First Mapping the Canopy Height of Primeval Forests in the Tallest Tree Area of Asia

Опубликована: Янв. 1, 2024

We have developed the world's first canopy height map for primeval forest located within distribution area of world-level giant trees, where tallest tree in Asia (102.3m) was recently discovered. This mapping is crucial identifying more individual and community as well analyzing quantifying effectiveness biodiversity conservation measures Yarlung Tsangpo Grand Canyon (YTGC) National Nature Reserve under conditions global warming. proposed a method to using deep learning base on fusion spaceborne LiDAR satellite imagery (GEDI, ICESat-2, Sentinel-2). customized depthwise separable convolutional (DSC) neural network—PRFXception, which incorporates pyramid receptive fields. PRFXception, tailored specifically forest, efficiently integrates multi-size field features infer at footprint level GEDI ICESat-2 from Sentinel-2 optical with 10-meter spatial resolution. To validate our approach, we conducted survey 227 permanent plots stratified sampling measured several including "Asia's tree" their communities UAV-LS, predicted compared validation data (RMSE=7.56m, MAE=6.07m, ME=-0.98m, R2=0.58m), UAV-LS point clouds (RMSE=5.75m, MAE=3.72m, ME=0.82m, R2=0.65m), ground (RMSE=6.75m, MAE=5.56m, ME=2.14m, R2=0.60m). mapped potential trees discovered two previously undetected an 89% probability having 80-100m tall, potentially taller than Asia's tree. The multi-source Earth observation data-driven PRFXception integrated framework propose expected achieve operational forward-looking monitoring dynamics forests worldwide. Combined surveys, it provides promising tool discovering trees. paper scientific evidence confirming southeastern Tibet—northwestern Yunnan fourth center supporting climate sustainable development initiatives promoting inclusion YTGC scope China's national park conservation.

Язык: Английский

First Mapping the Canopy Height of Primeval Forests in the Tallest Tree Area of Asia

SSRN Electronic Journal, Год журнала: 2024, Номер unknown

Опубликована: Янв. 1, 2024

We have developed the world's first canopy height map for primeval forest located within distribution area of world-level giant trees, where tallest tree in Asia (102.3m) was recently discovered. This mapping is crucial identifying more individual and community as well analyzing quantifying effectiveness biodiversity conservation measures Yarlung Tsangpo Grand Canyon (YTGC) National Nature Reserve under conditions global warming. proposed a method to using deep learning base on fusion spaceborne LiDAR satellite imagery (GEDI, ICESat-2, Sentinel-2). customized depthwise separable convolutional (DSC) neural network—PRFXception, which incorporates pyramid receptive fields. PRFXception, tailored specifically forest, efficiently integrates multi-size field features infer at footprint level GEDI ICESat-2 from Sentinel-2 optical with 10-meter spatial resolution. To validate our approach, we conducted survey 227 permanent plots stratified sampling measured several including "Asia's tree" their communities UAV-LS, predicted compared validation data (RMSE=7.56m, MAE=6.07m, ME=-0.98m, R2=0.58m), UAV-LS point clouds (RMSE=5.75m, MAE=3.72m, ME=0.82m, R2=0.65m), ground (RMSE=6.75m, MAE=5.56m, ME=2.14m, R2=0.60m). mapped potential trees discovered two previously undetected an 89% probability having 80-100m tall, potentially taller than Asia's tree. The multi-source Earth observation data-driven PRFXception integrated framework propose expected achieve operational forward-looking monitoring dynamics forests worldwide. Combined surveys, it provides promising tool discovering trees. paper scientific evidence confirming southeastern Tibet—northwestern Yunnan fourth center supporting climate sustainable development initiatives promoting inclusion YTGC scope China's national park conservation.

Язык: Английский

Enhanced Corn Mapping with Height-Spectral Gaussian Mixture Modeling

Опубликована: Янв. 1, 2024

Язык: Английский