What question did this study set out to answer?

The aim is to improve accuracy in estimating aboveground biomass (AGB) in tropical forests using integrated Sentinel-1 and Sentinel-2 data.

April 22, 2026Trees Forests and PeopleOpen Access

Aboveground biomass mapping and uncertainty estimation in tropical forests using machine learning with multispectral and SAR data

View Full Paper

Ask AI

Bookmark

Authors

Evaluates machine learning approaches to predict biomass in tropical forests, highlighting improved accuracy with integrated data sources.

The aim is to improve accuracy in estimating aboveground biomass (AGB) in tropical forests using integrated Sentinel-1 and Sentinel-2 data.
Compared three machine learning techniques: Random Forest, Support Vector Machine, and Extreme Gradient Boosting.
Utilized multispectral and radar data from Sentinel-1 and Sentinel-2.
Evaluated texture parameters and their impact on AGB predictions.
Random Forest achieved the highest accuracy with an R² of 0.55 and a mean absolute error of 57.79 Mg/ha.
Predicted AGB ranged from 16.64 to 375.73 Mg/ha, averaging 215.37 Mg/ha.
Significant AGB differences were noted among forest types, with high AGB in semievergreen forests.

Kanda et al. (2026) studied this question.

View Full Paper

Ask AI

Bookmark

Evaluates machine learning approaches to predict biomass in tropical forests, highlighting improved accuracy with integrated data sources.