MELT INCLUSION RECORDS OF MONOGENETIC ERUPTIONS: A BIMODAL EXAMPLE FROM DOTSERO VOLCANO, COLORADO

In order to better understand the magmatic evolution of monogenetic volcanic centers, when long-lived centers with well defined geochemical trends are not available, we have turned to olivine-hosted melt inclusions to provide a detailed history of the petrogenesis of individual eruptive events. However, recent investigations have observed that while melt inclusions may record a more primitive magmatic composition in monogenetic centers, they tend to record discrete compositions rather than a progressive record of magmatic evolution. In the case of Dotsero volcano, an alkali basalt which erupted in NW Colorado ~4200 years ago, two melt inclusions populations are preserved in olivine grains in the distal lava. The first population records a more primitive magmatic composition with SiO₂ ranging from ~42-44 wt%. The second population more closely resembles the whole rock composition (47.9 wt% SiO₂), with SiO₂ ranging from 46-50 wt%. The first population is a more typical undegassed melt inclusion, with S ranging from 0.10 to 0.24 wt% and Cl from 0.05 to 0.08 wt%. The second population has essentially degassed all of its volatiles, with S ranging from 0.04 wt% to below detection limit and Cl from 0.06 wt% to below detection limit. The second population likely records late stage crystallization and in some cases embayed but not fully enclosed melt pockets. This is supported by significantly lower olivine Fo# adjacent to inclusions, comparable to olivine rim compositions, and morphologically by the presence of skeletal grains. Despite the significant difference in major element compositions, the low SiO₂ population of inclusions has a trace element signature nearly identical to that of the whole rock, indicating this inclusion population is petrogenetically related to the host lava. The bimodality of the inclusions clearly indicates that melt inclusions in these short lived centers are recording a biased sampling of magma evolution. This may be explained by a need for some degree of undercooling for inclusions to form in basaltic systems, thus relating to the thermal history of the magma. Additionally, this allows us to also begin to evaluate magma chamber dynamics if the thermal and chemical histories can be integrated into a realistic model of magma chamber processes.