Vol48/04/2005def 787 ANNALS OF GEOPHYSICS, VOL. 48, N. 4/5, August/October 2005 Key words volcanology – geochemistry – soil de- gassing – monitoring – risk 1. Introduction The Azores Archipelago comprises nine volcanic islands in the Atlantic Ocean where the American, African and Eurasian lithospher- ic plates meet (Searle, 1980; fig. 1). The Mid- Atlantic Ridge clearly defines the northern and southern branches of the Azores Triple Junction (Lourenço et al., 1997) but the location and ge- odynamic relations of the boundary between the Eurasian and African plates, corresponding to the so-called Terceira Rift (Machado, 1959), is still controversial (Madeira and Ribeiro, 1990). The complexity of this geological set- ting, reflected by the seismic and volcanic ac- tivity, may bear on the inferred existence of a mantle plume (White et al., 1976). During more than 500 years of history about thirty important destructive earthquakes and twenty-eight volcanic eruptions have been re- ported in the archipelago, causing thousands of Monitoring of fumarole discharge and CO2 soil degassing in the Azores: contribution to volcanic surveillance and public health risk assessment Teresa Ferreira (1), João Luís Gaspar (1), Fátima Viveiros (1), Márcio Marcos (1), Carlos Faria (1) and Fernando Sousa (2) (1) Centro de Vulcanologia e Avaliação de Riscos Geológicos, Observatório Vulcanológico, Universidade dos Açores, Ponta Delgada, Açores, Portugal (2) Delegação Regional do Instituto de Meteorologia, Relvão, Ponta Delgada, Açores, Portugal Abstract Fluid geochemistry monitoring in the Azores involves the regular sampling and analysis of gas discharges from fumaroles and measurements of CO2 diffuse soil gas emissions. Main degassing areas under monitoring are as- sociated with hydrothermal systems of active central volcanoes in S. Miguel, Terceira and Graciosa islands. Fu- marole discharge analysis since 1991 show that apart from steam these gas emissions are CO2 dominated with H2S, H2, CH4 and N2 in minor amounts. Mapping of CO2 diffuse soil emissions in S. Miguel Island lead to the conclusion that some inhabited areas are located within hazard-zones. At Furnas village, inside Furnas volcano caldera, about 62% of the 896 houses are within the CO2 anomaly, 5% being in areas of moderate to high risk. At Ribeira Seca, on the north flank of Fogo volcano, few family houses were evacuated when CO2 concentra- tions in the air reached 8 mol%. To assess and analyse the CO2 soil flux emissions, continuous monitoring sta- tions were installed in S. Miguel (2), Terceira and Graciosa islands. The statistical analysis of the data showed that some meteorological parameters influence the CO2 flux. The average of CO2 flux in S. Miguel stations ranges from 250 g/m2/d at Furnas volcano to 530 g/m2/d at Fogo volcano. At Terceira Island it is about 330 g/m2/d and at Graciosa 4400 g/m2/d. Mailing address: Dr. Teresa Ferreira, Centro de Vul- canologia e Avaliação de Riscos Geológicos, Observatório Vulcanológico da Universidade dos Açores, Complexo Científico, Ala Sul, 3º Piso, 9501-801 Ponta Delgada, Açores, Portugal; e-mail: tf@notes.uac.pt 788 Teresa Ferreira, João Luís Gaspar, Fátima Viveiros, Márcio Marcos, Carlos Faria and Fernando Sousa deaths and severe damage. The July 9, 1998 Fa- ial earthquake (Senos et al., 1998) and the 1998-2001 submarine eruption west of Terceira Island (Gaspar et al., 2003) were the most re- cent events. Present-day volcanic activity in the Azores is marked by highly active fumarolic fields, hot springs and soils diffuse degassing phenomena. This paper reports the fluid geochemistry mon- itoring programme that is being undertaken for volcanic surveillance and public health risk as- sessment in the region. Special attention is giv- en to the CO2 diffuse emissions whose signifi- cance in the total degassing of volcanic areas is of major importance (Allard et al., 1991). 