More Like this

1. Predicting snag fall in an old-growth forest after fire

   https://doi.org/10.1186/s42408-023-00225-z  

   Becker, Kendall M. L. ; Lutz, James A. ( November 2023 , Fire Ecology)

   Snags, standing dead trees, are becoming more abundant in forests as tree mortality rates continue to increase due to fire, drought, and bark beetles. Snags provide habitat for birds and small mammals, and when they fall to the ground, the resulting logs provide additional wildlife habitat and affect nutrient cycling, fuel loads, and fire behavior. Predicting how long snags will remain standing after fire is essential for managing habitat, understanding chemical cycling in forests, and modeling forest succession and fuels. Few studies, however, have quantified how fire changes snag fall dynamics.

   We compared post-fire fall rates of snags that existed pre-fire (n= 2013) and snags created during or after the fire (n= 8222), using 3 years of pre-fire and 5 years of post-fire data from an annually monitored, 25.6-ha spatially explicit plot in an old-growthAbies concolor–Pinus lambertianaforest in the Sierra Nevada, CA, USA. The plot burned at low to moderate severity in the Rim Fire of 2013. We used random forest models to (1) identify predictors of post-fire snag fall for pre-existing and new snags and (2) assess the influence of spatial neighborhood and local fire severity on snag fall after fire. Fall rates of pre-existing snags increased 3 years after fire. Five years after fire, pre-existing snags were twice as likely to fall as new snags. Pre-existing snags were most likely to persist 5 years after fire if they were > 50 cm in diameter, > 20 m tall, and charred on the bole to heights above 3.7 m. New snags were also more likely to persist 5 years after fire if they were > 20 m tall. Spatial neighborhood (e.g., tree density) and local fire severity (e.g., fire-caused crown injury) within 15 m of each snag barely improved predictions of snag fall after fire.

   Land managers should expect fall rates of pre-existing snags to exceed fall rates of new snags within 5 years after fire, an important habitat consideration because pre-existing snags represent a wider range of size and decay classes.

    

   more » « less
2. Fixing a snag in carbon emissions estimates from wildfires

   https://doi.org/10.1111/gcb.14716  

   Stenzel, Jeffrey E. ; Bartowitz, Kristina J. ; Hartman, Melannie D. ; Lutz, James A. ; Kolden, Crystal A. ; Smith, Alistair M. S. ; Law, Beverly E. ; Swanson, Mark E. ; Larson, Andrew J. ; Parton, William J. ; et al ( July 2019 , Global Change Biology)

   Wildfire is an essential earth‐system process, impacting ecosystem processes and the carbon cycle. Forest fires are becoming more frequent and severe, yet gaps exist in the modeling of fire on vegetation and carbon dynamics. Strategies for reducing carbon dioxide (CO₂) emissions from wildfires include increasing tree harvest, largely based on the public assumption that fires burn live forests to the ground, despite observations indicating that less than 5% of mature tree biomass is actually consumed. This misconception is also reflected though excessive combustion of live trees in models. Here, we show that regional emissions estimates using widely implemented combustion coefficients are 59%–83% higher than emissions based on field observations. Using unique field datasets from before and after wildfires and an improved ecosystem model, we provide strong evidence that these large overestimates can be reduced by using realistic biomass combustion factors and by accurately quantifying biomass in standing dead trees that decompose over decades to centuries after fire (“snags”). Most model development focuses on area burned; our results reveal that accurately representing combustion is also essential for quantifying fire impacts on ecosystems. Using our improvements, we find that western US forest fires have emitted 851 ± 228 Tg CO₂(~half of alternative estimates) over the last 17 years, which is minor compared to 16,200 Tg CO₂from fossil fuels across the region.

    

   more » « less
3. SMLFire1.0: a stochastic machine learning (SML) model for wildfire activity in the western United States

   https://doi.org/10.5194/gmd-16-3407-2023  

   Buch, Jatan ; Williams, A. Park ; Juang, Caroline S. ; Hansen, Winslow D. ; Gentine, Pierre ( January 2023 , Geoscientific Model Development)

   Abstract. The annual area burned due to wildfires in the western United States (WUS) increased bymore than 300 % between 1984 and 2020. However, accounting for the nonlinear, spatially heterogeneous interactions between climate, vegetation, and human predictors driving the trends in fire frequency and sizes at different spatial scales remains a challenging problem for statistical fire models. Here we introduce a novel stochastic machine learning (SML) framework, SMLFire1.0, to model observed fire frequencies and sizes in 12 km × 12 km grid cells across the WUS. This framework is implemented using mixture density networks trained on a wide suite of input predictors. The modeled WUS fire frequency matches observations at both monthly (r=0.94) and annual (r=0.85) timescales, as do the monthly (r=0.90) and annual (r=0.88) area burned. Moreover, the modeled annual time series of both fire variables exhibit strong correlations (r≥0.6) with observations in 16 out of 18 ecoregions. Our ML model captures the interannual variability and the distinct multidecade increases in annual area burned for both forested and non-forested ecoregions. Evaluating predictor importance with Shapley additive explanations, we find that fire-month vapor pressure deficit (VPD) is the dominant driver of fire frequencies and sizes across the WUS, followed by 1000 h dead fuel moisture (FM1000), total monthly precipitation (Prec), mean daily maximum temperature (Tmax), and fraction of grassland cover in a grid cell. Our findings serve as a promising use case of ML techniques for wildfire prediction in particular and extreme event modeling more broadly. They also highlight the power of ML-driven parameterizations for potential implementation in fire modules of dynamic global vegetation models (DGVMs) and earth system models (ESMs). 

