ACCUMULATION AND LEAKAGE OF CO2 IN VOLCANIC TERRAINS

Over the past 20 years improvements in instrumentation and refinement of field and statistical methods have advanced the detection and quantification of CO₂ emissions from leaking reservoirs of magmatic gas. When coupled with geochemical or geophysical data the results from studies of diffuse CO₂ emissions at volcanoes provide a means to assess changes in the state of the underlying magmatic system and better evaluate periods of unrest.

One of the most well studied sites of quiescent degassing is Mammoth Mountain in eastern CA, where large areas of tree-kill developed following a period of seismicity in 1989. Measurements of CO₂ flux over the past decade show episodic changes in emission rates and indicate that gas discharge may continue for years to come. The high CO₂ emission rates provide opportunities to improve the ability to detect and map upflows of gas, model the delay between seismicity and increased discharge, and better delineate external forces that produce variations in emissions rates. Results enable us to develop better tools to monitor less obvious areas of CO₂ leakage.

Accumulation chamber and eddy covariance flux measurements, ¹⁴C analysis of tree cores, and measurement of the dissolved carbon content and discharge of cold springs have been used to identify, map and quantify magmatic CO₂ discharge at Mammoth Mountain. Each method provides information over different time scales and areal extents. Accumulation chambers give an immediate one-time measure of the CO₂ efflux from point sources. The discharge and carbon content of spring waters also give a one-time measurement, but provide an integrated record over a larger footprint. Eddy covariance gives a nearly continuous integrated measure of CO₂ efflux after an initial period of record that might be many days long. Tree core analysis gives multi-year records of relative emission strength averaged over the growing season. The tree core method, while less quantitative, offers the chance to look back in time (annually) and detect the onset of CO₂ emissions. Although each of these methods are used to study CO₂ discharge at other volcanic systems, use of all methods in tandem is rare and not always feasible.