Jean-Paul Toutain

Soil gas profiles as a tool to characterise active tectonic areas: the Jaut Pass example (Pyrenees, France)

A new method to investigate active tectonic structures, using soil gas composition at faults, suggests relevant information about regional stress conditions which can be obtained rapidly and at relatively low cost. In 1995, we carried out geochemical profiles around the epicentre of the M=5.1 1980 earthquake near Arudy in the French Pyrenees, where the presence of minor fractures is evident in the field, and confirmed using satellite SPOT imagery. Fractures are conduits facilitating fluid migrations in the crust, and are also pathways for the release of deep-seated gases to the atmosphere. In order to investigate the implication of these fractures in the present deformation, i.e. if they are connected to the Hercynian substratum at a depth of about 1800 m, soil gases were measured along four traverses crossing the observed structures. Gases determined were 222Rn, CO2 and 4He, each of them for their characteristic source: 222Rn has essentially a shallow origin due to its short half-life, whereas CO2 is the major soil gas component with a mainly surficial biogenic source. However, there could additionally be CO2 from crustal or mantle degassing, which would also be the principal sources of He. Data analysis clearly reveals anomalous values for each gas at specific positions along the traverses. Two sets of fractures corresponding to different observed trends are distinguished: the one characterised by He anomalies accompanied by other gases, and the second with few identified He anomalies. The agreement between the geochemical data and the field observations leads us to propose a deformation model for the area studied, analogous to a pull-apart system located in a right lateral shear zone.

Continuous monitoring of distal gas emanations at Vulcano, southern Italy

The increasing activity of Vulcano Island (Italy) since 1985 led to the initiation of continuous geochemical monitoring of the lateral soil gas emissions. On the basis both of their relative geochemical characteristics and of local considerations, three gaseous components were selected for monitoring, namely CO2, He and 222Rn. Monitoring has been performed by means of specific analysers. Gases extracted from a water well located at the foot of the active cone were selected for monitoring, on the basis of their geochemical and isotopic characters that indicate their genetic link with central high temperature fumarolic gases emitted at the crater. Very strong variations of gas composition can be observed within one day (from 1 to about 94% for CO2). Some variations display a daily character and can be correlated with that of atmospheric pressure. The three monitored gases are highly correlated, suggesting very high kinetics of gas transfer in the system. Because of these considerable variations of chemical composition, bulk concentrations obviously are not suitable for monitoring at Vulcano. However, the evolution with time of ratios such as 222Rn/CO2 and He/CO2 (the latter being corrected for atmospheric contamination) supplies numerical parameters that the expected to characterize the intensity of the degassing process. A new input of magmatic gases, that would lead to an increase in the 222Rn/CO2 and He/CO2 ratios, should therefore be detected by such a monitoring station.

Indirect gold trap–MC-ICP-MS coupling for Hg stable isotope analysis using a syringe injection interface

Gold traps are part of standard instrumentation to quantitatively pre-concentrate gaseous elemental mercury, Hg0(g), and are a pre-requisite for atmospheric Hg0(g) analysis. High precision mercury stable isotope ratio analysis by on-line coupling of gold traps to multi collector-inductively coupled plasma-mass spectrometry (MC-ICP-MS) has up to now been problematic due to the transient nature of the signal. We present a new off-line technique, where amalgamated elemental Hg0 is volatilized from the gold trap in an argon gas stream and temporarily stored in a large volume gas tight syringe. Hg0(g) contained in the syringe is then injected on-line into the MC-ICP-MS interface using a syringe pump at a continuous flow rate and resulting in a steady state Hg signal. Different syringe injection schemes are discussed, as well as matrix effects. The 2SD external precision on an in-house Hg0 vapor standard was found to be 0.24‰ for δ202Hg. The method is illustrated for fumerole Hg0(g) emissions at Vulcano (Italy) with δ202Hg of −1.74‰ relative to NIST SRM 3133.

Vapor deposition of trace elements from degassed basaltic lava, Piton de la Fournaise volcano, Reunion Island

Abstract Silica-tube sublimates resulting from the cooling of high-temperature volcanic gases from degassed lava in lava tunnels have been sampled at Piton de la Fournaise volcano. Mineral phases deposited in the tube are mainly halides (Na, K and Cu chlorides, Si and K fluorides). The distribution of most of the chemical elements as a function of the temperature is mainly controlled by their volatility. Volatile elements such as Rb, Cs, Se, Pb and Tl are associated with low-temperature deposits, whereas refractory elements such as R.E.E., alkaline earth, Sc, Ta, Hf, Th and Zr are not detected. The mineral assemblages are characterized by the lack of sulfides, sulfates, native sulfur and oxides. In the same way, numerous transition and semi-metallic elements are lacking. This is interpreted as the result of the magma transport in surface lava tunnels, leading to extensive degassing upstream of the collection site.

Chemistry and emission rate of volatiles from White Island Volcano (New Zealand)

Gases and a water condensate have been sampled at White Island volcano in two selected fumaroles (100 and 495°C). They have been analysed for major, minor and trace elements. Both the chemical composition of gases and thermodynamic calculations suggest that the fluids feeding high and low temperature fumaroles have the same origin, but that they follow different evolutions while ascending to the surface. Very low CO/CO2 ratios with respect to White Island previous results suggest that White Island is now in stage of reduced activity. According to the very low solubility of CO in silicate melts, an increase in activity due to a magmatic pulse should cause a significant increase of CO in the released gas phase. Long term fluxes of gases and metals have been estimated on the basis of COSPEC SO2 flux measurements performed during a medium activity stage. Our data concerning some heavy metals are similar to previous data, suggesting a constancy of the emissions during quiescent periods.

Earthquake-related elemental and isotopic lead anomaly in a springwater

Abstract Elemental and isotopic analyses of lead present at ultra-trace levels were obtained with a double focusing inductively coupled plasma mass spectrometer on a Pyrenean springwater. The sampling was carried out across a time series of 598 days encompassing an earthquake of magnitude 5.2. Lead concentrations show a tenfold increase and isotopic compositions shift toward anthropogenic values four days prior to the quake. The time lag between these geochemical anomalies and the earthquake, combined with hydrogeological constraints, are then used to infer where and when strain changes during earthquake preparation processes went beyond the threshold required to temporarily mix waters from normally independent aquifers. The occurrence of pre-seismic centrifugal strain waves propagating from the epicenter area is proposed on the basis of the observed geochemical anomalies.

Runoff control of soil degassing at an active volcano. The case of Piton de la Fournaise, Reunion Island

Abstract Airborne atmospheric and in-situ soil gas analyses demonstrate that distal gas leaks are non-existent at Piton de la Fournaise (PF) despite frequent eruptions. Airborne traverses made at a constant altitude indicated that no detectable large-scale CO 2 degassing occurs from the volcano. In-situ soil gas traverses (CO 2 , 4 He, 222 Rn) across the active rift-zones, the summit crater and the caldera rims, show both a lack of positive anomalies and evidence of strong depleted concentrations for He and CO 2 with respect to air. Depletions in soil He is explained by intense rain water infiltration. A more unexpected result lies in the very low CO 2 concentrations (down to 92 ppm) in bare soils in the summit region. Such a depletion was not previously reported and is attributed to considerable basalt weathering operating as the result of huge runoff. A model has enabled both the runoff and the CO 2 consumption occurring within the basaltic layers to be calculated. Data obtained are consistent with values expected from large-scale erosion models and with the local calculated diffusive flux of CO 2 in soils. The huge runoff, together with the basaltic lithology of the volcano, are found to be the main factor controlling the soil gas signatures at PF.