Map Projections

Have you ever added a layer to your map, only to find it showing up incorrectly - shapes distorted, the scale way off, or sometimes not displaying at all? Or maybe you've looked at a world map and wondered why Greenland looks as big as Africa, or why Antarctica seems to stretch across the entire bottom of the map?

These quirks, and many more, can be explained by Map Projections: the methods we use to flatten the curved surface of the Earth onto a two-dimensional map. Understanding projections is key to making sense of how spatial data appears (or misbehaves) in your GIS app.

Map projections can be one of the most confusing topics for developers new to GIS. In this post, we’ll break down the complexity and bring some clarity to how projections work and why they matter. We'll also answer key questions and walk through practical code samples that show you step-by-step how to reproject your data using the ThinkGeo API:

• What is a Map Projection?

• Key Terminology and Common Spatial Reference Systems

• How do I know when to reproject my data?

• How do I reproject my data using the ThinkGeo Api?

• Advanced Topics

What is a Map Projection?

Image 1: Courtesy of whuber and Directions Magazine

Map Projection is a method of representing a spherical body (such as the Earth) on a two-dimensional surface (such as a computer screen). There are many different projections available, each of which represents the spherical curves of Earth in different ways.

Because every projection distorts the data in some areas to a certain degree, choosing the right projection for the maps you want to view is important. For example, many datasets for specific local region of the United States will use a State Plane projection which focuses on that particular region. This will minimize distortion in the region of your local dataset at the expense of distorting other areas of the world. Because you will not need to display those other areas of the world, the distortion there is acceptable. For more information about the concept of projecting map data, we recommend this Wikipedia article.

Image 2: Courtesy of thetruesize.com

A good visualization of these projection ‘distortions’ is shown here in Image 2. This world map is projected in the ubiquitous Spherical Mercator projection - EPSG:3857 - which is used on most web-based world maps and is a good compromise that preserves shape and area for the more populated regions of the world. But it does this at the expense of polar regions appearing stretched and enlarged.

Here in Image 2, Greenland (in pink) appears about same size as the continent of Africa. However, if Greenland were moved to the equator (in blue), you can see that its actual size when compared to Africa is much smaller.

Image 3: Example of the Arctic Polar Stereographic projection

If a cartographer needs a more accurate projection to show the true size of Greenland, they could choose the Arctic Polar Stereographic projection - EPSG:3995 - which is shown here in Image 3. This is a more accurate representation of shapes in the northern polar region, but enlarges areas nearer the equator. In this projection, you can see that Greenland is much smaller while Africa appears much larger.

Key Terminology and Common Spatial Reference Systems

Before we proceed, there are a few key definitions that we need to go over:

• Spatial Reference System (SRS) or Coordinate Reference System (CRS) - Refers to the overall system used to define locations on Earth. An SRS/CRS specifies how coordinates are measured and represented, including the datum, ellipsoid, and potentially a map projection.   SRS and CRS are synonyms, but for simplicity this article will use SRS.

• SRID (or EPSG) - is a unique ID that identifies a SRS/CRS. SRIDs are used to by the ThinkGeo API to determine how to reproject your data layers. In some documentation, you will see EPSG used interchangeably or prepended on the SRID. The SRID is the official ID of the SRS, and EPSG is a database created by the European Petroleum Survey Group that catalogs many SRIDs and related information.

• Map Projection - is a mathematical transformation that converts the Earth's 3D spherical surface into a 2D plane, allowing for the creation of maps and charts.

While we often use the term ‘map projections’ in a very broad manner, it’s important to know that a projection is an optional subset of a Spatial Reference System (SRS). When reprojecting your datasets with the ThinkGeo Api, the SRID of the SRS will always be used.

It is also helpful to learn a few of the most common Spatial Reference Systems that your basemaps and datasets may be using.

• Web Mercator (or Spherical Mercator) - EPSG:3857 - Most web map providers use this projection for their basemap tiles. Google, Bing and ThinkGeo Cloud are all examples of providers that provide their basemaps in Web Mercator. Coordinates in this projection are measured in Meter.

• Geographic or Latitude/Longitude – EPSG:4326 - also known as WGS84 or ‘Decimal Degree’, this is commonly used by GPS satellite navigation and by NATO for geodetic surveying. Coordinates in this projection are in the Decimal Degree unit of measure. Note: While this article and code samples treat EPSG:4326 as a projection, it’s technically an Unprojected Geographic Coordinate System. You can browse the Advanced Topics section below for a more in-depth explanation.

• State Plane – a set of 125 geographic zones designed for specific local regions of the United States. Coordinates in State Plane projections are usually measured in Feet. More information can be found here.

• UTM – a set of 60 geographic zones covering the entire Earth, each of which consists of a six-degree band of longitude. Coordinates in this projection are measured in Meter. More information can be found here.

Knowing the SRS for your basemap and each of your data layers will be important later when building your map. For example, if your basemap is in Web Mercator (EPSG:3857) and your dataset is in Lat/Lon (EPSG:4326), then the Lat/Lon dataset will need to be reprojected to the basemap’s projection. We’ll talk more about this below and walk through code samples showing you exactly how to accomplish this.

