This study integrated ground-based plot surveys and multi-source remote sensing data to develop a canopy height and aboveground biomass estimation model for Hainan Jianfengling National Nature Reserve from 2003 to 2023. To support scientific transparency and model reproducibility, a portion of the research data has been organized and made publicly available. The dataset includes the following three categories:1、Survey data of tropical rainforest plots：Biomass data of 140 sample plots: Includes 140 sample plots (10 m × 10 m), with information on plot ID, geographic coordinates, forest type, and biomass of tree and understory layers.Biomass data of 64 historical plots: Contains biomass data from 64 historical plots (25.8 m × 25.8 m), including plot ID, location, and aboveground biomass.Data of individual trees from 140 sample plots: Covers detailed measurements of 4,732 individual trees, including species name, diameter at breast height (DBH), tree height, crown volume, and estimated biomass per tree.2、Forest type distribution data：Based on the natural forest classification dataset, GLC_FCS30D land cover data, and ALOS DEM elevation data, this shapefile represents forest type distribution in Hainan Jianfengling National Nature Reserve from 2003 to 2023 at a 30-meter spatial resolution (WGS 1984 coordinate system).3、Remote sensing estimation results of biomass and canopy height of Hainan Jianfengling National Nature Reserve from 2003 to 2023：Includes annual maps of forest aboveground biomass density and canopy height distribution in Hainan Jianfengling National Nature Reserve from 2003 to 2023. All data are provided in GeoTIFF format with a spatial resolution of 30 meters, suitable for GIS-based analysis and visualization.

Biomass carbon sequestration and sink capacities of tropical rainforests are crucial for addressing climate change. However, accurate canopy height estimation is necessary to determine carbon sink potential and implement effective forest management. This study compares the performance of four advanced machine learning algorithms—Random Forest (RF), Gradient Boosting Decision Tree (GBDT), Convolutional Neural Network (CNN), and Backpropagation Neural Network (BP)—in predicting forest canopy height in Hainan Tropical Rainforest National Park. The study uses 140 field survey samples and 315 unmanned aerial vehicle photogrammetry samples, along with multi-modal remote sensing datasets, including GEDI and ICESat satellite LiDAR data, Landsat imagery, and environmental information.The following is a description of the data package:The data in the folder (Model) includes the trained models of the four machine learning algorithms (BP, CNN, GBDT, RF).The data in the folder (Figures) includes the technology roadmap of the research, distribution map of the study area, and data point maps used in the study, along with scatter plots of RH80, RH85, RH90, and RH95 internal and external validation for the four machine learning algorithms (BP, CNN, GBDT, RF).The data in the folder (Datasets) includes the filtered modeling data and external validation data from GEDI and ICESat.The data in the folder (Code) includes the training code and prediction code for four machine learning algorithms (BP, CNN, GBDT, RF).If you require raw high-resolution raster data from other periods, please contact Dr. Qiu Zixuan at [email protected].

Biomass carbon sequestration and sink capacities of tropical rainforests are crucial for addressing climate change. However, accurate canopy height estimation is necessary to determine carbon sink potential and implement effective forest management. This study compares the performance of four advanced machine learning algorithms—Random Forest (RF), Gradient Boosting Decision Tree (GBDT), Convolutional Neural Network (CNN), and Backpropagation Neural Network (BP)—in predicting forest canopy height in Hainan Tropical Rainforest National Park. The study uses 140 field survey samples and 315 unmanned aerial vehicle photogrammetry samples, along with multi-modal remote sensing datasets, including GEDI and ICESat satellite LiDAR data, Landsat imagery, and environmental information.The following is a description of the data package:The data in the folder (Model) includes the trained models of the four machine learning algorithms (BP, CNN, GBDT, RF).The data in the folder (Figures) includes the technology roadmap of the research, distribution map of the study area, and data point maps used in the study, along with scatter plots of RH80, RH85, RH90, and RH95 internal and external validation for the four machine learning algorithms (BP, CNN, GBDT, RF).The data in the folder (Datasets) includes the filtered modeling data and external validation data from GEDI and ICESat.The data in the folder (Code) includes the training code and prediction code for four machine learning algorithms (BP, CNN, GBDT, RF).If you require raw high-resolution raster data from other periods, please contact Dr. Qiu Zixuan at [email protected].

