GNGTS 2017 - 36° Convegno Nazionale

158 GNGTS 2017 S essione 1.2 FIRST RESULTS OF A TRI-AXIAL FIBER BRAGG GRATING STRAIN SENSOR U. Giacomelli 1 , D. Carbone 2 , S. Gambino 2 , E. Maccioni 1 , M. Orazi 3 , R. Peluso 3 , F. Sorrentino 4,5 1 Dipartimento di Fisica, Università di Pisa, Italy 2 INGV - Osservatorio Etneo, Sezione di Catania, Italy 3 INGV - Osservatorio Vesuviano, Napoli, Italy 4 Marwan Technology Srl, Pisa, Italy 5 Istituto Nazionale di Fisica Nucleare, Sezione di Genova, Italy Introduction. Rock strains detection is one of the principal ways to monitor geohazards. Stress and strain changes are among the best indicators of impending volcanic activity. In volcano geodesy, borehole volumetric strain-meters are mostly utilized. However, they are not easy to install and involve high implementation costs. Classic strainmeters are cumbersome, hard to install and very expensive. Opto-electronics devices based on Fiber Bragg Grating (FBG) allows an answer to the request of having good sensitivity and easiness of installation, together with reduced overall cost. Moreover fiber optic based devices offer small size, wide frequency band, and even the possibility of creating a local network with several sensors linked in an array. In the framework of the MED-SUV project (MEDiterranean SUpersite Volcanoes, 2013) with the aim of Etna volcano activity monitoring, we have realized, tested and installed a prototype of a tri-axial strainmeter using FBGs. The installation site is Serra La Nave (Catania) about 7 km SW far from the mountain peak, at the premises of the Istituto Nazionale di Astrofisica (INAF) observatory at an elevation of about 1780 m. The device is installed in a 8.5 meters deep borehole. The main goal of our work is the realization of a tri-axial device having a high resolution and accuracy in static and dynamic strain measurements, paying attention to the trade-off among resolution, cost and power consumption. The sensor structure and its read- out system are innovative in their assembly and offers practical advantages in comparison with traditional strain meters. As a demonstration of the performances of our device, the data of the first 15 months of operation are shown. Sensor description. The sensor is formed by a concrete pillar (40 cm x 10 cm height x diameter) containing three independent orthogonal FBG strain sensors and a temperature probe (Fig. 1). Once cemented in a well, the pillar is deformed by the stress of the surrounding rock and each embedded Bragg probe senses the respective axial strain. The optical signal from the Bragg sensors is linked by an optical fiber cable to a surface opto-electronic read-out system and then acquired. As verified in laboratory tests and confirmed in situ by regional and teleseismic events, this sensor has shown a sensitivity of the order of 10 nanostrains on its vertical axes; 30 nanostrains on the East axes and some hundreds nanostrains on the North axes. The different sensitivity of each axes comes from the non-optimised nature of the individual sensors and of the read-out system. However this first sensor prototype has the aim to trace the road towards second generation systems and now we know which parameters must be tuned to reach the resolution of few nanostrains on each axes. Because of the shallow depth of the well, the performance is limited by the thermoelastic traction effect, that for the three axes ranges from hundreds nanostrains to some microstrain on a daily timescale. Moreover hard-rain events largely affect the signal. To avoid Fig. 1 - Left: assembled sensor. Right: the FBGs before embedding in the concrete pillar.