Biomass Estimation for Four Common Grass Species in Northern Arizona Ponderosa Pine

Abstract

Listen

Translate article

Vegetation allometric relations were examined for 4 important grass species in southwestern ponderosa pine (Pinusponderosa). Logarithmic regressions were developed relating aboveground biomass to basal area, height, and number of seedheads, as well as 3 factors: overstory type (pole, yellow pine), burning treatment (unburned, prescribed burn 2-, 5-, and 7-yr previously), and site (3 locations). Basal area was defined as longest basal diameter multiplied by the widest perpendicular diameter. Of the metric variables, basal area proved to be the best predictor of biomass. Height and number of seedheads did little to increase Rl values. Burning treatment was a significant factor for Sitanion longejolium and Muhlenbe&a mon&na. Overstory type significantly affected Poa fenderiana and Festuca ankonica equations. Site effects were important for all but SuMion lonsefolfum When biomass regressions are used for species such as these, sampling efficiency can be improved by including factors such as overstory type, burning history, and locale. Final regression equations relating biomass of each species to basal area and significant factors were significant at 60.05 and had adjusted Rz values ranging from 0.81 to 0.87. A validation test using 20% of the data not used in developing the regressions indicated that these equations are adequate predictors. When used with double sampling, weight prediction based on basal area indices should provide a more objectively measured predictor than percent cover. Allometric relationships are often used to estimate understory biomass (e.g., Brown and Marsden 1976, Ohmann et al. 1981, Olson and Martin 1981). When compared with harvesting, the application of regression equations based on these relationships can result in lower sampling costs (or higher precision) due to relatively greater sampling efficiency (Reese et al. 1980). Most understory biomass equations are developed using fairly subjective ocular estimates of percent cover as the independent variable. Ocular estimates have the disadvantage of varying between observers as well as over time for a single observer, introducing unknown, but potentially substantial, error into the biomass estimates. We felt that using some combination of basal area, plant height, or number of seedheads might overcome these problems. A recent study of the effect of prescribed burning on understory production in northern Arizona (Andariese 1982), provided us with the opportunity to test this technique for a ponderosa pinebunchgrass type and to determine if overstory conditions, site, or burning treatment significantly alter the allometric relationship.