GPlates GPGIM

Ideas and working model for GPlates GPGIM ↔ Shapefile interoperability and extension of GPGIM. Alternatively the File geodatabase format (`*.gdb`) is obviously another alternative as Spatialite falls through when working with ArcGIS.

In general, there's a painful barrier between the ESRI and the open source GIS world, especially when it comes to formats. The Shapefile format, for the time being, seems still one of the most robust containers.

GPGIM Elements

Below a working document for ideas/suggestions to change/add feature types to the existing GPGIM.

Rock units

For the rock units, the current GPGIM is not very useful. I am reverting to the following concept as used in the USGS WEP global compilation:

* Rockclass can be either Magmatic, Sedimentary, Metamorphic * Rocktype can be subclasses: such as Intrusive, Extrusive, * Lithology is the actual lithology of the rock unit such as Sandstone, Diorite, Gneiss

These classifications should be encapsulated in the files under the above attribute names.

Well data

It would be good to have a well GPGIM feature type to indicate ODP/industry drill sites and associated information. I somehow thought that one incarnation of the GPGIM at some stage had a wellsite or odpwell feature type. Could not find anything like this - hold on - here we go - There is a ''gpml:site'' which is linked to a ''gpml:OceanDrillSite''. This is specified in GPGIM Version: 1.6.0327 as

Description:-

An feature to hold data related to a single dilling site.

Comments:-

A single Ocean Drilling Site, with single point geometry, and various measured properties at the site.

FIXME This feature type should be renamed to gpml:drillsite as it should not be exclusively limited to ODP/DSDP sites.

FIXME This feature type is not found in the pygplates.FeatureType list??

Translating the GPGIM across formats

While GPlates internally uses the GPGIM, it would be good to make this accessible in other GISs or applications which cannot read the native GPML files. As most of the common spatial data formats do not allow for nested attributes here are some thoughts on how to translate between them.

One important consideration is the use of a “standardised format” for Shapefile attributes. Together with John Cannon and Michael Chin, we have established such a standard which is available on the EarthByte website. Similarily, a new rotation file format was created (*.grot files).

Shapefile attributes

When exporting a shp file from GPlates, it adds the following fields (Case sensitive when reading):

* PLATEID1: PlateID feature is located on * PLATEID2: Conjugate plate id * TYPE: Old Plates feature type * FROMAGE: appearance * TOAGE: disappearance * NAME: Name * DESCR: Description * FEATURE\_ID: Feature ID from GPlates in the form of GPlates-[UUID] * L\_PLATE: Left plate * R\_PLATE: Right plate * RECON\_METH: Reconstruction method * SPREAD_ASY: Spreading asymmetry

Workarounds

* Use GPGIM\_TYPE as the key attribute for the main GPlates GPGIM types such as BasicRockUnit. * Use G_SubCat as field for nested subcategories such as StrikeSlip:LeftLateral

New developments