Analyzing VOCs from the Great Barrier Reef

November 25, 2016

Australia’s Great Barrier Reef (GBR) stretches over 2300 km and is composed of over 3000 individual reef systems. The health of the reef therefore often comes under international scrutiny. Hilton Swan from Southern Cross University in Australia has been investigating volatile organic compound (VOC) emissions from the Great Barrier Reef using gas chromatography–mass spectrometry (GC–MS). He recently spoke to us about this work.

Australia’s Great Barrier Reef (GBR) stretches over 2300 km and is composed of over 3000 individual reef systems. The health of the reef therefore often comes under international scrutiny. Hilton Swan from Southern Cross University in Australia has been investigating volatile organic compound (VOC) emissions from the Great Barrier Reef using gas chromatography–mass spectrometry (GC–MS). He recently spoke to us about this work.

Q. Your group recently investigated volatile organic compound (VOC) emissions from coral across the Great Barrier Reef. What led your group to begin this study?A: The desire to understand the significance of coral reefs as sources of atmospheric dimethylsulfide (DMSa) led to this study. In comparison to the knowledge of oceanic DMS emission mechanisms, little is known about the mechanisms that promote coral reefs to emit DMS. As a starting point, laboratory chamber studies were used to identify conditions that trigger coral to release DMS. It was found that emission of DMS to the chamber headspace, while slowly bubbling instrument-grade air through filtered seawater containing coral branches, was associated with the release of coral mucus. Other sulfur VOCs and isoprene were also emitted in lesser amounts by mucus-coated coral (1). Corals release mucus in response to environmental stresses; for example, mucus provides protection against desiccation and solar radiation when coral is aerially exposed at low tide. Field studies were subsequently undertaken quantifying DMSa every 20–30 min over 2–3 week campaigns at Heron Island on the southern Great Barrier Reef (GBR) to identify coral reef-derived DMS emissions, while also monitoring tidal cycles, rainfall, wind speed, and other meteorological variables.

Q. What were the main analytical challenges you encountered and how did you overcome them?A: It is challenging to accurately quantify DMSa because it is present in nmol/m3 concentrations or part per trillion (pmol/mol) mixing ratios, so it is “very” trace analysis. In order to obtain detectable quantities for chromatographic analysis it is necessary to preconcentrate DMSa onto a suitable adsorbent, or directly capture it in a cryogenically cooled trap (cryotrap). We used gold-coated glass wool adsorbent to collect DMSa grab samples, which are subsequently thermally desorbed to the cryotrap to focus the DMS for gas chromatography (GC) analysis. For higher time-resolution sampling we automated a gas chromatograph fitted with a pulsed flame photometric detector (PFPD) to quantify DMSa every 20–30 min after drawing 3.5–4 L of air directly into the cryotrap (2).

A significant challenge to overcome is the reaction of DMSa with atmospheric oxidants during sampling prior to GC analysis. The most efficient oxidant scrubber we have devised is a glass-fiber filter impregnated with a mixture of sodium ascorbate and glycerol. The filter is dried and placed at the head of the intake line to minimize oxidation of DMSa during the sampling stage. An internal standard of ethyl methyl sulfide (EMS) is added as a surrogate with each sample to account for any oxidative loss of DMS. The EMS is introduced from a permeation tube in nanogram amounts, and is also used as a calibrant. Isoprene is more challenging to quantitatively sample than DMSa because it reacts with a wide range of atmospheric oxidants at a rate approximately 10x faster than DMS, putting even greater demands on efficient oxidant scrubbing during sampling.

DMS and other VOCs present in the headspace of chambers containing coral branches in filtered seawater were present at parts per billion (nmol/mol) mixing ratios. The significantly higher chamber headspace mixing ratios allowed samples to be collected, via the oxidant scrubber, into evacuated stainless steel canisters for off-site GC coupled to mass spectrometry (MS) analysis. This sampling approach also presented challenges because of the reactivity of sulfur VOCs, which required the use of internally surface deactivated canisters to obtain satisfactory recoveries.

Q. Are there any external factors that must be taken into consideration when conducting this kind of investigation?A: In order to collect coral from the GBR Marine Park for research purposes, it is first necessary to apply to the GBR Marine Park Authority for a permit to do so. Coral can only be taken from the scientific research zone of the coral reef surrounding Heron Island after obtaining a permit from the GBRMPA, which specifies the species and quantities permitted. It can take up to 16 weeks to obtain a permit after application. Permits are required from this Australian Government agency to manage the impacts on high-use and sensitive areas of the GBR. The GBRMPA manages the conservation and sustainable use of the Great Barrier Reef’s state and Commonwealth Marine Parks.

