Skip to main content
. 2022 Dec 22;13(1):91–97. doi: 10.1038/s41558-022-01544-w

Extended Data Table 2.

Evaluation

graphic file with name 41558_2022_1544_Tab2_ESM.jpg

a, Evaluation of the crown diameter to stem diameter conversion. Four CD–DBH equations were established, each using a random subset of 50% of the data. Tree biomass was predicted with each equation and aggregated to the hectare level. We then calculated the RMSE, rRMSE and bias between the 4 predictions for each 1 ha grid. Results shown here are aggregated over land cover classes. b, Tree crown model evaluation at pixel level. Confusion matrix between predicted and 6,591 manually labelled trees. These data are independent and were not used to establish or validate the model and the results.