Eruption Dynamics on the Global Mid-Ocean Ridge System

Abstract:

Due to the similar tectonic stress fields along the entire ridge system, theoretically constrained rates of magma supply, and empirically defined magma storage reservoir geometries, the 65,000km-long global mid-ocean ridge provides an excellent model system for examining the primary controls on eruption rates in basaltic volcanic systems. In this study we examine variations in eruption rate along the ridge system as a function of spreading rate using independent and global compilations of (1) volcanic morphology and (2) degassing and vesiculation in erupted basalts. Volcanic morphology (i.e., the formation of pillows and sheets) is thought to reflect the timescales of cooling/crust formation relative to that of flow advance. The timescales of cooling are equally fast along the entire ridge system due to the uniformly large (roughly one order of magnitude) temperature difference between seawater and erupted lavas. Thus, flow advance timescales exert the primary control on the distribution of lava morphology, with pillows forming at low eruption rates and sheet flows at high eruption rates. Data compiled from seafloor imaging studies spanning three decades show remarkable consistency and indicate a strong positive correlation between eruption rate and spreading rate. Degassing and vesiculation provide an independent method for evaluating eruption rates based on the kinetics of degassing and diffusion of CO₂ into vesicles at timescales relevant to magma ascent and lava emplacement. A global compilation of MORB volatile concentrations indicates greater degrees of supersaturation at fast spreading rates, signifying that ascent rates exceed degassing rates by a larger margin than at slow and ultraslow spreading rates. Vesicularity also differs significantly between ridges of different spreading rates, providing absolute constraints on the maximum ascent rates given known initial volatile concentrations and further supporting a correlation between spreading rate and eruption rate. Given this apparent correlation, we examine the potential mechanisms controlling this relationship. We suggest that much of the variation can be explained by the known dependence of melt-lens geometries (i.e., depth and volume) on spreading rate and resulting differences in the rates and magnitudes of depressurization.