Transcript GPlates and GPML

GPlates and GPML:
Open software and
standards for telling
the Fortune of the
World... Backwards
QuickTime™ and a
Cinepak decompressor
are needed to see this picture.
Dietmar Müller
School of Geosciences and
University of Sydney Institute of Marine Science (USIMS)
The University of Sydney
www.gplates.org
PLATE TECTONICS
Most of us have a “static” view of
the Earth and nearly all geodata
we store in Geographic Information
Systems (GIS) are associated with
present-day coordinates only.
However, the most fundamental,
large-scale process occurring in
the Earth’s interior is convection of
the mantle, responsible for the
continual reshaping of the surface
through plate tectonics
’Everything’ on Earth is
controlled by Plate Tectonics









Resources (hydrocarbons, minerals)
Geothermal energy (mostly granites and active volcanism)
Tourism (landscape, beaches, ocean)
Climate past and present (distribution of continents and oceans)
Agriculture (limestone, weathered basalt)
Wine & beer (beer: magnesium limestone, terroir)
Civil engineering (stability of slopes, tunnels, dams, hazards)
Evolution of life and biodiversity (distribution of continents)
Very important in planetary research
Oil & Gas
Religion/Politics
Wood
Climate
Building Materials
QuickTime™ and a
Cinepak decompressor
are needed to see this picture.
Tourism
Goals
 Need open “plate tectonic GIS”: information model, database,
reconstruction engine, visualisation, map making, web portal service
 Create standard for interactive and web-based plate reconstr.
 Handle point, line, and gridded data
 Allow deformation of plates
 Interface plate reconstructions with
– GIS data bases
– Paleogeographic map making module (GUI + scripting)
– Geodynamic/paleoceanographic/-climate modelling
 Open source (GPL) & cross-platform (Linux, OSX, Windows)
(C++/OpenGL/wxWidgets)
How are we addressing our goals?
 GPlates functionality:
– Display point and line data on the globe
– Rotate data according to Euler rotations
– Animate the rotations over time
 GPlates Highlights:
– Portability (Linux, Unix, Windows, MacOS)
– Open Source (licensed under the GPL)
– Stable and robust
– Active development community.
How are we addressing our goals?
 User-friendliness a priority
– Interactive interrogation/modification of data
– Point-and-click oriented interface
• GUI is written using wxWidgets.
• Data display done via OpenGL.
– Enabling researchers and students alike.
 Rotation of gridded data
– Attach grids to plates
 GPML (the GPlates Markup Language) is the native information model.
– http://www.geosci.usyd.edu.au/pmwiki/pmwiki.php?n=GPlatesGMLDev.GPlatesG
MLDev
 Has recently been linked to the CitComS software
– To provide a set of closed plate polygons
– To provide a plate velocity field as a boundary condition.
What areas of research will
benefit from GPlates?








