The vision for this carbon risk work began in 2009 with a group of researchers at the US Forest Service Rocky Mountain Research Station’s Missoula Fire Sciences Lab and the Pacific Northwest Research Station. To accomplish this work, we needed a number of components. One of these components was information on fire likelihood and intensity at the scale of the continental US. We collaborated with Pyrologix on calibrating and running the fire simulations, which take a number of months to complete on high performance computing.
We also required a tree-level model of the forests of the US with accompanying information on fuels (litter, duff, and woody debris on the forest floor). This multi-year effort consisted of an innovative use of random forests modeling (a form of artificial intelligence) to convey the complex relationships between existing vegetation, climate, and disturbance that contribute to forest structure. This effort uses nationally consistent information gathered at a network of forest plots across the country by the Forest Service’s Forest Inventory and Analysis program as well as gridded landscape data from the LANDFIRE project. We named the resulting datasets TreeMap and FuelMap.
A third set of models was needed to estimate the amount of carbon that would be emitted as smoke and which trees would die during fires of various flame lengths. We used another Forest Service model called the Forest Vegetation Simulator (link here: https://www.fs.usda.gov/fvs/) for this work.
We needed separate models to estimate biomass and emissions in non-forested, or rangeland, ecosystems. This was accomplished in a new model under development which estimates total fuels of various types (live shrub and herbaceous vegetation biomass, litter, and duff) in non-forested ecosystems. Carbon emissions in rangelands were then simulated in the SpatialFOFEM model.
We also needed expertise in web design and visualization of geospatial datasets, which our collaborators at Spatial Informatics Group provided.
This carbon risk work exists due to a long-term collaboration between researchers in three different research stations within the Forest Service and the private and nonprofit sectors..
