Spatial variability of wildland fuel characteristics in northern Rocky Mountain ecosystems

Abstract

We investigated the spatial variability of a number of wildland fuel characteristics for the major fuel components found in six common northern Rocky Mountain ecosystems. Surface fuel characteristics of loading, particle density, bulk density, and mineral content were measured for eight fuel components-four downed dead woody fuel size classes (1, 10, 100, 1000 hr), duff, litter, shrub, and herb-on nested plots located within sampling grids to describe their variability across spatial scales. We also sampled canopy bulk density, biomass, and cover for each plot in the grid. The spatial distribution and variability of surface and canopy fuel characteristics are described using the variance, spatial autocorrelation, semi-variograms, and Moran's I. We found that all fuels had high variability in loading (two to three times the mean), and this variability increased with the size of fuel particle. We also found that fuel components varied at different scales, with fine fuels varying at scales of 1 to 5 m, coarse fuels at 10 to 150 m, and canopy fuels at 100 to 500 m. Findings and data from this study can be used to sample, describe, and map fuel characteristics, such as loading, at the appropriate spatial scales to accommodate the next generation of fire behavior prediction models.Keywords: wildland fire, biomass, landscape ecology, woody debris, spatial scale, mineral content, particle density, bulk density, loading AuthorsRobert E. Keane is a Research Ecologist with the U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station at the Missoula Fire Sciences Laboratory in Montana. Since 1985, Keane has developed various ecological computer models for research and management applications. His recent research includes (1) developing ecological computer models for exploring landscape, fire, and climate dynamics; (2) describing, classifying, and mapping fuel characteristics; (3) investigating the ecology and restoration of whitebark pine; and (4) conducting fundamental wildland fuel science. He

Bayesian Techniques for Surface Fuel Loading Estimation

“…Their study entailed creating fuelbeds from fuels collected in the field of four known woody fuel loadings (.05, .10, .15, and .20 kg m -2 ) and distributing these fuels over a 20 x 25 m plot. Trained technicians estimated loadings for the fine woody fuels using the photoload (Keane & Dickinson 2007), fixedarea (Keane, Gray, & Bacciu, 2012) and planar intercept (Brown 1974) methods. The photoload method uses calibrated photos of known loadings pointing toward the forest floor to visually estimate fuel loadings (Keane & Dickinson, 2007).…”

“…It is the most recently developed method and the simplest to implement, however, it relies heavily on proper training and is consequently subject to human error. The fixed-area method uses a 1 x 1 m plot to define a sample frame with a fixed area and the dimensions (length, diameter) of all fuels within the plot boundary are measured to calculate volume that is then multiplied by field-estimated particle densities to estimate fuel loadings (Keane et al, 2012). The fixed-area plots are more time consuming and expensive than other methods (Keane & Gray, 2013).…”

Smoke consequences of new wildfire regimes driven by climate change

“…Each of these spatial layers has strengths and weaknesses [Keane et al, 2013b], but all share an overarching limitation, in that as coarse-scale data layers they cannot replicate fuels exactly for particular points 10.1002/2013EF000180 on a landscape, because of the scaling issue noted earlier [Keane et al, 2012a[Keane et al, , 2012b. This scale mismatch needs to be acknowledged in future projections of smoke.…”

“…Much of the fuel that contributes to smoke production comes from dead surface fuels, particularly duff and coarse wood, whose consumption mainly occurs in the smoldering phase Prichard et al, 2007]. Typically, loadings of these fuels cannot be inferred from live vegetation [Keane et al, 2012a]; attempts to establish predictive relationships have largely failed [Brown and Bevins, 1986;Raymond et al, 2006;Keane et al, 2012a], and this problem is magnified when species are not known, as in DGVMs. Moreover, different classes of dead fuel loadings are rarely correlated because each has unique decomposition and deposition rates, meaning that each class must be modeled, or derived heuristically, independently from the others [Keane et al, 2012a].…”

“…Estimates of variation in fire severity, in particular, are critical both for quantifying the timing and amount of smoke produced by combustion of both surface and canopy fuels [Keane et al, 2012a[Keane et al, , 2012b and for estimating the fire-produced energy that lofts smoke into the atmosphere where it can be transported downwind. In forests in particular, species composition introduces further variability because tree-species' adaptations to fire vary widely [Agee, 1993].…”