Rates and processes of crystallization in on-axis and off-axis MOR basaltic melts

Abstract

Residence times of olivine and plagioclase phenocrysts and xenocrysts in mid-ocean ridge (MOR) basaltic melts have been studied since the mid 1980s using geospeedometric techniques (i.e. using diffusion of major and trace elements) in order to constrain the processes of melt ascent and differentiation in this important magmatic setting. Residence times range from a few hours to several years, but potential links between these timescales and specific tectonomagmatic variables such as spreading rate and relative locations of eruption site and ridge axis have remained elusive. Here we demonstrate how incomplete chemical diffusion of Sr within plagioclase crystals from MOR basalts erupted in on- and off-axis settings on a number of ridges with variable spreading rates provide geospeedometric constraints. We combine electron probe microanalytical crystal maps with detailed laser ablation profiles of almost 70 plagioclase crystals from the fast spreading East Pacific Rise (EPR) at 9–10°N, the intermediate spreading Gorda and Juan de Fuca (JdF) ridges, and the ultraslow spreading Gakkel ridge to calculate crystal residence times. These range from a few days to several months. The scarcity of residence times exceeding years corroborates previous data indicating that most of the growth of plagioclase phenocrysts occurs within the conduit at the onset of and during eruption on the sea floor, and extends this result to the fast-spreading EPR. Further, statistical analysis is employed to show for the first time that residence times are systematically longer at slower spreading rates, in off-axis samples, and samples sourced from laterally distal axial melt lenses. Plagioclase textures and residence time variations appear to be linked to differences in the dynamics of late-stage, pre-eruptive magma storage and ascent in the different tectonomagmatic settings investigated. In the future, geospeedometric work on MOR samples will be required to assess if the effect of spreading rate on crystallization timescales are globally applicable, and to investigate potential variations in magma plumbing systems within individual ridge segments.