Thread by @tamarahking

A paper I'm really proud to co-author recently got published in JGR! (doi.org/10.1029/2019JB…)

We produced 168 (!) CSC models to test what scenarios rupture all 7 faults involved in the 2010 Darfield NZ earthquake (and hence, what scenarios create larger earthquakes) [thread]

The Darfield earthquake was a magnitude 7.1, and it ruptured across seven (known) faults. It created a surface rupture on the Greendale Fault (pictured), but seismological data shows it started on the Charing Cross Fault

So, the earthquake initiated on a mis-orientated fault (relative to tectonic stress orientations) in the interior of a multi-fault network, and rupture propagated onto more optimally orientated faults on the outer edges (the gif misses the final step, where all faults rupture)

Our modelling found that if the earthquake where to start on an OPTIMALLY orientated fault on the exterior of the network, it doesn't propagate into the network, it only ruptures that single fault (which means a smaller magnitude earthquake)

That brings up questions like, do those optimally orientated faults rupture frequently in small earthquakes, AND in multi-fault large earthquakes? That would have implications for modelling earthquake hazard for Christchurch, and other places with known multi-fault systems

Our work supports previous research on multi-fault systems, where it's thought that certain faults can regulate and control a critically stressed multi-fault network based on their geometry within the system (i.e. Fletcher et al 2016)

What we don't know, is whether the fault that ruptured in 2010, and caused a cascade of ruptures to the fault network, controls ALL earthquakes within the system, or whether those faults on the exterior can rupture by themselves as well

The point of ALL those models was to investigate how different CSC input values affect the model output. We have a good idea of what order the faults ruptured in the actual earthquake, so our models had to accurately simulate that order

We prefer a model scenario that integrates as much geological and seismological data as possible, including assigning different CSC threshold values (1 or 5 MPa) depending on orientation of the fault with tectonic stresses, based on fault slip tendency anaylis

We made all our models available in an @EarthArXiv methods paper, because we want to be transparent about how different CSC model inputs change the results, and to show how we ended up with the parameters used in the paper (eartharxiv.org/v8t3n/)

There are rupture influences and uncertainties that aren't considered in our models, which we discuss in our paper. Despite any limitations, our interpretations have important implications for hazard that we hope will be investigated, and/or interrogated, by other research

All the authors put so much work into this paper, and we think it's a really important contribution to how we think about and model hazard from multi-fault systems! (Good place to tag @structuregeo who did the fault slip tendency analysis!)

If you follow my work on Aus earthquakes, you may be confused about how I ended up a co-author...my supervisor Mark Quigley bought me on after the paper received extensive reviewer comments, which ultimately made the paper much stronger, but took a lot of time to work through

I drafted the methods paper, got alllllll of the results (404 static images from the 168 models, raw data, etc) into a more accessible form, produced a few extra figures for the paper, and helped rewrite sections of the manuscript that were previously complicated and inaccessible

It was a great learning curve about CSC modelling, multi-fault systems, earthquake science in more active areas, etc, but also about the publishing/reviewing process, and how to condense complex ideas into simpler language and figures. Really good experience as a PhD student!

You can access the manuscript here : drquigs.com/?page_id=122

And you can find the results on the EarthArxiv document (though, I just noticed some small inaccuracies in the text....and the videos don't seem to be working properly...so I need to find time to update that document)

Thanks for reading! The manuscript is still pretty dense with info, so if you have any questions get in touch with Mark, Brendan (@structuregeo), Abi or myself 😄

(cc: @SciMelb and @UnimelbESPG)