Figure 2 | Scientific Reports

Figure 2

From: Leveraging Signatures of Plant Functional Strategies in Wood Density Profiles of African Trees to Correct Mass Estimations From Terrestrial Laser Data

Figure 2

Bias in volume to mass conversion due to vertical WD gradients induced by the use of WDGWD: (A) Density plot of the relative errors (bi) in tree biomass estimation computed from combinations of tree volume (destructive data: black line, n = 822; LiDAR data: red line, n = 58) and species mean wood density extracted from the Global Wood Density database (WDGWD). (B) Local regression (loess function) representing the relationship between tree relative vertical WD profile (characterized by tree PCA1 score) and bi of panel A. Bias is reduced when using a tree-level estimate of WD, the volume-weighted wood density (VWWDm5): (C) Density plot of bi for tree biomass estimation computed from combinations of tree volume and the predicted VWWD from WDGWD and tree DBH (model 5, see Table 2). (D) Local regression (loess function) representing the relationship between the relative vertical WD profile and bi of panel C.

Back to article page