Measuring Volcanic Gases


Volcanic gases from Kilauea are
analyzed using a mass spectrometer
at the Hawaiian Volcano Observatory


There are three primary ways that gas geochemists collect data: 

Estimates from Rocks, Minerals, and Inclusions

Scientists can determine the amount and types of gas in a rock, in the minerals within a rock, or in the gas inclusions in minerals or glass (Ihinger and others, 1994). The methods fall into four classes: bulk extraction, energetic particle bombardment, vibrational spectroscopic techniques, and phase equilibrium studies. These method are used in experimental studies and for rocks from all tectonic settings, historic eruptions, and large pre-historic eruptions. Some methods have been found to underestimate the observed amount of gas released during some historic eruption. Thus, the methods may give an estimate of the minimum amount of gas released. For example, Gerlach and McGee (1994) used melt inclusion data to estimate 0.08 Mt of SO2 emitted by the Mount St. Helens eruption. TOMS, COSPEC and ash leachate data provide an estimate of 2 Mt. Gerlach and McGee suggested a vapor phase carried most of the SO2.



Direct Sampling

The easiest, but often the most difficult, way to collect a sample is by hand, placing a container directly in the gases. This method was perfected by the late Werner Giggenbach. Difficulty arises because of the high temperatures, dangers associated with being close to vents, and the possibility of contamination of the sample by the atmosphere. Left diagram: Evacuated-bottle sampling scheme. Direct samples are most commonly collected in solution-filled bottles (4N NaOH in a titanium or silica tube) and then returned to the lab for analyses. In the drawing on the left, the enlargement shows an evacuated glass bottle. The solution contains NaOH and collected H20), H2S, SO2, HCl, HF, and CO2. The headspace contains collected H20, H2, CH4, O2, CO, N2, Ar, Ne. Right diagram: Flow-through sampling scheme. Figure from Sutton and others, 1992).

Right:   Volcanic gases from Kilauea collected using the flow-through sampling scheme.   Photo by Steve Mattox.

Another way to measure gases released by a volcano is to collect fresh ash samples (before it rains) and pour distilled water through the ash. Then after the liquid passes through the ash it is collected. This is the leachate. The leachate is analyzed for Cl (chloride), F, SO4 and pH. The ratio of Cl to S increases prior to eruptions.

Leachates were measured prior to the 1980 eruption at Mount St. Helens (Nehring and Johnston, 1981). Cl and SO4 were measured in the field and the S/Cl ratio was observed to increase gradually from March 28 to May 18.

The S/Cl ratio increased 30 times over its initial value prior to an eruption of Asama volcano in Japan. At Fuego, in Guatemala, the S/Cl ratio increased 5 times over its initial value and the size of the change was proportional to eruption size.

Continuous direct sampling is a relatively new method to monitor gases. Results of measurements are telemetered to safe locations off the volcano. At Mount St. Helens, upward-moving fresh magma was detected 12 to 60 hours before it was extruded into the dome (McKee and Sutton, 1994).




Remote Sensing

The COSPEC (correlation spectrometer)

Volcanologist found an application for an instrument used to measure pollution. A correlation spectrometer (COSPEC) is designed to measure the amount of sulfur dioxide in a passing air mass (or volcanic plume). The spectrometer compares the amount of solar ultraviolet light absorbed by sulfur dioxide in the plume to an internal standard. Numerous measurements are made to achieve reliable results.

COSPEC used to measure SO2 released by Merapi Volcano.


Use of vehicle-mounted (left) and tripod-mounted (right) ground-based COSPEC.
A. Side view. B. Front view. C. Typical data. From Sutton and others (1992).



Balloons are used to take samples in or to carry instruments into aerosol layers. For example, Sheridan and others (1992) collected samples of atmospheric particles from the Pinatubo eruption by releasing balloons from Laramie, Wyoming. Most of the fine particles were H2SO4 droplets. Other larger particles were supermicrometer sulfate particles and composite sulfate/crustal particles. Deshler and others (1992) also used balloons to study the aerosols in the Pinatubo layer. Their study showed that 90% of the SO2 had been converted to H2SO4 aerosol within one month of the eruption. They also measured pressure, temperature, ozone, and particle density.


