• Kristina Barclay

New Paper: Underestimation of air-sea CO2 fluxes due to freshwater stratification in Hudson Bay

The exchange of carbon dioxide between the atmosphere and the surface of the ocean plays an important role in the global carbon budget, (i.e., what happens to carbon dioxide emissions). Differences in air-sea carbon dioxide concentrations mean that as carbon dioxide emissions increase in the atmosphere, they are taken up by the ocean, which acts as a major carbon sink. In Arctic regions, these air-sea carbon dioxide fluxes are also influenced by freshwater input from river runoff and melting sea ice. Measurements of marine carbon dioxide that are used to determine the global carbon budget each year are typically collected via automated underway systems onboard research vessels around the world. These underway systems are usually measuring carbon dioxide in the waters that are located below 5 meters from the water’s surface. Typically, these underway measurements are based on the assumption that surface waters (the top ~10 metres) are well mixed enough to get accurate carbon dioxide measurements that are used to air-sea carbon dioxide fluxes at regional and global scale.

However, a recent study has suggested that this top surface layer in Arctic waters might not be as well mixed as previously thought. Instead, there might be a gradient of carbon dioxide within the top few metres of the ocean caused by the stratification of freshwater at the ocean’s surface. We spoke with Dr. Mohamed Ahmed, a Postdoctoral Associate at the University of Calgary, who recently published another study where he and his co-authors set out to determine exactly how much freshwater stratification occurs in the Arctic, and how this might influence air-sea carbon dioxide flux estimates used to determine global carbon budgets.

Mohamed on a sampling adventure using the Zodiac (Photo credit: Vickie Irish)

Dr. Ahmed is a marine biogeochemist specializing in the marine carbon cycle, air-sea carbon dioxide fluxes in the Canadian Arctic, and geospatial technologies like remote sensing. He obtained his Bachelor’s degree in Geology from Beni-Suef University in Egypt where he explored his interest in fieldwork, and then attained a M.Sc. degree in Geomatics from Lund University in Sweden that allowed him to develop strong technical skills in geospatial technologies. His Ph.D. (University of Calgary, Alberta, Canada) allowed him to combine his love of field work with his technical and programming skills, where he studied marine carbon cycles in the Arctic using both observational and remote sensing data.

It was while conducting fieldwork in the Arctic for his Ph.D. research that Dr. Ahmed made an observation that inspired his most recent paper. As he observed sea ice melting and collecting on the surface of the water from a Zodiac while on a research cruise in the Canadian Arctic Archipelago (also known as the Northwest Passage), he started thinking about what was actually being measured when determining air-sea carbon dioxide fluxes. Underway systems on ships are below 5 metres depth, but this freshwater from sea-ice melt appeared to form a shallow stratified layer in near the surface of the water. Other researchers (including his co-author, Dr. Lisa Miller) had also recently suggested that these surface layers might not be as well mixed as we had assumed. So, Dr. Ahmed and his co-authors set out to answer two questions:

1) Is there actually a shallow stratified freshwater layer in the top few metres of the ocean? As in, is this layer widespread across large basins like Hudson Bay in Canada?


2) How might this shallow layer of freshwater on the surface of Arctic waters be influencing estimates of air-sea carbon dioxide fluxes?

Conducting a CTD (Conductivity-Temperature- Depth) cast from the ice-edge (Photo credit: David Barber)

To answer these questions, Dr. Ahmed took two sets of seawater chemistry measurements, as well as measurements from the ship’s underway system, on a research cruise of Hudson Bay onboard the CCGS Amundsen in 2018. The first set of water samples were of the top 30-50 cm of seawater collected in specialized horizontal Niskin bottles with the second set taken at 7 m depth to correspond with the ship’s underway system. Water samples were taken either from the foredeck of the ship or a small boat within 2 km of the ship. The research cruise allowed them to collect data representative of Hudson Bay, with over 45 stations across the bay. The timing of the cruise was also important, as they were able to capture the peak window of sea-ice melt in Hudson Bay from late May – mid July when there is the greatest potential for possible stratification of surface freshwater.

Drilling an ice hole to sample water below (Photo credit: Lucette Barber)
Sampling from an ice floe in Hudson Bay (Photo credit: Lucette Barber)
Sampling from the foredeck using the Niskin bottle (Photo credit: Sarah Schembri)

They found that there was stratification of freshwater near the surface, with surface waters containing having lower carbon dioxide values than those at 7 m depth (differences can reach to ~30 μatm in some stations). This means that underway measurements are likely underestimating air-sea carbon dioxide fluxes by as much as 50% in Hudson Bay.

