The Granite Aqueduct and Advection of Water and Heat Through Plutonic Terranes
Abstract
Although water plays a critical role in the genesis and movement of magma, it is largely lost from rocks upon crystallization. Observation of active volcanoes, analysis of magmatic inclusions, and experimental petrology indicate that intermediate magmas in subduction zones are water-rich, containing 5 wt% or more H2O. Carmichael (2002) wrote of the "andesite aqueduct" that conveys copious amounts of water from magma source regions in subduction zones to the surface and atmosphere. We suggest that this water plays a significant role in the thermal and textural history of the plutonic rocks through which it passes. A dacite magma with 5 wt% H2O crystallizes to granodiorite with ~0.5 wt% H2O, releasing >100 kg of H2O per m3. Field and geochronological data indicate that many sheet-like plutons are constructed from the top down, typically over 1 m.y. or more, likely bathing earlier pulses in ascending water released from later pulses. For a 5 km thick pluton, this release amounts to a condensed-water equivalent depth of ~500 m per unit of horizontal area, a truly vast amount. Plutons preserve abundant evidence for late-stage fluid transfer via a "granite aqueduct." For example, the Tuolumne Intrusive Suite of California is cut by myriad hydrothermally altered pipes that are typically found within or near aplite-pegmatite dikes (Mustart & Horrigan, 2000). The pipes attest to focused fluid flow, and the dikes themselves are the crystallized remnants of late-stage magmatic liquids. Upward advection of heat through dikes and pipes transfers thermal energy from newly crystallizing magma increments to older ones above much more efficiently and rapidly than thermal conduction, and could account for the widespread and profound recrystallization that produces the large grain size and low-temperature mineral assemblages of many granitic rocks. Although the concept that plutons represent the frozen record of huge, highly liquid magma chambers is losing favor, some recent studies invoke large-volume, long-lived areas of interconnected melt in an attempt to keep alive traditional ideas regarding processes such as magma flow, stoping, and crystal fractionation. However, thermal modeling consistently demonstrates that without continual input of new magma, such volumes cannot be maintained for times greater than a few hundred ka. Furthermore, advective heat loss via the granite aqueduct, coupled with fluid convection in wall rocks, will cool plutons far faster than conductive cooling alone. Models demonstrating long-lived interconnected melt without continued magma input require highly unrealistic and contrived assumptions, such as instantaneous emplacement of huge volumes of magma with no vertical heat transport (Memeti et al., 2010).
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2011
- Bibcode:
- 2011AGUFM.V14B..05G
- Keywords:
-
- 3653 MINERALOGY AND PETROLOGY / Fluid flow