Magmas observed on the Earth’s surface originate from depth, either through adiabatic melting of the
mantle beneath mid-ocean ridges and hotspots or via hydrous melting in subduction zones. Despite
significant advances, the processes governing fluid transport in subduction zones and magma
migration through the lithosphere, particularly in intraplate volcanism, remain relatively poorly
understood. While magma transport in the asthenosphere is generally associated with porous flow
mechanisms, this process appears too slow to explain magma migration through the lithosphere. The
strong thermal gradients characterizing the lithosphere pose significant challenges, as rising magmas
would cool and crystallize before reaching the surface. In this thesis, we address these challenges by
integrating petrological constraints with numerical modeling to investigate the dynamics of magma
and fluid transport within the lithosphere. Our models account for the lower lithosphere's ductile
rheology, which influences the migration of magma and fluids. By combining these approaches, we
aim to better understand the factors enabling magmas to traverse the lithosphere without crystallizing,
shedding light on the mechanisms behind intraplate volcanism and fluid transport in subduction zones.
This thesis is composed of three distinct parts:
The first study examines the role of melt migration in shaping the distribution of melt beneath
oceanic plates, providing critical insights into mantle dynamics and geophysical anomalies.
Geophysical studies suggest that low-degree melts contribute to the seismic Low-Velocity Zone (LVZ)
observed at 80 km depth under the Pacific plate. However, thermal models highlight a transition
between conductive and convective geotherms at depths exceeding 100 km for older oceanic plates.
To reconcile these contrasting observations, the study explores mechanisms that stabilize melts within
the LVZ, focusing on their volatile content and the thermal and geochemical conditions governing melt
behavior.
Using parameterizations for the solidus temperature of melts as a function of H₂O and CO₂ content,
the analysis demonstrates that volatile content in mantle partial melts is influenced by the
temperature difference between the geotherm and the mantle anhydrous solidus. Simulations show
that small amounts of volatile-rich melt (~0.1–0.2%) are plausible near the asthenosphere-lithosphere
thermal transition for ancient oceanic plates (>60 million years old). However, attributing seismic
discontinuities below 80 k m s olely t o m elts requires a ccounting f or u pward melt m igration f rom
deeper sources. Migrating melts, concentrated near the top of the asthenosphere, could reach melt
fractions of ~1%, sufficient to explain reductions in seismic wave velocity at ∼100 km depth. But this
depth is too deep to represent the seismic position of the LAB suggesting that further melt migration
across the base of the lithosphere is required.
Simulations reveal that melt migration stabilizes at shallower depths than previously anticipated,
aligning with LVZ observations beneath the Pacific plate. Seismic anomalies in regions of older oceanic
plates, where temperatures are insufficient to sustain melting, may instead represent traces of earlier
melt migration producing mantle metasomatism including the formation of hydrous cumulates
including carbonates. These findings highlight the role of small melt fractions in explaining seismic and
electrical anomalies, as well as their contribution to intraplate volcanism, such as Petit-spot volcanoes
near Japan. Melt migration also leads to mantle chemical re-enrichment, emphasizing its significance
as a chemical reservoir and its impact on intraplate magma petrogenesis and mantle chemical cycling.
The second study investigates the formation of Leucite Hills lamproites by combining petrographic
analysis of eruptive lavas and mantle xenoliths, supplemented by numerical simulations of magma
transport. Leucite Hills lamproites are notable for their ultrapotassic chemistry and unique
emplacement, characterized by small magma volumes emitted over an extended period (3–0.9 Ma).
These lamproites exhibit a compositional spectrum ranging from high-MgO diopside-madupitic to low-
MgO, high-SiO₂ phlogopite-lamproites varieties. A three-stage process is hypothesized for their origin:
2
lithospheric mantle depletion, metasomatic enrichment via ancient subduction fluids during the
Archean-Proterozoic, and remobilization of the enriched mantle, leading to the formation of
phlogopite-rich veins. Partial melting of these veins explains the lamproite formation, and potential
mineralogical variability in the vein mineralogy, melting pressure, and wall-rock contributions may
explain their chemical variability.
Physical magma transport mechanisms, however, remain less understood. Analysis of a peridotite
xenolith from South Table Mountain reveals interactions between K-rich melts and peridotite, causing
orthopyroxene dissolution and the formation of olivine, phlogopite, clinopyroxene, apatite, carbonate,
and residual melt. This interaction highlights the CO₂-rich nature of lamproitic melts, with residual melt
compositions aligning with high-SiO₂, low-MgO lamproites. Combining this observation with
experimental constraints suggest that MgO-rich lamproites originate from the melting of phlogopiterich
metasomatic cumulates, while compositional diversity reflects varying interactions between
lamproitic melts and peridotites during ascent.
To constrain the formation of metasomatic cumulates and magma transport, we developed a
numerical model based on continuum mechanics and conservation laws, incorporating porosity wave
dynamics. Simulations show that successive waves stabilize along pathways, facilitating melt transport
and mantle enrichment. Interactions enhance porosity and enrich incompatible elements, leading to
phlogopite formation. With repeated percolation, melts penetrate further, progressively enriching the
mantle. This mechanism explains the creation of the Leucite Hills magma source, supported by xenolith
evidence. These findings underscore the importance of melt-peridotite reactions in shaping lamproite
chemistry and the extraction of small volumes of ultrapotassic melts.
In the third article, we focus on the role of fluids to transport elements and metasomatized the
mantle. Fluid-driven metasomatism may play a more significant role in both subduction zone and
intraplate magmatism than traditionally recognized, acting as a primary agent of mantle
transformation. In subduction zones, volatile-rich fluids released from dehydrating slabs infiltrate the
mantle wedge, reducing the solidus temperature and enabling flux melting. In intraplate settings,
hydrous fluids or melts introduce incompatible elements and hydrous minerals, profoundly altering
mantle fertility and geochemistry.
Specifically, we examine the chemical and physical dynamics of metasomatic processes, with a focus
on fluid-mediated element transport and mineral formation. We investigate four key metasomatic
reactions associated with magmatism using thermodynamic and reactive transport models to reveal
their effects on mantle porosity and mineralogy: (1) Dunitization enhances porosity, promoting melt
transport and forming high-permeability pathways such as dunite channels; (2) Serpentinization
reduces porosity, potentially clogging transport pathways, though its reverse reaction releases
volatiles, contributing to arc magmatism; (3) Amphibolitization decreases porosity and stabilizes
amphibole, offering insights into the role of fluid-driven metasomatism in the mantle; (4)
Phlogopitization highlights the significance of high-pressure metasomatic processes in modifying thick
cratonic lithospheres and generating protoliths for alkaline and potassic magmatism.
Building on these findings, the study proposes a two-stage model for intraplate volcanism. The first
stage involves repeated fluid infiltration, leading to metasomatic enrichment of the lithosphere. This
is followed by a second stage where melting occurs within the modified lithosphere, driven by these
accumulated changes. The findings emphasize the central role of water in Earth's magmatic systems,
extending beyond its well-known function as a melting facilitator. Water emerges as a key driver of
metasomatic enrichment, mantle transformation, and mass transport, underscoring its importance in
shaping Earth's geodynamic and chemical evolution.