   more » « less
4. Coupled fire-atmosphere-smoke forecasting: current capabilities and plans for the future

   https://doi.org/10.14195/978-989-26-16-506_104  

   Kochanski, Adam K. ; Mandel, Jan ; Vejmelka, Martin ; Burke, Dalton ; Hearn, Lauren ; Haley, James ; Farguell Caus, Angel ; Schranz, Sher ( August 2018 , Advances in Forest Fire Research 2018)

   Viegas, Domingos Xavier (Ed.)

   In this paper, we present an integrated wildland fire forecasting system based on combining a high resolution, multi-scale weather forecasting model, with a semi-empirical fire spread model and a prognostic dead fuel moisture model. The fire-released heat and moisture impact local meteorology which in turn drives the fire propagation and the dead fuel moisture. The prognostic dead fuel moisture model renders the diurnal and spatial fuel moisture variability. The local wind and the fuel moisture variation drive the fire propagation over the landscape. The sub-kilometer model resolution enables detailed representation of complex terrain and small-scale variability in surface properties. The fuel moisture model assimilates surface observations of the 10h fuel moisture from Remote Automated Weather Stations (RAWS) and generates spatial fuel moisture maps used for the fire spread computations. The dead fuel moisture is traced in three different fuel classes (1h, 10h and 100h fuel), which are integrated at any given location based on the local fuel description, to provide the total dead fuel moisture content at the fire-model grid, of a typical resolution of tens of meters. The fire simulations are initialized by a web-based control system allowing a user to define the fire anywhere in CONUS as well as basic simulation properties, such as simulation length, resolution, and type of meteorological forcing for any time meteorological products are available to initialize the weather model. The data is downloaded automatically, and the system monitors execution on a cluster. The simulation results are processed while the model is running and displayed as animations on a dedicated visualization portal. 

   more » « less
5. Simulating dynamic fire regime and vegetation change in a warming Siberia

   https://doi.org/10.1186/s42408-023-00188-1  

   Williams, Neil G. ; Lucash, Melissa S. ; Ouellette, Marc R. ; Brussel, Thomas ; Gustafson, Eric J. ; Weiss, Shelby A. ; Sturtevant, Brian R. ; Schepaschenko, Dmitry G. ; Shvidenko, Anatoly Z. ( May 2023 , Fire Ecology)

   Climate change is expected to increase fire activity across the circumboreal zone, including central Siberia. However, few studies have quantitatively assessed potential changes in fire regime characteristics, or considered possible spatial variation in the magnitude of change. Moreover, while simulations indicate that changes in climate are likely to drive major shifts in Siberian vegetation, knowledge of future forest dynamics under the joint influence of changes in climate and fire regimes remains largely theoretical. We used the forest landscape model, LANDIS-II, with PnET-Succession and the BFOLDS fire extension to simulate changes in vegetation and fire regime characteristics under four alternative climate scenarios in three 10,000-km²study landscapes distributed across a large latitudinal gradient in lowland central Siberia. We evaluated vegetation change using the fire life history strategies adopted by forest tree species: fire resisters, fire avoiders, and fire endurers.

   Annual burned area, the number of fires per year, fire size, and fire intensity all increased under climate change. The relative increase in fire activity was greatest in the northernmost study landscape, leading to a reduction in the difference in fire rotation period between study landscapes. Although the number of fires per year increased progressively with the magnitude of climate change, mean fire size peaked under mild or moderate climate warming in each of our study landscapes, suggesting that fuel limitations and past fire perimeters will feed back to reduce individual fire extent under extreme warming, relative to less extreme warming scenarios. In the Southern and Mid-taiga landscapes, we observed a major shift from fire resister-dominated forests to forests dominated by broadleaved deciduous fire endurers (BetulaandPopulusgenera) under moderate and extreme climate warming scenarios, likely associated with the substantial increase in fire activity. These changes were accompanied by a major decrease in average cohort age and total vegetation biomass across the simulation landscapes.

   Our results imply that climate change will greatly increase fire activity and reduce spatial heterogeneity in fire regime characteristics across central Siberia. Potential ecological consequences include a widespread shift toward forests dominated by broadleaved deciduous species that employ a fire endurer strategy to persist in an increasingly fire-prone environment.

    

   more » « less