Observation of Passive and Explosive Emissions at Stromboli with a Ground-based Hyperspectral TIR Camera

Abstract:

Scientific imaging techniques have progressed at a fast pace in the recent years, thanks in part to great improvements in detector technology, and through our ability to process large amounts of complex data using sophisticated software. Broadband thermal cameras are ubiquitously used for permanent monitoring of volcanic activity, and have been used in a multitude of scientific applications, from tracking ballistics to studying the thermal evolution lava flow fields and volcanic plumes. In parallel, UV cameras are now used at several volcano observatories to quantify daytime sulfur dioxide (SO₂) emissions at very high frequency. In this work we present the results the first deployment of a ground-based Thermal Infrared (TIR) Hyperspectral Imaging System (Telops Hyper-Cam LW) for the study of passive and explosive volcanic activity at Stromboli volcano, Italy. The instrument uses a Michelson spectrometer and Fourier Transform Infrared Spectrometry to produce hyperspectral datacubes of a scene (320x256 pixels) in the range 7.7-11.8 µm, with a spectral resolution of up to 0.25 cm^-1 and at frequencies of ~10 Hz. The activity at Stromboli is characterized by explosions of small magnitude, often containing significant amounts of gas and ash, separated by periods of quiescent degassing of 10-60 minutes. With our dataset, spanning about 5 days of monitoring, we are able to detect and track temporal variations of SO₂ and ash emissions during both daytime and nighttime. It ultimately allows for the quantification of the mass of gas and ash ejected during and between explosive events. Although the high price and power consumption of the instrument are obstacles to its deployment as a monitoring tool, this type of data sets offers unprecedented insight into the dynamic processes taking place at Stromboli, and could lead to a better understanding of the eruptive mechanisms at persistently active systems in general.