Articles | Volume 9, issue 11
https://doi.org/10.5194/amt-9-5487-2016
https://doi.org/10.5194/amt-9-5487-2016
Research article
 | 
16 Nov 2016
Research article |  | 16 Nov 2016

Improving global detection of volcanic eruptions using the Ozone Monitoring Instrument (OMI)

Verity J. B. Flower, Thomas Oommen, and Simon A. Carn

Abstract. Volcanic eruptions pose an ever-present threat to human populations around the globe, but many active volcanoes remain poorly monitored. In regions where ground-based monitoring is present the effects of volcanic eruptions can be moderated through observational alerts to both local populations and service providers, such as air traffic control. However, in regions where volcano monitoring is limited satellite-based remote sensing provides a global data source that can be utilised to provide near-real-time identification of volcanic activity. This paper details a volcanic plume detection method capable of identifying smaller eruptions than is currently feasible, which could potentially be incorporated into automated volcanic alert systems. This method utilises daily, global observations of sulfur dioxide (SO2) by the Ozone Monitoring Instrument (OMI) on NASA's Aura satellite. Following identification and classification of known volcanic eruptions in 2005–2009, the OMI SO2 data, analysed using a logistic regression analysis, permitted the correct classification of volcanic events with an overall accuracy of over 80 %. Accurate volcanic plume identification was possible when lower-tropospheric SO2 loading exceeded ∼ 400 t. The accuracy and minimal user input requirements of the developed procedure provide a basis for incorporation into automated SO2 alert systems.

Download
Short summary
Volcanic eruptions pose a threat to human populations around the globe, but many active volcanoes are poorly monitored. This paper details the development of an automated volcanic plume detection method with focus on diffuse plumes, utilising daily, global observations of volcanic gases measured by satellites. The developed technique consistently distinguished volcanic plumes over 400 t, from control samples, indicating the potential for implementation within a volcanic alert system.