[CIG-SHORT] short-term tectonics and the CIG 5-year strategic plan

[Brad Aagaard]

Hi all:

The SSC is putting together the rolling 5-year CIG Strategic Plan for 
NSF. I received only a couple of comments from people on my list of 
priorities. Attached are portions of the strategic plan related to the 
short-term tectonic working group.

Now is your chance to give feedback before the strategic plan is 
submitted to NSF. We will have some time to discuss these priorities at 
the CFEM workshop but that will be too late to incorporate changes into 
this year's strategic plan.

Brad
-------------- next part --------------
3. CIG Accomplishments

3.5. Short time-scale tectonics. 

Working group members have held a workshop each of the last five years
cosponsored by various combinations of the Southern California
Earthquake Center, NASA, Los Alamos National Laboratory, NSF, and
CIG. These workshops have served to (1) establish a suite of
benchmarks for testing codes and comparing modeling techniques, (2)
train students, postdocs, and others in the use of a variety of
modeling tools (including mesh generators and modeling codes), and (3)
facilitate an exchange of ideas among modelers from academia, national
laboratories, and government agencies.

While the proposal to form CIG was in its infancy, two members of the
working group, Brad Aagaard (USGS) and Charles Williams (RPI) began
working towards integrating their modeling codes (EqSim and a version
of Tecton) into the Pyre framework with the ultimate goal of
developing highly modular codes for the simulation of earthquake
dynamics. A significant amount of commonality was identified between
the codes and since then Aagaard and Williams, with help from Matthew
Knepley, have coordinated their development with a plan to merge their
codes into a single suite of modules, PyLith. PyLith 1.0 was released
in June, 2007 with binary executables for several common platforms and
source code available from the SVN repository. This version can solve
1-D, 2-D, and 3-D quasi-static and dynamic elastic and viscoelastic
problems with prescribed slip on faults or prescribed displacements on
the boundaries of the modeling domain with parallel processing
afforded through the use of Sieve and PETSc. 

Prior to the release of PyLith 1.0, CIG released two updates to PyLith
0.8.0 with additional viscoelastic material models, traction boundary
conditions, efficiency improvements, and bug fixes. CIG plans
additional releases of PyLith (1.1 and 1.2) to transfer all features
present in PyLith 0.8 and EqSim to PyLith 1.0. These updates will
permit modeling dynamic kinematic and spontaneous earthquake ruptures
and quasi-static deformation with prescribed boundary tractions or
displacements and kinematic or frictional fault conditions.  PyLith
0.8.0 and 1.0 have been used extensively in the 2006 and 2007
"Workshops on Community Finite Element Models for Fault Systems and
Tectonic Studies" held on the campus of the Colorado School of Mines
which CIG partially supported.


4. Details of our short, intermediate, and long-term goals 

4.5 Short time-scale tectonics.

Short-term crustal dynamics focuses on simulating crustal deformation
associated with the accumulation and release of strain over the
earthquake cycle. The spatial scales range from meters to thousands of
kilometers and the temporal scales range from tens of milliseconds to
hundreds of thousands of years. At the larger length scales and longer
time scales, models blend into those of the long-term crustal dynamics
working group. An important continuing goal is the development of
software for the simulation of multiple earthquake cycles with
sufficient resolution to capture the buildup of strain in the crust,
strain release in propagating ruptures that radiate seismic waves, and
postseismic relaxation of the crust. Additionally, infrastructure is
needed to couple crustal dynamics software to other models of Earth
processes, as well as allow data assimilation into crustal dynamics
software. Easy data assimilation in crustal dynamics software will
promote integration of the wide spectrum of EarthScope data now being
collected.

During the June, 2006, workshop the short-term crustal dynamics
working group set forth four priorities for the year: (1) implement
additional viscoelastic material models and traction boundary
conditions in PyLith 0.8, (2) development of PyLith 1.0 with increased
modularity and functionality over PyLith 0.8, (3) improved support for
computation of 3-D Green's functions using PyLith, and (4)
infrastructure for archiving and comparing output associated with the
suite of working group benchmarks.

Updates to PyLith 0.8 (0.8.1 and 0.8.2) were released in October 2006,
and May 2007. Release 0.8.1 featured primarily efficiency improvements
and bug fixes. Release 0.8.2 included addition of several commonly
used viscoelastic models and traction boundary conditions. The PyLith
user manual was updated to document the additional functionality and
provide more detailed examples and installation instructions.

