spatial data

Spatial data is any type of data that directly or indirectly references a specific geographical area or location. Sometimes called geospatial data or geographic information, spatial data can also numerically represent a physical object in a geographic coordinate system. However, spatial data is much more than a spatial component of a map.

Users can save spatial data in a variety of different formats, as it can also contain more than location-specific data. Analyzing this data provides a better understanding of how each variable impacts individuals, communities, populations, etc.

There are several spatial data types, but the two primary kinds of spatial data are geometric data and geographic data.

Geometric data is a spatial data type that is mapped on a two-dimensional flat surface. An example is the geometric data in floor plans. Google Maps is an application that uses geometric data to provide accurate direction. In fact, it is one of the simplest examples of spatial data in action.

Geographic data is information mapped around a sphere. Most often, the sphere is planet earth. Geographic data highlights the latitude and longitude relationships to a specific object or location. A familiar example of geographic data is a global positioning system.

Georeferencing and geocoding

Similar processes, georeferencing and geocoding, are important aspects of geospatial analysis.  Both geocoding and georeferencing involve fitting data into the real world by using appropriate coordinates, but that is where the similarity ends.

Georeferencing concentrates on assigning data coordinates to vectors or rasters. This approach helps accurately model the planet's surface.

Geocoding, on the other hand, provides address and location descriptors. These can include information about cities, states, countries, and so on. Each exact coordinate references a specific location on the earth's surface.

spatial data for a restaurant
Geospatial customer data displayed on a map for a fictitious restaurant.

Vector and raster

Vector and raster are common data formats used to store geospatial data.

Vectors are graphical representations of the real world. There are three main types of vector data: points, lines and polygons. The points help create lines, and the connecting lines form enclosed areas or polygons. Vectors often represent the generalization of features or objects on the planet's surface. Vector data is usually stored in shapefiles, sometimes referred to as .shp files.

Raster represents information presented in a pixel grid. Each pixel stored within a raster has value. This can be anything from a unit of measurement, color or information about a specific element. Typically, raster refers to imagery, but in spatial analysis it frequently refers to an orthoimage or the photos taken from aerial devices or satellites

There is also something called an attribute. Whenever spatial data contains additional information or non-spatial data, it is called an attribute. Spatial data can have any number of attributes about a location. For example, this may be a map, photographs, historical information or anything else that may be deemed necessary.

What is spatial data science?

Spatial data science concentrates on extracting deeper insights from data by leveraging a comprehensive set of spatial algorithms and analytical methods. These include using machine learning and deep learning to identify hidden patterns within the data. This approach helps improve predictive models. 

Spatial data may also include attributes that provide more information about the entity it represents. This helps users understand where things happen and why they happen there. For example, geographic information systems (GIS) and other specialized software applications help access, visualize, manipulate and engage in spatial analysis.

Experts expect spatial data science will become more important as government agencies and enterprises strive to make smarter data-driven decisions.

Other aspects of spatial data science include spatial data mining and data visualization.

Spatial data mining describes the process of discovering hidden patterns in large spatial data sets. As a key driver of GIS application development, spatial data mining allows users to extract valuable data on contiguous regions and investigate spatial patterns. In this scenario, spatial variables like distance and direction are taken into account. 

Data visualization software, such as Tableau, allows data scientists and marketers to connect different spatial data files like Esri File Geodatabases, GeoJSON files, Keyhole Markup Language (KML) files, MapInfo tables, Shapefiles and TopoJSON files. Once connected, users can create points, lines and polygon maps using the information in spatial data files, lidar data files and geospatial data files.

Spatial data is important for the internet of things (IoT). It helps IoT protocols use remote sensing to collect data for spatial analysis. Spatial data is also used in transportation and logistics to help companies understand which machine would work best at a specific location, make accurate time estimations for deliveries and track deliveries in real time.

IBM PAIRS Geoscope mobile app
Example of a mobile app created with IBM's PAIRS service, which enables data scientists and developers to produce complex applications using enormous geospatial-temporal data sets from IoT data, maps, weather, drones, satellite, etc.

Environmental technologies also use spatial data to monitor temperature patterns, tidal patterns and more. The ability to track at-risk areas in combination with historical data, weather data and geospatial data gives scientists better information to predict natural disasters.

This was last updated in June 2021

Continue Reading About spatial data

Dig Deeper on Data Visualization Techniques for SQL Server