Rethinking landscape-level flux estimates

Insights into landscape-level C flux

Landscape-level carbon fluxes and their connectivity between the aquatic and terrestrial realm are important to quantify at large scales in order to be able to understand the effects climate and land use changes have on carbon cycling.


An eddy covariance flux tower in the field. Photo: Plum Island Ecosystems LTER

Terrestrial, marine and peatland carbon cycling have all been well studied, while standing and fluvial waters remain a missing piece in the global carbon cycling picture. Recent carbon upscaling studies have yielded interesting insights into inland waters and their ability to act as hotspots for carbon cycling. Inland waters have been known to be highly variable with respect to their role in regional and global carbon cycling, acting as either a sink or a source depending on inputs from allochthonous sources or from carbonate weathering, and dissolved organic carbon (DOC) concentrations.

Eddy covariance flux measurements are essentially limited to quantifying homogeneous ecosystem carbon dioxide (CO2) fluxes at a landscape scale, as pointed out by Premke et al. (2016). While eddy covariance is great for covering large areas (hundreds to thousands of meters squared) and can collect flux measurements for both terrestrial and aquatic ecosystems, this method of measurement is less robust when ecosystem heterogeneity and small scale flux variability become important to quantify. This is because the eddy covariance method averages data collected over these large footprint areas. In their study, Premke et al. (2016) highlight the importance of accounting for spatial heterogeneity and small scale variability when studying landscape-level carbon fluxes.

In order to investigate their hypothesis, Premke and colleagues analyzed spatial data for a temperate moraine landscape in north east Germany that is heavily dotted with water bodies and peatlands, and has been altered by agricultural activity. They looked at five different scenarios of increasing landscape diversity in order to assess the CO2 and methane (CH4) flux dynamics for each.

What Premke and colleagues found was that by including small water bodies in their simulation, landscape-level fluxes changed from net negative, to net positive sources of carbon efflux. When they simulated a homogeneous landscape of cropland only, CHspecifically was slightly net negative. When small lakes were included, and the landscape diversity increased to the most complex simulation, CHflux increased by more than 3 orders of magnitude (Figure 1) when compared to the cropland uptake scenario. This highlights the importance of accounting for this diversity and variability in order to build an accurate picture of what is actually happening with carbon cycling in these environments.

Figure 1. Average estimated methane flux for five different landscape diversity simulations (Premke et al. 2016). From left to right: homogeneous simulated landscapes to heterogeneous.

Figure 1. Average estimated methane flux for five different landscape diversity simulations (Premke et al. 2016). From left to right: homogeneous simulated landscapes to heterogeneous.

It turns out that previous estimates using only eddy covariance flux measurements may have grossly underestimated flux magnitudes largely due to the method of measurement. Considering the prevalence of eddy covariance flux measurements, this could add up to be a significant global underestimation in ecosystem-scale carbon balances.

The results of this study highlight the importance of combining multiple methods of measurement. For example augmenting eddy covariance towers with on-the-ground flux chambers to both get an estimate of the spatial heterogeneity of the systems as well as to provide a ground-truth measurement to compare to the eddy covariance flux estimates.


Premke, K., Attermeyer, K., Augustin, J., Cabezas, A., Casper, P., Deumlich, D., Gelbrecht, J., Gerke, H. H., Gessler, A., Grossart, H-P., Hilt, S., Hupfer, M., Kalettka, T., Kayler, Z., Lischeid, G., Sommer, M., and Zak, D. 2016. The importance of landscape diversity for carbon fluxes at the landscape level: small-scale heterogeneity matters. WIREs Water DOI: 10.1002/WAT2.1147