentation of iLand as well as the model source code can be found at
http://iland-model.org/.
We selected plausible but contrasting future climate scenarios that differed in
three ways relevant for fire and forest dynamics: whether precipitation increased with
warming, how the timing and intensity of drought varied, and whether anthropogenic C
emissions continued unabated. We chose three GCMs (CanESM2, HadGEM2-CC and HadGEM2-ES)
and two representative concentration pathways (RCP 4.5 and 8.5) from the Coupled Model
Intercomparison Project 5 (CMIP5) that provided these contrasts.
To model fire, we developed new statistical models that predict the timing, location
and maximum potential size of fires based on climate, then integrated them with the
dynamic fire module in iLand to spread fire spatially in response to fuels, topography
and weather. For each climate scenario, statistical climate-fire models were used to
generate 20 iterations for the locations, timing, and maximum potential size (all burned
severities) of large fires (≥ 400 ha) across the GYE from 2017 to 2099. Each of the fire
iterations per climate scenario provided input to iLand’s fire-spread algorithm.
In sum, we simulated 20 iterations of future fire and forest trajectories for each
GCM x RCP x landscape (n = 120 per landscape, but 599 simulations in total because one
iteration from the sgye landscape under HadGEM2-ES is missing) from 2017 to 2100. Forest
attributes were tracked annually for every iteration on every landscape.
Cited references:
Braziunas, K.H., W.D. Hansen, R. Seidl, R. W. Rammer, and M.G. Turner. 2018. Looking
beyond the mean: Drivers of variability in postfire stand development of conifers in
Greater Yellowstone. Forest Ecology and Management 430:460-471.
Braziunas, K. H., R. Seidl, W. Rammer, and M. G. Turner. 2021. Can we manage a
future with more fire? Effectiveness of defensible space treatment depends on housing
amount and configuration. Landscape Ecology 36:309-330.
Hansen, W. D., K. H. Braziunas, W. Rammer, R. Seidl, and M. G. Turner. 2018. It
takes a few to tango: Changing climate and fire regimes can cause regeneration failure
of two subalpine conifers. Ecology 99:966-977.
Hansen, W. D., D. Abendroth, W. Rammer, R. Seidl, and M. G. Turner. 2020. Can
wildland fire management alter 21st-century fire patterns and forests in Grand Teton
National Park? Ecological Applications 30(2):e02030.
Seidl, R., W. Rammer, R.M. Scheller, and T.A. Spies. 2012. An individual-based
process model to simulate landscape-scale forest ecosystem dynamics. Ecological
Modelling 231:87–100.
Seidl, R., W. Rammer, and T.A. Spies. 2014a\. Disturbance legacies increase the
resilience of forest ecosystem structure, composition, and functioning. Ecological
Applications 24:2063–2077.
Turner, M. G., K. H. Braziunas, W. D. Hansen, and B. J. Harvey. 2019. Short-interval
fire erodes the resilience of subalpine lodgepole pine forests. Proceedings of the
National Academy of Sciences 116:11319-11328.
