Carbon isotope and abundance systematics of Icelandic geothermal gases, fluids and subglacial basalts with implications for mantle plume-related CO2 fluxes

We report new carbon dioxide (CO2) abundance and isotope data for 71 geothermal gases and fluids from both high-temperature (HT > 150 degrees C at 1 km depth) and low-temperature (LT < 150 degrees C at 1 km depth) geothermal systems located within neovolcanic zones and older segments of the Icelandic crust, respectively. These data are supplemented by CO2 data obtained by stepped heating of 47 subglacial basaltic glasses collected from the neovolcanic zones. The sample suite has been characterized previously for He-Ne (geothermal) and He-Ne-Ar (basalt) systematics (Furi et al., 2010), allowing elemental ratios to be calculated for individual samples. Geothermal fluids are characterized by a wide range in carbon isotope ratios (delta C-13), from -18.8 parts per thousand to +4.6 parts per thousand (vs. VPDB), and CO2/(3) He values that span eight orders of magnitude, from 1 x 10(4) to 2 x 10(12). Extreme geothermal values suggest that original source compositions have been extensively modified by hydrothermal processes such as degassing and/or calcite precipitation. Basaltic glasses are also characterized by a wide range in delta C-13 values, from -27.2 parts per thousand to -3.6 parts per thousand, whereas CO2/He-3 values span a narrower range, from 1 x 10(8) to 1 x 10(12). The combination of both low delta C-13 values and low CO2 contents in basalts indicates that magmas are extensively and variably degassed. Using an equilibrium degassing model, we estimate that pre-eruptive basaltic melts beneath Iceland contain similar to 531 +/- 64 ppm CO2 with delta C-13 values of -2.5 +/- 1.1 parts per thousand, in good agreement with estimates from olivine-hosted melt inclusions (Metrich et al., 1991) and depleted MORB mantle (DMM) CO2 source estimates (Marty, 2012). In addition, pre-eruptive CO2 compositions are estimated for individual segments of the Icelandic axial rift zones, and show a marked decrease from north to south (Northern Rift Zone = 550 +/- 66 ppm; Eastern Rift Zone = 371 +/- 45 ppm; Western Rift Zone = 206 +/- 24 ppm). Notably, these results are model dependent, and selection of a lower delta C-13 fractionation factor will result in lower source estimates and larger uncertainties associated with the initial delta C-13 estimate. Degassing can adequately explain low CO2 contents in basalts; however, degassing alone is unlikely to generate the entire spectrum of observed delta C-13 variations, and we suggest that melt-crust interaction, involving a low delta C-13 component, may also contribute to observed signatures. Using representative samples, the CO2 flux from Iceland is estimated using three independent methods: (1) combining measured CO2/He-3 values (in gases and basalts) with He-3 flux estimates (Hilton et al., 1990), (2) merging basaltic emplacement rates of Iceland with pre-eruptive magma source estimates of similar to 531 +/- 64 ppm CO2, and (3) combining fluid CO2 contents with estimated regional fluid discharge rates. These methods yield CO2 flux estimates from of 0.2-23 x 10(10) mol a(-1), which represent similar to 0.1-10% of the estimated global ridge flux (2.2 x 10(12) mol a(-1); Marty and Tolstikhin, 1998). (C) 2014 Elsevier Ltd. All rights reserved.