Abstract: | Predictive models of aboveground biomass of nonnative Tamarix ramosissima of various sizes were developed using destructive sampling techniques on 50 individuals and four 100- m 2 plots. Each sample was measured for average height (m) of stems and canopy area (m 2 ) prior to cutting, drying, and weighing. Five competing regression models ( P T. ramosissima using average height and/or canopy area measurements and were evaluated using Akaike's Information Criterion corrected for small sample size (AIC c ). Our best model (AIC c = –148.69, ΔAIC c = 0) successfully predicted T. ramosissima aboveground biomass (R 2 = 0.97) and used average height and canopy area as predictors. Our 2nd-best model, using the same predictors, was also successful in predicting aboveground biomass (R 2 = 0.97, AIC c = –131.71, ΔAIC c = 16.98). A 3rd model demonstrated high correlation between only aboveground biomass and canopy area (R 2 = 0.95), while 2 additional models found high correlations between aboveground biomass and average height measurements only (R 2 = 0.90 and 0.70, respectively). These models illustrate how simple field measurements, such as height and canopy area, can be used in allometric relationships to accurately predict aboveground biomass of T. ramosissima . Although a correction factor may be necessary for predictions at larger scales, the models presented will prove useful for many research and management initiatives. |