2. Degassing areas Degassing areas in the Azores are related to hydrothermal systems (Ferreira, 1994). The most important zones of degassing are located in S. Miguel, Terceira and Graciosa islands and at some submarine volcano-tectonic structures along the Terceira Rift. Less significant de- gassing fields exist in Faial, Pico, S. Jorge and Flores islands (Ferreira, 1994; Viveiros, 2003). There are no reported gas emanations in Santa Maria or Corvo islands. The monitored degassing areas in S. Miguel Island are associated with the Furnas and Fogo active central volcanoes. Furnas village (M- FV) and Furnas Lake (M-FL) fumarolic fields are located inside the Furnas caldera (fig. 2) defining a WNW-ESE path compatible with a major regional tectonic trend that traverse the island (Gaspar et al., 1995). Pico Vermelho (M- PV), Caldeira Velha (M-CV) and Caldeiras da Ribeira Grande (M-CRG) fumaroles are on the northern flank of the Fogo volcano associated with the NW-SE Ribeira Grande graben (fig. 3). In Terceira Island monitoring is focussed in the Furnas do Enxofre (T-FE) fumarolic field (fig. 4). The area covers a part of the Galhardo trachytic lava dome and extends beyond its crater being located in the intersection of NW-SE, E-W and NNE-SSW faults (Lloyd and Collis, 1981). At Graciosa Island the main hydrothermal area under surveillance is within the Furna do Enxofre lava cave (G-FE). This volcanic struc- ture was formed at the end of an intracaldera la- va lake episode and is part of the main conduit of the Caldera volcano, controlled by NE-SW and NW-SE faults (Gaspar, 1996; fig. 5). Fig. 1. Geographic location of the Azores Archipelago. 789 Monitoring of fumarole discharge and CO2 soil degassing in the Azores Fig. 2. Furnas volcano, S. Miguel Island, fumarolic fields (M-FL – Furnas Lake; M-FV – Furnas Village). Fig. 3. Fogo volcano, S. Miguel Island, fumarolic fields (M-PV – Pico Vermelho; M-CRG – Caldeiras da Ribeira Grande; M-CV – Caldeira Velha). 790 Fig. 4. Furnas do Enxofre fumarolic field at Terceira Island (T-FE – Furnas do Enxofre). Fig. 5. Furnas do Enxofre fumarolic field at Graciosa Island (G-FE – Furnas do Enxofre). Teresa Ferreira, João Luís Gaspar, Fátima Viveiros, Márcio Marcos, Carlos Faria and Fernando Sousa 791 Monitoring of fumarole discharge and CO2 soil degassing in the Azores 3. Fumaroles regular monitoring programme 3.1. Sampling and analytical processes The procedures for fumarole gas sampling are described in Ferreira (1994) and Ferreira and Oskarsson (1999) and follow the so-called Gig- genbach methodology (Giggenbach and Gogel, 1989) using a known volume of a strong KOH solution (12 M) for adsorption of the acid gases. The samples are collected by placing a funnel, connected to a silicone tubing sampling train, over the steam vents. Gas flowing through the tubing system is collected into an evacuated borosilicate sample-bottle containing a measured weight of the KOH solution. CO2 and H2S are adsorbed in this solution, which becomes diluted with the condensed steam, while the fixed gases H2, Ar, CH4, N2 and O2 fill the remaining space. After being equilibrated to atmospheric pres- sure, gases in the sample-bottle are analysed with a gas chromatograph (Perkin-Elmer) with thermal conductivity detector, automated fixed sample loop, molecular sieve column and He as carrier gas. Ar is analysed in a separate run after washing the gas sample with basic pyrogallol so- lution. CO2 and H2S dissolved in the basic solution are quantified by titrimetric methods. CO2 is de- termined by automated titration with hydrochlo- ric acid and H2S is titrated with mercury acetate using dithizone for end point detection. A small volume of the basic solution is weighted for den- sity determination, needed for mass balance in- volved in total gas analysis quantification. 