Michel Condomines

Timescales of Magma Chamber Processes and Dating of Young Volcanic Rocks

One of the initial goals of measuring U-series disequilibria in young volcanic rocks was the development of new dating methods in an age range (about 0–300 ka), where few other methods were applicable at that time (Cerrai et al. 1965; Cherdyntsev et al. 1967; Kigoshi 1967; Taddeucci et al. 1967; Allegre 1968). In the following years, several studies showed that the 238U-230Th method, the first to be applied, was indeed able to give reliable ages on rocks from various volcanoes (e.g., Allegre and Condomines 1976; Condomines et al. 1978). But it was soon realized (e.g., Oversby and Gast 1968; Allegre and Condomines 1976; Capaldi et al. 1976) that U-series disequilibria could not only be used to date volcanic rocks, but also to infer the timescales of magma transfer and evolution in the crust, and that these methods could give rather unique information on such important parameters. During the last twenty years, besides continuing efforts to develop and discuss the U-series dating methods, many studies have addressed the problem of the timescales of magmatic processes, from melting in the mantle, to crystallization and differentiation of the magmas in their reservoir(s), and magma degassing near the surface. Moreover, as far as studies on volcanic minerals are concerned, determining the eruption age and the timescales of crystallization are intimately linked. It is the purpose of this chapter to review some of the recent advances in the fields of magma chamber processes and dating (the timescales of partial melting in the mantle are treated in other chapters of this book: Lundstrom 2003; Bourdon et al. 2003; Turner et al. 2003). General reviews of U-series disequilibria in volcanic rocks have been published by Condomines et al. (1988), Gill et al. (1992), …

Long-term behaviour of continental hydrothermal systems:: U-series study of hydrothermal carbonates from the French Massif Central (Allier Valley)

U-series dating of hydrothermal carbonates, deposited by CO2-rich thermo-mineral waters of the French Massif Central, provides new insights on the long-term behaviour of a continental hydrothermal system. Dating of aragonite veins and travertines impregnating old terrace levels of the Allier river allowed identification of the main episodes of surface deposition, and thus factors likely to influence these circulations. Fifteen layers from three veins and four travertines from two separate areas were analysed. Sr isotope compositions were also measured on vein samples. The results show that this region was subjected to at least 3 main episodes of surface or near surface hydrothermal deposition: 253 to 208 ky, 135 to 100 ky and less than 8 ky. Comparison of these ages with a global climatic curve indicates that preferential deposition of carbonates occurs during warm periods, suggesting a strong influence of climatic conditions on the hydrothermal system. It is suggested that this climatic influence does not necessarily imply the absence of carbonate deposition during cold and dry periods, but rather that carbonate precipitation might occur at depth before the geothermal fluids reach the surface. In addition, the isotope compositions of fluids recorded by the 87Sr/86Sr and (234U/238U) initial ratios in the aragonite veins from Coudes remained remarkably constant over 250 ky, ranging from 0.71360 to 0.71371 and from 3.10 to 3.39 respectively. The two samples coming from Saladis show a slightly higher (234U/238U) initial ratio around 3.95. The constancy of these ratios over such a long period suggests a hydrothermal system in a near steady state with respect to water-rock interaction. We thus propose a possible model allowing a conservative steady state despite variations in the water recharge rates, in response to the climatic variations. The difference between (234U/238U) initial ratios measured in the Coudes and Saladis systems suggests the existence of two separate reservoirs and constrains their lateral extension to a few km at most.

Holocene Variations of Radiocarbon Reservoir Ages in a Mediterranean Lagoonal System

To obtain a precise radiocarbon Holocene chronology in coastal areas, it is necessary to estimate the modern 14C reservoir age R(t) and its possible variations with time in relation to paleoenvironmental changes. The modern reservoir 14C age was estimated by comparing AMS 14C ages of 2 recent mollusk shells found in sediment cores sampled in the Pala- vasian lagoonal system (south of France) with ages derived from 210Pb and 137Cs data and historical accounts of identifiable storm events. The calculated modern R(t) value of 943 ± 25 14C yr is about 600 yr higher than the global mean sea surface res- ervoir age. This high value, probably due to the relative isolation of the lagoon from marine inputs, is in good agreement with other R(t) estimates in Mediterranean lagoonal systems (Zoppi et al. 2001; Sabatier et al. 2008). 14C ages were also obtained on a series of Holocene mollusk shells sampled at different depths of the ~8-m-long core PB06. Careful examination of the 14C ages versus depth relationships suggests that R(t) in the past was lower and similar to the value presently measured in the Gulf of Lion (618 ± 30 14C yr, Siani et al. 2000). The change in R(t) from 618 to 943 yr is thought to result from final closure of the coastal lagoon by the sandy barrier, due to the along-shore sediment transfer.