“These are interesting results as the impact of freshwater stratification may not be limited to polar waters, as many coastal areas around the world host many freshwater sources,” says Dr. Ahmed.

This figure shows the presence of freshwater stratification layer close to the sea-ice melt and riverine input. Contour plots of salinity, temperature, density anomaly (σθ), and Brunt–Väisälä frequency (BVF) measured by the ship’s Conductivity-Temperature- Depth (CTD) instrument in a transect from the sea-ice edge to the Nelson River in Hudson Bay. The dotted lines indicate locations of the CTD casts.

However, this gradient of carbon dioxide in the surface waters only persisted for about 5 weeks, suggesting that sea-ice melt and river runoff are likely to influence the accuracy of underway system measurements only within the first 5 or so weeks of peak freshwater input during the summer months. Not only that, but after the first week, they were able to apply a correction to the underway system measurements to correct for this shallow freshwater layer.

An example of vertical pCO2 gradients at two different stations in Hudson Bay. Temperature, salinity, and density anomaly (σθ) in the upper 16 m of the water column at stratified (stn 34) and un-stratified (stn 21) station. Surface pCO2 values at both stations were derived from samples collected by a horizontal Niskin bottle, while pCO2 values at 7-m depth were measured via the underway pCO2 system.

Conducting fieldwork onboard a research vessel in the Arctic is not without its challenges. Dr. Ahmed had originally planned to conduct this research in 2017, but the research cruise was cancelled when the CCGS Amundsen (operated by the Canadian Coast Guard) was called for search and rescue missions. But, Dr. Ahmed took the opportunity to discuss his ideas with other researchers, including co-author Dr. Miller, who also had similar observations and was working on a separate study based on similar observations. When the research cruise happened the following year, another challenge Dr. Ahmed faced was having the ship maneuvered carefully towards the sample stations so as not to disturb the surrounding waters, which was not always possible if the captain had other responsibilities and the ship’s schedule did not allow for the possible delay. A final obstacle was the sea-ice itself, which would sometimes impede collection of bottle samples from the small inflatable Zodiac boat, which unlike the CCGS Amundsen, is not capable of breaking through ice.

Amundsen icebreaker escorting a cargo ship stuck in ice in 2017 (Photo credit: Mohamed Ahmed)

The take-home of the new study by Dr. Ahmed and co-authors is that we are likely underestimating carbon dioxide sinks in the Arctic, meaning that seawater in at least some areas of the Arctic, such as Hudson Bay, is taking up more carbon dioxide than we are currently estimating. Dr. Ahmed emphasized that these results are important for creating accurate carbon budgets and ensuring that underway system measurements are representative of the surface seawater conditions where air-sea carbon dioxide exchange actually occurs.

“It will be interesting to see if our findings are matching other observations in different locations in the Arctic waters as the air-sea gas exchange of other gases might be impacted by the presence of freshwater stratification,” he says.

Dr. Ahmed hopes that researchers can use these data to determine freshwater stratification in other areas of the Arctic, and ultimately allow us to estimate global carbon budgets more accurately, especially in the vulnerable Arctic region.

Mohamed on another adventure using Zodiac (Photo credit: Catherine Mundy)

Read Ahmed et al., 2020 here:

Underestimation of surface pCO2 and air-sea CO2 fluxes due to freshwater stratification in an Arctic shelf sea, Hudson Bay. Elementa: Science of the Anthropocene 8:084

To learn more about Dr. Ahmed and his research, please follow his updates via Twitter (@MohamedMMAhmed).


Thanks to Dr. Mohamed Ahmed for his interview with OA CoP Coordinator, Kristina Barclay, and for providing edits on this article. All photo/figure captions were provided by Dr. Ahmed.


Ahmed, Mohamed M. M., Brent G. T. Else, David Capelle, Lisa A. Miller, and Tim Papakyriakou. 2020. Underestimation of surface pCO2 and air-sea CO2 fluxes due to freshwater stratification in an Arctic shelf sea, Hudson Bay. Elementa: Science of the Anthropocene 8:084.


Miller, Lisa A., Tonya M. Burgers, William J. Burt, Mats A. Granskog, and Tim N. Papakyriakou. 2019. Air‐Sea CO2 Flux Estimates in Stratified Arctic Coastal Waters: How Wrong Can We Be? Geophysical Research Letters 46:235 – 243.


Recent Posts

See All

Marine Environmental Observation Prediction and Response (MEOPAR) Network

Ocean Acidification Community of Practice 

Proudly created with Wix.com