LIDAR is a ground-based remote sensing method that is used to measure the distribution and amounts of several gases in the atmosphere. NOAA has a fine page that explains LIDAR data.


TOMS image of the Mt. Pinatubo SO2 plume two days after the June 15th eruption The red dot marks the location of the volcano. The concentration of sulfur dioxide is expressed in units of milli-atmosphere centimeters, which gives the total column abundance in the atmospheric column Image created by Gregg Bluth and Arlin Krueger, NASA Goddard Space Flight Center.



The Total Ozone Mapping Spectrometer(TOMS) is used for high resolution mapping and measurements of the ozone layer. TOMS also detects volcanic eruptions and measures the amount of sulfur dioxide released. TOMS was used to measure the sulfur dioxide clouds from three explosive eruptions of the Crater Peak vent of Mount Spurr during the Summer of 1992.



The Advanced Very High Resolution Radiometer (AVHRR) is used to track the aerosol layer produced by eruptions. For example, during the Pinatubo eruption, AVHRR showed that the layer circled the Earth in 21 days and that it had inhomogeneities that persisted for more than two months. The layer covered 42% of the Earth after only two months, over twice the area of El Chichon in the same amount of time. Data gathered allowed an estimate of the net global cooling effect of 0.5 degree C for a period of 2-4 years after the eruption (Stowe and others, 1992). Maps show thickness of aerosol optical products prior to and 20 days after the 1991 eruption of Pinatubo. Photo credit: G.J., Orme, Department of the Army.

Monitoring Volcanic Gases

(By S. Grocke, 2010)

Understanding volcanic gases is essential to understanding how and why volcanoes erupt. First, it is important to realize that gases can be both dissolved in a magma chamber at depth and can be emitted from volcanoes at the surface. It is dissolved gases that cause volcanoes to erupt and it is gases emitted at the surface that can cause hazards and changes in climate.

Eyafallajokul volcano plume

Satellite imagery from NASA's Terra satellite, the Moderate Imaging Spectroradiometer instrument captured image on May 11 at 12:15 UTC (8:15 a.m. EDT), and shows a dark brown ash plume streaming south from Iceland's Eyjafjallajokull Volcano and over the waters of the North Atlantic Ocean. Credit: NASA Goddard / MODIS Rapid Response Team

Dissolved Gases   

A magma chamber at high pressures below the surface contains dissolved gases or volatiles. The density contrast between the magma and the surrounding rock, will allow the more buoyant magma to rise to the surface. As the magma ascends, dissolved gases come out of the liquid, or exsolve as tiny bubbles. Bubbles will grow and increase in volume, making the magma more buoyant and able to ascend even closer to the surface. As the magma travels upward, the overlying pressure decreases and the bubbles will expand and create a magmatic foam. It is not until the pressure in the bubbles becomes greater than the pressure of the overlying rock that the chamber will burst and produce a volcanic eruption.

The viscosity, temperature and composition of the magma will determine whether the eruption is explosive or effusive. If the rapidly expanding gas bubbles remain in contact with the liquid and cause the magma to fragment into volcanic rock, an explosive eruption will occur, like the 1980 eruption of Mount St. Helens. If the bubbles can rise through the liquid and escape, then the eruption will be more effusive and generate lava flows like those we see in Hawai’i.  It is the expansion of gases as they rise to the surface that drives volcanic eruptions.

The image to the right, from Oleg Melnik, shows a magma chamber at depth with dissolved volatiles. As the magma rises, the bubbles come out of solution producing vesiculated magma. The magma fragments into volcanic rock at the fragmentation level. Gas-particle dispersion occurs at the top of the conduit just below the surface and pyroclasts erupt onto the surface during a volcanic eruption.  


Emitted Gases

On a global scale, volcanic gases actually created the atmosphere and oceans that enable life to exist here on Earth. Once a volcano erupts, gases are emitted and released into the atmosphere. During large eruptions, gases have actually caused global climate change. In other systems, gases are emitted continuously into the atmosphere from soils, volcanic vents, fumaroles and hydrothermal deposits. The gradual release of gas acts as an irritant and may pose long-term health-hazards.  It is therefore important to monitor gases emitted from active volcanoes not only because they can pose very serious health risks and cause climate change but also because they are indicators of what is happening inside the volcano.