Determination of the four natural Ra isotopes in thermal waters by gamma-ray spectrometry

Our method for the simultaneous determination of the four natural Ra isotopes ((226)Ra, (228)Ra, (224)Ra and (223)Ra) in thermal waters involves a separation of Ra on a selective filter (3M EMPORE Radium Rad disk), and a single counting using a broad-energy HPGe detector (BE Ge manufactured by CANBERRA). The calculation of (223)Ra and (228)Ra activities requires interference and cascade summing corrections. The (226)Ra activities in CO(2)-rich thermal waters of the Lodève Basin (South of France) range from 530 to 2240mBq/l. The low ((228)Ra/(226)Ra) activity ratios (0.19-0.29) suggest that Ra is mostly derived from the aquifer carbonates. The short-lived (224)Ra and (223)Ra are probably added to the water through recoil or desorption processes from Th-enriched coatings on the fracture walls.

New archeomagnetic and 226Ra-230Th dating of recent lavas for the Geological map of Etna volcano

This work deals with the dating of Mount Etna lava flows and eruptive fissure deposits to the last four millennia following field investigations and stratigraphic data (BRANCA et alii, 2011a). We have studied 24 of these volcanic products, including 301 large samples, through high precision archeomagnetic dating checked by 226Ra-230Th radiochronology, thus providing additional material to the previous paper by TANGUY et alii (2007). In most cases our results allow attributing ages to the historical period, although two flows are shown to be prehistoric. For the historic lavas, archeoma - gnetic ages can be defined within decades, except for three of them that erupted during a time span (Greco-Roman epoch) when the geomagnetic field underwent little variation. Although 60% of these volcanics exhibit ages comprised between 700 AD and 1850, only one (1285) is mentioned by contemporary written accounts. We conclude that i) historical documents alone are insufficient to reconstruct a coherent sequence of eruptions, and ii) a multidisciplinary approach is necessary to obtain a comprehensive eruptive history of such a very active volcano, useful for both scientific and civil protection purposes, even for such a geologically recent period as that of the last 10 or 20 centuries. Thanks to these new archeomagnetic and 226Ra-230Th data coupled with stratigraphic data, a comprehensive volcanic history of the still-outcropping Mount Etna volcanics is now available for the last 2,400 years. KEY WORDS: Etna volcano, archeomagnetic dating, 226Ra- 230Th dating, stratigraphy, historical account.

Determination of low-level Radium isotope activities in fresh waters by gamma spectrometry

A new portable sampling system was developed to extract Radium isotopes from large volumes (up to 300L) of fresh surface- and ground-waters of low Ra-activities (<5mBq/L). Ra is quantitatively adsorbed on a small amount (6.5g) of MnO2-coated acrylic fibers, which are then dried and burned at 600°C in the laboratory. The resulting Mn-oxide powder (about 2cm3 when compacted) is then analyzed through gamma-ray spectrometry which allows measurement of the whole Ra quartet (226Ra, 228Ra, 224Ra and 223Ra) in a single counting of a few days. The usual relative standard combined uncertainties (1σ) are 2-3% for 226Ra, 228Ra and 224Ra; and less than 10% for 223Ra. This method was applied to the analysis of Ra in karstic waters of the Lez aquifer, and surface- and ground-waters of the upper and middle Vidourle watershed (South of France). The analyzed waters have relatively low 226Ra activities (1-4mBq/L) in both cases, regardless of the contrasted geology (Mesozoic limestone vs crystalline Variscan basement), but clearly distinct (228Ra/226Ra) ratios in agreement with the differences in Th/U ratios of the two drained areas. Short-lived Ra isotopes (224Ra and 223Ra) appear to be mainly influenced by near-surface desorption/recoil processes for most of the sampling sites.