7. Conclusions
Our numerical analysis of laboratory experiments suggest that in magmatic systems at intermediate melt fractions (ca. 0.3 – 0.6), melt loss accommodated by repacking is likely a more efficient process than grain-boundary diffusion-controlled creep until the melt fraction diminishes towards the maximum packing fraction. We find that particle-particle friction exerts a dominant control over the dissipation of energy associated with compaction in this regime. At melt fractions below the maximum packing fraction, individual grains comprising the matrix are no longer able to rotate or translate freely, and further melt redistribution should be accommodated by grain-boundary diffusion-controlled creep or possibly other intracrystalline creep processes. However, the low T + P experiments of Hoyos et al.(2022) demonstrate that force chains can in some cases reduce the ability of granular media to repack and introduce ephemeral, intermediate jamming states. A consequence of jamming under high enough T + P conditions could be to augment momentarily the melt fraction at which the rheology of compacting columns transition from repacking to grain-boundary diffusion-controlled creep. Constraining the rheology of granular media when deformation is accommodated by repacking and when it is accommodated by grain-boundary diffusion-controlled creep is crucial to understand how melt is extracted in magmatic systems. Furthermore, the control exerted by particle shape and size distributions on jamming may have important ramifications for how melt is accommodated by repacking versus grain-boundary diffusion-controlled creep.
Geologically, in regions of partially molten rock where melt is largely redistributed by grain-boundary diffusion-controlled creep, its fingerprint is recorded by virtue of crystal overgrowths, zoning truncations, etc. (Holness, 2018). In regions where, instead, melt is redistributed largely by repacking, such signatures may be absent (Holness, 2018), weak, or in isolated regions where ephemeral jamming states persisted due to the buildup of force chains. Several plutons in the geologic rock record have been investigated and show evidence of significant phase separation (Cornet et al. , 2022, Fiedrichet al. , 2017, Hartung et al. , 2017, Lee & Morton, 2015, Tavazzani et al. , 2020). Such evidence includes a plethora of chemical compositions along unmixing lines between chemical endmembers (cumulate and extract) (Cornet et al. , 2022, Hartung et al. , 2017, Schaen et al. , 2018, Tavazzani et al. , 2020), shape-preferred orientation in matrix forming crystals that increases in intensity towards the base of the pluton (Fiedrich et al. , 2017, Garibaldi et al. , 2018), and gradients in melt fraction obtained by trace elements or modal mineral abundance (Fiedrich et al. , 2017, Gelman et al. , 2014, Hartung et al. , 2017, McKenzie, 2011, Tegner et al. , 2009). Future work integrating numerical models, textural and chemical sample analysis, and experiments will be crucial to understand how melt is segregated across a wide spectrum of melt fraction, the timescales required to construct large eruptible magma-rich horizons, and to interpret chemical signatures of igneous bodies.