Processed data supporting "Apparent vegetation recovery masks slow-state legacy under recurrent warm-dry extremes"

This dataset contains processed data products and workflow documentation supporting the manuscript “Apparent vegetation recovery masks slow-state legacy under recurrent warm-dry extremes”. The archived materials include the processed compound warm-dry event catalogue, E1–R–E2 recurrent-event chains, slow-state legacy diagnostics, M_slow classifications, matched-control outputs, balance and robustness summaries, GPP diagnostic results, table source data and figure source data used to support the analyses reported in the manuscript. The dataset was prepared for reproducibility and peer-review purposes. Original remote-sensing and reanalysis inputs were derived from publicly available products described in the manuscript, including NIRv, ERA5-Land, SMAP soil moisture, VODCA vegetation optical depth and GPP products. Large intermediate gridded data cubes and raw public input products are not redistributed here. These processed files support the manuscript results on apparent NIRv recovery, incomplete slow-state recovery, external matched-control inference and conditional response-state boundaries under recurrent warm-dry extremes on the Tibetan Plateau.