Direct mapping of plantation forest aboveground biomass change with deep learning and SAR-optical fusion
深層学習とSAR-光学融合による植林地上部バイオマス変化の直接マッピング (AI 翻訳)
Lee, Brian, Rich, Alex, Thomas, Nathan, Stovall, Atticus, Fatoyinbo, Temilola, Olmedo, Guillermo, Quijado, Pablo, Ramirez, Pablo, Heilmayr, Robert
🤖 gxceed AI 要約
日本語
本研究は、Sentinel-1、ALOS PALSAR、Sentinel-2のマルチセンサーデータとMixture-of-Experts（MOE）機械学習を用いて、森林の地上部バイオマス変化（ΔAGB）を直接推定する手法を開発した。9,087プロットの大規模反復調査データで訓練し、R2=0.90、RMSE=26.89 Mg/haと高精度を達成し、間接法よりRMSEを57%以上削減した。不確実性も提供し、森林炭素モニタリングの信頼性向上に貢献する。
English
This study develops a Mixture-of-Experts (MOE) machine learning framework using multi-sensor fusion (Sentinel-1, ALOS PALSAR, Sentinel-2) to directly estimate forest aboveground biomass change (ΔAGB). Trained on 9,087 repeat inventory plots, it achieves high accuracy (R2=0.90, RMSE=26.89 Mg/ha) and reduces RMSE by at least 57% compared to indirect methods, while providing per-pixel uncertainty. This enhances forest carbon monitoring for climate mitigation.
Unofficial AI-generated summary based on the public title and abstract. Not an official translation.
📝 gxceed 編集解説 — Why this matters
日本のGX文脈において
日本の森林炭素モニタリングやJ-クレジット制度におけるバイオマス推定の精度向上に寄与する可能性がある。ただし、直接的な開示要件や政策連動は限定的で、研究としての価値が中心。
In the global GX context
Globally, this work supports climate mitigation by improving forest carbon accounting accuracy, relevant to REDD+ and national GHG inventories. It demonstrates the value of combining remote sensing and ML for transparent carbon monitoring, which can inform climate disclosure and sustainability reporting.
👥 読者別の含意
🔬研究者:Provides a novel MOE approach for direct ΔAGB estimation with uncertainty, outperforming indirect methods.
🏢実務担当者:Offers a method for more accurate forest carbon monitoring, useful for carbon credit projects and sustainability reporting.
🏛政策担当者:Highlights the potential of remote sensing and ML for national forest carbon inventories and climate policy.
📄 Abstract（原文）
Accurately tracking changes in forest aboveground biomass (ΔAGB) is necessary for understanding global carbon dynamics. Traditional approaches estimate ΔAGB indirectly by differencing two independent biomass predictions, compounding uncertainty and reducing accuracy. Here, we develop a Mixture-of-Experts (MOE) machine learning framework that uses multi-sensor fusion of Sentinel-1 C-band SAR, ALOS PALSAR L-band SAR, and Sentinel-2 optical data to directly estimate ΔAGB and prediction uncertainty. Our training and validation data are drawn from a large, repeat forest inventory (9,087 plots resampled between 2016-2021) that characterizes a variety of management conditions causing both increases and decreases in AGB (e.g. planting, growth, pruning, thinning and harvest). Using this dataset, we trained an ensemble of three component models at 30-meter resolution: (1) a classifier that identifies change type, (2) a regression model for AGB growth, and (3) a regression model for AGB loss. This MOE approach of ΔAGB achieves high accuracies (R2 of 0.90, RMSE of 26.89 Mg/ha, global NRMSE of 4.5%), and dramatically outperforming the indirect approach applied to both internal baselines and existing global products (reducing RMSE by at least 57%). A heteroscedastic Gaussian Negative Log-Likelihood loss function and Monte Carlo dropout provide per-pixel predictive uncertainty alongside each prediction, offering a transparent and operationally useful measure of model confidence. This study demonstrates how multi-sensor fusion, large-scale repeat inventories, and MOE modeling can improve the accuracy and reliability of forest carbon monitoring for climate mitigation and sustainable forest management.
🔗 Provenance — このレコードを発見したソース
- EarthArXiv https://eartharxiv.org/repository/object/13334/download/23514/first seen 2026-06-17 04:19:20
🔔 こうした論文の新着を逃したくない方は キーワードアラート に登録（無料・3キーワードまで）。
gxceed は公開メタデータに基づく研究支援データセットです。要約・翻訳・解説は AI 支援で生成されています。 最終的な解釈・検証は利用者が原典資料に基づいて行うことを前提とします。