ICTJA PhD presentation award 2019 - Ajay Kumar (Oral Presentation)

Opposite polarity subduction zone architecture in the Alboran and Algerian Basins, Western Mediterranean

Ajay Kumar, Manel Fernàndez, Jaume Vergés, Montserrat Torne, Ivone Jimenez-Munt 

Understanding the subduction zone cycle is fundamental to the tectonics and recycling of oceanic/continental rocks. Now the question is why one should even bother about it. Well, the obvious reason is human nature called “curiosity” which has paved the ways for our 21-century lifestyle. Moreover, subduction works as an engine recycling material ranging from large oceanic plates to volatiles, like Carbon, the primary building block of the life on Earth. And the most necessity comes from the direct societal impacts like mega-thrust earthquakes, tsunami and deposition of natural mineral resources.

Figure 1. Paleo-reconstruction of the West Mediterranean taken from Vergés et al. 2019).

Subduction starts from consumption of an oceanic plate, the formation of new oceanic lithosphere in the space made by an extension provided by retreat/back-falling of the subducting plate under its own weight, and finally the end of a retreating subduction zone. End of an oceanic subduction zone involves subduction of a buoyant continental margin and leads to the exotic geodynamic processes such as slab delamination, slab detachment, slab tear and consequences as surface uplift, and extensive volcanism. Further, this geological setting has also been reported to be a candidate for initiation of a new subduction zone involving the newly formed oceanic lithosphere in the back-arc. 

The Alpine Himalayan Collision zone starting from Burma in the east and ending in Gibraltar strait in the west is a perfect natural laboratory to study the subduction processes shaping our planet. Closure of the Tethys Ocean produced the roof of the Earth ”Tibetan Plateau” in the East but left new back-arc oceanic basins in the Mediterranean. The Alboran and Algerian basins are among them and lie in the Western Mediterranean. You must be wondering why oceanic basin and nice holiday beaches for Europe and tough terrane in the Tibet. Well, I do not know the answer exactly but it has something to do with the big Sumatra subduction to the east of Tibet. 

In the western Mediterranean from the reconstructions of the paleo-positions of the Spain and Africa using magnetic field of the Earth which changes with time, we know that the Central Atlantic ridge during the Early Jurassic (Dinosaurs world!) produced an ocean, a part of the Tethys Ocean, in between Spain and Africa (Fig. 1). This ocean was composed of several segments less dense (i.e. continental like) in the west and denser to the east (i.e. oceanic lithosphere like). Then, for some reasons, I do not know (maybe someone knows), the Central Mid-Atlantic ridge decided to go northwards along the Newfoundland-Iberia margin, which at present lies at the eastern coast of Canada. Then the northward motion of Africa relative to Eurasia since the Late Cretaceous pushed this ocean and it started to subduct (go down, it’s denser and been pushed) and produced the Alboran and Algerian Basins by retreating/falling-back under its own weight. We know that there was subduction but in which direction it was operating is still under debate. 

Figure 2. Anatomy of a typical subduction system. (Wikimedia Commons user MagentaGreen/ licensed under CC).

To know the direction of subduction we need to look at the anatomy of typical subduction as shown in Figure 2. As you can see (Fig. 2) oceanic plate it going down to the left and further to left you have a volcanic arc (like the one in the Andes). Volcanic rocks in the Arc have a typical signature from the water and sediment which is going down and are called orogenic volcanic rocks. In the leftmost part, you have another type of volcanism which occurs because of the flow generated by the dense slab in the right (retreating or falling back under its own weight). This flow manifests itself as an extension and you know if you decompress something it is likely to change phase (solid-liquid), hence melting and another volcano. To understand this you can do a simple experiment with an ice cube, maybe in your home from the freezer or in a bar from your drink. Press the ice-cube with your finger what you will see is ice is turning into the water underneath your pressed finger; right there you have just performed a melting experiment using pressure. There is catch here, ice melts when you increase the pressure but mantle melts when you decrease it! Something to think about, I will give you a hint- it is to do with the density: ice is less dense than liquid water and solid mantle is denser than the melt it generates. If the magnitude of the release of pressure in the Earth is enough an entirely new ocean can be generated, like the one being generated in mid-oceanic ridges in the Atlantic Ocean, and are called back-arc basins. Another effect of subduction is that near the point where an oceanic plate is going down the solid outermost layer of Earth, lithosphere, thickens and further to the back-arc it thins (because of stretching).