Plate Tectonic reconstructions
Paleomagnetism (polar wander curves)
Paleogeographic reconstructions
Global geodynamics – model boundary constraints/validation
Global climate modeling (GCM) boundary constraints/validation
Provide constraints for basin modeling, sea level studies
Plate margin tectonics (regional continental geology)
Education
What are some examples of data we
want to play with in GPlates?
Point Data
Stress maps
Magnetic anomaly
picks
Line Data
Plate boundaries
Magnetic/tectonic
lineations
Grid Data
Topography
Crustal/sediment
thickness
GPS Velocities
Hotspots
Point histories
Heatflow
Rock samples
Well locations
Paleomagnetics
Coastlines
COBs
LIPs
Fracture zones
Isochrons
Contours
Dikes
Gravity
Magnetics
Crustal age
Ice thickness
Geoid
Remotely sensed
images
What’s the holdup to
implement GPML?
Need straightforward way to turn UML into GML
schema
Need open source GML3.1 compliant parser (there
isn’t one yet) (use Xerxes for now)
Parser needs to be written in C or C++ - why?
Because we need it both for web feature serving
(ErathByte ande Chronos portal) as well as
bundling with interoperable software (ie GPlates)
Age of the ocean basins and continental
paleogeography through time - outlines
dynamic plate boundaries
QuickTime™ and a
Video decompressor
are needed to see this picture.
We talk the talk, but do we walk the walk?
 Judge for yourself:
Key international connections
Norwegian Geological Survey/Trondheim (Trond Torsvik)
Caltech GeoFramework (Mike Gurnis)
Comp. Infrastructure for Geodynamics (CIG) (Mike Gurnis)
1st GPlates Project Workshop
NGU/Trondheim 19/09-23/09 2002
CHRONOS
Recent CHRONOS workshop, co-chaired by Müller and
Cervato (CHRONOS manager) at EGU meeting in Vienna
(March 2005)
Identified a global group of “Power users” to lead an effort
of minimal standardization required for a web/grid-based
system of plate reconstructions linked to modelling efforts
NSF is keen to provide funding for US-based node
EarthByte e-research pilot proposal
Prototype EarthByte software and database system aims to
connect the open source and architecture-independent
GPlates (Müller et al., 2005) and GMT (Generic Mapping
Tools) software (Wessel and Smith, 1991) into a unified
software system embedded in the APAC Geoscience Grid
infrastructure, and to implement GML-based service
interfaces and databases
Paul Wessel (Univ. Hawaii primary GMT architect) is on a 1year sabbatical at Sydney Univ. from June 2005-July 2006.
EarthByte e-research pilot
proposal member institutions
EarthByte Partners
QuickTime™ and a
TIFF (LZW) decompressor
are needed to see this picture.
Data to plate encoder via
global plate polygon file to
GPML
EarthByte System
Plate Tectonic GIS
Interactive
manipulation
of plate models
GPlates map making module
(based on GMT software),
interactive or scripting-based
Geodynamic/
paleoclimate
modelling
applications
Also need to design Gplates database
Pilot study for a project on global sedimentary
basin analysis
Have experimented with PostgreSQL/PostGIS
Python scripts to access database and generate
GMT readable files for data processing and
plotting
SQL database (PostGIS)
query basin polygon configuration
query basin classification
query basins which are not member of Sedbac project
Python - Example
Accesses
SQL
database
extracts
polygons
parses data
to desired
output format
(ascii xy,
GPML/XML)
QGIS / GIS
Used to display
and interrogate
SQL database
(graphical
OpenDX visualization of CitcomS
mantle convection simulation
From Mike Gurnis (2005)
Depends on interoperability between plate tectonic GIS, plate
kinematic model, 3D parallel convection code and visualization tool
Dynamic surface topography due to vertical component
of mantle convection, Jurassic to present
Depends on interoperability between plate kinematic model, 3D
convection code, and visualization via GMT software
Modeling flow of deep water currents
QuickTime™ and a
YUV420 codec decompressor
are needed to see this picture.
Currently no tools exist to create paleo-depth maps for oceans
to properly constrain deep water flow in paleoceanographic
models
Challenges
 Standard information model development
– GPlates Markup Language - GPML - still in its infancy
– Geological timescale model (S. Cox, CHRONOS) - relatively advanced
 Embedding of gridded data into GPML - platform independent binary
encodings essential (netcdf - based on CSML?)
 Design and “populate” data base (import legacy data)
 Link data to variety of dynamic modelling tools - workflow issues:
– Make data available based on standard info model
– Extract data need for model boundary conditions and transform the data into a suitable
form required for providing model constraints
– Benchmark and execute model
– Visualize model outputs with constraining data superimposed
 Need to satisfy requirements of web services, workflow (data-model
connection) and standalone application
Where can I get a copy of
GPlates?
 From our webpage:
– http://www.gplates.org
 GPlates0.7 will be released in November
 Join the mailing list to be notified of future releases