ThinkGeo supports the reprojection of thousands of Spatial Reference Systems. A comprehensive list of the currently supported SRIDs can be found here. If you want to find more details on a specific SRID or find out what projections are commonly used in certain geographic areas, check out the spatialreference.org website as well.

How do I know when to reproject my data?

In a perfect world, your dataset will be in the same SRS as your basemap. But this is not always the case, and you may try to add the dataset to your map and it shows up in the wrong location, or is missing all-together.

Image 4: Frisco, Tx city limit polygon being rendered in Poland - before reprojection.

An example of this can be seen in the images to the right which were taken from our WPF ‘How Do I’ samples. Image 4 shows the Frisco, Tx city limits polygon (in blue) after it has been initially added to the map. You will notice that this polygon renders incorrectly in Poland south of Kaliningrad.

Image 5: Frisco, Tx city limit polygon properly reprojected and displaying near Dallas, Tx

To understand what’s happening, it’s important to know that the basemap is using the Web Mercator SRS(EPSG:3857) while the Frisco, Tx. city limits polygon is in a N. Texas State Plane SRS (EPSG:2276). In Web Mercator, all coordinates are measured in x and y values based on the distance (in meters) from the origin(0,0) located where the equator and prime meridian intersect. However, in the N. Texas State Plane projection, the coordinates are measured in x and y values based on the distance(in feet) from a false origin located in an arbitrary location in southwest Texas.

When the city limits polygon is added as a new layer to the basemap, this causes the polygon to show up in Poland like Image 4! To render the polygon in the proper location like Image 5, a ProjectionConverter must be used to convert the city limit layer from EPSG:2276 to EPSG:3857.

How do I reproject my data using the ThinkGeo Api?

To continue the example with the Frisco city limits polygon discussed above, let’s walk through some code snippets that demonstrate setting up the basemap, adding the city limits layer, and finally reprojecting the city limits layer so it is rendered in the appropriate location and scale.

The first code block below sets up the map and adds a new basemap using the ThinkGeoCloudVectorMapsOverlay. Note: other basemap providers such as Google, Bing or OpenStreetMaps can be used in a similar manner.

  
    // Set the map's unit of measurement to meters(Spherical Mercator)
    MapView.MapUnit = GeographyUnit.Meter;

    // Add Cloud Maps as a background overlay
    var thinkGeoCloudVectorMapsOverlay = new ThinkGeoCloudVectorMapsOverlay
    {
        ClientId = SampleKeys.ClientId,
        ClientSecret = SampleKeys.ClientSecret,
    };
    MapView.Overlays.Add(thinkGeoCloudVectorMapsOverlay);
  

In the code above, it’s very important to note that the first line sets the MapView.MapUnit property. By default, the ThinkGeoCloudVectorMapsOverlay renders the basemap in Web Mercator (SRID:3857). The unit of measure for SRID 3857 is in meter, which is why the code below sets the MapUnit to GeographyUnit.Meter. If your basemap were in a state plane projection, you would typically use Feet for your MapUnit. Likewise, if your basemap were in SRID 4326, you would set the MapUnit to DecimalDegree. You can learn more about Units in the ‘Advanced Topics’ section below.

The next block of code below creates a new ShapeFileFeatureLayer that renders the Frisco, Tx city limits polygon from the City_ETJ.shp shapefile. The code then sets the styling of the layer to the blue/gray colors you saw in the screenshots in the previous section.

  
    // Load the Frisco data to a layer
    var friscoCityBoundary = new ShapeFileFeatureLayer(@"./Data/Shapefile/City_ETJ.shp");

    // Style the data so that we can see it on the map
    friscoCityBoundary.ZoomLevelSet.ZoomLevel01.DefaultAreaStyle = AreaStyle.CreateSimpleAreaStyle(new GeoColor(16, GeoColors.Blue), GeoColors.DimGray, 2);
    friscoCityBoundary.ZoomLevelSet.ZoomLevel01.ApplyUntilZoomLevel = ApplyUntilZoomLevel.Level20;
  

However, with just the code above, the Frisco city limit will be incorrectly displayed in Poland (see Image 4 above). This is because the basemap is projected in EPSG:3857 by default, and when the friscoCityBoundary layer is added, the map will try to render the layer in this projection instead of the EPSG:2276 projection that the shapefile was created in. To successfully reproject the friscoCityBoundary layer from N. Texas(2276) to Web Mercator (3857), the following single line of code is required:

  
    // Reproject the friscoCityBoundary layer from the N. Texas (SRID:2276) to Web Mercator(SRID:3857)
    friscoCityBoundary.FeatureSource.ProjectionConverter = new ProjectionConverter(2276, 3857);
  

That’s all there is to it! After adding the correct ProjectionConverter to the FeatureSource, the map will properly render like Image 5 in the section above.

