Dataset for article: "Multi-output deep learning models enhance the reliability of simultaneous above- and belowground biomass predictions in tropical forests"
Description
The dataset is used to develop the article entitled: "Multi-output deep learning models enhance the reliability of simultaneous above- and belowground biomass predictions in tropical forests". The dataset was developed based on a destructive sample of 175 trees collected from 27 purposively selected plots distributed in the Central Highlands ecoregion of Vietnam and was used to develop and cross-validate multi-output Deep Learning (DL) models as an alternative to the conventional Weighted Nonlinear Seemingly Unrelated Regression (WNSUR) method for simultaneous predictions of aboveground (AGB), belowground (BGB) and total tree biomass (TB) in two main tropical forest types - Dipterocarp Forest (DF) and Evergreen Broadleaf Forest (EBLF). As a result, multi-output DL models improved simultaneous predictions compared to conventional WNSUR models, and they can incorporate many complex ecological variables to enhance reliability of the forest biomass predictions.