This study integrated ground-based plot surveys and multi-source remote sensing data to develop a canopy height and aboveground biomass estimation model for Hainan Tropical Rainforest National Park from 2003 to 2023. To support scientific transparency and model reproducibility, a portion of the research data has been organized and made publicly available. The dataset includes the following three categories:1、Survey data of tropical rainforest plots：Biomass data of 140 sample plots: Includes 140 sample plots (10 m × 10 m), with information on plot ID, geographic coordinates, forest type, and biomass of tree and understory layers.Biomass data of 64 historical plots: Contains biomass data from 64 historical plots (25.8 m × 25.8 m), including plot ID, location, and aboveground biomass.Data of individual trees from 140 sample plots: Covers detailed measurements of 4,732 individual trees, including species name, diameter at breast height (DBH), tree height, crown volume, and estimated biomass per tree.2、Forest type distribution data：Based on the natural forest classification dataset, GLC_FCS30D land cover data, and ALOS DEM elevation data, this shapefile represents forest type distribution in Hainan Tropical Rainforest National Park from 2003 to 2023 at a 30-meter spatial resolution (WGS 1984 coordinate system).3、Remote sensing estimation results of biomass and canopy height of tropical rainforests in Hainan from 2003 to 2023：Includes annual maps of forest aboveground biomass density and canopy height distribution in Hainan Tropical Rainforest National Park from 2003 to 2023. All data are provided in GeoTIFF format with a spatial resolution of 30 meters, suitable for GIS-based analysis and visualization.

This study integrated ground-based plot surveys and multi-source remote sensing data to develop a canopy height and aboveground biomass estimation model for Hainan Tropical Rainforest National Park from 2003 to 2023. To support scientific transparency and model reproducibility, a portion of the research data has been organized and made publicly available. The dataset includes the following three categories:1、Survey data of tropical rainforest plots：Biomass data of 140 sample plots: Includes 140 sample plots (10 m × 10 m), with information on plot ID, geographic coordinates, forest type, and biomass of tree and understory layers.Biomass data of 64 historical plots: Contains biomass data from 64 historical plots (25.8 m × 25.8 m), including plot ID, location, and aboveground biomass.Data of individual trees from 140 sample plots: Covers detailed measurements of 4,732 individual trees, including species name, diameter at breast height (DBH), tree height, crown volume, and estimated biomass per tree.2、Forest type distribution data：Based on the natural forest classification dataset, GLC_FCS30D land cover data, and ALOS DEM elevation data, this shapefile represents forest type distribution in Hainan Tropical Rainforest National Park from 2003 to 2023 at a 30-meter spatial resolution (WGS 1984 coordinate system).3、Remote sensing estimation results of biomass and canopy height of tropical rainforests in Hainan from 2003 to 2023：Includes annual maps of forest aboveground biomass density and canopy height distribution in Hainan Tropical Rainforest National Park from 2003 to 2023. All data are provided in GeoTIFF format with a spatial resolution of 30 meters, suitable for GIS-based analysis and visualization.

This study integrated ground-based plot surveys and multi-source remote sensing data to develop a canopy height and aboveground biomass estimation model for Hainan Tropical Rainforest National Park from 2003 to 2023. To support scientific transparency and model reproducibility, a portion of the research data has been organized and made publicly available. The dataset includes the following three categories:1、Survey data of tropical rainforest plots：Biomass data of 140 sample plots: Includes 140 sample plots (10 m × 10 m), with information on plot ID, geographic coordinates, forest type, and biomass of tree and understory layers.Biomass data of 64 historical plots: Contains biomass data from 64 historical plots (25.8 m × 25.8 m), including plot ID, location, and aboveground biomass.Data of individual trees from 140 sample plots: Covers detailed measurements of 4,732 individual trees, including species name, diameter at breast height (DBH), tree height, crown volume, and estimated biomass per tree.2、Forest type distribution data：Based on the natural forest classification dataset, GLC_FCS30D land cover data, and ALOS DEM elevation data, this shapefile represents forest type distribution in Hainan Tropical Rainforest National Park from 2003 to 2023 at a 30-meter spatial resolution (WGS 1984 coordinate system).3、Remote sensing estimation results of biomass and canopy height of tropical rainforests in Hainan from 2003 to 2023：Includes annual maps of forest aboveground biomass density and canopy height distribution in Hainan Tropical Rainforest National Park from 2003 to 2023. All data are provided in GeoTIFF format with a spatial resolution of 30 meters, suitable for GIS-based analysis and visualization.

This study integrated ground-based plot surveys and multi-source remote sensing data to develop a canopy height and aboveground biomass estimation model for Hainan Jianfengling National Nature Reserve from 2003 to 2023. To support scientific transparency and model reproducibility, a portion of the research data has been organized and made publicly available. The dataset includes the following three categories:1、Survey data of tropical rainforest plots：Biomass data of 140 sample plots: Includes 140 sample plots (10 m × 10 m), with information on plot ID, geographic coordinates, forest type, and biomass of tree and understory layers.Biomass data of 64 historical plots: Contains biomass data from 64 historical plots (25.8 m × 25.8 m), including plot ID, location, and aboveground biomass.Data of individual trees from 140 sample plots: Covers detailed measurements of 4,732 individual trees, including species name, diameter at breast height (DBH), tree height, crown volume, and estimated biomass per tree.2、Forest type distribution data：Based on the natural forest classification dataset, GLC_FCS30D land cover data, and ALOS DEM elevation data, this shapefile represents forest type distribution in Hainan Jianfengling National Nature Reserve from 2003 to 2023 at a 30-meter spatial resolution (WGS 1984 coordinate system).3、Remote sensing estimation results of biomass and canopy height of Hainan Jianfengling National Nature Reserve from 2003 to 2023：Includes annual maps of forest aboveground biomass density and canopy height distribution in Hainan Jianfengling National Nature Reserve from 2003 to 2023. All data are provided in GeoTIFF format with a spatial resolution of 30 meters, suitable for GIS-based analysis and visualization.

