Page 173 - Caldera Volcanism Analysis, Modelling and Response

148 Gerardo J. Aguirre-Dı ´az et al.

Fredrikson, 1987; Aguirre-Dı ´az and McDowell, 1991), petrogenetic-oriented studies
(Bagby et al., 1981; Cameron et al., 1980; Cameron and Cameron, 1986; Ruiz et al.,
1988; Wark et al., 1990), and many on the geology of particular areas with important
Au–Ag ore deposits (Gross, 1975; Damon et al., 1981; Clark et al., 1982; Goodell,
1981; Lyons, 1988; Scheubel et al., 1988; Randall et al., 1994).
Some researchers have drawn attention to the problem of the lack of
observed conventional caldera structures for the extraordinary volume of silicic
ignimbrites (Swanson and McDowell, 1984; Aguirre-Dı ´az and Labarthe-Herna ´ndez,
2003). It has also been difficult to explain how this large volume of silicic products
originated. Two models have been proposed to explain the magma genesis, one
favouring crystal fractionation from mantle-derived basalt (Cameron and Cameron,
1986), and the other favouring crustal anatexis (Ruiz et al., 1988). Newer work
basically focuses on either one of these two hypotheses. The former hypothesis has
the problem of requiring a large volume of a basaltic magma parent, for which
evidence is not abundant, and the latter requires melting of large quantities of
continental crust. Wark et al. (1990) proposed a crystal fractionation evolution from
an andesitic magma parent, based on their work in the Tomochic caldera,
Chihuahua, a model that at least would reduce the large amounts of basaltic parental
magma required, but with the problem of still needing a petrogenetic history for the
parental andesite; i.e., either crystal fractionation from a basaltic parent, a magma-
mixing derived andesite, or an andesite formed from crustal melting by basalt
underplating of the intermediate composition batholithic roots of the SMO. Ferrari
et al. (2002) propose an elaborate hypothesis involving detachment of the extinct
Farallon oceanic plate beneath the North American plate, which caused basalt
underplating through this slab window and assimilation of the crust. In our opinion,
a plausible petrogenetic model for the origin of the large mass of rhyolitic magma
that the SMO ignimbrites represent remains unsolved.
McDowell et al. (1990) referred to the ‘‘Ignimbrite Flare-up’’ in the SMO, in a
similar way as Lipman et al. (1970, 1972) used to describe a large-volume
ignimbrite event in the western U.S. More recently, Aguirre-Dı ´az and Labarthe-
Herna ´ndez (2003) define the Ignimbrite flare-up as ‘‘a period of intense explosive
volcanic activity that produced enormous volumes of silicic ignimbrite sheets,
which took place mainly between 38 and 23 Ma in Mexico.’’ Aguirre-Dı ´az and
McDowell (1991, 1993) report that Eocene volcanism in the SMO was as extensive
as that in the Oligocene, and that Basin and Range block faulting in Mexico started
at least 29 Ma, coincident with peaks in volcanic activity of regional scale.
McDowell and Mauger (1994) confirm this model for the northern sector of
the SMO. Ferrari et al. (1999) summarise the space and time distribution patterns
of the volcanism in western Mexico, since the SMO to the Mexican Volcanic Belt,
on the basis of the space distribution of a compilation of radiometric ages. In
summary, it is generally accepted that the SMO was formed during several volcano-
tectonic episodes, in which peaks in volcanism coincided with peaks in a long-term
extensional regime that includes the formation of the Basin and Range province
(Aguirre-Dı ´az and McDowell, 1991; Henry and Aranda-Go ´mez, 1992; McDowell
and Mauger, 1994; Aranda-Go ´mez et al., 2000, 2003; Ferrari et al., 2002;
Aguirre-Dı ´az and Labarthe-Herna ´ndez, 2003).