Task Aware Modulation Using Representation Learning for Upsaling of Terrestrial Carbon Fluxes
2026-03-10 • Machine Learning
Machine Learning
AI summaryⓘ
The authors address the problem of estimating how much carbon is exchanged between the land and atmosphere, which is hard because ground measurements are limited and unevenly spread. They developed a new method called TAM-RL that combines learning patterns over time and space with math rules based on carbon balance. When tested on over 150 sites worldwide, their method predicted carbon flows more accurately than previous approaches. This shows that using both data patterns and physical knowledge helps make better and more reliable global carbon estimates.
terrestrial carbon fluxglobal carbon budgetflux towerupscalingspatio-temporal representation learningencoder-decoder architecturecarbon balance equationpredictive performanceroot mean square error (RMSE)explained variance (R^2)
Authors
Aleksei Rozanov, Arvind Renganathan, Vipin Kumar
Abstract
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 ($R^2$) 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.