SOLAS SAGE - sea-air gas exchange experiment

NIWA contacts: 


In 2004, scientists added iron salts to surface Sub-Antarctic waters east of New Zealand's South Island to determine whether this stimulated phytoplankton growth, and to examine the exchange of carbon dioxide (CO2) and trace gases between the atmosphere and ocean.

Our climate and weather are heavily dependent on the interaction between the oceans and the atmosphere. As an example, the world's oceans absorb 30-40% of the CO2 produced by mankind. Without this very effective buffer, climate change would be happening far quicker, and with far worse effects.

This interaction between the ocean and the atmosphere is especially important in the vast expanses of the Southern Ocean, where gas exchange is highly efficient as a result of low temperatures and turbulence at the sea surface generated by high winds and large waves.


Our understanding of the climate feedback processes which are driven by the ocean is still incomplete. This makes it harder to predict the future state of the climate. While we understand in theory how gases move in and out of the ocean, there haven't been many experiments which actually measured this exchange with the atmosphere.

Another factor that influences gas exchange in the open ocean is phytoplankton. These tiny plants – many of which are too small to be seen with the naked eye – are incredibly important, accounting for about half of all the photosynthesis that happens on earth. Photosynthesis is important not only because it allows plants to grow, but also because it removes carbon dioxide from the atmosphere.

Phytoplankton in the ocean play another important role in climate: they take up mineral sulfur from sea water and some of this is converted into the gas dimethyl sulphide, or DMS. When DMS transfers into the atmosphere it stimulates the creation of aerosols – tiny particles suspended in the air – that influence marine cloud cover, and how reflective it is. Even modest changes in how much DMS is emitted could have significant effects on global temperature.

If we understand the physical, biological and chemical processes that control how gases move between the ocean and atmosphere we will be able to better predict the impact of future changes in our climate.


The aim of the SOLAS Sea-Air Gas Exchange (SAGE) experiment was to study the exchange of gases that influence the climate. This was carried out by stimulating the growth of plankton, through the addition of iron sulphate to the surface of some of the Southern Ocean's waters.

To mount the SAGE experiment, NIWA brought together scientists from 20 organisations in New Zealand and abroad – the experiment required a wide range of expertise, and the Southern Ocean is a particularly challenging place to work.

The team identified the Bounty Trough, east of New Zealand's South Island, as the best place to conduct the experiment. In March 2004, iron sulphate was added to the surface area waters over an area of 50 square kilometres.

They also added very small amounts of two dissolved gases: sulphur hexafluoride (SF6) and an isotope of Helium, Helium-3. These dissolved gases were used as 'tracers', that allowed the team to track the movement of the patch of ocean, and also estimate the rate of the gas exchange between the atmosphere and the surface of the ocean.

Iron sulphate was added another three times over the 13-day experiment to maintain a level of iron sulphate which would stimulate phytoplankton growth. Throughout the experiment, the team took measurements at the centre of the patch of fertilised ocean, and compared them with stations outside the patch. This allowed them to monitor the response to the iron addition.


There were two significant results that came out of the experiment.

1. The influence of strong winds on gas exchange was determined.

The rate of gas exchange between the ocean and the atmosphere has not often been measured under such strong winds as that experienced during SAGE. This resulted in the development and publication of the SAGE wind-speed parameterisation, a relationship that provides calculation of gas exchange based upon windspeed. The SAGE parameterisation is particularly valuable to modelling CO2 uptake by the ocean, as it was obtained under high winds typical of the Southern Ocean.

2. Phytoplankton response to the addition of iron sulphate.

The Southern Ocean is iron-limited, and so the team expected to see a response from phytoplankton to the addition of iron sulphate.

However Chlorophyll-a, an indicator of phytoplankton, only doubled, despite the addition of iron four times over the experiment.

The reason that there wasn't more of a response was thought to be due to a range of factors, including:

  • high mixing and dilution of both dissolved iron and phytoplankton
  • zooplankton (microscopic animals which also live at in surface waters) grazing upon the phytoplankton
  • the phytoplankton lacked the nutrient of silicate which is required by some types of phytoplankton, called diatoms, for their cell walls.

In 2007, an international group of experts looked at the responses of 12 similar experiments in which iron was added to a patch of ocean, and found that the SAGE experiment produced the smallest response. Understanding the minor response observed during SAGE is important in determining what controls phytoplankton growth in New Zealand waters

Read more about the experiment in Water & Atmosphere 14

Eleven papers detailing the SAGE results were published in this special Issue of Deep-Sea Research 


  • Boyd, P.W et al (2007). Mesoscale Iron Enrichment Experiments 1993-2005: Synthesis and Future Directions. Science 315(5812): 612-617 doi: 10.1126/science.1131669.
  • Harvey, M.J.; et al (2011). The SOLAS air-sea gas exchange experiment (SAGE) 2004. Deep Sea Research Part II: Topical Studies in Oceanography. 58: 753-763 doi: 10.1016/j.dsr2.2010.10.015.
  • Ho, D. al (2006). Measurements of air-sea gas exchange at high wind speeds in the Southern Ocean: Implications for global parameterizations, Geophys. Res. Lett., 33, L16611, doi:10.1029/2006GL026817.
  • Law, C.S. et al (2011). Did dilution limit the phytoplankton response to iron addition in HNLCLSi Sub-Antarctic waters during SAGE ? Deep-Sea Res II, 58:786–799.
  • Peloquin, J. et al (2010). Control of the phytoplankton response during the SAGE experiment: a synthesis. Deep-Sea Res II, 58:824-838.

SAGE participants

NIWA, LDEO (USA), WHOI (USA), PML (UK), Bureau Meteorology (Australia), University Colarado (USA), Southern Cross University (Australia), University Rhode Island (USA), University Helsinki (Finland), RSMAS (USA), VIMS (USA), AAD (Australia), Dalhousie University (Canada).

SAGE funding

The New Zealand component of this work was sponsored by the then Foundation for Research, Science and Technology with support for collaboration through RSNZ ISAT Linkages. U.S. and U.K. collaborators were supported by National Science Foundation and Natural Environment Research Council Grants, respectively.


  • Ho et al, (2006). Measurements of air-sea gas exchange at high wind speeds in the Southern Ocean: Implications for global parameterizations. Geophysical Research Letters. DOI: 10.1029/2006GL026817, 2006 
  • Harvey et al, (2011). The SOLAS air–sea gas exchange experiment (SAGE). 2004 Deep Sea Research Part II: Topical Studies in Oceanography. DOI: 10.1016/j.dsr2.2010.10.015
  • Boyd et al., (2007). Mesoscale Iron Enrichment Experiments 1993-2005: Synthesis and Future Directions. Science. DOI:10.1126/science.1131669
Page last updated: 
25 February 2014
Preparation for release: the deck of RV Tangaroa with the iron tanks on the left and the SF6 tracer tanks on the right. (Photo: Matt Walkington)
The SAGE logo, with the voyage underway sea surface temperature plot incorporated in the “G”. Credit: SAGE
The SAGE addition set-up, with dissolved tracer tanks in the foreground, and dissolved iron solution tanks in the background. Credit: NIWA
The location of the SAGE experiment. The colour contours represent depth in meters. Credit: NIWA
Plot comparing the response of the phytoplankton (as maximum chlorophyll-a) with the response in other iron experiments. Red bars indicate the chlorophyll-a inside the iron-enriched patch and the blue bars the chlorophyll-a in waters outside the patch (Harvey et al, 2011).