Hardwood understories can contribute significantly to total ecosystem biomass and fuel loads, but few models are available to directly quantify this component. In part, this is due to the small size of the hardwoods. Many understory trees simply do not reach the height required to determine diameter at breast height (d.b.h.), so conventional models (e.g., the National Biomass Estimators [NBE]) that rely on this predictor are unavailable. Further, understory hardwoods can be present in such numbers or have inconvenient growth forms such that biomass estimates based on diameters are impractical. However, a quick and easily measured attribute, stem length, can be used instead of diameter to facilitate understory hardwood biomass estimation. We destructively sampled 513 small hardwood shrubs and trees in Arkansas, Louisiana, and Mississippi and oven dried their aboveground live biomass (stems, branches, leaves) to a constant weight. The high degree of variability in plant form, branch patterning, and wood density among the 31 different taxa sampled suggested that a single hardwood grouping would be as effective as more specific equations. Nonlinear ordinary least squares regression was then used to predict aboveground live biomass with a modified version of the NBE (using stem length rather than d.b.h.). The coefficient of determination of the resulting model was reasonably high (R2
= 0.71), particularly for data comprising such varied individuals. Further confirmation of the utility of this understory biomass model followed a comparison of several species with varying wood density.
Bragg, Don C.; Scott, D. Andrew. 2014. A preliminary aboveground live biomass model for understory hardwoods from Arkansas, Louisiana, and Mississippi. In: Groninger, John W.; Holzmueller, Eric J.; Nielsen, Clayton K.; Dey, Daniel C., eds. Proceedings, 19th Central Hardwood Forest Conference; 2014 March 10-12; Carbondale, IL. General Technical Report NRS-P-142. Newtown Square, PA: U.S. Department of Agriculture, Forest Service, Northern Research Station: 251-260.