Toward error analysis of large-scale forest carbon budgets

Quantification of forest carbon sources and sinks is an important part of n ational inventories of net greenhouse gas emissions. Several such forest ca rbon budgets have been constructed, but little effort has been made to anal yse the sources of error and how these errors propagate to determine the ov erall uncertainty of projected carbon fluxes. We performed an error analysi s for estimates of tree volume and volume change determined by repeated mea surements of permanent sample plots for the South-eastern United States as a step toward assessing errors in the carbon budget constructed by the USDA Forest Service. Error components recognized were: sampling error for sampl e plot selection; measurement error for tree height and diameter; and regre ssion error for tree volume. Most of the propagated error in volume and vol ume change estimation was due to sampling error. Error estimates depended o n the size of the area examined (single state to region) and the degree to which tree growth and recruit-ment balanced mortality and harvesting. Appro ximate regional 95% confidence intervals were 3 455 073 000 +/- 39 606 000 (1.1%) m(3) for current growing-stock volume, and 10 616 000 +/- 4210 000 ( 39.7%) m(3) years(-1) for growing-stock volume change. These methods should be useful in further analysis of the sources of error and overall uncertai nty in national efforts to quantify carbon fluxes associated with forests a nd land cover dynamics.