Skip to main content
Work in Progress

This page is currently under development. Content may be incomplete or contain inaccuracies. If you notice any errors or have suggestions, please contact us.

Geospatial Data Types & File Formats

Geospatial data comes in two fundamental categories — raster (grid-based) and vector (geometry-based) — plus several specialized formats for point clouds, time series, and tabular spatial data. Understanding these formats is essential for choosing the right tools and workflows.


Raster Data

Raster data represents the world as a regular grid of cells (pixels), where each cell holds a value. This is the primary format for satellite imagery, elevation models, and continuous surfaces.

GeoTIFF (.tif, .tiff)

The most widely used raster format. A standard TIFF image extended with georeferencing metadata (CRS, extent, resolution) embedded in the file header.

  • Use: Satellite imagery, elevation models, land cover maps
  • Tools: GDAL, Rasterio, QGIS, Google Earth Engine
  • Spec: OGC GeoTIFF Standard

Cloud Optimized GeoTIFF (COG)

A GeoTIFF with internal tiling and overviews that allows efficient partial reads over HTTP, enabling cloud-native workflows without downloading entire files.

  • Use: Cloud-hosted imagery, web map services
  • Tools: GDAL, Rasterio, STAC catalogs
  • Spec: COG Specification

NetCDF (.nc)

Network Common Data Form — a self-describing, machine-independent format for array-oriented scientific data. Widely used in climate and atmospheric science.

  • Use: Climate model output, reanalysis data, oceanographic measurements
  • Tools: xarray (Python), ncdf4 (R), CDO, NCO
  • Spec: Unidata NetCDF

HDF5 (.h5, .hdf5) / HDF-EOS

Hierarchical Data Format version 5 — a flexible format for large, complex datasets with internal groups and metadata. HDF-EOS is NASA's extension for Earth observation data.

  • Use: Satellite sensor data (e.g., MODIS, Landsat Collection 2), multi-dimensional arrays
  • Tools: h5py (Python), GDAL, HDFView
  • Spec: HDF Group

JPEG2000 (.jp2)

A wavelet-based compressed image format used by the European Space Agency for Sentinel-2 imagery.

  • Use: Sentinel-2 Level-1C and Level-2A products
  • Tools: GDAL, SNAP, Rasterio
  • Spec: ISO/IEC 15444

Zarr (.zarr)

A cloud-native, chunked, compressed array format designed for parallel read/write access. Increasingly used as a modern alternative to NetCDF/HDF5.

  • Use: Analysis-ready climate data, large raster time series
  • Tools: xarray + Zarr (Python), Dask
  • Spec: Zarr Specification

Vector Data

Vector data represents geographic features as points, lines, and polygons with associated attribute tables.

GeoPackage (.gpkg)

An open, SQLite-based format that can store multiple vector layers, raster tiles, and attribute data in a single file. The modern replacement for Shapefiles.

  • Use: Administrative boundaries, infrastructure networks, point observations
  • Tools: QGIS, GDAL/OGR, GeoPandas, sf (R)
  • Spec: OGC GeoPackage

Shapefile (.shp + .shx + .dbf + .prj)

The legacy vector format from ESRI, still widely used despite limitations (2 GB size limit, 10-character field names, no NULL values). Always consists of multiple associated files.

  • Use: Legacy datasets, interoperability with ArcGIS workflows
  • Tools: QGIS, GDAL/OGR, GeoPandas, sf (R)
  • Note: Consider migrating to GeoPackage or GeoJSON for new projects

GeoJSON (.geojson)

A lightweight, text-based vector format using JSON syntax. Human-readable and web-friendly, but not suitable for large datasets.

  • Use: Web mapping, API responses, small to medium datasets
  • Tools: Any JSON parser, GeoPandas, Leaflet, Mapbox
  • Spec: RFC 7946

KML / KMZ (.kml, .kmz)

Keyhole Markup Language — an XML-based format developed for Google Earth. KMZ is a zipped KML with embedded resources.

  • Use: Visualization in Google Earth, simple data sharing
  • Tools: Google Earth, QGIS, GDAL

GeoParquet (.parquet)

A columnar storage format combining Apache Parquet's efficiency with geospatial metadata. Designed for large-scale analytical queries.

  • Use: Big data analytics, cloud-native spatial data processing
  • Tools: GeoPandas, DuckDB, Apache Spark
  • Spec: GeoParquet

Point Cloud Data

LAS / LAZ (.las, .laz)

The standard format for LiDAR point cloud data. LAZ is the compressed version. Each point contains XYZ coordinates plus attributes like intensity, classification, and return number.

  • Use: Terrain modeling, forestry, urban 3D mapping
  • Tools: PDAL, CloudCompare, LAStools, lidR (R)
  • Spec: ASPRS LAS Specification

Metadata and Catalog Formats

STAC (SpatioTemporal Asset Catalog)

A JSON-based specification for describing geospatial assets (imagery, point clouds, etc.) to make them searchable and discoverable.

  • Use: Satellite data catalogs, data discovery
  • Tools: pystac (Python), rstac (R), STAC Browser
  • Spec: STAC Spec

ISO 19115 / INSPIRE

International metadata standards for geographic information, widely used in European spatial data infrastructures.

  • Use: Official metadata records, INSPIRE-compliant data portals
  • Spec: ISO 19115

Format Selection Guide

Use CaseRecommended Format
Satellite imagery (local)GeoTIFF
Satellite imagery (cloud)Cloud Optimized GeoTIFF (COG)
Climate / atmospheric dataNetCDF or Zarr
Vector features (new projects)GeoPackage
Vector features (web)GeoJSON
Large-scale vector analyticsGeoParquet
LiDAR point cloudsLAS/LAZ
Data catalogsSTAC