Figure 3. Different models purposed for the evolution of the Western Mediterranean. (Compilation by Chertova et al. 2014).

At present these subduction zones are gone and the subducted plates are sitting in the mantle/inside of the Earth. So, now we have our chance to be “Sherlock Homes”. Let us look for clues/observation that we have:
1. Orogenic volcanism has been observed in the Valencia trough and later on it changed to anorogenic.
2. In the Alboran basin change from orogenic to anorogenic volcanism is also observed.
3. Another set of observations comes from right beneath our feet and from the CT-scan of Earth, known as seismic tomography models. These models reveal that there are two cold anomalies sitting beneath the Alboran basins and the North Africa margin in Algeria. From the seismic tomography, all we know is that there is something cold but how cold, what composition (i.e. less dense or denser)?.

There are different models proposed for the geodynamic evolution for the Alboran and Algerian basins (Fig. 2). In this work, we test the third scenario where two subduction zones, with opposite direction, are purposed and see if this model can explain the present-day structure on the surface and inside the Earth.

But, first, we have to image the inside of the Earth. To do so we make two cross-sections marked in Figure 3 in the Alboran (Fig. 4) and Algerian basin (Fig. 5). We model for the temperature, density and seismic velocities constrained by gravity—sensitive to density, elevation—sensitive to density, surface heat flow—sensitive to temperature and composition data—controls the density (from the fragments of the mantle brought to the surface by volcanoes).

Figure 4. Crust and upper mantle structure across the Alboran basin.

Figure 4 shows the cross-section along the Alboran basin. We can see the lithosphere is thick beneath the Betics (in the NW) and is thinning in the Alboran basin (i.e. back-arc basin, in the SE) which continues into the north margin of Africa. Now, the older orogenic volcanism in the Alboran basin has the component of water/sediments (i.e. orogenic) from the initial subducting Alboran slab. Later due to the high density of the hanging slab it retreated/moved backwards (i.e. NW) under its own weight and provide space/by-stretching for the mantle and produced present-day Alboran back-arc basin. During this stretching mantle decompressed and produced anorogenic volcanism in the Alboran Basin. Further, the composition of the cold slab sitting in the mantle is depleted, means not pure oceanic lithosphere or in simple terms not dense enough, which hints towards the stretched continental like segment produced during Jurassic. 

Figure 5. Crust and upper mantle structure across the Algerian basin. 

Figure 5 shows the cross-section along the Algerian basin. Here, we see the thick lithosphere beneath the north margin of Africa (in the southeast) and thinning towards the northwest in the Algerian basin (i.e. back-arc basin). In this case, stretching from the retreating slab was enough to produce a new oceanic lithosphere. Further, the composition of the slab sitting beneath the north margin of Africa has a composition similar to an oceanic lithosphere which again hints towards oceanic/more-dense nature of the segment produced during the Jurassic rift to the east. The orogenic volcanism in the Valencia trough can be explained from initial stages of the subduction of this slab which was not yet moving backwards/retreating. Later, during its evolution, it started to move backwards and stretched the Valencia trough. This stretching lead to decompression of the mantle beneath and resulted in anorogenic volcanism in the Valencia Trough and this same stretching produce oceanic Algerian basin.

So, in conclusion, the model proposed in the third scenario can explain the present-day structure along our modeled cross-sections. However, some questions which immediately comes to mind are: where is the boundary between opposite dipping subduction zone at present, why there is no oceanic lithosphere in the Alboran Basin. Well, I do not know the answers yet but I am “curious”.

- Ajay Kumar

Further reading:

Vergés, J., and M. Fernàndez (2012), Tethys–Atlantic interaction along the Iberia–Africa plate boundary: The Betic–Rif orogenic system, Tectonophysics, 579, 144–172, doi:10.1016/j.tecto.2012.08.032.


Schettino, A.; Turco, E. Tectonic history of the western Tethys since the Late Triassic. Geol. Soc. Am. Bull. 2011, 123, 89–105.


Chertova, M. V., W. Spakman, T. Geenen, A. P. van den Berg, and D. J. J. van Hinsbergen (2014), Underpinning tectonic reconstructions of the western Mediterranean region with dynamic slab evolution from 3-D numerical modeling, J. Geophys. Res. Solid Earth, 119, 5876–5902, doi:10.1002/
2014JB011150.