The above example works well for reprojecting data that’s already loaded into a Layer. But what if you have individual features that need to be reprojected? Luckily, the ProjectionConverter class also has many helper methods for converting Shapes, Features, Enumerations, etc. You can check out these ‘ConvertTo’ helper functions in the ProjectionConverter api docs. And, since the ProjectionConverter class is in ThinkGeo.Core, all your projections will work the exact same whether you’re using Desktop, Mobile or Web!

While the examples above have all been using Vector data, it’s also a simple process to reproject raster data using the GdalProjectionConverter class. Note: to use this class you have to first reference the ThinkGeo.Gdal Nuget package.

To see the raster reprojection in action, just check out the ‘How Do I’ samples, clone the repo and search the solution for ‘GdalProjectionConverter’ and you’ll find multiple samples that reproject raster data. Whether you’re converting raster files, mbtiles or web-based tiles, we have you covered.

Advanced Topics

Earlier, we went over some basic definitions for Spatial Reference Systems(SRS), SRIDs and Projections. And for most users, knowing the SRID is all that is necessary to get up and running. But some scenarios call for a deeper understanding of Spatial Reference Systems.

Here are a few more advanced definitions that you’ll want to understand before we go any further:

• Ellipsoid - An ellipsoid (or spheroid) is a mathematically defined surface that approximates the shape of the Earth. It's slightly flattened at the poles and bulged at the equator. Ellipsoids are used as the geometric basis for datums. Common examples of Ellipsoids include: GRS 1980, WGS 84 Ellipsoid, etc.

• Datum - A datum is a reference framework that defines the origin and orientation of latitude and longitude lines on the Earth. It combines an ellipsoid with an origin point that anchors the ellipsoid to a specific location on Earth’s surface. Common examples of Datum include: WGS 84, NAD83, NAD27, etc.

• Coordinate System - A coordinate system is a method for defining positions on the Earth. It consists of a datum, an ellipsoid, and a map projection (if it's a projected coordinate system). It allows geographic data to be located and interpreted spatially. There are 2 main types of Coordinate Systems in GIS:

  • Geographic Coordinate System (GCS) – Uses latitude and longitude on a 3D surface. One common GCS is EPSG:4326 or Lat/Lon. EPSG:4326 is often mistaken for a Projected Coordinate System, it’s actually an unprojected GCS using a Cartesian Grid with its coordinates in Decimal Degree Unit of Measure..

  • Projected Coordinate System (PCS) – The main focus of this article, a PCS converts the curved surface to a 2D map using a projection method (e.g., Spherical Mercator, UTM, State Plane).

• Unit of Measure - The unit of measure defines how distances or angles are quantified in a coordinate system. The unit can also be a helpful way to try and determine what SRS is used for your data. If you see decimal values, your SRS is likely 4326, while seeing larger integer values often points to your SRS being 3857, or a UTM or State Plane projection.

  • In a GCS, units are typically degrees (angular)

  • In a PCS, units are typically meters, feet, or US survey feet (linear)

An SRS is made up of a datum (which includes an ellipsoid) and, for projected coordinate systems, a map projection. For example, an SRS could be defined as "WGS 84, UTM Zone 10 North," which specifies the datum (WGS 84) and the projected coordinate system (UTM Zone 10 North). 

It’s also important to note that an SRS is typically identified using its SRID (i.e. EPSG:2276), you can alternatively use a proj-string to define your SRS.

A proj-string, also known as a PROJ.4 string, is a text string format used by the PROJ library to describe coordinate reference systems (CRSs) and coordinate transformations, including parameters like projection type, units, and datum shifts.  An example of a proj-string for EPSG:2276 is shown below:

+proj=lcc +lat_0=31.6666666666667 +lon_0=-98.5 +lat_1=33.9666666666667 +lat_2=32.1333333333333 +x_0=600000 +y_0=2000000.0001016 +ellps=GRS80 +towgs84=0,0,0,0,0,0,0 +units=us-ft +no_defs +type=crs

You can find details on each component of the proj-string here. For the purposes of this article it’s just important to know that if your dataset does not use a standard SRID, or if you’re unsure of the SRID, you can also use the ThinkGeo ProjectionConverter class with any proj-string. The ProjectionConverter constructor will accept either an SRID or a proj-string in parameters, and you can see examples of using both methods in our ‘How Do I’ samples.

Summary

We hope you find this post useful. If you have questions, or if you have a topic you'd like us to explore further, we’d love to hear from you! Email us at sales@thinkgeo.com or schedule a meeting to talk in person. We always enjoy connecting with our customers, learning how you’re using ThinkGeo, and finding new ways to improve our products.

About ThinkGeo

We are a GIS software company founded in 2004 and located in Frisco, TX. Our clients are in more than 40 industries including agriculture, energy, transportation, government, engineering, IT, and defense. We pride ourselves on our excellent service and transparency. ThinkGeo offers a variety of products and services to meet almost any GIS application need. We can even help you develop your next project - anywhere from a few hours of consulting to outsourcing an entire project. To learn more, email us at sales@thinkgeo.com, or call us direct at 1-214-449-0330.