Principles for satellite monitoring of vegetation carbon uptake

Abstract

Remote-sensing-based numerical models harness satellite-borne measurements of light absorption by vegetation to estimate global patterns and trends in gross primary production (GPP) — the basis of the terrestrial carbon cycle. In this Perspective, we discuss the challenges in estimating GPP using these models and explore ways to improve their reliability. Current models vary substantially in their structure and produce differing results, especially regarding temporal trends in GPP. Many models invoke the light use efficiency principle, which links light absorption to photosynthesis and plant biomass production, to estimate GPP. However, these models vary in their assumptions about the controls of light use efficiency and typically depend on many, poorly constrained parameters. Eco-evolutionary optimality principles can greatly reduce parameter requirements, improving the accuracy and consistency of GPP estimates and interpretations of their relationships with environmental drivers. Integrating data across different satellites and sensors, and utilizing auxiliary optical band retrievals, could enhance spatiotemporal resolution and improve model-based detection of vegetation physiology, including drought stress. Extending and harmonizing the eddy-covariance flux-tower network will support systematic evaluation of GPP models. Improved reliability of GPP and biomass production estimates will better characterize temporal variation and advance understanding of the response of the terrestrial carbon cycle to environmental change. Global patterns and trends in primary production are estimated using remote-sensing-based models. This Perspective outlines ways to ensure that the next generation of model predictions robustly characterizes how this key element of the terrestrial carbon cycle is changing.