A new strategy for improving the accuracy of forest aboveground biomass estimates in an alpine region based on multi-source remote sensing

ABSTRACT

Spatially explicit information on the distribution of dominant tree species groups and aboveground biomass (AGB) in forested areas is essential for developing targeted forest management and biodiversity conservation measures, as well as assessing forest carbon sequestration capacity. There is a shortage of continuously updated 30-m spatial resolution products for mapping dominant tree species groups. The vast majority of remote sensing-based AGB estimation approaches have relatively low accuracy for dominant tree species groups or forest types and are unsuitable for AGB modeling. Therefore, this study aims to develop an integrated framework that considers the phenological characteristics of different tree species to improve the mapping accuracies of forest dominant tree groups and corresponding AGB estimates. Thirty-meter resolution maps of dominant tree species groups were created using machine learning algorithms and phenological parameters. Features extracted from optical and radar images and phenological characteristics were used to construct AGB estimation models in a temporally consistent manner to improve the AGB estimation accuracy and perform dynamic AGB monitoring. The proposed method accurately characterized the dynamic distribution of the dominant tree species groups in the study area. The traditional AGB model that does not consider different forest types or species had an R² value of 0.52, whereas the proposed model that considers phenology and forest types had an R² value of 0.67. This result indicates that incorporating information on phenology and dominant species improves the accuracy of AGB estimations. The AGB in most regions was 30–55 t/ha, showing that the majority of the forests were young or middle-aged stands, and the areal percentage of AGB greater than 30 t/ha increased during the study period, suggesting an improvement in forest quality. Furthermore, the oak AGB was the highest, indicating that oak afforestation should be encouraged to enhance the carbon sequestration capacity of future forest ecosystems. The results provide new insights for researchers and managers to understand the trends of forest development and forest health, as well as technical information and a database for formulating more rational forest management strategies.

KEYWORDS:

• AGB
• forest species groups
• phenology
• remote sensing

Acknowledgments

Special thanks are extended to the USGS EROS Data Center and ESA for providing the Landsat and Sentinel-2 images used in this work.

Disclosure statement

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper

Data availability statement

The data that support the findings of this study (i.e. the land cover and probability maps) are available from the corresponding author and USGS (https://glovis.usgs.gov/).

Additional information

Funding

This work was jointly funded by the National Natural Science Foundation of China under grant number 31971577, the Foundation of Anhui Province Key Laboratory of Physical Geographic Environment under grant number 2022PGE011, the General Items of Support Plan for Outstanding Young Talents in Colleges and Universities under grant number gxyq2021217, the Priority Academic Program Development of Jiangsu Higher Education Institutions (PAPD), and the Open Fund of State Laboratory of Information Engineering in Surveying, Mapping, and Remote Sensing, Wuhan University, under grant number 21R04.