Plinian  Eruption of Lascar Volcano (Chile) and Schematic of Physical Processes

Image from:

The most abundant gas is water vapor (H2O), followed by carbon dioxide (CO2), and sulfur dioxide (SO2). Secondary gases are also commonly emitted from volcanoes and include hydrogen sulfide (H2S), hydrogen (H), carbon monoxide (CO), hydrogen chloride (HCl), hydrogen fluoride (HF), and helium (He).  The greatest potential hazards to humans, animals and agriculture are SO2, CO2 and HF. 



Additional References:

Symonds, R.B., Rose, w.I., Bluth, G., and Gerlach, T.M., 1994, Volcanic gas studies: methods, results, and applications, in Carroll, M.R., and Holloway, J.R., eds., Volatiles in Magmas: Mineralogical Society of America Reviews in Mineralogy, v. 30, p. 1-66.

Gerlach, T.M., 1991, Present-day CO2 emissions from volcanoes: Eos, Transactions, American Geophysical Union, Vol. 72, No. 23, June 4, 1991, pp. 249, and 254-255. 

Measuring Volcanic Gases

Because volatiles play an important role in the generation, evolution and eruption of magma, it is critical that we use various tools to monitor gases both within and emitted from a volcano. Advanced analytical techniques have been employed for measuring dissolved volatiles in volcanic rocks and in remote sensing technology used for analyzing volcanic emissions. These advancements have led to our most recent understanding of volatiles fluxes from volcanic eruptions.

There are three primary ways that gas geochemists collect data

1.) Measuring Dissolved Volatiles in Rocks/Minerals/Inclusions

2.) Direct Sampling

3.) Remote Sensing


Cleveland Volcano in the Aleutian Islands, Alaska, erupts; image courtesy of NASA


Dissolved Volatiles in Rocks/Minerals/Inclusions


Pre-eruptive dissolved volatile contents can be measured directly through the study of melt inclusions (MI) and fluid inclusions (FI).

A melt inclusion is defined as containing glass or crystallized glass.

A fluid inclusion contains no glass but rather one or more fluids at room temperature.

Because gases exsolve or come out of the liquid during eruption, the tephra erupted on the surface reveals little information on the original volatile content that existed in the magma chamber at depth. The tools that we use to measure preeruptive volatile contents directly include Fourier Transform Infrared spectrometry (FTIR) and Secondary Ion Mass spectrometry (SIMS). 

Melt Inclusions

Five naturally glassy, bubble-free MI of various sizes from Plinian fallout deposit of the Tara Ignimbrite, Chile. Image is 700 microns. 

Phenocrysts often trap small (1-300mm) fractions of silicate melt at magmatic temperatures and pressures during crystallization. Melt inclusions, incorporated into relatively incompressible phenocryst hosts like quartz, are able to retain the pre-eruptive volatile signature of the melt during eruption. This makes melt inclusions very useful tools in determining dissolved volatile concentrations directly. Through analytical techniques of the quenched inclusions, we can quantify the pre-eruptive concentrations of volatile gases such as H2O, CO2, S and Cl. Melt inclusion analyses can also provide a history of the evolution of a particular magmatic system.

Fluid Inclusions

Fluid inclusions are small droplets (<1mm) of fluid that like melt inclusions, are trapped within a phenocryst host. They are primarily two-phase, consisting of a liquid and a gas or vapor bubble. Fluid inclusions are remnant samples of very recent to ancient fluids that existed within the magmatic system. Studies of fluid inclusions allowed geologists to reconstruct the past history of the host rocks within which they are found.

 Image from USGS

Thanks to several decades of strong efforts to improve analytical techniques, we can now identify and analyze the volatile content of these tiny inclusions. MI data can faithfully record magmatic processes, which other petrological tools could not reveal. We have discovered that H2O and CO2 are the most abundant volatiles in most silicate magmas and that they play a major role in controlling the rheological properties of magmas, including the viscosity and density.

It is interesting to note that much of what we know about km-sized magma bodies

comes from analyses of micrometer-sized features!



Additional References:

de Vivo, B. A. Lima, & J. D. Webster, Volatile in Magmatic-Volcanic Systems, Elements, 1, 19-24. 

