What question did this study set out to answer?

To develop and evaluate a hybrid machine learning and geostatistical modeling framework for accurate forest aboveground biomass (AGB) estimation using multi-source data.

August 19, 2026ForestsOpen Access

Hybrid Machine Learning and Geostatistical Approaches for Forest Aboveground Biomass Estimation in a Subtropical Region of China

View Full Paper

Ask AI

Bookmark

Authors

BABirhanie AlemayehuInstitute of Urban Environment YZYang ZhangState Forestry and Grassland Administration XLXin LiuInstitute of Urban Environment

Discussion

Member takes

Implication

Comparative modeling reveals combining Random Forest with Empirical Bayesian Kriging yields optimal forest biomass estimation, highlighting the value of spatial residual correction.

Key Points

To develop and evaluate a hybrid machine learning and geostatistical modeling framework for accurate forest aboveground biomass (AGB) estimation using multi-source data.
Integrated 48 predictors from six heterogeneous datasets, including Landsat-8, Sentinel-1 SAR, topography, meteorology, soil, and the 2014 National Forest Inventory.
Applied a two-stage variable selection framework using Pearson correlation screening (reducing to 36 variables) and recursive feature elimination (RFE) with 5-fold spatial block cross-validation.
Evaluated individual and hybrid models including Random Forest (RF), eXtreme Gradient Boosting (XGB), Empirical Bayesian Kriging Regression Prediction (EBKRP), RF_EBKRP, and XGB_EBKRP.
Stand age and stand density were the primary predictors, contributing 33.6% and 24.0% in RF, and 36.5% and 17.3% in XGB, respectively.
RF_EBKRP achieved the highest predictive accuracy with an R2 of 0.77.
RF_EBKRP reduced RMSE by 17.20% compared to standalone RF and by 43.75% compared to standalone EBKRP.

Cite This Study

Alemayehu et al. (2026) studied this question.

synapsesocial.com/papers/6a85640f03308d306e2d799d https://doi.org/10.3390/f17080976

View Full Paper

Ask AI

Bookmark

Implication

Comparative modeling reveals combining Random Forest with Empirical Bayesian Kriging yields optimal forest biomass estimation, highlighting the value of spatial residual correction.

Hybrid Machine Learning and Geostatistical Approaches for Forest Aboveground Biomass Estimation in a Subtropical Region of China

Discussion

Member takes

Implication

Key Points

Cite This Study

Hybrid Machine Learning and Geostatistical Approaches for Forest Aboveground Biomass Estimation in a Subtropical Region of China

Discussion

Member takes

Implication

Key Points

Cite This Study

Also Consider

Also Consider