What Is Spatial Data? Definition, Types, and Examples

Spatial data (also called geospatial data) is information that has a geographic or coordinate component, such as latitude and longitude, addresses, or boundaries. Examples include GPS tracks, weather grids, census tracts, store locations, and satellite imagery. Spatial data is the foundation of mapping, logistics, environmental analytics, and location-based applications.

This guide explains what spatial data is, the two main types, how it is stored and queried, and how modern data platforms support it alongside traditional rows and columns.

Two Types of Spatial Data

Spatial data divides into two broad categories: vector and raster. Each is suited to different kinds of analysis and stored in different formats.

Common Spatial Data Formats

Spatial data has its own ecosystem of file formats, most of which predate the cloud. Modern warehouses now support spatial types directly, so you can query GeoJSON or WKT data in SQL without leaving the warehouse.

• •GeoJSON — JSON-based vector format, good for web apps
• •Shapefile — legacy ESRI format, still common in GIS
• •WKT/WKB — text and binary representations of geometries
• •GeoTIFF — raster format for satellite and elevation data
• •Parquet + GeoParquet — modern columnar format with spatial extensions

Spatial Queries

Spatial databases support a set of operations that traditional SQL does not: distance, intersection, containment, buffering, and nearest neighbor. PostGIS (the spatial extension to PostgreSQL) and BigQuery GIS are the two most common environments for spatial SQL.

Example queries: "find all stores within 5 km of this customer," "count census tracts that intersect this polygon," "compute the area of every park in the city." Each is a one-line SQL query in a spatial-aware warehouse but would take hundreds of lines without spatial functions.

Use Cases for Spatial Data

Spatial data shows up in surprisingly many domains. Logistics companies optimize routes. Retailers pick store locations. Insurance companies underwrite climate risk. Public health teams track disease spread. Environmental scientists monitor deforestation. Autonomous vehicles navigate roads. The common thread is that geographic context unlocks decisions that would be invisible from non-spatial data alone.

What used to require specialized GIS software (ArcGIS, QGIS) now runs in the warehouse next to your business data. This convergence makes it easy to join spatial and non-spatial data — for example, joining customer locations with revenue to find under-served regions.

Spatial Data in Modern Warehouses

All three major cloud warehouses now ship spatial support. BigQuery has GEOGRAPHY type and a full GIS function library. Snowflake added GEOGRAPHY and GEOMETRY types in 2022. Databricks supports spatial via H3, Sedona, and Mosaic libraries. Postgres remains the gold standard via PostGIS for OLTP spatial workloads.

Data Workers connects to all four through standard MCP connectors. Spatial tables appear in the catalog like any other dataset, with their geometry types tagged so AI agents know to use spatial functions when querying. See the docs for spatial connector setup.

Spatial Data Quality Issues

Spatial data has its own quality pitfalls. Coordinate systems mismatch (lat/lon vs projected). Geometries are invalid (self-intersecting polygons). Addresses fail to geocode. Coordinate precision is too coarse for the use case. Each issue silently produces wrong answers — distances are off, joins miss matches, areas are wrong.

A good spatial data pipeline includes validation: check that all geometries are valid, normalize coordinate systems on ingest, and report geocoding success rates. Read our companion guide on how to ensure data integrity for general patterns that apply to spatial data too.

To see how Data Workers handles spatial data alongside the rest of your warehouse, book a demo.

Spatial data is data with a where. Vector for discrete features, raster for continuous fields. Modern warehouses support both natively, making it possible to combine geographic context with business data in a single query — and unlocking decisions that pure tabular data cannot inform.