A medley of methods: Air-sea CO2 flux in seagrass

To understand how natural and anthropogenic processes will impact Earth’s current and future climate, climatologists rely heavily on complex computer models. However, these models are only as good as the data that is fed into them, making field based measurements and the instrumentation and methodologies used to gather them of vital importance.

Estimates of natural and human-driven greenhouse gas emissions are key inputs to these models. Depending on the ecosystem, greenhouse gases (GHGs) can be produced naturally through a variety of different pathways. Coastal ecosystems are interesting because of their unique combination of high productivity, abundant nutrients, tidal action, and frequent mixing which all allow for a very productive food web and therefore high GHG emissions. They are important contributors to the global carbon cycle and are well known for their ability to sequester carbon in their sediments which can be trapped for millennia, known as ‘Blue Carbon’.


Temperate seagrass species Zostera marina. Photo: the Marine Life Information Network (MarLIN)

Seagrass meadows are a subset of these coastal ecosystems, producing a net average of 817 g C m-2 per year. To put that value into context, open oceanic waters have a net production of 130 g C m-2 per year, whereas terrestrial ecosystems sit around 200 g C m-2 per year (Mateo et al. 2006). Despite their high productivity, seagrass meadows are one of the lesser studied ecosystems in the coastal realm, partly due to technical challenges involved in measuring gas flux from seagrass beds including water inundation, corrosion, biofouling, power sources in the field, and so on.

In this blog post we will take you through a few examples of how researchers in the field have tackled the technological challenge of making these measurements in order to better answer questions around the importance of seagrass beds in the global carbon balance and global change.

Seagrass Sampling Method 1

The Do-It-Yourself Chamber: Make your own method (Bahlmann et al. 2015)

It seems that more often than not you need an instrument with a certain set of specifications that just isn’t on the market for the price that you would like (…sound familiar?). Bahlmann et al. (2015) were in a similar situation when deciding on how best to measure trace gas flux in a seagrass meadow (Zostera noltii). Their solution was to build their own gas flux chamber out of a 10 l glass bottle with the bottom cut off. If you are curious for more information on their chamber design they have a detailed description in their paper.


Figure 1. Gas flux chamber design by Bahlmann et al. (2015) depicted in both air exposed (left) and immersed (right) periods.

Finding: The automated chamber designed by Bahlmann and colleagues allowed them to sample during full tidal cycles, which gave them surprising results regarding when the highest greenhouse gas fluxes occurred in the seagrass meadow. Previously it had been thought that low tide yielded the highest greenhouse gas fluxes, however Bahlmann et al. (2015) found that high tide inundation of the seagrass meadow had the highest emissions of CO2, CH4, and other Volatile Organic Compounds (VOCs).


Figure 2. Bahlmann et al. (2015) CO2 flux data over a four day period in April in a Z. noltii meadow. Tidal inundation is indicated as either blue (tidal immersion) or green (air exposure), and yellow bars indicate periods of daylight.

The ability of their instrumentation to be inundated with saltwater in addition to collecting high temporal resolution data (every 15 minutes) provided them with the necessary means to be able to decipher these tidally-influenced trends. They concluded that advective transport, or “skin circulation”, was strongly controlled by tidal immersion and is what governs trace gas flux transport dynamics in intertidal sediments. During tidal inundation, trace gas fluxes were significantly enhanced as compared to fluxes during air exposure of the sediments.

Seagrass Sampling Method 2

The Permanent Fixture: Eddy covariance (Tokoro et al. 2014)

Eddy covariance is a commonly used method for sampling gas fluxes in terrestrial environments, however Dr. Tokoro and colleagues decided to use it for gas flux in the coastal environment. One of the greatest advantages of the eddy covariance method is the ability to directly and autonomously sample over broad spatial and temporal scale. The eddy covariance towers for this study were deployed in two different coastal seagrass meadow locations: a boreal site (dominant species: Zostera marina), and subtropical site (dominant species: Cymodocea serrulata, Thalassia hemprichii, and Enhalus acoroides) where the area covered, or the footprint of the tower, ranged from hundreds of m2 to several km2. As is the case with eddy covariance, there are many factors that need to be considered when taking measurements such as the height from which measurements are being taken: area sampled, wind speed and direction, atmospheric stability and spatial uniformity of the study sites.


Figure 3. Tokoro et al. (2014) eddy covariance air-sea CO2 flux data at boreal site in summer (a) and winter (b) and at subtropical site during summer (c).

