Bull Volcanol (2006) 68: 328–332 DOI 10.1007/s00445-005-0013-x SHORT SCIENTIFIC COMMUNICATION Glyn Williams-Jones · Keith A. Horton · Tamar Elias · Harold Garbeil · Peter J. Mouginis-Mark · A. Jeff Sutton · Andrew J. L. Harris Accurately measuring volcanic plume velocity with multiple UV spectrometers Received: 11 June 2003 / Accepted: 18 May 2005 / Published online: 10 December 2005 C  Springer-Verlag 2006 Abstract A fundamental problem with all ground-based remotely sensed measurements of volcanic gas flux is the difficulty in accurately measuring the velocity of the gas plume. Since a representative wind speed and direction are used as proxies for the actual plume velocity, there can be considerable uncertainty in reported gas flux values. Here we present a method that uses at least two time- synchronized simultaneously recording UV spectrometers (FLYSPECs) placed a known distance apart. By analyzing the time varying structure of SO 2 concentration signals at each instrument, the plume velocity can accurately be de- termined. Experiments were conducted on K¯ ılauea (USA) and Masaya (Nicaragua) volcanoes in March and August 2003 at plume velocities between 1 and 10 m s −1 . Con- current ground-based anemometer measurements differed from FLYSPEC-measured plume speeds by up to 320%. This multi-spectrometer method allows for the accurate remote measurement of plume velocity and can therefore greatly improve the precision of volcanic or industrial gas flux measurements. Keywords FLYSPEC . Plume velocity . Volcanic emissions . Ultraviolet correlation spectrometer Editorial responsibility: A. Woods G. Williams-Jones () Department of Earth Sciences, Simon Fraser University, 8888 University Drive, Burnaby, British Columbia, V5A 1S6, Canada e-mail: glynwj@sfu.ca Tel.: 604-291-3306 Fax: 604-291-4198 K. A. Horton · H. Garbeil · P. J. Mouginis-Mark · A. J. L. Harris Hawaii Institute of Geophysics and Planetology, University of Hawaii at Manoa, 1680 East-West Rd., Honolulu, Hawaii 96822, USA T. Elias · A. J. Sutton U. S. Geological Survey, Hawaiian Volcano Observatory, Hawaii National Park, P.O. Box 51, 51 Crater Rim Drive, Hawaii 96718-0051, USA Introduction Correlation spectrometry (COSPEC) has been used for over 30 years to determine and monitor the emission rates of in- dustrial plumes (SO 2 , NO 2 ) and degassing volcanoes (SO 2 ) (e.g., Moffat and Millan 1971; Stoiber et al. 1983). While there are a number of methods of making COSPEC mea- surements (stationary scanning, mobile ground or airborne traverses, etc.), the most commonly used technique involves ground-based vehicular traverses at some distance down- wind of the gas source (e.g., Elias and Sutton 2002; Stoiber et al. 1983; Williams-Jones et al. 2000). Gas flux is typically calculated by multiplying the average concentration-pathlength (ppm-m) of SO 2 by the plume width and average plume velocity. This calculation is thus strongly dependant on accurate knowledge of plume velocity. However, direct measurement of plume speed or direction is often exceedingly difficult and therefore, one generally measures a representative wind speed and direction as a proxy. Ideally, this information would be obtained from instruments in the gas plume (e.g., using a radiosonde, tethered balloon or in situ airborne measure- ments; Doukas 2002). However, for ground-based gas flux measurements, these data are often not available and thus other methods are required. Should the volcano be near an airport or large city, it is sometimes possible to obtain the wind speed and direction over a range of elevations from a local meteorological station (from radiosondes or ap- proaching aircraft). If this information is unavailable, wind speed measurements may be made using an anemometer (handheld or mast-mounted) and then factored into numerical relationships (log or power law) to estimate the wind speed at a given height up to ∼200 m above the ground level (e.g., Strataridakis et al. 1999). At volcanoes where the plume is close to the ground (e.g., K¯ ılauea, USA; Masaya, Nicaragua), this technique is useful (Elias and Sutton 2002; Williams-Jones et al. 2003). However, while the effects of ground layer shear and topographic effects may be reduced somewhat by mounting the instrument on a 10-m-high mast downwind by 10 times the height of