Answers are grounded in this paper and cite their sources.
Satellite greening is widely used to judge whether mined land recovers, but around mines recorded only as points, it leaves unclear where and under which conditions recovery stalls. In Ganzhou, China, a global source of ionic rare earths for low carbon technologies, hundreds of such sites are scattered across nearly 40,000 km 2 of subtropical hills, too dispersed for repeated field surveys. Here we followed all 401 sites in 2000–2025, pairing an absolute ecological condition index (ECI) from Landsat with relative ecological gain (REGAIN), which rescales ECI to each 500 m cell's observed range and needs no dated disturbance or baseline. Condition improved significantly in 71.3% of cells after correction for multiple testing, while nearly one fifth stalled or declined. Regionally, condition rose and fell with interannual climate, but the land nearest the sites started lower and gained more, an imprint of mining that faded within 3–5 km. On mapped footprints in the core mining area, heavily mined cells kept pace with neighbors in REGAIN yet remained 0.27 lower in ECI. Spatially validated machine learning ranked land cover far above mining location, and even slight cropland or built-up expansion was associated with slower recovery. Stage-by-stage sensitivity analyses showed these results to be internally robust. Recovery around dispersed mines may therefore hinge on what becomes of the land, and regional greening can conceal stalled locations and a recovery debt on mined ground. As demand for critical minerals grows, such debts could become an uncounted cost of the energy transition.
No takes yet. Share an insight, caveat, or question.
Wang et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: