Challenges and solutions for reducing bias in Forest inventory-based estimates of aboveground biomass in arid zones

Forest inventories are fundamental instruments for estimating the aboveground biomass density (AGBD) of forests and for assessing their contribution to climate change mitigation. However, inventories may entail errors that generate uncertainty in the estimates, the magnitude of which varies according to vegetation type. Few studies have addressed the sources of error and bias in AGBD estimates based on forest inventories in arid zones. In this study, three major sources of error or bias were analyzed: the selection of inadequate dendrometric variables during sampling, the omission of small trees and shrubs due to inclusion criteria, and the lack of allometric equations for some of the most abundant species, such as columnar cacti. The results reveal an alarming level of underestimation due to the omission of smaller individuals in xeric shrubland and tropical dry forest, whose contribution can increase the average AGBD value by up to 307% and 180%, respectively. In this study, we propose a methodology to mitigate such underestimation and provide new allometric equations to estimate the biomass of the columnar cactus Pachycereus pringlei . Improvements of this kind are essential to determine the actual contribution of arid and semiarid vegetation to carbon storage, which appears to be seriously underestimated.