Stability of rift axis magma reservoirs: spatial and temporal evolution of magma supply in the Dabbahu rift segment (Afar, Ethiopia) over the past 30 kyr

Unravelling the volcanic history of the Dabbahu/Manda Hararo rift segment in the Afar depression (Ethiopia) using a combination of cosmogenic (36Cl and 3He) surface exposure dating of basaltic lava-flows, field observations, geological mapping and geochemistry, we show in this paper that magmatic activity in this rift segment alternates between two distinct magma chambers. Recent activity in the Dabbahu rift (notably the 2005–2010 dyking crises) has been fed by a seismically well-identified magma reservoir within the rift axis, and we show here that this magma body has been active over the last 30 kyr. However, in addition to this axial magma reservoir, we highlight in this paper the importance of a second, distinct magma reservoir, located 15 km west of the current axis, which has been the principal focus of magma accumulation from 15 ka to the subrecent. Magma supply to the axial reservoir substantially decreased between 20 ka and the present day, while the flank reservoir appears to have been regularly supplied with magma since 15 ka ago, resulting in less variably differentiated lavas. The trace element characteristics of magmas from both reservoirs were generated by variable degrees of partial melting of a single homogeneous mantle source, but their respective magmas evolved separately in distinct crustal plumbing systems. Magmatism in the Dabbahu/Manda Hararo rift segment is not focussed within the current axial depression but instead is spread out over at least 15 km on the western flank. This is consistent with magneto-telluric observations which show that two magma bodies are present below the segment, with the main accumulation of magma currently located below the western flank, precisely where the most voluminous recent (<15 ka) flank volcanism is observed at the surface. Applying these observations to slow spreading mid-ocean ridges indicates that magma bodies likely have a lifetime of a least 20 ka, and that the continuity of magmatic activity is maintained by a system of separate relaying reservoirs, which could in return control the location of spreading. This long term (>105 yr>105 yr) alternation between distinct crustal reservoirs located broadly at the same location relative to the segment appears to be a key feature for organizing and maintaining active spreading centres over stable soft points in the mantle