Near Real-Time Isotopic Measurements of Carbon Dioxide from Outgassing Volcanoes

Abstract:

For the past several years we have been using a Picarro G1101-i isotopic mass analyzer to study the behavior of carbon dioxide emanating from active volcanoes. Because of its portability (it weighs about 30 kg), the instrument accompanies us on our field campaigns. Typically, we collect gas samples during the day and analyze them in the evening. The result is near-real-time isotopic measurements of CO2, and we are thus able to plan and adjust our field campaigns according to the results that we obtain on a continual basis. This is the primary advantage of the instrument. The G1101-i requires about 350 watts of power, typically provided by wall current with an uninterruptible power supply between the wall and instrument to deal with power fluctuations and outages. We calibrate the instrument every 2-5 days with a series of four well-characterized gas standards which we bring with us into the field in evacuated glass containers. Calibrations are typically robust and highly linear, with sub per mil precision. We also normally obtain a few samples which we analyze both by the G1101-i and later by mass spectrometry, in order to provide an independent means of checking our accuracy. Standards and samples are typically analyzed at similar CO2 concentrations to minimize any concentration-dependent effects on the isotopic analysis, even though these are generally small to negligible. Our applications so far have been focused at one caldera system and one subduction-related stratovolcano. We have analyzed soil gases at Long Valley caldera, California, to study the interplay of volcanic and tectonic controls upon diffuse CO2 release. We have analyzed CO2 in the the plume of Turrialba volcano, Costa Rica, to identify the volcanic isotopic signal and understand the mixing of the plume with surrounding atmosphere. At both localities, with appropriate dilutions as needed, we have been able to analyze the isotopic signal for CO2 concentrations ranging from atmospheric (400 ppm) to 100 per cent, demonstrating the large flexibility of our measurements. In summary, this approach is highly cost-effective, robust, and accurate for field campaigns at active volcanoes.