King P. L., T. W. Vennemann, J. R. Holloway, R. L. Hervig, J. B. Lowenstern, and J. F. Forneris, Analytical techniques for volatiles: A case study using intermediate (andesitic) glasses, American Mineralogist, 87, 2002. 

 Lowenstern, 1995. Melt Inclusions Come of Age: Volatiles, Volcanoes, and Sorby’s Legacy.  

Lowenstern, J.B.(1995) Applications of silicate melt inclusions to the study of magmatic volatiles. In: Thompson, J.F.H. (ed.) Magmas, Fluid and Ore Deposits. Mineralogical Association of Canada Short Course 23, 71-99. 

Volcanic Emissions - Plumes

How much gas is emitted from a volcano during a certain time period is directly related to the volume of magma that sits in the subsurface reservoir. Measuring the rate at which a volcano releases gas or degasses, typically reported in metric tons per days, allows scientists to get a glimpse of what is happening below the surface. Changes in gases like sulfur dioxide and carbon dioxide are important to monitor in active volcanic systems since they can be indicative of activity occurring in the volcano’s magma reservoir and hydrothermal system. Emission rates can be measured either from the ground or from an aircraft. Gas ejected high into the atmosphere during a volcanic eruption requires satellites to measure the emitted gas. 

Looking south across Halema`uma`u Crater at the gas plume rising from the Overlook vent. From USGS, HVO

Direct Sampling

The easiest, but perhaps the most dangerous way to collect a gas sample is by hand, placing a container directly in the gases. This technique is used to produce a detailed chemical analysis of a specific fumarole or vent, where a scientist can actually insert a tube into a hot opening. This method is ideal for long term study of volcanoes rather than for monitoring rapidly changing conditions. The technique requires days to weeks of laboratory analysis following sampling in order to get data.


Gas sampling from Baker, 1981


Direct sampling requires a scientist to insert a chemically inert and heat resistant tube into a hot opening like a fumarole or vent. It takes about 5 minutes for the tube to heat up to a point where any condensation within the tube has reached equilibrium with the escaping gases. Then, either by attaching an evacuated-sample bottle or a flow-through sample bottle to the collection tube, the gases will be gathered for analysis. 

USGS geologists collect gas samples around the dome of Mount St. Helens  


Evacuated-bottle method

The evacuated-bottle method is shown in the image to the right. The device includes a glass bottle with a sample port and a high vacuum-stopcock. Before arriving at the collection site, the bottle must be partially filled with concentrated aqueous sodium hydroxide (NaOH) that has been carefully weighted and evacuated with a vacuum pump. Once the tube is inserted into the fumarole or vent, the gases will bubble through the solution and gases like CO2, H2S, SO2, HCL and HF will dissolve into the liquid. Those gases that remain like N2, O2, H2, CO and He will rise further and collect in the headspace of the bottle.

This technique involves collecting the gases at the site where the gases are being emitted and then returning to the laboratory for analysis. This method is used because of its good analytical precision that stems from its ability to concentrate the gases in the solution and the headspace. Those gases that rise into the headspace are analyzed by gas chromatography. Those that dissolve into the liquid are analyzed by ion chromatography or traditional wet-chemical techniques.



Flow-through bottle method

The flow-through bottle method is shown in this image to the left. The device includes a glass bottle but with a stopcock at each end and a hand-operated pump attached to the sampling tube. The purpose of the hand pump is to flush out the air while entraining the gases into the bottle. This method is not as precise as the evacuated-bottle method, but is utilized in situations where sampling must be done rapidly due to hazardous environments and conditions. 




A detailed analysis has that advantage that it can provide the information necessary to reconstruct the conditions of the magma at depth, which is the source region for the emitted gases.



Additional References:

USGS website: Direct gas sampling and laboratory analysis 

Sutton, A.J., McGee, K.A., Casadevall, T.J., and Stokes, B.J., 1992, Fundamental volcanic-gas-study techniques: an integrated approach to monitoring: in Ewert, J.W., and Swanson, D.A. (eds.), 1992, Monitoring volcanoes: techniques and strategies used by the staff of the Cascades Volcano Observatory, 1980-90: U.S. Geological Survey Bulletin 1966, p. 181-188.

