As you know, the larger the magnitude, the larger the deformation, and the displacement. So, for smaller earthquakes, would we be able to recognize all the deformation, specially some time after the quake? Surface deformation should be very subtle, and in consequence it's very difficult to say if earthquakes around Mw5 or even less are capable to produce surface rupture. There is one publised case "Surface rupture associated with a 5.3-mb earthquake: The 5 April 1986 Cuzco earthquake and kinematics of the Chincheros-Quoricocha faults of the High Andes, Peru", but I really don't know the work, so I can not say that the displacement identified is really from that earthquake. you can read the abstract here "http://www.bssaonline.org/content/88/1/242.abstract" and you might take a look to this as well "http://libra.msra.cn/Publication/5347437/minimum-earthquake-magnitude-associated-with-coseismic-surface-faulting"

The displacement and rapture on the ground is depends on the how shallow the event takes place for example a 4mb magnitude may cause cracks in the house and other buildings. where greater than that if it happens at deeper levels just a rumble it ends. So it cannot be specifically say only this big or so much of magnitude will get deformation on the surface, it depends how shallow the event happened nearer to the surface. But bigger magnitude at the deeper level also will get deformation or cracks on the surface.

I agree with both the opinions.

I am unaware of specific papers addressing proved coseismic rupture (either primary or secondary) associated with events of M < 5.5 or nearby - given all M estimation uncertainties, of course. As Paula reminds us, surface rupture are highly volatile and tend to very rapidly disappear due to surface runoff, weathering,..

On Mt. Etna cosesimic ruptures are documented even for seismic events with M

Thank you for the interesting replies and suggestions for literature, I am familiar with the Bonilla, 1988 paper and I have also found a nice paper by Brune and Allen, 1967 http://www.bssaonline.org/content/57/3/501.short , claiming to have found surface rupture of 1,5 cm of a M 3.6 earthquake. The offset is revealed by the white center line of a highway. However, this conclusion has been corretcted later and suggested that the observed offset has probably been the result of postseismic creep. And of course, such a small displacemet is easily disguised by wheathering processes so one can really only observe such processes right after the actual event. Maybe someone knows a recent study where this has actually been observed!?

The smallest earthquake upon which we have observed co-seismic changes using SG is Mw5.0 recorded at a distance of 60 km.

Thank you, Salvatore, for pointing this out. Wrongly, I had overlooked events that occur in volcanic areas where, as you aptly highlighted, rheology strongly controls rupture propagation.

As for recent studies concerning coseismic rupture. i guess something interesting is out there concerning the Christchurch earthquake. Otherwise, the 6 April 2009 earthquake in Italy has been addressed by so many authors; you may start from Boncio et al. (Boll. Geofis. Teor. Appl., 52, 3, 2011), Blumetti et al. (Quat. Intern., 2012), Falcucci et al., (GRL, 80, 2009), EmerGeo (Terra Nova, 2009) and so many more..

In 2010, on Etna (Guglielmino et al., 2011, EPSL), coseismic ruptures have been observed for a sequence with M4.3 and M3.6 earthquakes (I have the photos taken a few hours after). In that case, hypocentres were very shallow and probably this is the reason why we observed surface rupture. In 2002, ruptures were observed for a deeper sequence with a M3.7 (Acocella et al., 2003, GRL), but they were probably due to fault scarp instability (Bonforte et al., 2007, Bull Volcanol). It is true that we observe continuous and episodic surface ruptures even without seismicity, or without registered seismicity, due to contiunuous creep. In my opinion, this means that there is not a defined threshold; it depends on the depth of the hypo.

This discussion is evolving somehow about co-seismic surface rupture also in the sense that surface = topography. Although, yes it is correct to assume that interpretation, when I first read the question, I associated it with the direct evidences of a fault rupturing the surface. Not topography deformation related somehow with an event. Thinking about surface as the general topography, then yes, you have a lot of other possibilities. For instances, at Christchurch events as far as I am aware, the fault rupture did not reach the surface (was a blind fault) however there was a lot of topography deformation mainly due to liquefaction which is considered a secondary effect. So this opens another branch of possibilities, when the deformation is not directly related with the fault. When it’s mostly due to the materials proprieties and gravity driven effects induced by the shaking...

In the 2010 El Mayor- Cucapah Mw 7.2 in north Mexico, some faults in southern California accommodated very small displacements, and echelon cracks were identified far away from the epicentre zone, I think this published already (I can check..). Why am I bringing this up? Imagine that you would have an Mw 3 or Mw 4 and afterwards walking in the field, you find preserved surface deformation (in deserts environments, this is easily possible). You can assume that this specific deformation would be related with the more recent local shaking which in this case would be just an Mw 3 or 4. But in fact...their genesis could be due to a larger and distant earthquake...

Volcanic areas are a different story ...

I guess this would be a good research theme, to evaluate how small an earthquake could be to produce surface deformation, but probably it would integrate InSAR and other techniques, rather than just the human eye.

