Effect of measurement errors on the estimation of tree biomass

Abstract

Diameter at breast height (DBH) is commonly used to predict the aboveground biomass (AGB) of forests and to derive biomass models for single trees; however, there is evidence that measurement errors of DBH have not been previously considered. In this study, two types of measurement errors were evaluated: errors in national forest inventory data (NFID) and errors in a calibration data set (CDS). Using Monte Carlo simulations, the uncertainties arising from these two measurement errors were quantified. In addition, the effects of measurement errors on estimates under different error assumptions were analyzed to determine how these two uncertainties change with increasing errors. The results show that CDS measurement error contributes more to the total uncertainty, whereas NFID measurement error has a negligible effect on estimating the biomass of regional forests. The uncertainties of both types of measurement error increased with increasing error assumptions; however, the uncertainties caused by CDS measurement error were noticeably larger than those caused by NFID measurement error. Thus, the greatest potential for reducing uncertainties caused by measurement error lies in increasing the accuracy of DBH measurements in CDS.