Earth-prints repository, logo   DSpace

About DSpace Software
|earth-prints home page | roma library | bologna library | catania library | milano library | napoli library | palermo library
Please use this identifier to cite or link to this item: http://hdl.handle.net/2122/10041

Authors: Coltelli, M.*
Cristaldi, A.*
Mangiagli, S.*
Nunnari, G.*
Pecora, E.*
Title: A new approach for evaluating the strombolian type activity
Issue Date: 7-Sep-2003
Keywords: Strombolian Activity, Automatic Monitoring System
Abstract: Volcanoes represent one of the most important application fields for risk mitigation techniques (Mc Guire, Kilburn, and Murray, Eds, 1995) due to the ingent damages and casuality. This paper deals with a monitoring system that could be successful used for risk mitigation related to forecast paroxysmal explosions of strombolian type. The typical activity of Stromboli volcano consists of intermittent mid explosions lasting a few seconds, which take place at different vents and at variable intervals, the most common time interval being 10-20 minutes (Chouet, Hamisevicz, and McGetchin, 1974; Blackburn, Wilson, and Sparks, 1976). However the routine activity can be interrupted by more violent paroxysmal explosions, that eject m-sized scoriaceous bombs and lava blocks to a distance of several hundreds of meters from active vents. On average, one or two paroxysmal explosions occurred per year over the past century, buth this statistic may be underestimated in absence of continuous monitoring. For this reason from summer 1996 a remote surveillance camera works on Stromboli recording continuously the volcanic activity. It is located on Pizzo Sopra la Fossa, 100 metres above the crater terrace where are the active vents. Moreover, from September 2001 an on-line image analyzer system referred to as VAMOS (Volcanic Activity MOnitoring System) operates detection and classification of explosive vents in real-time. Using VAMOS it is possible to identify changes in the explosive activity trend that could precede a particular eruptive event, like paroxysmal explosions, fire fountains, lava flows. Since the camera installation up to present, several explosions occurred at the different craters and the parameterization in classes of intensity for each explosion on the base of tephra dispersion and kinetic energy have been made. The analysis include the counting of the explosions occurred at the different craters and the parameterisation in classes of intensity for each explosion on the base of tephra dispersion and, in generic sense, of the kinetic energy released. The plot of dissipated energy by each crater versus time seems to exhibit a cyclic behaviour with max and min explosive activity ranging from a few days to a month. Often the craters show opposite trends so when the activity decreases in a crater, increases in the other. Before every paroxismal explosion recorded, the crater that produced the event decreased and then stopped its activity from a few days to weeks before. The other crater tried to compensate increasing its activity and when it declined the paroxysmal explosion occurred suddenly at the former site (Coltelli, and Cristaldi, 2003). The system has automatically recorded and analysed the change in energetic trend that proceded the 20 October 2001 paroxysmal explosion that killed a woman and the strong explosive activity that preceded the onset of 28 December 2002 lava flow eruption.
Appears in Collections:Conference materials
04.08.06. Volcano monitoring

Files in This Item:

File Description SizeFormatVisibility
1c-7.pdfMain article856.66 kBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.


Share this record
Del.icio.us

Citeulike

Connotea

Facebook

Stumble it!

reddit


 

Valid XHTML 1.0! ICT Support, development & maintenance are provided by CINECA. Powered on DSpace Software. CINECA