Spatiotemporal evolution and prediction of land use change and carbon storage in ionic rare earth mining areas based on the YOLOv11–SegFormer–InVEST–PLUS integrated model
Rare earth resources are strategic minerals vital to high-tech and defense sectors globally, playing a crucial role in sustainable development and technological innovation worldwide. Rising demand has intensified the exploitation of ionic rare earth mines, potentially undermining dual carbon goals....
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-09-01
|
| Series: | Ecological Indicators |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S1470160X25009136 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Rare earth resources are strategic minerals vital to high-tech and defense sectors globally, playing a crucial role in sustainable development and technological innovation worldwide. Rising demand has intensified the exploitation of ionic rare earth mines, potentially undermining dual carbon goals. These mines often use open-pit leaching, which severely degrades vegetation and weakens regional carbon sink capacity. Yet, the spatiotemporal dynamics of carbon storage in such areas, along with the effects of policy interventions, remain insufficiently quantified. Existing assessments often rely on coarse-resolution land use data, failing to detect micro-scale infrastructure like sedimentation ponds or to capture the complex interactions of restoration and management efforts. This study investigates a representative mining area in Jiangxi Province using a multi-source remote sensing framework. High-resolution imagery is analyzed with the YOLOv11 model to detect sedimentation pond clusters, while SegFormer enables fine-scale land use classification. These outputs serve as inputs for the InVEST model, enabling it to evaluate variations in carbon storage between 2015 and 2025. The PLUS model further simulates future trajectories under varying scenarios. Results show: (1) carbon storage displays spatial heterogeneity linked to forest cover and mining intensity; (2) carbon storage rose 6.12% over the decade; (3) By 2030, compared to the naturally developed scenario, ecological protection scenario may increase carbon storage by 0.3%, while the economic development scenario may reduce it by 0.8%. By examining the relationship between the policy index and changes in carbon storage, it can be concluded that the increase in carbon storage is positively correlated with the strength of policy implementation. |
|---|---|
| ISSN: | 1470-160X |