南平市の森林垂直構造の高解像度データセットを開発(High-Precision Canopy Height Dataset Illuminates Forest Structure Dynamics in Nanping City, China)

2026-03-13 中国科学院(CAS)

中国学院信息創新研究院(AIRCAS)中心する研究チームは、福建省南平森林垂直構造把握する10m解像度樹冠データセット(2022–2023年)開発した。GEDI衛星LiDAR、Sentinel-1/2、UAV LiDAR観測、ASTER GDEM地形データ、MODIS表面温度、CHIRPS降水量などデータ統合し、バイアス補正モデルランダムスト回帰組み合わせ作成した。さらにSHAP解析により地形・気候・スペクトル要因影響評価した。検証ではRMSE11.80mから1.70m大幅改善し、0.29から0.80向上した。データ森林バイオマス炭素蓄積評価、森林攪乱監視、持続可能森林管理活用できる。

High-Precision Canopy Height Dataset Illuminates Forest Structure Dynamics in Nanping City, China<関連情報>

中国福建省南平市向け高精度樹冠高測定器(10m) A 10 m High-precision Canopy Height Product for Nanping City, Fujian Province, China

Ling Yi,Xiaojing Yao,Aixia Yang,Liqiang Zhang,Dacheng Wang,Yue Jiao,Yaoliang Chen,Shufu Liu,Gang Chen & Yalan Liu
Scientific Data  Published:10 February 2026
DOI:  https://doi.org/10.1038/s41597-026-06767-6  Unedited version

Abstract

As a key ecological barrier and core carbon sink area in Southern China, Nanping City presents significant challenges for high-accuracy forest canopy height mapping due to its unique vegetation structure and complex topography. The current products for canopy height couldn’t meet the requirements for accurate carbon stock estimation. This study calibrated the biases in GEDI data by proposing an innovative Bias Calibration Model and combined with multi-source data including Sentinel-1, 2, UAV LiDAR, and topographic data, produced 10-meter canopy height products for 2022 and 2023 for Nanping City using a random forest model. The generated data products achieved R² = 0.62 for both 2022 and 2023 when validated against independent UAV data, with RMSE = 2.88 m and 3.09 m for 2022 and 2023 respectively, accurately characterizing the vertical structural attributes of the forest This study provides a reliable data foundation for the estimation of subsequent forest biomass and carbon stock.

1304森林環境
ad
ad
Follow
ad
タイトルとURLをコピーしました