3.2. Results Occasional monitoring of S. Miguel (M-FL, M-FV, M-CV, M-CGR), Terceira (T-FE) and Graciosa (G-FE) fumaroles started in 1991 but regular measurements were started within the present programme. The available data show that CO2 is the main constituent of the total gas in the steam discharges, representing 94 to 99.6 mol%. H2S and H2 are the other major compo- nents contributing for 0.1 to 3 mol% of the to- tal gas. Nitrogen concentration is below 1.5 mol% in all samples, the highest values being found at T-FE. CH4 values are in general lower than 0.08 mol% except at T-FE where they can be 10 times higher. All monitored fumaroles have discharges of broadly similar composition. However, on Ter- Fig. 6. H2/CH4 relation for the fumaroles under monitoring at the Azores Archipelago. References as in text. 792 Teresa Ferreira, João Luís Gaspar, Fátima Viveiros, Márcio Marcos, Carlos Faria and Fernando Sousa ceira Island fumaroles are richer in H2 and CH4. The differences between gas compositions are better displayed using the ratio H2/CH4. In fact these gases show the most regular abundances of all the analysed components allowing a dis- crimination between the fumarolic fields (fig. 6). Based on equilibrium relations of H2 and CH4 and assuming a deep origin for these gases a simple model was applied to S. Miguel fuma- role discharges considering a binary carbon dioxide-steam system (Ferreira and Oskarsson, 1999). According to this model these gases are derived from supercritical conditions, presum- ably equilibrated within hot rocks below the hy- drothermal systems. 4. Mapping of CO2 soil concentrations 4.1. Methodology Analysis of CO2 soil concentrations at the Azores follows the methodology applied by Baubron et al. (1994) and Ferreira and Gaspar (1997). An iron probe 1 m long is buried in the soil at a depth of 50-70 cm and the gas is pumped into a portable CO2 analyzer with an infrared detector. The instrument is calibrated to measure concentrations between 0 and 100 mol%. Soil temperature is measured at the same sites as the CO2 using a digital thermome- ter with 0.1ºC precision. The distance between sampling sites de- pends on the observed values, being shorter, in the order of a few meters, where the concentra- tions are above the ground noise level (0 to 1.5 mol%). The collected data are inserted in a GIS and the anomaly grid maps are produced for CO2 and temperature by means of the kriging statistical method. 4.2. Results CO2 soil concentration maps have been pro- duced in the Azores for hazard and risk assess- ment. Baubron et al. (1994) showed the exis- tence of an important anomaly in Furnas vil- lage, within the Furnas Caldera in S. Miguel Is- land, and Baxter et al. (1999) discussed the public health risk associated with such hy- drothermal area. Fig. 7. CO2 soil concentrations at Furnas village, S. Miguel Island (modified from Sousa, 2003). 793 Monitoring of fumarole discharge and CO2 soil degassing in the Azores A new survey recently performed in the same zone led to the conclusion that the general shape and magnitude of the anomaly remain al- most identical 10 years later (fig. 7). Soil CO2 concentrations reach 95 mol% in some areas and soil temperatures can be as high as 100ºC. These observations indicate that on a large tem- poral scale the Furnas village anomaly map can be used as a reference level concerning Furnas volcano surveillance. With respect to public health risk assessment, Sousa (2003) calculate that 62% of the 896 houses at Furnas village are within this geochemical anomaly, 5% being lo- cated in areas of moderate to high risk according to the classification of Baubron et al. (1994). Public health risk at S. Miguel Island was also reported by Ferreira and Gaspar (1997) at Ribeira Seca village, about 3 km NW of the Pi- co Vermelho geothermal area. In April 1997 a CO2 geochemical anomaly affecting an area of approximately 250 m2 was detected but regular observations until the end of 1997 revealed that the anomaly pattern remained unchanged. At that time the maximum CO2 soil concentration was about 13.5 mol% and measured soil tem- peratures reached 50°C. In March 1998 a new survey carried out in the zone showed that the anomaly increased to an area of about 10000 m2 (fig. 8). CO2 soil concentration levels of 24 mol% and soil temperatures of 50°C were then detected and four families were evacuated from their residences because CO2 concentrations in- side houses reached values of 8 mol% (Gaspar et al., 1998). Regular monitoring surveys done since then show a slight decrease of the anom- aly area but the maximum observed values re- main similar. CO2 soil concentration maps are also useful for identification of major active tectonic faults (Sugisaki et al., 1983; Giammanco et al., 1998; Baubron et al., 2002). In the Azores, this tech- nique has proved to be very efficient consider- ing that young pyroclastic deposits often bury the active fault areas of the main trachytic cen- tral volcanoes (Faria et al., 2003; Marcos et al., 2003). Fig. 8. CO2 soil concentrations at Ribeira Seca in 1998, S. Miguel Island (modified from Ferreira, 2000). 