Machine Learning Techniques Improve Accuracy of Aboveground Biomass Estimation in Temperate Forests

Strategic Insights -

Current research from the National Institute of Forestry, Mexico, has shown that machine learning techniques can significantly improve the accuracy of aboveground biomass estimation in temperate forests. The study modelled AGB in central Mexico using active and passive remote sensing data combined with machine learning techniques (Random Forest and XGBoost) and compared the estimations against a traditional method, such as linear regression. The models obtained acceptable performance in all cases, but the machine learning algorithm Random Forest outperformed the regression method.

Key Takeaways:

  • The study found that machine learning techniques can improve the accuracy of aboveground biomass estimation in temperate forests.
  • The Random Forest algorithm outperformed the regression method, with R2 values of 0.54 and RMSE values of 19.17.
  • The variables that made significant contributions to the models were NDVI, kNDVI, and the HV polarisation from ALOS-Palsar.
  • The study found that machine learning ensemble had a higher Spearman correlation (r = 0.68) than the linear regression (r = 0.50).
  • The findings highlight the potential of integrating machine learning techniques with remote sensing data to improve the reliability of AGB estimation in temperate forests.
  • The study provides a new approach for AGB estimation in temperate forests, which can be useful for forest management and climate change research.
  • The model's performance can be improved by incorporating additional variables, such as climate and soil data.
  • The study suggests that machine learning techniques can be useful for other applications, such as land cover classification and crop yield prediction.
  • The study's findings can be applied to other regions with similar temperate forest ecosystems.

Statistics:

  • R2 values for Random Forest model: 0.54
  • RMSE values for Random Forest model: 19.17
  • R2 values for regression method: 0.41
  • RMSE values for regression method: 25.76
  • Spearman correlation coefficient for machine learning ensemble: 0.68
  • Spearman correlation coefficient for linear regression: 0.50

Sources:

  • Can Combining Machine Learning Techniques and Remote Sensing Data Improve the Accuracy of Aboveground Biomass Estimations in Temperate Forests of Central Mexico?. Geomatics, 2025,5(3):30. MDPI AG.
  • Martin Enrique Romero-Sanchez, National Institute of Forestry, Agriculture and Livestock Research, Mexico City 04010, Mexico.
  • Antonio Gonzalez-Hernandez, Efrain Velasco-Bautista, Arian Correa-Diaz, Alma Delia Ortiz-Reyes, Ramiro Perez-Miranda.