Accurately upscaling terrestrial carbon fluxes is central to estimating the global carbon budget, yet remains challenging due to the sparse and regionally biased distribution of ground measurements. Existing data-driven upscaling products often fail to generalize beyond observed domains, leading to systematic regional biases and high predictive uncertainty. We introduce Task-Aware Modulation with Representation Learning (TAM-RL), a framework that couples spatio-temporal representation learning with knowledge-guided encoder-decoder architecture and loss function derived from the carbon balance equation. Across 150+ flux tower sites representing diverse biomes and climate regimes, TAM-RL improves predictive performance relative to existing state-of-the-art datasets, reducing RMSE by 8-9.6% and increasing explained variance (R2) from 19.4% to 43.8%, depending on the target flux. These results demonstrate that integrating physically grounded constraints with adaptive representation learning can substantially enhance the robustness and transferability of global carbon flux estimates.
Figures & tables
Figure 1: Spatial distribution of the sites used in the current study where color represents climate type according to Köppen–Geiger climate classification ( Cui et al. 2021 ) . The sites are derived from the FLUXCOM, AmeriFlux, ICOS, and JapanFlux networks.
Figure 2: TAM-RL Architecture
Figure 3: Barplots illustrating mean RMSE of CT-LSTM, TALMSTL, TAM-RL, XGBoost and FLUXCOM-X-BASE aggregated by IGBP and Köppen climate types. The error bars are computed for classes with more than one site presenting them. NEE_VUT_USTAR50_QC=1.
Figure 4: Relative RMSE heatmap of XGBoost, CT-LSTM, TAMLSTM, and TAM-RL derived in comparsion to FLUXCOM-X-BASE with NEE_VUT_USTAR50_QC=1.
Figure 5: Scatter plots of RMSE and R2 of TAM-RL and FLUXCOM-X-BASE computed for every site and colored by the Koppen climate type.
Model
GPP
NEE
RMSE
R2
RMSE
R2
CT-LSTM
2.03
0.42
1.67
0.21
TAMLSTM
2.04
0.40
1.63
0.21
TAM-RL
1.97
0.43
1.62
0.23
XGBoost
2.17
0.34
1.76
0.16
FLUXCOM-X-BASE
2.18
0.36
1.76
0.16
Table 1: Model performance comparison across targets. Metrics are averaged across all sites. NEE_VUT_USTAR50_QC=1 and NEE_VUT_REF_QC=1.
Eddy-covariance (EC) flux towers provide in situ measurements of CO2 flux and serve as the ground-truth data for predictive `upscaling' models derived from satellite products. However, many satellites now resolve spatial scales smaller than an EC tower's footprint. We show theoretically that upscaling models trained on high-resolution data in heterogeneous landscapes must account for an EC tower's footprint to avoid bias in pixel-level predictors. To address this problem, we introduce Footprint-Aware Regression (FAR), a deep-learning framework that simultaneously predicts spatial footprints and pixel-level estimates of CO2 flux, and show it yields unbiased pixel-level predictions given sufficient training data. We demonstrate FAR on our AMERI-FAR25 dataset, which combines 205 site-years of tower data with corresponding Landsat scenes, and show that FAR outperforms non-footprint-aware models. FAR increased half-hourly R2 from approximately 0.575 to 0.634 and reduced RMSE by about 7% relative to the best fixed-footprint baseline on a dataset of withheld sites. Gains are larger for monthly and yearly averages relative to a coarser 990 m baseline. Site-level analyses show that these gains extend across multiple ecosystem types. These performance gains are observed whether footprints are learned jointly or estimated independently using an established footprint model, despite substantial variation in footprint size between methods. In a regional comparison over the Western Cascades, FAR produces flux estimates with a distribution comparable to that of existing high-resolution process-based models.
Jacob Searcy, Anish Dulal, Courtney Mathers +6
Department of Data Science, University of Oregon · Department of Computer Science, University of Oregon · Institute of Ecology and Evolution, University of Oregon +3
Accurately quantifying terrestrial carbon exchange is essential for climate policy and carbon accounting, yet models must generalize to ecosystems underrepresented in sparse eddy covariance observations. Despite this challenge being a natural instance of zero-shot spatial transfer learning for time series regression, no standardized benchmark exists to rigorously evaluate model performance across geographically distinct locations with different climate regimes and vegetation types. We introduce CarbonBench, the first benchmark for zero-shot spatial transfer in carbon flux upscaling. CarbonBench comprises over 1.3 million daily observations from 567 flux tower sites globally (2000-2024). It provides: (1) stratified evaluation protocols that explicitly test generalization across unseen vegetation types and climate regimes, separating spatial transfer from temporal autocorrelation; (2) a harmonized set of remote sensing and meteorological features to enable flexible architecture design; and (3) baselines ranging from tree-based methods to domain-generalization architectures. By bridging machine learning methodologies and Earth system science, CarbonBench aims to enable systematic comparison of transfer learning methods, serves as a testbed for regression under distribution shift, and contributes to the next-generation climate modeling efforts.
Aleksei Rozanov, Arvind Renganathan, Yimeng Zhang +1
Department of Computer Science & Engineering, University of Minnesota – Twin Cities, Minneapolis, MN, USA
We introduce FLUXtrapolation, a benchmark for extrapolating ecosystem fluxes under progressively harder distribution shifts. Ecosystem fluxes are central to understanding the carbon, water, and energy cycles, yet they can only be measured directly at sparsely located measurement towers. Producing global flux estimates therefore requires training models on observed sites using globally available covariates and predicting in unobserved regions, that is, upscaling. Flux upscaling is a challenging domain generalization problem that is affected by a shift in covariate distribution across climates, ecosystem types, and environmental conditions, as well as by conditional shift: important drivers remain unobserved at global scale. We provide a quantitative analysis of both these shifts in PX and PY∣X. FLUXtrapolation is designed based on domain expertise on flux upscaling: it defines temporal, spatial, and temperature-based extrapolation scenarios and evaluates performance across held-out domains, temporal aggregations, and tail errors. In a pilot study, we find that baselines perform similarly under median hourly RMSE, but separate under the proposed tail-focused and multi-scale evaluation. FLUXtrapolation therefore poses a realistic and thus relevant challenge for machine learning methods under distribution shift; at the same time, progress on this benchmark would directly support the scientific goal of improving flux upscaling.
Anya Fries, Jacob A Nelson, Martin Jung +2
Seminar for Statistics, ETH Zürich Zürich, Switzerland · Max Planck Institute for Biogeochemistry Jena, Germany