Q. DMS and isoprene are produced by mucus released by stressed coral. How significant are these VOCs and how wide are their implications?A: Gas phase DMS, isoprene, and atmospheric oxidants constitute a potent combination for the formation and rapid growth of new particles, commonly referred to as secondary organic aerosols. The hygroscopic properties of this aerosol allow it to grow to the point where it can act as cloud condensation nuclei. Therefore, these VOCs can ultimately assist formation of low-level clouds, influence the Earth’s radiative balance, and potentially impact the regional climate of the GBR.

A multiorganizational campaign was recently conducted on the GBR during September and October 2016 to examine the extent to which the GBR contributes to atmospheric particle production, and its capacity to influence rainfall and regional climate. Further general information about this study, known as the “Reef-to-Rainforest” campaign, is available from reference 3.

Q. There has been much in the media about the “death” of the Great Barrier Reef. Do you think this statement was premature? What is your view on this?A: I expect the recent sensational media obituary for the GBR was an unwarranted attempt to solicit widespread awareness of the issues facing the survival of the GBR. Climate change-induced stresses imposed by increasing sea surface temperatures, together with coral CaCO3 dissolution by ocean acidification (4), make the GBR extremely vulnerable. However, reports claiming the GBR is now dead are not helpful because uninformed readers are likely to believe it’s true and lose hope in initiatives such as the Paris accord to rein in carbon emissions.

The 2015–16 summer delivered the most extreme mass coral bleaching event yet observed throughout the GBR, affecting almost all of the reef system. The most northerly part of the GBR suffered the most prolonged heat stress and unfortunately a significant amount of that section of the GBR has not recovered from the severe conditions, but the entire GBR has not died. The GBRMPA official media release states that overall coral mortality throughout the GBR resulting from this extreme event is 22%, where there is a gradation of coral mortality ranging from highest in the far north to virtually none on the southern GBR (5). Surveys assessing the survival rates and recovery of coral across the GBR are ongoing. What matters now is that the GBR is not subject to another prolonged underwater heat wave in the coming summer, which could provide the impetus for more extensive coral mortality across the GBR.

Q. What were your main findings on the origins of atmospheric dimethylsulfide at Heron Island in the southern Great Barrier Reef (6)?

A: The automated GC–PFPD deployed to the Heron Island Research Station detected occasional coral reef-derived spikes of DMSa above the background continuum oceanic DMSa signal derived from phytoplankton and other marine biota. When these DMSa spikes were detected, they were generally observed at low tide under low wind speeds, indicating they were derived from the platform reef surrounding the island. At low tide parts of the reef can become aerially exposed, allowing direct exchange of DMS from coral to the atmosphere. A recent laboratory study has shown that air exposure of coral leads to an increase in DMS gas phase concentrations (6), providing support for our field observations (7). The Heron Island Research Station, which is a world-class research and teaching facility operated and managed by the University of Queensland, has provided an ideal location to study DMS emissions from the GBR.


1. H.B. Swan et al., Journal of Atmospheric Chemistry 73(3), 303–328 (2016).
2. H.B. Swan et al., Analytical Methods7(9), 3893–3902 (2015).
4. B.D. Eyre et al., Nature Climate Change4, 969–976 (2014).
6. F.E. Hopkins et al., Scientific Reports6, 36031 (2016).
7. H.B. Swan et al., Biogeosciences Discussions doi: 10.5194/bg-2016-387 (2016).

Hilton Swan completed a B. App. Sc. at the University of Technology Sydney (UTS) and an M.Sc. from University of New South Wales (UNSW) Australia, that involved identifying and quantifying VOCs associated with high strength wastewaters to assist engineering odour abatement solutions. Hilton has been involved with DMSa research since the 1990s, which involved work with the Australian Antarctic Division to measure DMSa at Prydz Bay, and assisting the CSIRO to obtain DMSa data at the Cape Grim Baseline Air Pollution Station. Hilton also undertook DMSa measurements for the US led Aerosol Characterization Experiments over the Southern Ocean, and Eastern Asia. Hilton is currently completing his Ph.D. with Southern Cross University (SCU) investigating the significance of the GBR as a source of DMS. He recently obtained a DMSa dataset at Mission Beach in North Queensland for the “Reef-to-Rainforest” campaign.