Year: 2024
Type: article
Abstract: The Qinghai-Xizang Plateau (QXP) is a crucial ecological system on the globe, and its forest cover status plays a significant role in global climate change and ecosystem protection. To obtain more accurate forest cover information of the QXP, this study generated a 10-meter resolution forest cover product of the QXP in 2021 based on Sentinel-1, Sentinel-2 remote sensing data, and DEM data, employing an optimized random forest model. The main characteristic of the proposed method lies in the optimization of the random forest model′s parameters and training features tailored to the complex land surface and vegetation types of the QXP, which effectively improves the model′s accuracy and generalization ability. Accuracy verification results demonstrate that the overall accuracy and Kappa coefficient of the product are 98.29% and 0.9496, respectively, outperforming the existing ESA WorldCover 2020 product with 95.04% and 0.7712. This product provides high-resolution forest cover information for the QXP, which can serve as basic data for forest resource management, research on carbon storage and carbon cycle of forest ecosystems, and other studies in the QXP.
Source: China Scientific Data
Authors: Tengfei LONG, Yantao GUO, Weili JIAO, Xiaomei ZHANG
Language: English
Subfield: Media Technology
Field: Engineering
Domain: Physical Sciences
SDGs: Life on land, Climate action
Open Access status: diamond
OpenAlex ID: W4411591787