Remark

Please be aware that these lecture notes are accessible online in an ‘early access’ format. They are actively being developed, and certain sections will be further enriched to provide a comprehensive understanding of the subject matter.

1.5. Sources of Geospatial Data

Geospatial data describes objects, events, or features with a location on or near the Earth’s surface, combining location information with attribute and often temporal data. This data is crucial for various applications, from urban planning and environmental management to disaster response and business analytics.

This chapter maps the territory; later chapters work with the data themselves. Chapter 3 covers file formats and acquisition, including OpenStreetMap in Section 3.5 and the workflow from acquisition to delivery in Section 3.8, Chapter 4 covers operations on vector data, and Chapter 5 covers street networks and routing.

Table 1.5 demonstrates the diverse range of geospatial data sources available to researchers, practitioners, and the general public. The table illustrates the rich ecosystem of geospatial data sources:

1. Government Agencies provide authoritative, often freely accessible data covering a wide range of topics from topography to environmental monitoring; many US federal products are in the public domain [Cornell Law School, 2026].

2. Commercial Providers offer specialized, often high-resolution data products, particularly useful for business applications and detailed analyses.

3. Open-Source Platforms democratize access to geospatial data, often relying on community contributions to create comprehensive global datasets, as OpenStreetMap does [Haklay and Weber, 2008], or on public-domain compilations such as Natural Earth [Natural Earth contributors, 2026].

4. Environmental Organizations focus on ecological and climate-related data, crucial for understanding and addressing global environmental challenges. Examples include global forest-change maps built from Landsat imagery [Hansen et al., 2013] and the terrestrial ecoregions map [Dinerstein et al., 2017].

5. Academic Institutions contribute valuable research data, often covering unique or specialized topics not available from other sources, such as the gridded population estimates of WorldPop [Tatem, 2017].

6. Remote Sensing Sources provide Earth observation data from satellites and aerial platforms, offering broad coverage and regular updates. Landsat [Wulder et al., 2019] and Sentinel-1 [Torres et al., 2012] are two widely used programs with free and open data access.

7. Specialized Data Providers cater to specific industries or applications, offering in-depth data for particular sectors.

8. Crowd-Sourced Platforms leverage user-generated content to create dynamic, real-time datasets, particularly useful for rapidly changing information [Goodchild, 2007].

Table 1.5 Examples of Geospatial Data from Diverse Sources

Source Type	Examples
Government Agencies	- USGS topographic maps - Census boundary files - Annual National Land Cover Database (NLCD) - NOAA climate and weather data (NCEI) - EPA environmental monitoring data - GeoGratis (NRCan) Canadian geospatial data
Commercial Providers	- SafeGraph Points of Interest (POI) data - Esri demographic data - Here Technologies street network data - Mapbox boundaries data - DMTI Spatial Canadian location data
Open-Source Platforms	- OpenStreetMap road and building data - Natural Earth vector and raster data - GADM global administrative boundaries - OpenAerialMap imagery - OpenAddresses global address data
Environmental Organizations	- Global Forest Watch deforestation data - ClimateCheck climate risk assessments - World Wildlife Fund ecoregions (RESOLVE Ecoregions 2017) - UNEP environmental indicators - iNaturalist species observations
Academic Institutions	- CIESIN socioeconomic data - WorldPop population estimates - ESDAC soil data - Global Earthquake Model (GEM) seismic hazard maps - University of Waterloo Geospatial Centre
Remote Sensing Sources	- Landsat satellite imagery - MODIS land surface temperature data - Sentinel-1 radar data (Copernicus Data Space Ecosystem) - USGS 3D Elevation Program (3DEP) lidar and elevation products - Canada Centre for Mapping and Earth Observation
Specialized Data Providers	- Zillow real estate data - Foursquare venue data - AccuWeather historical weather data - TomTom traffic data - Dun & Bradstreet business locations
Crowd-Sourced Platforms	- Waze user-reported traffic incidents - OpenWeatherMap weather observations - Mapillary street-level imagery - OpenSky Network flight tracking data

1.5.1. Authoritative and Volunteered Data

Authoritative sources, such as government agencies, mapping authorities, and many academic and environmental programs, produce data under documented specifications, usually with stated accuracy and a regular update cycle. Volunteered geographic information, such as OpenStreetMap, is contributed by the public [Goodchild, 2007, Haklay and Weber, 2008], so its coverage and detail depend on who contributes. Comparisons of OpenStreetMap with national mapping data, and assessments of road-network completeness worldwide, find that completeness varies strongly from place to place [Barrington-Leigh and Millard-Ball, 2017, Haklay, 2010]. Neither type is simply better: the right choice depends on the question, and analyses often combine them, as the example below does.

