GNGTS 2018 - 37° Convegno Nazionale

274 GNGTS 2018 S essione 1.3 Italiano and Nuccio, 1991, Maugeri et al. , 2010, Merkel et al., 2011, Monecke et al., 2012, Tassi et al., 2002). To assess variations in chemical composite, Heinicke et al. on in the very short period of the emitted gas, the fumarole with the highest flow was analyzed with an automated micro- gaschromatograph (m-GC) which offers high frequency and high sensitivity of gas analysis, allowing a possible alternative to the typical multigas monitoring of volcanic areas. The chromatographic monitoring station (CMS) is an in-situ micro-gaschromatograph implemented with a computer, a sampling system and a router to allow the remote complete control of the instruments and the automatic transmission of the data. Some specific parts are manufactured to adapt the characteristics of the m-GC to the sampling of gases from fumaroles, or from the ground or dissolved in the water. The CMS was placed on the beach about 50 m far from the main underwater gas emission (Fig. 1). The major perplexities concerned the methods of collecting the gas without the sea water entering in the system. To prevent this, a ballasted metallic funnel (similar to the one described by Italiano and Nuccio in 1991) was connected to the CMS with a plastic pipe for the first 30 meters, and then with a vacuum-type silicone hose near the shoreline to prevent tourists from inadvertently crushing it or producing creases, in order to allow a regular gas flow to the m-GC. The fumarole gas arrived almost dry from steam and the presence of liquid water in the sampling system was never observed. As a precaution, anyway, a small filter system was installed before the CMS to protect the m-GC introduction system. Data acquisition was carried out from 17 to 23 of September 2016 automatically performing an analysis every 10 minutes for a total numer of 963 gas analyses, without detecting any particular problem. Table 1 shows the average chemical composition of the gases and the relative standard deviation for the acquisition period. Relative standard deviation is very good (RSD ≤ 5 %) for H 2 , CH 4 , H 2 S and CO 2 . Helium RSD is higher because the measured gas concentrations (0.9 to 2.0 ppm) are slightly above the limit of detection of He for the micro-GC (1 ppm). The high RSD of O 2 and N 2 could be explained either by a variable amount of dissolved air in the sea water that interacts with the fumarolic gases, or by a small air contamination occurred between the sampled fumarole and the CMS. Tab. 1 - Main statistics of the chromatographic analyses. Gas component He (ppm) H 2 (ppm) O 2 (%) N 2 (%) CH 4 (ppm) H 2 S (ppm) CO 2 (%) Average 1,30 352,11 0,09 0,61 1092,77 28577,01 98,80 composition Relative Standard 11,82 1,55 39,86 13,39 2,58 5,06 0,12 Deviation (%) The concentrations of O 2 and N 2 (fig. 2) are stable during the acquisition time, except for a definite period of time (from 19 september 23:30 GMT to 20 september 7 a.m.) where a small air contamination has likely occurred. N 2 /O 2 average ratio is 7.45 in the not-contaminated interval (4.15 during 19 th late night to 20 th morning) much higher than the ration in air (~3.73) because a part of the oxygen continuously reacts with H 2 S leading to the formation of colloidal sulfur. It is necessary to check whether the N 2 /O 2 ratio undergoes significant changes following different wave and sea-stream conditions. A longer data acquisition would allow us to understand if interaction between the hydrothermal system and seawater is a very superficial process or if there is a slightly deeper circulation of fluids in which seawater is involved too. The concentration of H 2 during the acquisition period spanned 337 to 372 ppm, with very evident oscillations, not due to changes in the observed low-to-high atmospheric pressure variation (Fig. 3). During the period of acquisition no earthquake, even if of low magnitude,