The research area of this study is National Park of Hainan Tropical Rainforest. From February 20, 2022 to July 30, 2022, the investigators went to National Park of Hainan Tropical Rainforest to conduct a forest resources survey. The scope of the investigation includes Jianfengling Branch of Hainan Tropical Rainforest National Park Administration, Maorui Branch of Hainan Tropical Rainforest National Park Administration, Wuzhishan Branch of Hainan Tropical Rainforest National Park Administration, Diaoluo Mountain Branch of Hainan Tropical Rainforest National Park Administration, Limushan Branch of Hainan Tropical Rainforest National Park Administration, Bawangling Branch of Hainan Tropical Rainforest National Park Administration and Yingge Ling Branch of the Hainan Tropical Rainforest National Park Administration. In order to be consistent with the size of forest sample plots in the Hainan Continuous Forest Inventory, the forest sample plots investigated in this study was 25.8m × 25.8m. The distance between two forest sample plots was about 4 kilometers. A total of 98 forest sample plots were investigated, of which 64 were reviewed and 34 were supplementary. The investigators mainly investigated the tree species, DBH, tree height of trees with diameter above 5 cm in the forest sample plots. At the same time, data such as latitude, longitude, climate, slope gradient, slope direction and altitude of the forest sample plots were collected. In addition, this study combined with the data of the sixth, seventh, eighth and ninth Hainan Continuous Forest Inventory in the National Park of Hainan Tropical Rainforest included 64 forest permanent sample plots. In principle, the National Continuous Forest Inventory should be conducted every five years. Every year, the competent forestry authorities under the State Council shall uniformly arrange provinces to carry out National Continuous Forest Inventory. The review shall be conducted in the same year and the results shall be reported to the competent forestry authority under The State Council two years later.

The research area of this study is National Park of Hainan Tropical Rainforest. From February 20, 2022 to July 30, 2022, the investigators went to National Park of Hainan Tropical Rainforest to conduct a forest resources survey. The scope of the investigation includes Jianfengling Branch of Hainan Tropical Rainforest National Park Administration, Maorui Branch of Hainan Tropical Rainforest National Park Administration, Wuzhishan Branch of Hainan Tropical Rainforest National Park Administration, Diaoluo Mountain Branch of Hainan Tropical Rainforest National Park Administration, Limushan Branch of Hainan Tropical Rainforest National Park Administration, Bawangling Branch of Hainan Tropical Rainforest National Park Administration and Yingge Ling Branch of the Hainan Tropical Rainforest National Park Administration. In order to be consistent with the size of forest sample plots in the Hainan Continuous Forest Inventory, the forest sample plots investigated in this study was 25.8m × 25.8m. The distance between two forest sample plots was about 4 kilometers. A total of 98 forest sample plots were investigated, of which 64 were reviewed and 34 were supplementary. The investigators mainly investigated the tree species, DBH, tree height of trees with diameter above 5 cm in the forest sample plots. At the same time, data such as latitude, longitude, climate, slope gradient, slope direction and altitude of the forest sample plots were collected. In addition, this study combined with the data of the sixth, seventh, eighth and ninth Hainan Continuous Forest Inventory in the National Park of Hainan Tropical Rainforest included 64 forest permanent sample plots. In principle, the National Continuous Forest Inventory should be conducted every five years. Every year, the competent forestry authorities under the State Council shall uniformly arrange provinces to carry out National Continuous Forest Inventory. The review shall be conducted in the same year and the results shall be reported to the competent forestry authority under The State Council two years later.

Rubber trees are the primary source of natural rubber and are extensively planted in tropical regions. Hainan Province is one of China's main rubber-planting areas. The rubber planting industry is multifunctional and multi-beneficial. It is not only a pillar of Hainan's agriculture but also a significant source of income for farmers. The benefits and impacts of its ecosystem services are substantial. In September 2023, we conducted a survey of 102 rubber forest plots in Hainan Province. We recorded the coordinates of each plot and performed three-dimensional LiDAR scanning and biomass sampling, obtaining three-dimensional point cloud information and biomass data for each plot. After collecting the plot survey data, we used high-resolution Google satellite imagery with a resolution of 0.59m, combined with the Res-Unet deep learning algorithm, to develop a rubber forest recognition model and obtained the distribution of rubber forests in Hainan Province for 2023. Additionally, here is the description of our data folder files:1.The folders 'Classification of Rubber Forest' and 'Figures' contain the results of our identified rubber forest distribution in Hainan Province for 2023 and the distribution of sampling points from the plot survey.2. The folder 'Rubber Forest Sample Plot Survey Data' contains our three-dimensional LiDAR scanning measurements and biomass data of the rubber forest plots.