Regional Forest Stock Volume Mapping for Jingdong County, Yunnan, China
by Zeyu Li·Updated 1mo ago
15.8 KB1files
Available on 1 platform
Sign in to view source links and access this dataset
Description
143 sample plots of Pinus kesiya var. langbianensis forests in Jingdong County, Yunnan, China, were used to develop a machine learning model for Forest Stock Volume (FSV) estimation. The model, created by Zeyu Li, combined GEDI LiDAR, Landsat-8 imagery, and topographic data, achieving an R² of 0.93 and estimating a total standing stock of approximately 8.66 × 10⁷ m³ for 2019. The dataset includes the resulting spatially explicit FSV map.
Use Cases
Update forest inventory databases based on the spatially explicit FSV map.
Monitor forest productivity and carbon storage based on the estimated volume per hectare.
Prioritize forest management areas based on the mapped stock volume distribution.
Validate and compare remote sensing interpolation methods for FSV estimation based on the comparison of six methods.
Strengths
Model performance is high, with a reported R² of 0.93 and RMSE of 9.50 m³/ha.
The dataset is based on 143 field sample plots for model training and validation.
The methodology integrates multiple data sources: GEDI LiDAR, Landsat-8 imagery, and topographic variables.
Limitations
The dataset is very small (15.8 KB), suggesting it may contain only summary results or a document, not the full underlying geospatial data.
Column-level documentation is absent; field semantics must be inferred after download.
Geographic scope is limited to a single county (Jingdong County, Yunnan, China).
Provenance
Source
Zeyu Li via figshare.
Collection Method
Combined field measurements from 143 sample plots with GEDI LiDAR metrics, Landsat-8 imagery, and topographic variables using a multi-level stacking ensemble model.
Time Range
2019
Freshness
Last updated 2026-04-30 05:41:41; freshness should be verified.
Geography
Jingdong County, Yunnan, China.
The primary file format is DOCX, which likely contains a research paper or report rather than raw data tables; actual data availability requires inspection after download.