Daniel MEGE







students & postdocs


hot BasaltAlteration VM PlutoTriton GSOgaden VolcOgaden DS DykeProp Termites cooling frozen






Ogaden volcanism and uplift

The geology of the Ogaden, an Ethiopian plateau gently dipping toward Somalia and the Indian ocean, is not well known; all the geological maps contain major errors. One of the reasons is the very small number of in situ rock outcrops, another is the long-lasting conflict between the Somali people of the Ogaden and the present Ethiopian government, a third one is rapting (the continental version of piracy offshore Somalia).

One of the most brilliant photo-interpretation works based on satellite imagery (Quickbird images provided by Google Earth) has been published in a report by Human Rights Watch in a June 2008, Ethiopia: Collective Punishment, War Crimes and Crimes against Humanity in the Ogaden Area of Ethiopia's Somali Regional State.

The ongoing work in the Ogaden includes

  • Geomorphological evolution (including gravitational spreading of western Ogaden topography and other aspects)
  • Cenozoic basaltic volcanism and dyking (see below an account of Cenozoic magmatism in the Ogaden from old and recent datasets, and details on the newly identified Ogaden dyke swarm!)
  • Cenozoic uplift
  • Spectral characterization of basalt alteration (in relation to the alteration of the Martian basalts, as detected by OMEGA and CRISM)
  • The surface and hydrothermal alteration mechanisms of sedimentary deposits, mainly the paleogene Jessomma and Auradu formations

These works are based on oil company and water drilling reports, aeromagnetic data, 2 field works carried out in western Ogaden in 2005 and 2006 and 2 field works in eastern Ogaden in 2008 within the framework of a cooperation between the university of Nantes and Pexco Exploration (East Africa) NV, and in collaboration with P&R Geological Consultants (Scarborough, Australia).

Me, breaking a Cenozoic basaltic flow into small pieces in eastern Ogaden, West of Mustahil

Photograph by Peter Purcell, September 2008


Cenozoic magmatism of the Ogaden and Somalia

Cenozoic magmatism is usually overlooked (see figure below); the Actions Marges symposium held at the Réunion annuelle des sciences de la Terre (RST) in Bordeaux on October 28th, 2010) was an opportunity to demonstrate that the Soamli plate needs re-evaluation. Meandering basaltic flows several hundreds of kilometers long are observed, as well as huge volcanic fields (meander overflows?) and a dyke swarm 400+ km long identified in 2008 and that we named the Ogaden Dyke Swarm.

Simplified geologic and kinematic map of the Afar triple junction and surrounding areas. Note that Tertiary volcanics (in violet) are almost absent at the surface of the Somalia plate. Map drawn by Peter Purcell.


The talk presented by D. Mège, P.G. Purcell, and F. Jourdan at RST 2010 is here, the corresponding abstract is below:

The three plates surrounding the Afar triple junction - Arabia, Nubia, and Somalia - have been the site of Cenozoic magmatism related to the Ethiopian plume, the opening of the Red Sea and Gulf of Aden, and the East African Rift System. The magmatism on the Somali Plate is by far the least documented: published geological maps of Somalia and southeast Ethiopia (the Ogaden) are imprecise regarding the extent, location, and age of magmatism. We present a synthesis of the existing evidence (outcrops, well logs, seismic, gravity, and magnetic data), and complemented by new field observations, geophysical data, and 40Ar-39Ar ag-dating. The salient features of northern Somali Plate magmatism are: (1) volcanic activity ranging from Eocene (prior to the onset of Ethiopian Traps volcanism at 31 Ma) to late Miocene, and almost exclusively basaltic; (2) individual lava flows of Oligocene age (but usually younger than most of the Trap Series) which can be traced over distances > 100 km much more in some instances), commonly defining paleodrainage networks; (3) the 400+ km long mafic Ogaden Dyke Swarm, contemporaneous with the Red Sea opening (~25 Ma), aligned with the Red Sea rift in Afar and the SE-trending Marda Fault Zone and volcanic range, and traceable to the Ethiopia-Somalia border; (4) widespread Eocene and Oligocene (perhaps Miocene) volcanism in southern and eastern Somalia, extending to the Indian Ocean margin, evidenced by vintage magnetic datasets and correlations of oil and water well logs; and (5) basaltic fanglomerate of Middle Miocene age along the Shebele River, pointing to significant volcanic topography upstream at that time, either along the Ethiopian Rift or the southern Afar margin. Cenozoic magmatism on the northern Somali Plate is far more diverse and abundant than previously recognized, and should be considered in any reconstructing of the evolution of the African horn.

Both this project and the Ogaden dyke swarm project have been partky supported by the INSU/Margins initiative.


The Ogaden dyke swarm

Abstract of a talk presented at IDC-6 in February 2010 by D. Mège and P.G. Purcell

The talk presentation is here (some proprietary figures were removed)

Cenozoic magmatism of eastern Ethiopia and Somalia formed the eastern region of the Oligocene Ethiopian flood basalt province but has attracted amazingly little attention to date. Remnants of Oligocene or younger lava flows of basaltic composition are observed across this vast region in the form of meandering ancestral valleys infillings and hills whose geometry is partly controlled by tectonics. Evidence of a formerly widespread mafic volcanic cover of probably similar age is also found at surface at the southern Ogaden-Somalia border east of the town of Dolo, as well as at depth in Somalia in wells and on aeromagnetic data. Well data also gives evidence of widespread mafic volcanism during the Eocene, as well as gabbroic bodies and mafic sills. This contribution reviews the lines of evidence that some of these flows were fed by a dyke swarm, that we name the Ogaden Dyke Swarm.

  1. Medium-resolution aeromagnetic data obtained by Whitestone in 1976 show a NW-SE trending segmented linear magnetic anomaly in north central Ogaden. The anomaly traces the southeastern end of the Marda Fault Zone, a major Precambrian structure that was apparently reactivated several times during the Phanerozoic. The anomaly is also the southward continuation of the Marda volcanic range, an alignment of basaltic hills that sits along the Marda Fault Zone. There is however no demonstrated connection between the volcanics at surface and the anomaly at depth. Two field transects along the Marda range could not establish fissure feeding, and a local meandering pattern in the volcanic outcrop demonstrates that at least part of the volcanics flowed horizontally by following an ancestral valley rather than spreading laterally from a dyke system.
  2. The dyke swarm can be seen on satellite imagery and topographic data extending to the southeast towards the Somalia border as a series of long, linear depressions a few meters deep associated with vegetation distribution anomalies, and probably denoting fracturing above dyke tip followed by hydrothermal alteration and dissolution of the overlying sediment cover.
  3. High-resolution aeromagnetic surveying in 2008 shows a series of high frequency anomalies co-incident with the satellite-defined dykes as well as parallel features.
  4. Medium-resolution vintage Bouguer gravity data also shows this apparent extension of the Marda Fault Zone trend to the southeast.
  5. Several basaltic hills are located along some of these dykes, with the geometry indicative of structural control.
  6. Helicopter-based field reconnaissance survey revealed that one of the hills was fed by fissure eruption, and future detailed field investigations is expected to reveal the existence of other outcropping dykes.

Some of the widespread Ogaden and Somalia volcanics of Oligocene or younger age were fed by this newly-identified Ogaden Dyke Swarm. There is also a possibility that some of the basalt flows, which are extremely (> 100 km) long, could have emanated from the Abyssinian Plateau prior to Miocene rifting. The mantle melting zone of the Ethiopian plume remains poorly defined, but two major crustal pathways have now been identified, the Ethio-Sudanese dyke swarm complex (see the companion contribution), and the Ogaden Dyke Swarm.