Geometric Data

The the overarching purpose of geometric_features is to facilitate access to and manipulation of the geometric data in the repository. This data is currently divided across 6 components, bedmachine, bedmap2, iceshelves, landice, natural_earth, and ocean.

Components

bedmachine and bedmap2

These components each define two “coastlines” between the Antarctic continent and the ocean. One excludes the cavity under ice shelves while the other includes them. The bedmachine component uses the BedMachine Antarctica dataset (Morlighem et al. 2020) to define the coastlines, while the bedmap2 dataset is based on Bedmap2 (Fretwell et al. 2016).

iceshelves

This component defines regions that extend the bounds of present-day Antarctic ice shelves both out to the edge of the continental shelf and into regions of grounded ice. These maps are used to partition maps of Antarctic melt into total and average melt fluxes over these ice shelves in ocean models that simulate this melting. The ice-shelf features have been divided across the IMBIE1 Basins so that melt rates can also be aggregated per basin.

landice

This component defines regions covering IMBIE1 Basins in Antarctica and Greenland.

natural_earth

This component defines a single, high-resolution region for global land coverage from Natural Earth.

ocean

This component contains regions, transects and points of significance for the global ocean. The regions mostly define individual seas from the International Hydrographic Organisation, but some are masking regions used to delineate larger basins such as those used for computing the Meridional Overturning Circulation. The transects include passages through which ocean transport is commonly measured; locations of so-called “critical passages”, though which ocean flow must be allowed at lower model resolution; and “critical land blockages” such as the Antarctic Peninsula, though which ocean should not be allowed to flow. The points in the database are mostly used as seed points for flood-filling ocean meshes to exclude inland seas that are not connected to the ocean. These points are also sometimes used to sample ocean properties to debug model runs.

Object Types

geometric_features supports three object types: regions, transects and points. Regions are by far the most common, defining areas on the globe. Transects define a connected series of line setments on the globe (typically with the implication that they will also include a depth component from the ocean model, though the features are defined only on the surface of the sphere). These could represent such features as ship tracks, passages, radar lines, or float trajectories. Points specify a single location on the globe.

Feature

An individual feature is defined by nested datastructure of python dictionaries. The outermost dictionary contains at least 3 keys: 'type', 'properties' and 'geometry'.

type

The value corresponding to the 'type' key of a feature is always 'Feature'.

properties

The properties of a feature are stored in another python dictionary, nested in the outer dictionary. Properties include: 'name' - the name of the feature; 'tags' - a list of tags describing the feature, separated by semicolons; 'object' - the object type of the feature (region, transect or point); 'component' - the component of the feature; and 'author' - a comma- or semicolon-separated list of authors (or URLs). Optionally, a feature can include any number of other user-defined properties.

geometry

The 'type' of the geometry ('Polygon' or 'MultiPolygon' for a region, 'LineString' or 'MultiLineString' for a transect and 'Point' or 'Multipoint' for a point) and the 'coordinates' defining the feature’s geometry. By far, the easiest way to create an manipulate the geometry is with the shapely package.

Examples

Here are two simple examples of features from the repository:

{
    "type": "Feature",
    "properties": {
        "name": "Southern_Ocean_Drake_Passage_W070.0_S55.0",
        "tags": "seed_point",
        "object": "point",
        "component": "ocean",
        "author": "Todd Ringler"
    },
    "geometry": {
        "type": "Point",
        "coordinates": [
            -70.000000,
            -55.000000
        ]
    }
}

{
    "type": "Feature",
    "properties": {
        "name": "Global Ocean 15S to 15N",
        "tags": "Equatorial_Basin",
        "object": "region",
        "component": "ocean",
        "author": "Anne Berres, Xylar Asay-Davis"
    },
    "geometry": {
        "type": "Polygon",
        "coordinates": [
            [
                [
                    180.000000,
                    15.000000
                ],
                [
                    180.000000,
                    -15.000000
                ],
                [
                    0.000000,
                    -15.000000
                ],
                [
                    -180.000000,
                    -15.000000
                ],
                [
                    -180.000000,
                    15.000000
                ],
                [
                    0.000000,
                    15.000000
                ],
                [
                    180.000000,
                    15.000000
                ]
            ]
        ]
    }