794 Teresa Ferreira, João Luís Gaspar, Fátima Viveiros, Márcio Marcos, Carlos Faria and Fernando Sousa 5. CO2 soil flux continuous monitoring 5.1. Monitoring network and sampling Continuous monitoring of CO2 soil diffuse degassing at the Azores started in October 2001. The first permanent station was installed on S. Miguel Island in a large degassing area that extends into Furnas village (station GFUR1) inside the caldera of the Furnas vol- cano. A second station was set up on the island on February 2002 in the northern flank of the Fogo volcano, within the Pico Vermelho geot- hermal area (GFOG1). On December 2002 two other stations were installed at Terceira Island, in Furna do Enxofre fumarolic field (GTER1), and at Graciosa Island, inside the Furna do Enxofre lava cave (GGRC1). All CO2 soil flux continuous monitoring stations are based on the accumulation chamber method (Chiodini et al., 1998). The station GGRC1 is also equipped with a sensor for H2S flux measurements. All the stations include ad- ditional sensors to record information on mete- orological and environmental parameters such as barometric pressure, air temperature, air hu- midity, wind speed and direction, rainfall, soil water content and soil temperature. Flux stations are configured to read data every hour. Each CO2 flux measurement is made during a time period of 1-3 min and the value is registered at the same time as the data from all the above-mentioned sensors. Addi- tional information on the operating status of the equipment is also available. Data transmission links from the field stations to the volcanologi- cal Observatory of the Azores University, on S. Miguel Island, were established using GSM (GFUR1), freewave (GFOG1 and GTER1) and UHF plus freewave (GGRC1). 5.2. Results The Azores climate is typically oceanic such that low pressure areas are followed by in- creased wind speed and increased precipitation that strongly affect the soil degassing. Knowl- edge of how meteorological parameters inter- fere with diffuse soil degassing at any particu- lar monitoring site is crucial to recognise changes due to deep processes. The influence of external factors on CO2 flux is gauged by ap- plying multiple regression analysis (Draper and Smith, 1981) and time series analysis model- ling (Box and Jenkins, 1976). For S. Miguel Island the results obtained show that the main influencing parameters do not act in the same way on each monitoring site (Viveiros et al., 2003). At GFUR1 barometric pressure, rainfall, soil water content, air tem- perature and soil temperature are the parame- ters that have significant statistical connection to CO2 flux fluctuations. At GFOG1 station wind speed, barometric pressure, soil tempera- ture, soil water content and rainfall are the sta- tistically relevant influencing parameters (Viveiros, 2003). The average of CO2 flux at GFUR1 is near 250 g/m2/d while at GFOG1 it is around 530 g/m2/d. The recent CO2 flux data of Terceira and Graciosa islands are still under analysis. How- ever, the preliminary study of the obtained data showed that the average CO2 flux is about 330 g/m2/d at GTER1 and 4400 g/m2/d at GGRC1. In a general way, the CO2 flux averages ob- served in the above mentioned hydrothermal zones are lower than the ones reported for oth- er volcanic areas, such as Solfatara and Vesuvio (Granieri et al., 2003). The Furna do Enxofre soil degassing area at Graciosa Island is the ex- ception. 