Improved estimation of net ecosystem CO 2 exchange over North America using LSTM-based flux upscaling (2001-2021)
Abstract
Accurate estimation of regional-scale terrestrial carbon budgets is of great importance but remains challenging. With particular advantages, the Long Short-Term Memory (LSTM) network method shows potential for improving regional carbon budget upscaling estimations. Here, using LSTM, we upscaled regional net ecosystem carbon exchange (NEE) with available flux tower measurements and satellite land surface observations in North America. With well-established ecosystem-specific LSTMs, we produced monthly NEE at a spatial resolution of 0.1°× 0.1°over 2001-2021 (labelled as MemoryFlux). Unlike existing upscaling estimates, our dataset properly identified the Midwest Corn Belt as a region of large seasonal carbon uptake during the peak growing season, a feature revealed by previous top-down studies and recognized as a model benchmark. Moreover, the seasonal variations in NEE estimated by MemoryFlux were strongly correlated with those from independent atmospheric inversions, including the ensemble mean of the Orbiting Carbon Observatory-2 Model Published by Copernicus Publications. 7368 C. Huang et al.: LSTM-based NEE upscaling over North America
Intercomparison Project (OCO-2 v10 MIP; r = 0.96, p < 0.001) and CarbonTracker2022 (CT2022) (r = 0.97, p < 0.001). The mean annual NEE was estimated at -1.27 ± 0.12 Pg C yr -1 , which was closer in magnitude to inversions (-0.83 to -0.70 Pg C yr -1 ) than existing upscaling estimates (-3.30 to -1.68 Pg C yr -1 ). In addition, MemoryFlux captured spatial NEE anomaly patterns associated with six selected severe drought and flood events. We further found that explicitly incorporating historical predictor information improved the representation of NEE interannual variability and spatial anomalies associated with climate extremes. MemoryFlux provides an improved bottom-up estimation of North American NEE and shows greater consistency in magnitude with independent atmospheric inversion estimates than several existing EC-based upscaling products.
Related
- A field observation method to characterize site soundscape and improve detection probability estimation for eastern bat species in North America
- Stochastic modelling and upscaling of hydrodynamic transport in geological fractures
- FineKarstAGB: a 30 m resolution aboveground biomass dataset for Southwest China derived by upscaling plot-level inventory using sub-meter GaoFen satellite data
- Rainfall isotopic variability in northern Central America driven by regionally-organized deep atmospheric convection and airmass rainout
- Tropospheric Ozone Assessment Report II: South America regional analysis
BibTeX
@article{huang2026improved,
title = {{Improved estimation of net ecosystem CO 2 exchange over North America using LSTM-based flux upscaling (2001-2021)}},
author = {Chengcheng Huang and Wei He and Brendan Byrne and Ngoc Tu Nguyen and Jingfeng Xiao and Hua Yang and Philippe Ciais and Songhan Wang and Xing Li and Han Ma and Peipei Xu and Mengyao Zhao and Hui Chen and Weimin Ju},
journal = {Earth System Science Data},
year = {2026},
month = oct,
doi = {10.5281/zenodo.20482274(},
url = {https://hal.science/hal-05820678v1},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Improved estimation of net ecosystem CO 2 exchange over North America using LSTM-based flux upscaling (2001-2021)
AU - Chengcheng Huang
AU - Wei He
AU - Brendan Byrne
AU - Ngoc Tu Nguyen
AU - Jingfeng Xiao
AU - Hua Yang
AU - Philippe Ciais
AU - Songhan Wang
AU - Xing Li
AU - Han Ma
AU - Peipei Xu
AU - Mengyao Zhao
AU - Hui Chen
AU - Weimin Ju
PY - 2026
DA - 2026/10/07
JO - Earth System Science Data
DO - 10.5281/zenodo.20482274(
UR - https://hal.science/hal-05820678v1
AB - Accurate estimation of regional-scale terrestrial carbon budgets is of great importance but remains challenging. With particular advantages, the Long Short-Term Memory (LSTM) network method shows potential for improving regional carbon budget upscaling estimations. Here, using LSTM, we upscaled regional net ecosystem carbon exchange (NEE) with available flux tower measurements and satellite land surface observations in North America. With well-established ecosystem-specific LSTMs, we produced monthly NEE at a spatial resolution of 0.1°× 0.1°over 2001-2021 (labelled as MemoryFlux). Unlike existing upscaling estimates, our dataset properly identified the Midwest Corn Belt as a region of large seasonal carbon uptake during the peak growing season, a feature revealed by previous top-down studies and recognized as a model benchmark. Moreover, the seasonal variations in NEE estimated by MemoryFlux were strongly correlated with those from independent atmospheric inversions, including the ensemble mean of the Orbiting Carbon Observatory-2 Model Published by Copernicus Publications. 7368 C. Huang et al.: LSTM-based NEE upscaling over North America Intercomparison Project (OCO-2 v10 MIP; r = 0.96, p < 0.001) and CarbonTracker2022 (CT2022) (r = 0.97, p < 0.001). The mean annual NEE was estimated at -1.27 ± 0.12 Pg C yr -1 , which was closer in magnitude to inversions (-0.83 to -0.70 Pg C yr -1 ) than existing upscaling estimates (-3.30 to -1.68 Pg C yr -1 ). In addition, MemoryFlux captured spatial NEE anomaly patterns associated with six selected severe drought and flood events. We further found that explicitly incorporating historical predictor information improved the representation of NEE interannual variability and spatial anomalies associated with climate extremes. MemoryFlux provides an improved bottom-up estimation of North American NEE and shows greater consistency in magnitude with independent atmospheric inversion estimates than several existing EC-based upscaling products.
SN - 1866-3508
SN - 1866-3516
ER -