Improved estimation of net ecosystem CO<sub>2</sub> exchange over North America using LSTM-based flux  upscaling (2001–2021)

Huang, Chengcheng; He, Wei; Byrne, Brendan; Nguyen, Ngoc Tu; Xiao, Jingfeng; Yang, Hua; Ciais, Philippe; Wang, Songhan; Li, Xing; Ma, Han; Xu, Peipei; Zhao, Mengyao; Chen, Hui; Ju, Weimin

doi:10.5194/essd-18-7367-2026

Articles | Volume 18, issue 10

https://doi.org/10.5194/essd-18-7367-2026

© Author(s) 2026. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/essd-18-7367-2026

© Author(s) 2026. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 18, issue 10

Data description article

|

07 Oct 2026

Data description article |

| 07 Oct 2026

Improved estimation of net ecosystem CO₂ exchange over North America using LSTM-based flux upscaling (2001–2021)

Chengcheng Huang, Wei He, Brendan Byrne, Ngoc Tu Nguyen, Jingfeng Xiao, Hua Yang, Philippe Ciais, Songhan Wang, Xing Li, Han Ma, Peipei Xu, Mengyao Zhao, Hui Chen, and Weimin Ju

Supplement

https://doi.org/10.5194/essd-18-7367-2026-supplement

Data sets

Improved estimation of net ecosystem CO2 exchange over North America using LSTM-based flux upscaling (2001–2021) Chengcheng Huang and Wei He https://doi.org/10.5281/zenodo.20482274

glass_vi_010_025 Changhao Xiong https://doi.org/10.6084/m9.figshare.22267048.v1

ERA5-Land monthly averaged data from 1950 to present J. Muñoz Sabater https://doi.org/10.24381/cds.68d2bb30

Model code and software

Improved estimation of net ecosystem CO2 exchange over North America using LSTM-based flux upscaling (2001–2021) Chengcheng Huang and Wei He https://doi.org/10.5281/zenodo.20482274

Download

Article (12728 KB)
Full-text XML

Short summary

We created MemoryFlux, a new dataset that showed how ecosystems across North America absorbed and released carbon from 2001 to 2021. Using a deep-learning approach trained on historical climate and environmental conditions, we improved estimates of land carbon exchange and captured carbon anomalies associated with major droughts and floods. MemoryFlux provides a more reliable view of carbon cycling and improves our understanding of how ecosystems respond to a changing climate.