This is a wonderful question that you should be able to answer at a basic level by consulting a simple phase diagram such as http://www.see.leeds.ac.uk/structure/dynamicearth/melt/melt6.jpg

Estimate the depth to the mantle before extension. Estimate the decrease in pressure necessary to move this point up to the liquids. Then estimate how much you have to thin the crust to achieve this decrease in pressure.... of course- now come the details. What is the density profile of the crust? What is the composition/temparature of the mantle? What is the volatile content of the mantle? Are there any preexisting temperature anomalies in the mantle (hot spot or other related. As with my questions like this, the order of magnitude answer is easy achieved  on the back of a envelope, the correct answer will take a lot more research :-) Love it!

Pablo,

Your is a primary question in earth sciences and I think that it is rather difficult, or even impossible, to reply with a more or less composite answer, since, as evidenced also by David, several constraints should be considered before trying to reply to your question.

And, first of all, I guess that you should take into consideration whole lithosphere extension, and not simply “crustal stretching”, and the evidence that slow-ultraslow rift systems are characterized by magmatic activity that is stored into the shallow mantle lithosphere without giving rise to subaerial volcanic activity.

But to obtain appropriated answers, the question should be posed with more precision. In fact, you write of “decompression melting (of the upwelling asthenosphere?) and recall that “such decompression results from the crustal stretching related with rifting”. Are you referring to a mechanism of passive rifting (driven by far field tectonic forces) with almost passive asthenosphere upwelling, without any active upwelling of deep/hot asthenosphere, as it happens in the case of asthenosphere diapirs or mantle plumes?

Your question (“how much of this crustal extension is needed for triggering the magmatic activity observed in rift settings”) is apparently simple, but it is solely posed in an oversimplified way. It needs that the estimates evidenced by David and, moreover, the driving forces of rifting and the geodynamic context are constrained.

Rift formation has long been the focus of attention for researchers, and an enormous number of studies have been carried out in order to understand causes and modes of whole lithospheric extension (e.g., Brun, 1999; Corti et al., 2003; Dewey and Hancock, 1987; Illies, 1981; Keen et al., 1987; Khain, 1992; Morgan and Baker, 1983; Neugebauer, 1983; Neumann and Ramberg, 1978; Palmason,1982; Ruppel, 1995; Whitmarsh et al., 2001; Ziegler, 1992; and many others).

Just as an example suggested by my personal experience, I can recall that in the case of passive rifting (where asthenosphere is considered to play no active role in passive lithosphere extension) experimental investigations evidenced that the lithosphere must be thinned to about half of its original thickness (i.e., thinned by a factor β≃2: Foucher et al., 1982; Corti et al., 2007; Ranalli et al., 2007) (considering a lithosphere away from mid-ocean ridges, mantle plumes and subduction settings), in order that the near-adiabatically upwelling asthenosphere undergo decompression melting under spinel-peridotite facies conditions.

FOUCHER, J.P., LE PICHON, X., SIBUET, J.C., ROBERTS, D.G., CHENET, P.Y., BALLY, A.W., OXBURGH, E.R., KENT, P., DEWEY, J.F., BOTT, M.H.P., JACKSON, J.A., OSMASTON, M.F. & TURCOTTE, D.L., 1982. The Ocean-Continent Transition in the Uniform Lithospheric Stretching Model: Role of Partial Melting in the Mantle [and Discussion]. Philosophical Transactions of the Royal Society A, 305, 27-43.

CORTI, G., BONINI, M., INNOCENTI, F., MANETTI, P., PICCARDO, G.B. & RANALLI, G., 2007. Experimental models of extension of continental lithosphere weakened by percolation of asthenospheric melts. Journal of Geodynamics, 43, 465-483.

RANALLI, G., PICCARDO, G.B. & CORONA-CHAVEZ, P., 2007. Softening of the subcontinental lithospheric mantle by asthenosphere melts and the continental extension/oceanic spreading transition. Journal of Geodynamics, 43, 450-464.

In conclusion, my suggestion is that you have to explore fundamental references on rifting, easily found in the international bibliography, and try to make your own understanding of the primary process you are interested in, on the basis of a good knowledge of the present state of art of researches on rifting.

Ciao

Giovanni

In a structural point of view extensional regime need much less deformation to create a huge structures.  E.g: few hundred meters of extension create a huge graben having surface size as kilometers of compressive deformation). But, to reach a point of having (alkaline passive margin) volcanic activities its clear that we need kilometers of horizontal extension.

Geodynamics researchers where not looking for extensional rate as same as melting parameters, temperature initial parameters and models... But based on their research we can estimate the shortening rate needed to have surface volcanoes.

For example, after Perderson and Skogseid (1989) work   {http://www-odp.tamu.edu/publications/104_SR/VOLUME/CHAPTERS/sr104_49.pdf} it seams that the horizontal shortening is around 2 to 3 km only. ==> Shortening Rate less than 20%.

Hi Pablo,

I agree with comments above, but I would like to emphasize the key role of the extension velocity (briefly mentioned by Giovanni). In short, rapid extension favors magmatism, while slow (cold) extension favors other phenomena such as mantle serpentinization. You can have a look at first-order estimations in the following paper, which models the effects of extension velocity, mantle temperature, and mantle composition on the amount of magma generated, and the relative timing between magmatism and serpentinization:

Pérez-Gusinyé, M., Phipps Morgan, J., Reston, T.J. and Ranero C. (2006), The rift to drift transition at non-volcanic margins: insights from numerical modelling, EPSL, 244: 458-473.

Regards,

David

A comment on the figure proposed by David. It is not correct to report a decompression path in this way.

All the solids, melts and gases involved in decompression cool. De degree of cooling is known as adiabatic gradient. For the solids it is relatively low ( 0.3-0.6 °C/km), for the melts it is higher (around 1 °C/km) and for the gases it is very high (up to 10 °C/km). The difference stays in the different compressibility of the three states.

This means that a piece of solid mantle upwelled from 200 to 100 km cools for 60 °C. It is not intuitive, but in this case cooling can be associated with partial melting. Of course all depends on the local geotherm and the local solidus. The solidus of the mantle ca be extremely variable, with a temperature range as large as 600 °C at 3-4 GPa in cases of carbonate-bearing alkali-rich compositions vs. volatiles-free alkali-poor compositions.

The solidus temperature of the mantle rock is important, but it is necessary to know the geotherm (i.e., how the temperature varies with increasing depth) in a given place. What I can say is that the classically accepted geotherm of McKenzie and Bickle, with a maximum Tp (Potential Temperature) of about 1300 °C is not correct. The maximum Tp at the base of the LID can be as high as 1500 °C in midplate conditions.

For people not experts, the Tp in igneous petrology represents the temperature of a piece of rock (mantle) or melt (basalt) would have if brought to a reference pressure (Earth's surface). In other word, with increasing the depth (pressure) the absolute T of a rock (or a melt) increases, but the Tp does not change. Decompression is always adiabatic, i.e., with no exchange of heat, but with diminishing of temperature.

michele

http://www.mantleplumes.org/Unzipping.html

http://www.searchanddiscovery.com/pdfz/documents/2013/120118rupke/ndx_rupke.pdf.html

http://www.cprm.gov.br/33IGC/1345044.html ...

and many others as http://www.cprm.gov.br/33IGC/Sess_317.html