EVALUATING POTENTIAL LITHOLOGIC SOURCES OF SYN-RIFT MAGMATISM IN THE LITHOSPHERE AND ASTHENOSPHERE USING GEODYNAMIC MODELS

Abstract

Rifting and related magmatism are key processes associated with the fragmentation of continents and supercontinents. However, the sources of syn-rift magmatism are often disputed as early syn-rift magmatism is generally attributed to either elevated mantle temperatures from mantle plumes or the presence of fusible lithologies in the lithosphere and/or asthenosphere. In most cases, the mantle lithologies are not known. Neither are the relationships between the magmatic evolution of a rift and the thermal and geodynamic history of the rift clearly understood. In this dissertation, I evaluate the dynamics of melt generation from various proposed lithologic sources of syn-rift magmatism and the impacts such melting may have on the magmatic, thermal, and structural evolution of a rift.To evaluate syn-rift melting behavior of fusible mantle sources I developed one-dimensional (1D) and two-dimensional (2D) geodynamic modeling packages. The 1D package, consisting of the MATLAB packages MELT1D and MELT1D2, computes pressure and temperature in an extending lithosphere and rising asthenosphere and calculates the resulting melt fraction, melting rates, and total melt produced during rifting for different common mantle lithologies. MELT1D does not include the latent heat of melting in the temperature calculation and thus calculates maximum melting scenarios. MELT1D2 includes the latent heat of melting in the temperature calculations and thus is a modification of MELT1D. The 2D package developed for this work consists of modifications of the finite-element geodynamic modeling code ASPECT (Advanced Solver for Planetary Evolution, Convection, and Tectonics). The modifications provided new functionality to evaluate melting/freezing and melt transport in a rheologically complex layered earth model. To assess the relative timing of decompression melting of different protoliths in a compositionally heterogeneous mantle during rifting and the potential impacts of latent heat consumed during early syn-rift magmatism, two groups of 1D models were developed. The first group used MELT1D to create a series of 1D models that simulate syn-rift melt production from i) dry and wet mantle that are compositionally similar to those beneath mid-ocean ridges, ii) dry and wet relatively fertile ultramafic compositions, representing plume or primitive mantle material, iii) pyroxenite, representing recycled ultramafic oceanic crust or magmatic metasomes, and iv) basalt, representing recycled mafic crust or metasomes. These models were evaluated for the relative timing of syn-rift melting of each protolith across a wide range of rift conditions (e.g., mantle temperatures, lithosphere thickness, crustal thickness, extension rates). Model results predicted sequential melting of different mantle compositions that are broadly consistent with observed basalt eruption histories in many Phanerozoic rifts. Results showed a progressive transition in magma sources as the lithosphere thins, beginning with melting of wet mantle and compositionally fertile mafic components near the lithosphere-asthenosphere boundary during the earliest stages of tectonic extension. This transitions to magmatism dominated by melting of relatively fertile ultramafic components (pyrolitic and pyroxenitic compositions) as extension progresses, and finally to melting of ambient lherzolite asthenosphere as lithosphere thinning approaches continental breakup. Results also showed that mantle composition, pre-rift lithosphere thickness, and mantle temperature exert the greatest controls on the timing and volumes of magmas produced from each lithology. These models suggest that a cool or thick lithosphere has a greater capacity to sequester fertile lithologies than thin or warm lithosphere and thus has a greater capacity to produce early syn-rift magmas without requiring a hot mantle plume. The second group of 1D models used MELT1D2 to evaluate three compositional models representing mantle composed of i) 100% lherzolite; ii) 85% lherzolite and 15% moderately fusible pyroxenite; and iii) 70% lherzolite, 15% moderately fusible pyroxenite, and 15% highly fusible pyroxenite. These models were used to determine the impact that latent heat consumed during melting has on the thermal/structural evolution and thus magmatic evolution of a rift. The second group of models showed that consumption of latent heat during melting of the most fusible source rocks will enhance syn-rift cooling of the lower lithosphere and upper asthenosphere sufficiently to inhibit later syn-rift melting of less fusible rocks. Such cooling delays or may prevent later melting of moderately fusible lithologies. In the models, melting is primarily concentrated near the lithosphere-asthenosphere boundary, so this boundary is where the thermal evolution of the rift is most impacted by latent heat consumption. However, temperatures in this part of the mantle remained sufficiently high during rifting to prevent latent heat consumption due to mantle melting from having a significant impact on lithospheric strength. Building on the results from the 1D geodynamic models which focused on fundamental rift processes, 2D models were applied to the West Antarctic Rift System (WARS) where various fusible mantle lithologies have been proposed as the sources of syn-rift magmatism. Five mantle compositional models were tested: i) 100% depleted lherzolite, ii) 100% fertile lherzolite, iii) 30% pyroxenite and 70% peridotite, iv) 30% basalt and 70% peridotite, and v) a reference model where no melting was permitted. The melting of each lithology is evaluated for a range of mantle potential temperatures Tp (1270 °C ≤ Tp ≤ 1330 °C) consistent with thermobarometric results and geodynamic models of previous WARS studies. The 2D WARS models consisted of a layered Earth with crust, lithosphere mantle, and upper asthenosphere layers which use flow laws for diorite, dunite, and lherzolite, respectively. Initial crust and lithosphere mantle thicknesses are 39 km and 141 km for East Antarctica and 45 km and 75 km for West Antarctica. A total extension rate of 7 mm yr-1 over 100 m.y. is simulated to reproduce the duration of rifting in the WARS from the Late Cretaceous to present. The models with cool mantle temperatures (Tp = 1300 °C) best reproduced the known structural and magmatic evolution of the WARS. Results also showed that of all lithologies evaluated, only basaltic sources in the mantle would melt during WARS rifting. During the modeled initial Cretaceous broad rifting phase (~100 – 50 Ma), broad melting of basaltic sources occurs in the asthenosphere spanning the Ross Sea but achieves only 7% maximum melt by the end of the rifting phase at ~50 Ma. During the modeled Cenozoic narrow rifting phase (50 Ma – present), the focus of melting shifted to the asthenosphere underlying the western flank of the Ross Sea region, while melting ceased under the rest of the Ross Sea. The models do not include high-temperature mantle plume effects, suggesting that syn-rift melting of basalt or other lithologies with high fusibilities are sufficient to produce observed Cenozoic volcanism along the western margin of the Ross Sea though this does not preclude the presence of an underlying plume beneath Ross Island. However, the absence of melting at the modeled eastern margin of the Ross Sea (i.e. Marie Byrd Land) suggested that elevated temperatures such as those associated with a mantle plume are necessary for Cenozoic magmatism.