14C and U-series disequilibria age constraints from recent eruptions at Sete Cidades volcano, Azores

Accurate dating of young (<10 ka) volcanic rocks poses numerous challenges yet determining eruption ages is critical to understanding magmatic evolution and hazards associated with active volcanic centers. C-14 dating is the preferred technique for dating young geologic materials, but a few studies have suggested that volcanic degassing can result in erroneously old C-14 dates (e.g. Pasquier-Cardin et al., 1999). Finding an isotopic system with which to accurately constrain eruptive recurrence intervals at recently active volcanic centers is critical to assessing future volcanic hazards. In this study, we have obtained C-14 data from modern terrestrial gastropods on Sao Miguel island, Azores to assess the current impact of volcanic degassing on C-14 ages near active and extinct volcanic centers, and for one active volcanic center, we have compared C-14 of paleosols with Ra-226-Th-230 maximum eruptive ages for pumices immediately overlying the respective paleosols. C-14 ages of the modem terrestrial gastropod Cornu aspersum demonstrate the effects of dilution from a range of current outgassing levels at three active trachytic stratovolcanoes on Sao Miguel island: Sete Cidades, Fogo, Furnas, as well as the extinct Povoacao volcano. A modem gastropod sampled at Furnas, where there is evidence of extensive present-day degassing, has a C-14 age of 394 +/- 97 cal BP. At Fogo volcano, also a location with active degassing, the C-14 age of a modem gastropod is 1324 +/- 60 cal BP. Both of these ages clearly show effects of dilution of atmospheric C-14 due to volcanic degassing. In contrast, there is currently no apparent outgassing occurring at Sete Cidades or Povoacao, and modem gastropod ages from these volcanoes are post-bomb. We also obtained C-14 data from paleosols and Ra-226-Th-230 ages associated pumices from the youngest eruptive sequence of Sete Cidades volcano, the Pepom P1-P17 deposits, the ages of which are poorly constrained but thought to be <5000 years (Queiroz, 1997). Ra-226-Th-230 disequilibrium in a P1 glass separate indicates a maximum eruptive age of 4440 +/- 195 years. C-14 ages of paleosols below P1 and P3 are older than those derived from Ra-226-Th-230 dating, at 7293 +/- 141 cal BP and 7698 +/- 172 cal BP, respectively. We interpret the C-14 ages as being erroneously old due to abundant volcanic degassing at the time of eruption. Similarly, but to a lesser extent, the Ra-226-Th-230 maximum eruptive age of glass separated from the P8 eruption is 2659 +/- 55 y, whereas the age of the paleosol sampled directly beneath P8 is older (3054 +/- 194 y BP). In contrast, paleosols sampled at multiple field localities from below the most recent eruptive deposit (P17) are within error of each other (476 +/- 56 cal y BP and 585 +/- 68 cal y BP), while Ra-226-Th-230 maximum eruptive ages in glass separates from two different P17 sampling sites are 762 +/- 14 y and 730 y +/- 15 y. This is consistent with the observation that Sete Cidades volcano is experiencing limited hydrothermal activity at the present time and likely had similarly low levels of degassing just prior to the last eruption. The average eruptive recurrence interval for the P1-P17 deposits, based on Ra-226-Th-230 maximum eruptive ages from P1 and P8 and C-14 ages from P17, is similar to 230 years. This recurrence interval is similar to the similar to 200 y eruptive recurrence interval at Furnas volcano and much shorter than the similar to 1200 y eruptive interval at Fogo volcano. This has important implications with respect to volcanic hazards and should be considered in any future volcanic hazard planning. (C) 2019 Elsevier B.V. All rights reserved.