Predicting Net Ecosystem Carbon Exchange of Typical Forest Ecosystems in China Based on ChinaFLUX

We created MemoryFlux, a new dataset describing how land ecosystems absorbed and released carbon between 2001 and 2021. Using an artificial intelligence approach that learns from past environmental conditions, we produced more accurate estimates and captured carbon anomalies caused by major droughts and floods. MemoryFlux offers a more reliable view of carbon cycling and helps improve our understanding of how ecosystems respond to a changing climate.

We studied how planting green manure between tea rows affects carbon dioxide release from tea fields. Mixing legume and grass species improved soil health, reduced emissions from tea rows, and stabilized carbon over time. Although inter-row areas released more carbon, overall emissions declined with continued intercropping. This shows that green manure can make tea farming more climate-friendly and sustainable.