Improving Measurements of CO2 Fluxes from Landscapes

Understanding the links between large swirling winds and CO2 fluxes is critical to closing the energy balance of land surface and interpreting heat and carbon exchanges.

The Science

Turbulent vertical fluxes of heat, water vapor, and carbon dioxide (CO2) occur constantly between land surfaces and the atmosphere. For decades, measuring such fluxes has relied on eddy covariance (EC), a complex statistical technique. However, most studies using EC are unable to close the surface-energy balance between sensible and latent heat fluxes. This widely reported gap, known as the nonclosure problem of the surface energy balance, is commonly attributed to the influence of large-scale eddies of swirling wind on both kinds of heat fluxes.

A new paper by scientists at Washington State University and Pacific Northwest National Laboratory provides new insights into EC by investigating two understudied issues: (1) how CO2 fluxes are influenced by large eddies and (2) the mechanistic links between CO2 fluxes and energy balance nonclosure.

The results demonstrate, in part, that reductions in the magnitude of CO2 fluxes associated with large turbulent eddies are mechanistically linked to nonclosure of the surface energy budget.

The Impact

The research findings improve the understanding of how non-closure of the surface-energy balance impacts measurements of CO2 fluxes. It also provides direct evidence that further studies are needed to investigate how landscape heterogeneity—sagebrush terrain, in the case of this work—influences CO2 fluxes.


The new study relies on a dataset collected by an EC flux system in a semi-arid sagebrush ecosystem in the Hanford Area of rural southeastern Washington. The research shows a link between non-closure and reduced CO2 fluxes associated with large turbulent eddies. It attributes that link to the simultaneous influence of low-frequency motions on sensible and latent heat fluxes and on CO2 fluxes.

The researchers used a recently developed approach, ensemble empirical mode decomposition, to extract large eddies from the turbulence time series. Then they analyzed the impacts of amplitude and phase differences on flux contribution.

One challenge in this work was identifying occasional spectral gaps, especially under unstable atmospheric conditions when convective motions tend to overlap the scales between large eddies and small eddies. Based on previous work by these scientists, the authors defined large eddies as the sum of a certain number of oscillatory components that are largely responsible for the run-to-run variations in fluxes. There was no surprise at the non-closure of the surface energy balance and therefore biases in COfluxes. However, the researchers found that the energy balance closure ratio decreased as atmospheric instability increased. The underlying causes of that remain unclear. Work on finding those causes, the authors say, is underway.

The authors, who also include researchers from Lanzhou University in China, collected their high-quality data from three eddy covariance flux sites within the Hanford area.

Principal Investigator

Heping Liu
Washington State University

Program Manager

Paul Bayer
U.S. Department of Energy, Biological and Environmental Research (SC-33)
Environmental System Science


This work was supported by the Office of Biological and Environmental Research (BER), within the U.S. Department of Energy (DOE), as part of BER’s Subsurface Biogeochemical Research Program (SBR) at the Pacific Northwest National Laboratory. This work was performed in collaboration with Washington State University.

Related Links


Gao, Z., H. Liu, J. E. C. Missik, J. Yao, et al. "Mechanistic links between underestimated CO2 fluxes and non-closure of the surface energy balance in a semi-arid sagebrush ecosystem." Environmental Research Letters 14 (4), 044016  (2019).