It is becoming a very interesting discussion. Volcanoes are a different story for dykes and rifts, but the only difference for faults are their "engine" (mainly flank instability) and then their depth. The depth of the top, because they can extend deep for kilometres involving also the basement. But they are always (or almost always) cutting the surface. For this reason we can have very small slip producing surface dislocation and, as already mentioned, significant depper seismicity without ground evidence. I attach a funny image showing the displacement that is continuously cumulated by the Pernicana fault, with or without (recordable) seismicity (Bonforte et a., 2007, JVGR). InSAR showed a number of active faults on Etna, even new faults (Solaro et al., 2010; Bonforte et al., 2011, G3) that were later searched and confirmed by soil gas surveys (Bonforte et al., 2012, JVGR), visible also without significant seismicity because they are all surface-cutting faults.

Regarding the minimum magnitude, you may refer to McCalpin's work that proposes that magnitude 5,5 is the boundary used by seismologists to define an earthquake associated with fault rupture. This is certainly not available for all case but only a general empirical observation. There is indeed no reason to exclude that some lower magnitude events do not reach the surface, but as underlined above, it is often quite tricky to determine whether small scarps are indeed directly related to the source of the earthquake.

McCalpin, J-P., 2009. Paleoseismology, 2nd Edition, Intern. Geophys. Ser., Academic Press, London, 613 pp.

Concerning my previous comment, there are two links to download two works on it http://pubs.usgs.gov/of/2010/1333/of2010-1333_ebook.pdf - Triggered Surface Slips in Southern California Associated with the 2010 El Mayor-Cucapah, Baja California, Mexico, Earthquake, Open File Report 2010 - 1333 and "Slip on faults in the Imperial Valley triggered by the 4 April 2010 Mw 7.2 El Mayor‐Cucapah earthquake revealed by InSAR" , GEOPHYSICAL RESEARCH LETTERS, VOL. 38, L01308, doi:10.1029/2010GL045235, 2011 https://sciencepolicy.colorado.edu/~bilham/WeiElMajor2011.pdf

enjoy the reading :)

For those at the INGV. is Giancarlo Scalera still at the INGV, and have you

read his book " Why Expanding Earth " ?

Hi Michael, yes Giancarlo still works here in Rome. I didn't have the chance to read his book, although we exchange a few views now and then - our offices are virtually nearby.

You should take the time to read his book. It has articles from some of the most

accomplished people. I am rereading it again for the third or fourth time and doing

lots of calculations in the wide margins to verify the various authors assertions.

Most are correct or very close to what I calculate. I would say his book is still the

best one out there, and I have them all except Terra Non-Firma Earth by

DR. James Maxlow.

You should take the time to talk at length with Giancarlo before he is gone for good,

and read the last few issues of NEW CONCEPTS IN GLOBAL TECTONICS on the web.

I like the picture with the helmet and light . It reminds me of how I started understanding the Earth by exploring caves ( 47 years ago ). To understand a planet,

you need to look at it from the inside to outside.

Another example of surface faulting measured after a small earthquake was the 1966 ML 3.6 earthquake in the Imperial Valley of California that caused up to 1.5 cm of slip on a section of the Imperial Fault. See the Brune and Allen (1967) paper: http://bssa.geoscienceworld.org/content/57/3/501.abstract

Thank you very much, Eric, for bringing up this case..

The following example, described in Demoulin EPSL 245 (2006) 630-641 ( http://www.sciencedirect.com/science/article/pii/S0012821X06002512 ), to underline how elusive it may be to ascribe observed tiny surface displacements to true coseismic surface rupture: 3 weeks after precise levelling measurements performed across the Feldbiss fault at Sittard (Netherlands), a Mw 4.6 earthquake occurred on the same fault at Eschweiler (Germany), at a distance of 35 km and about 16 km depth ( http://seismologie.oma.be/index.php?LANG=FR&CNT=BE&LEVEL=271&id=1306 ). Of course, I immediately relevelled the Sittard line and observed an anomalous 2.7 mm vertical ground displacement concentrated on the fault trace (localized between two adjacent topographic nails 50 m apart from each other). This motion was clearly related to the event. However, as the earthquake rupture area could not have extended that far and static stress change in Sittard could no more be invoked as a trigger for the observed displacement, I suggested to understand it as a creep event dynamically triggered by the passage of trapped seismic waves. But would have the conclusion been the same if the surface observation had been made at Eschweiler?...

Thanks for making this point, Alain. In general, the differentiation between coseismic rupture and aseismic or postseismic creep is often challenging, in particular in areas characterized by very slow deformation rates such as the Lower Rhine Graben from where you gave the example from. To make a reasonable assumption on whether a particular event has been produced coseismic rupture presumably depends to a large extend to the location where you make the observation. And this is not easy considering the high spatiotemporal variability of coseismic events, in particular in intraplate regions.

Ground Effects during the 9 September 1998,M w= 5.6, Lauria Earthquake and the SeismicPotential of the "Aseismic" Pollino Region inSouthern ltaly

Alessandro M Michetti, Luca Ferreli, Eliana Esposito, Sabina Porfido Gndt, Anna Maria, Blumetti, Eutizio Vittori, Leone Ilo Serva, Gerald P Roberts

Seismological Research Letters (impact factor: 3.01). 01/2000; 1.