A significant shift occurred in 2007 in the development of PyLith. The
focus moved from transforming old, entangled legacy simulation codes
to a new modular, implementation that leveraged Pyre, Sieve, and
PETSc. This new implementation, PyLith 1.0 (released in June 2007),
merges the dynamic earthquake rupture capabilities present in EqSim
with the quasi-static crustal deformation capabilities present in
PyLith 0.8. Aagaard, Williams, and Knepley employed unit and
regression testing throughout the entire development cycle, so that
the functionality of each module of the code received thorough testing
continuously from initial implementation to release. These 400+ tests
are run on multiple platforms whoever changes to the software source
code are committed to the subversion repository, aiding the SDT in
identifying bugs early in the development cycle. The modular
implementation of PyLith 1.0 with Pyre allows users to extend the code
by adding their own modules (e.g., bulk and fault constitutive models,
importers for additional mesh generators, exporters to additional
visualization packages, and interfaces with physical property models)
by writing simple, small modules.

Two additional releases of PyLith are planned in the coming months so
that the current version contains all of the features present in EqSim
and PyLith 0.8. PyLith 1.1, with an anticipated release in September
2007, will focus on including absorbing boundary conditions for better
support of dynamic modeling with kinematic earthquake ruptures,
reimplementation of the viscoelastic models present in PyLith 0.8
making use of the more modular approach used in PyLith 1.0, and
support for output of stress, strain, and displacements fields over
various subsets of the modeling domain. PyLith 1.2, with an
anticipated release in November 2007, will add fault constitutive
models for both dynamic and quasi-static modeling along with traction
boundary conditions. With this release PyLith will contain all of the
functionality present in its ancestors as well as many additional
features in the form of an open-source, modular, scalable, extensible
software package with comprehensive documentation and thorough unit
and regression testing.

In 2006, the working group also began using the CIG benchmarking
infrastructure for comparing output from PyLith, GeoFEST (a code
similar to PyLith 0.8), and COMSOL (formerly called Femlab) for
community established crustal deformation benchmarks. Luis Armendariz
(CIG) implemented quantitative metrics to compared solutions across
the codes in addition to solutions from semi-analytic methods. The
benchmark comparisons demonstrated PyLith 0.8 and GeoFEST produce
essentially identical results for elastic problems. PyLith 1.0 also
produces the comparable elastic solution. The benchmarking effort is
ongoing with an emphasis on identifying the source of differences in
the viscoelastic time-dependent solutions between PyLith (0.8 and 1.0)
and GeoFEST.

With the release of PyLith 1.0 satisfying a major priority for the
working group, focus will shift in the coming year towards satisfying
other previously identified priorities. (1) Improved support for
computation of 3-D Green's functions using PyLith in order to permit
inversion of geodetic data for crustal deformation using the latest
3-D Earth models. (2) Support for adaptive mesh refinement in PyLith
or a new code built reusing many of the geodynamics modules (fault
interfaces, constitutive models, etc.) written for PyLith.

The working group has also identified three longer term priorities (1)
Development of robust, validated, open-source versions of codes for
dislocations in layered elastic and layered viscoelastic
domains. While a variety of codes are in use to do these types of
computations, few are portable and none undergo continuous regression
testing. Furthermore, these codes would serve as
tutorials/introductions to the highly modular, portable software
developed by CIG for the short-term crustal dynamics community and
provide reference solutions for additional benchmarking. (2) Coupling
short-term crustal dynamics simulations to other simulations in order
to more accurately capture interactions among geodynamics
phenomena. (2) Integration of tools for formal data assimilation in order
to permit data from EarthScope and other sources to be included
directly into simulations.


Table 4.1


Short Term Goals

Complete merging of EqSim and PyLith 0.8 into PyLith 1.0.


Intermediate Term Goals

Direct support for computing Green's functions in PyLith 1.0. Adaptive
mesh refinement capabilities in PyLith or a code with similar
functionality.


Long Term Goals

Coupling short-term simulations to other simulations in order to
accurately capture interactions among geodynamics phenomena. Integrate
tools for formal data assimilation.