Fuel Classifications - Wildland Fuel Fundamentals and Applications - page 114

Fig. 7.3 The classification diagram showing the clustering of fire effects groups (e.g., EG1) on
gradients of smoke emissions production and soil heating. These fire effects groups were then
divided into finer groups to create the FLMs. (Lutes et al. 2006)
represent actual fuelbeds with measured loadings. As such, these classifications can
be used as (1) inventory techniques to quantify fuel characteristics (Sikkink et al.
2009 ); (2) descriptors of unique fuel types to facilitate communication between
managers, scientists, and other professionals (Sandberg et al. 2001 ); and (3) map
units in fuel mapping efforts (Keane et al. 2001 ). Effective classified fuel systems
contain dichotomous keys that can uniquely identify a class on the ground based on
qualities of the fuelbed (Sikkink et al. 2009 ). The loading information for a clas-
sified category can be used in fire applications, such as simulating fire effects and
validating fuel maps, and the variability of loadings within a category can be incor-
porated into the analyses. And since statistical classifications have low redundancy
between classes, class attributes may be used for quantifying loading in fire models,
as a field inventory technique (Chap. 7), and for identifying possible thresholds in
fire behavior and effects modeling (Lutes et al. 2009 ).
Directly classified fuel classifications, such as FLMs, also have drawbacks. All
fuel classifications, and especially those developed from direct classification tech-
niques, require extensive data sets to fully represent the diversity of fuelbeds in the
analysis. As a result, the depth, scope, and quality of the data sets used to create
the classification system are rarely comprehensive enough to represent all possible