Review: Geospatial concepts and terms

QGreenland Researcher Workshop 2023

Geospatial data

Vector vs Raster (PyGIS)

Two main representations of geospatial data: vector and raster.

Vector data

Choose the right feature type:

• Point
• Line
• Polygon

Vector map (Wikimedia Commons)

Vector attributes

Attributes add additional data to a geospatial element

• A point representing a weather station may have a temperature attribute
• a line representing a river may have a name attribute
• A polygon representing a state may have a population attribute

Attribute table for the “Human activity/Research sites/GEM research stations” point layer in QGreenland

Vector topology

• Spatial relationships between vector features.
• E.g,
  • “what points fall within a given polygon?”
  • “What polygons are adjacent to this one?”

Examples of topological spatial relations (Wikimedia Commons)

Raster data

RGB Raster Image (Wikimedia Commons)

The most ubiquitous form of raster data are RGB images! Rasters are arrays of values gridded in a regular way over an area of space.

Raster data: continuous and categorical

Continuous: temperature, sea ice concentration, wind speed

Continuous data image

Categorical: land cover, cloud type, storm category

US Land Cover Classification

• Rasters can sometimes contain a mix of continuous and categorical data (a measure plus flag values, e.g., sea ice concentration with “land”, “coast”, “lake” flags).

• The symbology of raster data may be categorical while the underlying data is continuous (e.g., high/medium/low sea ice concentration). We will talk more about symbology on day 3.

Metadata

Data about data

Metadata is “data about data”, and is how standards-compliant data files “self-describe”.

Geospatial metadata concepts

• Location: where does a feature occur in space/on the Earth.
• Time: when did the feature occur?
• Provenance: Where did the data come from and what kind of processing was applied?
• Measurement information
  • What units of measure is the feature recorded in?
  • “Missing/No Data” and other flags

Coordinate Reference System (CRS) information

Includes all of the information necessary to accurately locate features on Earth.

• Geodetic datum
• Coordinate system
  • Geographic (latitude/longitude)
  • Projected (planar coordinates, usually in meters)

Note

Coordinate Reference System (CRS) == Spatial Reference System (SRS)

Geodetic Datums

A model representation of the Earth serving as a reference for locating features.

• Often a spherical or ellipsoidal representation.
• WGS84 is a common global datum, but many others exist (including locally best-fitting models).
• Differences between datums can be significant (> 100 meters in some cases).

Global and regional ellipsoids (Wikimedia Commons)

When there’s a difference between datums, that’s called Datum shift.

Point out the different regions in the graphic. Illustrate how the different ellipsoids fit different areas of the bumpy Earth.

Geographic Coordinate System (GCS)

Coordinate system that represents locations on the Earth in spherical coordinates of latitudes and longitudes.

Diagram of the latitude ϕ and longitude λ angle measurements for a spherical model of the Earth (Wikimedia Commons)

By default in programs like QGIS, data referenced to a Geographic Coordinate System (unprojected) are shown in an ‘equirectangular’ projection in which latitudes are mapped to the Y axis and longitudes are mapped to the X axis of a planar coordinate system.

Projected Coordinate System (PCS)

Coordinate system that represents locations on the Earth in planar coordinates (X, Y).

• Projection is the transformation of spherical (lat/lon) coordinates to a planar surface.
• Benefits include visualization on flat surfaces (e.g., maps) and simplified distance calculations.
• Projection results in one or more distortions: shape, area, distance, direction

Comparison of different forms of Mercator projections (Wikimedia Commons)

Equal-area circles (Tissot’s Indicatrices) on the Mercator projection (Wikimedia Commons)

A projection is an algorithm (e.g., stereographic) and its parameters (e.g, lat/lon of origin, units, etc).

There are MANY map projections. Each has advantages and disadvantages depending on use-case and the part of the world being mapped.

https://www.jasondavies.com/maps/transition/

Website demonstrates many different projections with an interactive control to change the perspective. Note differences in where/how distortions are introduced!

These maps slowly change their origin coordinates to produce the animation effect. Try clicking and dragging!

Standard representations of CRS information

Well Known Text (WKT) 🌎

PROJCRS["WGS 84 / NSIDC Sea Ice Polar Stereographic North",
    BASEGEOGCRS["WGS 84",
        DATUM["World Geodetic System 1984",
            ELLIPSOID["WGS 84",6378137,298.257223563,
                LENGTHUNIT["metre",1]]],
        PRIMEM["Greenwich",0,
            ANGLEUNIT["degree",0.0174532925199433]],
        ID["EPSG",4326]],
    CONVERSION["US NSIDC Sea Ice polar stereographic north",
        METHOD["Polar Stereographic (variant B)",
            ID["EPSG",9829]],
        PARAMETER["Latitude of standard parallel",70,
            ANGLEUNIT["degree",0.0174532925199433],
            ID["EPSG",8832]],
        PARAMETER["Longitude of origin",-45,
            ANGLEUNIT["degree",0.0174532925199433],
            ID["EPSG",8833]],
        PARAMETER["False easting",0,
            LENGTHUNIT["metre",1],
            ID["EPSG",8806]],
        PARAMETER["False northing",0,
            LENGTHUNIT["metre",1],
            ID["EPSG",8807]]],
    CS[Cartesian,2],
        AXIS["easting (X)",south,
            MERIDIAN[45,
                ANGLEUNIT["degree",0.0174532925199433]],
            ORDER[1],
            LENGTHUNIT["metre",1]],
        AXIS["northing (Y)",south,
            MERIDIAN[135,
                ANGLEUNIT["degree",0.0174532925199433]],
            ORDER[2],
            LENGTHUNIT["metre",1]],
    USAGE[
        SCOPE["unknown"],
        AREA["World - N hemisphere - north of 60°N"],
        BBOX[60,-180,90,180]],
    ID["EPSG",3413]]

WKT is highly verbose and relatively readable to humans.

Point out DATUM and CONVERSION (contains projection algorithm, METHOD, and parameters, PARAMETER).

Standard representations of CRS information

Proj parameters

+proj=stere +lat_0=90 +lat_ts=70 +lon_0=-45 +x_0=0 +y_0=0 +datum=WGS84 +units=m +no_defs

Proj parameters encode CRS parameters as key-value pairs. As you can see, this is much more succinct than the WKT representation.

You may have heard of proj4, which is one version of “Proj”.

Standard representations of CRS information

European Petroleum Survey Group (EPSG) Geodetic Parameter Dataset codes

EPSG:3413

EPSG is extremely concise. A registry of short-codes that map to full CRS definitions enables tools like QGIS to work with short, easy-to-remember values to represent Coordinate Reference Systems.

This is not very human-readable (unless you’re familiar with a specific code). This also relies on the correctness of the EPSG database, which is sometimes a problem.

Note there are other standard representations as well, e.g., Geography Markup Language (GML). The three listed here (WKT, Proj, EPSG codes) are some of the most common.

Metadata standards

When a dataset follows of one or more common standards, GIS practitioners and tools can accurately parse information about CRS, date/time, etc.

See the Continued learning page for examples of common standards.

Standards for metadata are often context-dependent. Some standards are developed for particular data formats, others are focused on certain scientific domains (e.g, Ecological Markup Language, EML).

Exercise

💪 Select a dataset