The effects of uncertainty in individual tree volume model predictions on large area estimates of forest volume

Abstract

Greenhouse gas inventories for the forestry sector rely on large area eslimates of tree biomass. These estimates are calculated by predicting volumes of individual trees, multiplying by a biomass conversion factor; adding individual tree biomass estimates at the plot level, and then averaging over plots to obtain the large area estimates. However, the uncertainty in the model prediction is generally ignored with the result that the precision of the large area estimates is over-estimated. The study objetive was to estimate the effects of model-related uncertainties on large area volumes estimates for study areas in the state of Minnesota, USA, and the state of Santa Catarina, Brazil. Monte Carlo simulation approaches were used because of the complexities associated with multiple sources of uncertainty and the nonlinear nature of the models. The effects of uncertainty in model predictions on the large area volumes estimates were small, although the results depend heavily on sample size and quality of fit models.

Citation

​McRoberts, Ronald E. 2014. The effects of uncertainty in individual tree volume model predictions on large area estimates of forest volume. In: Shortridge, Ashton M.; Messina, Joesph P.; Finley, Andrew; Kravchenko, Sasha, eds. Accuracy 2014; Proceedings of the 11th international symposium on spatial accuracy assessment in natural resources and environmental sciences; 2014 July 8-11; East Lansing, MI. East Lansing, MI: Michigan State University: 137-140.