Page 59 - Caldera Volcanism Analysis, Modelling and Response
P. 59

34                                                              Fidel Costa


          although when errors of about 100 ky are considered, the ages of the three deposits
          overlap. Geochronological data on zircon (mainly single crystal) were also obtained
          by Bachmann et al. (2007b) and found a spread in ages from about 28.03 to
          28.67 Ma. The mean age of the zircons from the main tuff was determined to
          be 28.41 Ma (using  206 Pb/ 238 U) and overlaps with the zircon of the pre- and post-
          caldera deposits when errors are considered (Table 8).
             It is apparent from the foregoing discussion that there is an age difference
                     40
                          39
          between the Ar/ Ar results on several minerals and those of the U/Pb system on
          zircon of about 300–400 ky. There are two main interpretations for this
          discrepancy: (1) it represents the residence time of the Fish Canyon magma (e.g.,
                                                                          39
          Bachmann et al., 2007b) or (2) it is due to a systematic bias of the  40 Ar/ Ar and
          U–Pb methods (e.g., Schmitz and Bowring, 2001). Bachmann et al. (2007b) noted
          that even if one allows for the 1% of bias between the two radioactive systems, a
          residence time of about 300 ky still remains, and thus it seems that the prolonged
          residence time of the Fish Canyon magma is a robust conclusion (see Bachmann
          et al., 2007b for detailed discussion). The uncertainties of the most precise dates are
          in the order of 100 ky, and this remains a minimum estimate for the residence time.
             Insights into the time scales of the Fish Canyon magma were also given by
          Bachmann and Bergantz (2003, 2006) who tested the hypothesis of remobilisation
          of a crystal-rich magma with a thermal model where heat was transported by a
          H 2 O–CO 2 fluid phase. Their calculations show that to raise the temperature of the
          Fish Canyon by about 401C it requires 100–200 ky if the amount of mafic magma
          or intrusion time remained within reasonable limits. This implies heating rates
                                    1
          on the order of 3   10  4  Ky . Such time scales agree with age data that indicates
          residence times on the order of a few 100 ky, and can be taken as the time since
          remobilisation and eruption.

          3.7. Other calderas

          Other caldera-related magmas that have been investigated for their residence times
          include those of Crater Lake (Oregon), the Kos Plateu Tuff (Greece), and La Pacana
          (Chile). Other data on time scales of small and non-silicic caldera-related magmas
          can be found in Reid (2003) and Condomines et al. (2003).


          3.7.1. Crater Lake caldera
                                                   3
          It was formed during the eruption of ca. 50 km of rhyodacite magma from Mount
          Mazama about 7,700 years ago (Bacon, 1983). Together with the main tuff there are
          many xenoliths, including blocks of granodiorite that are thought to be the wall-
          rock of the magma reservoir of the climatic eruption. Bacon et al. (2000) found
          zircon ages of ca. 100 ka for the granodiorite and indicates the presence of a sizeable
          pluton beneath Mount Mazama at the depth of the caldera-forming magma. These
          ages coincide with those of some silicic flows preserved on the north caldera wall
          (between 130 and 100 ka, Bacon and Lanphere, 2006). More recent determinations
          from the granodiorite blocks vary from 20 ka to W300 ka, with peaks at ca 60, 100
          and 200 ka (Bacon and Lowenstern, 2005). Similar ages were found in pre-caldera
   54   55   56   57   58   59   60   61   62   63   64