Integrating MODIS-derived indices for eucalyptus stand volume estimation: an evaluation of MODIS gross primary productivity DOI Creative Commons
Manizheh Rajab Pourrahmati, Guerric Le Maire, Nicolas Baghdadi

et al.

Frontiers in Remote Sensing, Journal Year: 2025, Volume and Issue: 6

Published: May 14, 2025

Accurate estimates of stand volume dynamics in Eucalyptus plantations is critical for sustainable forest management and wood production. This study investigates the integration MODIS-derived indices, such as gross primary productivity (GPP), net photosynthesis (PSN) normalized difference vegetation index (NDVI), with traditional age-based methods to improve estimation plantations. MODIS GPP was first evaluated against flux tower measurements, showing moderate agreement systematic biases, particularly during periods highest lowest years after planting, an RMSE 19.65 gC m-2 8day-1 R2 0.38. Multiple linear regression (MLR) two machine learning models, including random (RF) stochastic gradient boosting (SGB), were used estimate by incorporating cumulative indices (Cgpp, Cpsn Cndvi) age. The SGB model showed best performance using full dataset, stands aged from 1.6 8.4 years, 22.63 m 3 ha-1, rRMSE 17.15% R 2 0.90. We that growth significantly improved model’s ability predict middle-aged mature stands. These results highlight utility products medium large-scale plantation management, providing scalable cost-effective monitoring volume.

Language: Английский

Integrating MODIS-derived indices for eucalyptus stand volume estimation: an evaluation of MODIS gross primary productivity DOI Creative Commons
Manizheh Rajab Pourrahmati, Guerric Le Maire, Nicolas Baghdadi

et al.

Frontiers in Remote Sensing, Journal Year: 2025, Volume and Issue: 6

Published: May 14, 2025

Accurate estimates of stand volume dynamics in Eucalyptus plantations is critical for sustainable forest management and wood production. This study investigates the integration MODIS-derived indices, such as gross primary productivity (GPP), net photosynthesis (PSN) normalized difference vegetation index (NDVI), with traditional age-based methods to improve estimation plantations. MODIS GPP was first evaluated against flux tower measurements, showing moderate agreement systematic biases, particularly during periods highest lowest years after planting, an RMSE 19.65 gC m-2 8day-1 R2 0.38. Multiple linear regression (MLR) two machine learning models, including random (RF) stochastic gradient boosting (SGB), were used estimate by incorporating cumulative indices (Cgpp, Cpsn Cndvi) age. The SGB model showed best performance using full dataset, stands aged from 1.6 8.4 years, 22.63 m 3 ha-1, rRMSE 17.15% R 2 0.90. We that growth significantly improved model’s ability predict middle-aged mature stands. These results highlight utility products medium large-scale plantation management, providing scalable cost-effective monitoring volume.

Language: Английский

Citations

0