THE ORIGIN OF CONTINENTAL ALKALINE MAFIC MAGMATISM: MELT INCLUSIONS PERSPECTIVE ON MESOZOIC LAVAS FROM MAKHTESH RAMON, ISRAEL

The erosional cirque of Makhtesh Ramon, Negev highlands, Israel, records several cycles of Mesozoic within-plate magmatism. While the late Triassic lavas were related to rifting at the margins of the Tethyan Levant basin, the more abundant, multi-phased, Early Cretaceous magmatism is thought to represent the activity of a hot spot beneath the African plate. However, the geochemical and isotope characteristics of the Cretaceous Ramon lavas, as well as the mantle xenoliths they carry, indicate a lithospheric source. The Makhtesh Ramon suites may thus document temporal variation in the mantle source, melting mechanism and lithospheric deformation in an evolving syn- to post- rift continental environment.

The cirque exposes well preserved alkaline mafic rocks and volcanic edifices of the regional Early Cretaceous ‘Levant’ intra-continental magmatic province, in particular of the latest, Aptian, Ramon volcanics. They constitute a highly diverse rock assemblage of alkali basalts through basanites and nephelinites to melilite nephelinites with a typical OIB-like signature and negligible crustal contribution. Melt inclusions are mostly of an evolved alkali basaltic melt, implying some low-pressure crystallization (2.2-5.7 kb) prior to entrapment. pMELTS-based modelling suggests partial melting at depths greater than 60 km to be the dominant process in which the rocks differentiated, with minor control by fractional crystallization. In Ga’ash Hill, a well preserved concentrically zoned volcano, subvolcanic nephelinite bodies intrude alkaline basalts, suggesting progressively lower amount of partial melting with time. However, partial melting of a homogenous source alone cannot explain some of the observations, mainly with regards to Rb and K. It is therefore suggested that a heterogenous source that is selectively melted or mixing of two distinct sources, combined with a process of batch partial melting, should provide the best explanation for the assemblage. The most likely source candidates are the EM (1 or 2) and HIMU mantle end members.