Home

img/muratlogo.jpg

1. What Is MuRAT?
2. What Is MuRAT For?
3. What Problem Does MuRAT Solve?
4. What Design Principles Underlie MuRAT?
5. How Does MuRAT Accomplish Its Goals?

What Is MuRAT?

For decades seismic imaging has been dominated by phase- and coherency-dependent techniques, like travel-time tomography. However, attenuation tomography has progressively taken over in heterogeneous media, where coherent waves get scattered and absorbed. MuRAT is an open-access attenuation tomography code that allow users to measure and map direct-wave attenuation, scattering, and absorption. It comes from the principl that an accurate description of the physics is useless without a correct tomographc framework, allowing anyone to test the user's results.

What Is MuRAT For?

MuRAT is designed to do the following:

• Assess the reliability of seismic datasets for amplitude-dependent imaging.
• Perform separate measurements of seismic scattering and absorption.
• Image total attenuation of direct waves correting source and site effects.
• Image scattering attenuation using the effect of heterogeneity on direct waves.
• Image absorption by measuring coda attenuation at late lapse timed.
• Enable the user to do so from single seismic waveforms, by building a simple input file.

What Problem Does MuRAT Solve?

MuRAT is especially designed for imaging crustal heterogeneous structures, like fault networks and volcanic areas. While proprietary and open-access codes that image seismic velocity are plenty, and some allow users to pair direct-wave attenuation maps, no open-access code exists that allows specifically to:

1. Calibrate th edataset for attenuation imaging;
2. Use coda waves to image the Earth in 3D;
3. Do it with compatibility to datasets downloaded from data servers, like IRIS.

Ideally, for an application to a well monitored volcano like Mount St. Helens, here is what you can expect:

1. Download some data and metadata from IRIS
2. Populate missing headers of the sac with earthquake and station coordinates, and picking of a phase.
3. Use a standard 1D of 3D velocity model - iasp91 if you cannot find anything better.
4. Get a map of direct-wave attenuation, scattering and absorption in 3D.

Reality is generally harder.

The user needs automatic benchmarks to assess the validity of the downloaded dataset. The average seismologist can populate a SAC header, but way less seismologists are able to assess the validity of tomographic maps. This is especially true for attenuation measurements, as amplitudes are way more sensitive to noise than phases. Even if all works fine from the start, the user must be aware of the complexities of working with ray-tracing algorithms and sensitivity kernels, and will fail if he is not supported throughout the application.

---

MuRAT does all this, allowing the user to output models of volcanic structures given by how seismic energy is lost in the medium:

img/muratcf.png

These images are incredibly meaningful once correct data processing and inversions are applied, as they exploit the sensitivity of propagating energy to fluids, fractures and melt.

## What Design Principles Underlie MuRAT?

Hystrix works by:

• Using Matlab, which, despite being a commercial language, is likely the easiest code to teach to inexperienced users;
• Being an open access code that focuses on data processing and inversion, allowing the user to take full control of the procedure leading to maps;
• Enabling standardization of procedures in attenuation imaging;
• Allowing the new user to work from the start with advanced theories going beyond ray propagation;
• Pairing and staying up-to-date with new developments in radiative transfer theory and inversion;
• Providing first-order 3D imaging tools.

How Does MuRAT Accomplish Its Goals?

Hystrix does this by:

• Focusing on a single data format, i.e., the Seismic Analysis Code;
• Embedding ray-bending strategies for mapping coherent waves;
• Embedding diffusive sensitivity kernels to map sesmic absorptions;
• Bringing official standard of computational science coding into seismology;
• Allowing to test the analysis at each step;
• Providing a single path from data to visualization while producing outputs readble by high-level visualization tools.

Learn more about How It Works and How To Use.