Remote Sensing

COSPEC (Correlation Spectrometer)  

A correlation spectrometer or COSPEC was initially designed to measure industrial pollutants and now has been applied to the field of volcanology to measure volcanic gas emissions. The spectrometer is designed to measure the concentration of sulfur dioxide (SO2) in the volcanic plume that is emitted from the volcano. The device requires a standard, from which to analyze the ultraviolet light absorbed by the SO2 molecules in the plume.

Multiple measurements are made to acquire reliable results. This COSPEC is used either from the ground where it is mounted on a vehicle or tripod that scans the plume, or the device can be attached to an aircraft that traverses underneath the plume. The best quality measurements are obtained when an aircraft flies at right angles to the direction of plume travel acquiring data with each flight.


These images show various ways that a COSPEC can be set up-on a tripod, in a vehicle or on an aircraft.

Images from USGS.




Average daily SO2 emission rates from Mount St. Helens from 1980-1988. COSPEC data were retrieved using a COSPEC mounted on an aircraft.

For data, see Open-File Report 94-212.




Infrared Carbon Dioxide Analyzer (LI-COR


An infrared carbon dioxide analyzer or Li-COR has become a standard method for measuring carbon dioxide (CO2) emission rates. It is employed in a similar manner to the COSPEC but requires data from the whole plume in order to calculate a carbon dioxide emission rate. The aircraft that hosts the device flies systematically through the plume creating a cross-section analysis of the gas emissions at different elevations.


This photo to the right is taken while flying under the volcanic plume to measure SO2- photo from USGS. 


The LI-COR can also be used to measure soil efflux emissions. These soil emissions are typically in areas where volcanic gases rise from depth and remain in the soil directly beneath the surface. To measure the rate of gas emissions into the atmosphere, a accumulation chamber is set up on the soil surface and connected to a LI-COR instrument. The gas enters the chamber and is measured for increasing CO2 concentrations. A soil efflux for that specific location is calculated based on other parameters that include, pressure, temperature. Additional efflux values at various locations must be measured to acquire reliable measurements that are representative of a volcanic system, from which a map can be constructed showing the elevated soil CO2 values.




From USGS- Map of CO2 concentration- constructed from data near Horseshoe Lake and Mammoth Mountain, California from


Gerlach, T.M., Doukas, M.P., McGee, K.A., and Kessler, R., 2001, Soil efflux and total emission rates of magmatic CO2 at the Horseshoe Lake tree kill, Mammoth Mountain, California, 1995-1999: Chemical Geology, v. 177, Issues 1-2, pgs. 101-116. 

USGS- Measuring volcanic gases; soil efflux  






Fourier Transform Infrared Spectrometer (FTIR)

The FTIR or Fourier Transform Infrared Spectrometer can be used to measure dissolved volatile concentrations as described above or can be used to measure several gases emitted from a volcano simultaneously. The device can be used both as an open-path or closed-path system. The open-path system aims the FTIR at a plume using an optical telescope. The closed-path system delivers gas from a plume or fumarole to a gas cell within the FTIR.  


Additional References:

Gerlach, T.M., Doukas, M.P., McGee, K.A., and Kessler, R., 2001, Soil efflux and total emission rates of magmatic CO2 at the Horseshoe Lake tree kill, Mammoth Mountain, California, 1995-1999: Chemical Geology, v. 177, Issues 1-2, pgs. 101-116. 

McGee, K.A., and Casacdevall, T.J., 1994, A Compilation of Sulfur Dioxide Emission-Rate Data from Mount St. Helens During 1980-1988. U.S. Geological Survey Open-File Report 94-212, Version 1.0 

Continuous Sampling

Continuous volcano monitoring stations can be used to gauge both short-lived degassing episodes that happen within minutes to hours as well as long-lived activity that happens over days to years. With advancing technology, scientists are able to set up a station to monitor gases from fumaroles, vents, soils, hydrothermal deposits, etc and transmit the data directly to an online directory or observation location. 

To monitor the activity at the Pu `u `O `o vent on the Big Island of Hawai’i, the Hawaiian Volcano Observatory (HVO) set up a monitoring station on the flanks of Kilauea’s east rift zone. Gas emissions as well as the wind speed and wind direction are periodically sampled and get transmitted every 10 minutes to HVO. This allows HVO to monitor degassing at the active vent almost instantaneously.