6. Future developments The Volcanological Observatory of the Azores University is responsible for the seismic and volcanic monitoring of the Azores region using geophysical, geodetic and geochemical techniques. Geophysical monitoring includes the management of a seismological network of more than 40 seismic stations, run in collabora- tion with the Meteorological Institute. Geodetic monitoring is based on a GPS network of sev- eral permanent receivers and complemented by regular GPS field surveys. As discussed, the ac- tual geochemistry monitoring programme in- volves the regular sampling and analysis of gas discharges in fumaroles and the study of CO2 795 Monitoring of fumarole discharge and CO2 soil degassing in the Azores diffuse soil gas emissions. To a minor extent, hot springs, geothermal and water wells are al- so under observation. The development of the Azores seismovol- canic network comprises the set up of warning and alert systems based on the integration of all the monitoring data. This strategy implies the extension of the monitoring programme to all the islands and will benefit from the data trans- mission system that is being installed in the ar- chipelago for civil protection activities. The recognition of hazardous zones due to the permanent diffuse soil degassing in several inhabited or tourist places requires the installa- tion of alarm sensors in order to reduce the pub- lic health risk. Such preventive measure is al- ready being implemented at Furna do Enxofre lava cave, where two people died in 1992 due to an increase in CO2 in the atmospheric air (Ferreira et al., 1993; Gaspar, 1996). Acknowledgements This work was performed in the scope of the project «ALERTA-INTERREG IIIB – Reducing Volcanic Risks in Islands» supported by the Eu- ropean Community and the project «SIMOVA2 – Risk assessment and emergency planning in the Azores Archipelago» supported by the Region- al Government through the SRPCBA. REFERENCES ALLARD, P., J. CARBONNELLE, D. DAJLEVIC, J. LE BRONEC, P. MOREL, M.C. ROBE, J.M. MAURENAS, R. FAIVRE- PIERRET, D. MARTIN, J.C. SABROUX and P. ZETTWOOG (1991): Eruptive and diffuse emissions of CO2 from Mt. Etna, Nature, 351, 387-391. BAUBRON, J.-C., P. BAXTER, R. COUTINHO, P. ALLARD, T. FERREIRA and J.L. GASPAR (1994): Gas hazard in dwellings Furnas map. Ec environment programme, Furnas laboratory volcano, health aspects, Open File Rep. No. 1, pp. 6. BAUBRON, J.-C., A. RIGO and J.P. TOUTAIN (2002): Soil gas profiles as a tool to characterise active tectonics areas: the Jaut Pass example (Pyrenees, France), Earth Plan- et. Sci. Lett., 196, 69-81. BAXTER, P., J.-C. BAUBRON and R. COUTINHO (1999): Health hazards and disaster potential of ground gas emissions at Furnas volcano, São Miguel, Azores, J. Volcanol. Geotherm. Res., 92, 95-106. BOX, G.E. and G.M. JENKINS (1976): Time Series Analysis. Forecasting and Control (Holden-Day, Inc.), pp. 575. CHIODINI, G., R. CIONI, M. GUIDI, B.E. RACO and L. MARI- NI (1998): Soil CO2 flux measurements in volcanic and geothermal areas, Appl. Geochem., 13, 543-552. DRAPER, N. and H. SMITH (1981): Applied Regression Analysis (John Wiley & Sons Inc.), 2nd edition, pp. 709. FARIA, C., J.L. GASPAR, T. FERREIRA and F. SOUSA (2003): Tectonic structures revealed by CO2 soil diffuse de- gassing anomalies at Faial island (Azores), in EGS- AGU-EUG Joint Assembly, 6-11 April, France, Nice (poster). FERREIRA, T. (1994): Contribuição para o estudo das em- anações gasosas associadas a processos de vulcanismo no Arquipélago dos Açores, Tese APCC (Universidade dos Açores), pp. 183. FERREIRA, T. and J.L. GASPAR (1997): Cartografia da anom- alia geoquímica da Ribeira Seca, Documento Técnico- Científico do Centro de Vulcanologia 06/DGUA/97, pp. 10. FERREIRA, T. and N. OSKARSSON (1999): Chemistry and iso- topic composition of fumarole discharges of Furnas caldera, J. Volcanol. Geotherm. Res., 92, 169-179. FERREIRA, T., J.L. GASPAR and G. QUEIROZ (1993): Consid- erações sobre as emanações gasosas da Furna do Enx- ofre (ilha Graciosa, Açores), Açoreana, 7 (4), 603-612. GASPAR, J.L. (1996): Ilha Graciosa (Açores): história vul- canológica e avaliação do hazard, Tese de Doutora- mento no ramo de Geologia, especialidade de Vul- canologia (Universidade dos Açores, Departamento de Geociências), pp. 361. GASPAR, J.L., T. FERREIRA, G. QUEIROZ, N. WALLENSTEIN, J. PACHECO, J. GUEST, A.M. DUNCAN and P. COLE (1995): Evolução