Using eddy covariance, and comparing this method with other air-sea flux measurement methods, Tokoro et al. (2014) were able to determine that air-sea CO2 flux had a significant positive correlation with measured Dissolved Inorganic Carbon (ΔDIC). These air-sea fluxes varied depending on the measurement timescale and season. Overall, all three methods found the boreal and subtropical sites had a negative air-sea CO2 flux, except in the case of diurnal timescale measurements, where the boreal site became a source of air-sea CO2 flux in winter. These findings indicate that air-sea CO2 flux is an important part of the carbon sequestration process and that it should be taken into consideration when assessing carbon cycling dynamics, in addition to carbon burial rates in seagrass meadows.

Seagrass Sampling Method 3

The Drop-N-Measure Method: The floating chamber (Podgrajsek et al. 2014; Tokoro et al. 2014)

Using the floating chamber method is attractive because of it’s ability to directly measure air-sea flux in the field at a low cost and with less effort than compared with installing eddy covariance towers at a site. The chamber floats on the water surface, and can be left in a stationary location or towed behind a vessel (hence Drop-N-Measure!). Changes in carbon dioxide concentration over time in the floating chamber are used to determine the air-sea gas flux. This method is best used in calm conditions, otherwise bubbles tend to end up inside the chamber or the chamber can be flipped over entirely. Logistically speaking, this method is typically carried out only during daylight hours, thus the data being collected is discontinuous.

Finding: Due to the differences in area coverage between the floating chamber (< 1 m2) and eddy covariance methods (several m2 to several km2), flux measurements will not be the same from each, however using the two methods in unison can help highlight the spatial dependence of air-sea flux. A similar study by Podgrajsek et al. (2014) highlights the drastic difference between the two methods in terms of temporal scales. Looking at CO2 and CH4 flux from a lake using both methods, they found that the floating chamber produces inherently discontinuous data and thus misses important episodic flux data. However, the two methods produced similar flux estimates in terms of magnitude. Podgrajsek and colleagues (2014) findings support the conclusions made by Tokoro et al. (2014): that the two methods are so different in terms of their spatial and temporal scales that they should be seen as complementary to one another, rather than as truly comparable methods.


Figure 4. Flux data time series of CO2 in freshwater lake over various sampling periods from Podgrajsek et al. (2014) comparing eddy covariance and floating chamber measurements. Eddy covariance tower 1 shown in black dots and eddy covariance tower 2 shown in blue; floating chamber data shown in red dots.


Methodologies and instrumentation all have their advantages and disadvantages in the field, as well as with data processing, data quality, and spatial and temporal coverage.

  • Eddy covariance towers offer large spatial and temporal coverage for gas flux measurements in both terrestrial and aquatic settings, however they are intensive in terms of installation, data processing and maintenance.
  • Floating chambers are relatively cheap to deploy and operate, but are limited in their spatial and temporal coverage, potentially biasing diurnal measurements.
  • The DIY setup can be cost-effective, though this is situation-dependent.

Regardless of methods used to obtain important data on GHG flux in seagrass meadows, it is important that the technology envelope is being pushed from all sides in order to obtain better and better measurements.

All of this measurement improvement will ultimately lead us in being better informed as a society on how to move forward when facing issues as large as future climate change and variability.


Bahlmann, E., Weinberg, I., Lavrič, J. V., Eckhardt, T., Michaelis, W., Santos, R., Seifert, R., 2014. Tidal controls on trace gas dynamics in a seagrass meadow of the Ria Formosa lagoon (Southern Portugal). Biogeosciences 12: 1683-1696

Mateo, M., Cebrián, J., Dunton, K., Mutchler, T.: Carbon flux in seagrass ecosystems, seagrasses: Biology, Ecology and Conservation, Springer Netherlands, 159-192, 2006

Podgrajsek, E., Sahlée, E., Bastviken, D., Holst, J., Lindroth, A., Tranvik, L., Rutgersson, A., 2014. Comparison of floating chamber and eddy covariance measurements of lake greenhouse gas fluxes. Biogeosciences 11: 4225-4233

Tokoro, T., Hosokawa, S., Miyoshi, E., Tada, K., Watanabe, K., Montani, S., Kayanne, H., Kuwae, T., 2014. Net uptake of atmospheric CO₂ by coastal submerged aquatic vegetation. Glob Change Biol 20: 1873-1884