1.5.2. Questions to Ask of Any Source

Whatever its type, a source should be checked before it is used. Table 1.6 lists seven questions; they overlap with the properties that make data findable, accessible, interoperable, and reusable [Wilkinson et al., 2016].

Table 1.6 Questions to ask of any geospatial data source

Question	What to look for
Who produced it, and how?	The agency, company, or community; a documented method; stated accuracy
What may I do with it?	The licence: public domain, open with attribution or share-alike, non-commercial, or paid
Where and when does it apply?	Spatial extent, reference date, and time span
How detailed is it?	Scale or spatial resolution, positional accuracy, and attribute detail
How current is it?	Update frequency, and the date of the version you downloaded
How do I get it?	Bulk download, web service or API, or cloud catalogue; file format and coordinate system
Is it documented?	Metadata, a data dictionary, lineage, and known limitations

Free to download is not free to reuse. Natural Earth is in the public domain [Natural Earth contributors, 2026], and OpenStreetMap is published under the Open Database License, which requires attribution and share-alike for derived databases [Open Data Commons, 2026]. GADM boundaries, by contrast, are free for academic and other non-commercial use, while redistribution or commercial use needs the project’s permission, so check the current terms on the GADM site before reusing them. Chapter 3 returns to licensing in Section 3.8.

1.5.3. How the Data Reach You

Data arrive in three main ways:

• File downloads are bulk files, such as boundary shapefiles or raster tiles, and suit a one-off analysis.

• Web services and APIs return only what you ask for, through standards such as OGC API – Features [Open Geospatial Consortium, 2022] or provider interfaces such as the Overpass API for OpenStreetMap [OpenStreetMap Contributors, 2026].

• Cloud archives and platforms hold large imagery collections that are increasingly used in place instead of downloaded, for example Sentinel data on the Copernicus Data Space Ecosystem [Copernicus Data Space Ecosystem, 2026].

Access points change. The Copernicus Open Access Hub (SciHub), long the usual place to download Sentinel data, stopped serving users in November 2023 and was replaced by the Copernicus Data Space Ecosystem, which is why the Sentinel-1 link in Table 1.5 points there. Record the source, the retrieval date, and the version of every layer you use.

1.5.4. Example

Example 1.14

Building the “which neighborhoods are more than 500 meters from a park” map from Section 1 draws on exactly this ecosystem. Park boundaries and pathways come from OpenStreetMap (an Open-Source Platform); neighborhood boundaries and population counts come from Census boundary files (a Government Agency source); and if the analysis needed to check whether a park was actually usable—say, overgrown or inaccessible—Landsat or Sentinel imagery (a Remote Sensing Source) would fill that gap. No single source in the table builds the map alone; acquisition means identifying which combination of sources covers which layer.

Each choice also carries its own terms and dates. The OpenStreetMap layer must credit “© OpenStreetMap contributors” [OpenStreetMap Contributors, 2026] and changes continuously, so the download date matters. The Census files are released as annual snapshots and, as US federal products, are generally in the public domain [Cornell Law School, 2026]. Recording these facts for each layer is what lets someone else rebuild the map later.

1.5.5. Summary

Geospatial data come from government, commercial, open, environmental, academic, remote-sensing, specialized, and crowd-sourced providers, and a real analysis usually draws on several of them. Choosing a source is a matter of asking who produced it, what the licence allows, how detailed, current, and well documented it is, and how it can be obtained. The chapters that follow take up the formats, operations, and workflows needed to turn such sources into analysis-ready data.