morfoestrutural do Vulcão das Furnas (ilha de S. Miguel, Açores), Memórias (Museu e Labo- ratório Mineralógico e Geológico, Faculdade de Ciên- cias, Universidade do Porto), 4, 999-1003. GASPAR, J.L., T. FERREIRA, G. QUEIROZ and J.M. PACHECO (1998): CO2 soil degassing and risk assessment in the in- habited area of Ribeira Seca (S. Miguel Island, Azores), in Volcanic Hazard Assessment, Monitoring and Risk Mitigation. Abstracts Book of the EC Advanced Study Course, edited by J.L. GASPAR, J. GUEST, G. SIGVALDA- SON, J.L. CHEMINÉE and F. FERRUCCI, pp. 98. GASPAR, J.L., G. QUEIROZ, J.M. PACHECO, T. FERREIRA, N. WALLENSTEIN, M.H. ALMEIDA and R. COUTINHO (2003): Basaltic lava balloons produced during the 1998-2001 Serreta Submarine Ridge Eruption (Azores), in Explosive Subaqueous Volcanism, edited by J. WHITE, D. CLAUGE and J. SMELLIE, Am. Geophys. Un., 140, 205-212. GIAMMANCO, S., S. GURRIERI and M. VALENZA (1998): Anomalous soil CO2 degassing in relation to faults and eruptive fissures on Mt. Etna (Sicily, Italy), Bull. Vol- canol., 60, 252-259. GIGGENBACH, W.F. and R.L. GOGUEL (1989): Collection and analysis of geothermal and volcanic water and gas sam- ples, DSIR New Zealand, Report No. CD 2387, pp. 53 GRANIERI, D., G. CHIODINI, W. MARZOCCHI and R. AVINO (2003): Continuous monitoring of CO2 soil diffuse de- gassing at Phlegraean Fields (Italy): influence of envi- ronmental and volcanic parameters, Earth Planet. Sci. Lett., 212, 167-179. 796 Teresa Ferreira, João Luís Gaspar, Fátima Viveiros, Márcio Marcos, Carlos Faria and Fernando Sousa LLOYD, E.F. and S.K. COLLIS (1981): Geothermal Prospec- tion – Ilha Terceira, Açores (Secretaria Regional do Comércio e Indústria, Laboratório de Geociências e Tecnologia, Ponta Delgada), pp. 96 (unpublished rep.). LOURENÇO, N., J. LUÍS and M. MIRANDA (1997): Azores triple junction bathimetry, Carta Editada pela Facul- dade de Ciências da Universidade de Lisboa e pela Universidade do Algarve. MACHADO, F. (1959): Submarine pits of the Azores plateau, Bull. Volcanol., Série II, Tome XXI, 109-116. MADEIRA, J. and A. RIBEIRO (1990): Geodinamic models for the Azores triple junction: a contribution from tecton- ics, Tectonophysics, 184, 405-415. MARCOS, M., J.L. GASPAR, T. FERREIRA and F. VIVEIROS (2003): Cartografia da zona de desgaseificação difusa de CO2 na área geotérmica do Pico Vermelho (ilha de S. Miguel, Açores), in Proceedings do 3º Simpósio de Meteorologia e Geofísica da APMG, 28-33. SEARLE, R. (1980): Tectonic pattern of the Azores spreading centre and triple junction, Earth Planet. Sci. Lett., 51, 415-434. SENOS, M.L., J.L. GASPAR, J. CRUZ, T. FERREIRA, J.C. NUNES, J. PACHECO, P. ALVES,G. QUEIROZ, P. DESSAI, R. COUTIN- HO, D. VALES and F. CARRILHO (1998): O terramoto do Faial de 9 de Julho de 1998, in Proceedings do 1º Sim- pósio de Meteorologia e Geofísica da APMG, 61-68. SOUSA, F. (2003): Cartografia de anomalias geoquímicas associadas a processos de desgaseificação difusa no Vulcão das Furnas (S. Miguel, Açores): contribuição para a avaliação de riscos vulcânicos, Tese de Mestra- do no Ramo de Geologia Especialidade Vulcanologia e Riscos Geológicos (Universidade dos Açores), pp. 106. SUGISAKI, R., M. IDO, H. TAKEDA, Y. ISOBE, Y. HAYASHI, N. NAKAMURA, H. SATAKE and Y. MIZUTANI (1983): Origin of hydrogen and carbon dioxide in fault gases and its relation to fault activity, J. Geol., 91 (3), 239-258. VIVEIROS, F. (2003): Contribuição para o estudo dos proces- sos de desgaseificação difusa nos Açores no âmbito da monitorização sismovulcânica e da avaliação do risco: discriminação de factores que influenciam a variação do fluxo de CO2, Tese de Mestrado no Ramo de Geolo- gia Especialidade Vulcanologia e Riscos Geológicos (Universidade dos Açores), pp. 140. VIVEIROS, F., T. FERREIRA, C. VIEIRA and J.L. GASPAR (2003): The influence of environmental parameters on CO2 soil diffuse degassing in S. Miguel Island (Azores), in EGS- AGU-EUG Joint Assembly, 6-11 April, Nice, France (poster). WHITE, M., J. SCHILLING and S. HART (1976): Evidence for the Azores mantle plume from strontium isotope geo- chemistry of the Central North Atlantic, Nature, 263, 659-663.