Global warming is accelerating the decomposition of organic carbon stored in permafrost regions, particularly during winter when photosynthesis ceases. However, due to the lack of high-quality and gridded time-series data on winter CO₂ fluxes, substantial uncertainty remains in assessing winter CO₂ emissions across the Northern Hemisphere permafrost regions. This study constructed a dataset of winter CO₂ flux at a spatial resolution of 0.1° for Northern Hemisphere permafrost regions from 1982 to 2022. The dataset integrates 2,487 monthly winter CO₂ flux observations with multi-source geospatial data, including climate, vegetation, snow cover, soil properties, and permafrost characteristics. Winter CO₂ fluxes were upscaled separately for the Arctic and the Tibetan Plateau using an Extremely Randomized Trees model. Validation results indicate excellent model performance, with an R² of 0.67, an MAE of 3.94 g C m⁻² month⁻¹, and an RMSE of 5.67 g C m⁻² month⁻¹ for the Arctic permafrost region; and an R² of 0.82, an MAE of 2.76 g C m⁻² month⁻¹, and an RMSE of 3.76 g C m⁻² month⁻¹ for the Tibetan Plateau. This dataset provides fundamental support for research on carbon cycling in permafrost ecosystems.
| # | number | name | type |
| 1 | 2024YFF0810900 | Impacts of Permafrost Degradation on Methane Emissions from Peatlands in the Qinghai-Tibet Plateau | National key R & D plan |
This work is licensed under
CC BY 4.0 (Creative Commons Attribution 4.0 International License).
CSTR:11738.11.ncdc.permafrost.db7432.2026
10.12072/ncdc.permafrost.db7432.2026
