MASS-DIAMETER REGRESSIONS FOR MOOSE BROWSE ON THE COPPER RIVER DELTA,ALASKA

Regression equations were developed to predict 3 mass components of 7 browse species important to moose (Alces alces gigas) on the Copper Ri ver Delta in southcentral Alaska. The accuracy of model predictions wa s the criterion for model selection. Model accuracy was evaluated usin g data splitting or jackknife procedures. Annual production of twigs a nd leaves and available twig mass on a stem were most accurately predi cted from stem basal diameter with zero intercept models, zero interce pt log-linear models, or log-log models. Twig mass eaten by moose was most accurately predicted from the diameter at the point of browsing o f a twig with zero intercept or full linear models. Heteroskedasticity was significant (P<0.05) in most of the data sets and could not be si gnificantly reduced with log transformations or use of weighted least squares models. Heteroskedasticity appeared to have a relatively minor effect on model predictions. Most of the models gave mean predictions within +/- 20% of the actual values, particularly for the most ubiqui tous species that were also the most important to moose. For each spec ies, there were few differences (P<0.05) in model coefficients between years and among habitat types. Differences in coefficient estimates a ppeared to be related to differences in stem morphology that were rela ted to both site conditions and past browsing by moose.