Every major GIS file format explained — 75+ formats covering vector, raster, elevation, database, cloud-native, CAD, temporal, project, cartographic, 3D, and indoor mapping. File parts, software support, usage rankings, pros & cons, and who created each one.
Detailed cards cover the top formats with full file components, software lists, pros/cons, and restrictions. Compact rows cover additional formats with descriptions and file parts. Color bands in the legend match the left border on each card. The quick reference table at the bottom covers all 73 formats at a glance.
Store geographic features as points, lines, and polygons with attribute data attached.
The Shapefile is the de facto standard for GIS vector data exchange and has been for 30 years. Despite its age and well-documented limitations, nearly every piece of GIS software on the planet reads and writes shapefiles. While newer formats (GeoPackage, GeoJSON, FlatGeobuf) are technically superior, shapefiles remain inescapable in government, academia, and enterprise workflows. ESRI published the specification openly in 1998, which is a key reason for its universal adoption. Note: a "shapefile" is actually a collection of multiple files — all must be kept together in the same folder.
Sharing vector GIS data between organizations, desktop GIS analysis, government data distribution, importing data into virtually any GIS tool. Most US federal agencies distribute data in shapefile format.
Open specification — freely readable and writable by any software. ESRI trademarks the name "Shapefile" but the format itself is unrestricted. The .dbf attribute format has a 255 field limit.
GeoJSON is the dominant format for web mapping applications. It encodes geographic data structures — points, lines, polygons, and their attributes — as JSON text, making it human-readable and directly parseable by JavaScript without any library. Always uses WGS84 (EPSG:4326) coordinate system as mandated by RFC 7946. A single .geojson file contains everything: geometry and attributes together. Perfect for REST APIs and web services.
Web mapping APIs, JavaScript applications, sharing small to medium datasets online, REST API responses, Leaflet and Mapbox maps, GitHub (renders GeoJSON files as maps automatically).
Fully open IETF standard. No licensing restrictions. Coordinate system is locked to WGS84 by the RFC specification.
KML is an XML-based format developed originally for Google Earth. It encodes placemarks, lines, polygons, image overlays, 3D models, and time-based animations. KMZ is simply a ZIP-compressed version of KML, often bundling embedded icons and images. KML became an OGC international standard in 2008. Uses WGS84 decimal degree coordinates. The primary visualization format for Google Earth — nothing else matches its styling richness (custom icons, animated paths, photo overlays, time sliders).
Google Earth projects, sharing maps with non-GIS users, animated and time-enabled visualizations, government alert feeds (NOAA hurricane tracks, USGS earthquakes), educational map sharing.
Open OGC standard. Google maintains the KML specification. No licensing fees. KMZ unpacks with any ZIP tool.
GML is the OGC standard XML grammar for expressing geographical features. It can represent points, curves, surfaces, topology, and coverage data, along with coordinate reference systems. GML is the underlying encoding used by many OGC web services (WFS responses). More verbose than GeoJSON — the same data results in a much larger GML file — but it carries richer semantic meaning and supports more geometry types including 3D and time.
OGC WFS service responses, government data exchange (especially European INSPIRE directive), detailed topographic data, interoperability between enterprise GIS systems.
Open OGC standard. No licensing fees. Multiple GML versions (2.1, 3.2, etc.) which can cause compatibility issues.
OSM is the native XML format of the OpenStreetMap project — the world's largest crowd-sourced geographic database with over 10 billion GPS points. .osm files are XML-based; .pbf (Protocol Buffer Binary Format) is a compact binary alternative that's typically 10–50× smaller. OSM data uses a node/way/relation model rather than features/attributes, making it semantically richer but requiring conversion for most GIS workflows.
Free base map data, routing and navigation, street-level analysis, building footprints, land use, points of interest globally. Download regional extracts from Geofabrik or BBBike.
ODbL (Open Database License) — free to use but derived products must also be released as open data with attribution. Attribution required: "© OpenStreetMap contributors."
GPX is the universal exchange format for GPS data — waypoints, tracks, and routes recorded by GPS receivers. It's XML-based and stores latitude, longitude, elevation, and timestamps. Nearly every GPS device, fitness tracker, and hiking app exports GPX. Not a GIS analysis format but essential for bringing real-world GPS data into GIS workflows. QGIS and ArcGIS both import GPX natively.
Importing GPS tracks from outdoor activities, field survey routes, drone flight paths, and vehicle GPS logs into GIS software. Convert to shapefile or GeoJSON for analysis.
Open standard. No restrictions. Free to use in any application.
TopoJSON is an extension of GeoJSON that encodes topology — shared boundaries between polygons are stored once rather than duplicated, dramatically reducing file sizes (often 80% smaller than equivalent GeoJSON). Ideal for web maps of complex political boundaries or census geography. Introduced by Mike Bostock as part of the D3.js visualization library. Not supported natively by most desktop GIS software — primarily a web format.
D3.js web maps, data visualization dashboards, Census boundary maps for web, reducing bandwidth for complex polygon datasets served over the web.
Open source, ISC license. No restrictions on use.
MapInfo TAB is the native vector format for MapInfo Professional (now Precisely MapInfo). Common in telecommunications, utilities, and government organizations that have historically used MapInfo software. Like shapefiles, it requires multiple component files. MapInfo MIF/MID (.mif, .mid) is a text-based exchange format for MapInfo data that is more widely supported.
Telecom network mapping, utility asset management, organizations migrating from MapInfo to modern GIS platforms. Convert to GeoPackage or Shapefile for broader compatibility.
Proprietary to Precisely (formerly MapInfo). Reading requires MapInfo Pro or GDAL. No open specification published.
Store data as a grid of cells (pixels), each with a value — used for imagery, satellite data, temperature grids, and continuous surfaces.
GeoTIFF is the gold standard for distributing georeferenced raster data. It extends the TIFF format by embedding geospatial metadata — coordinate reference system, spatial extent, and pixel size — directly into the image tags. Supports 8-bit to 32-bit data, multiple bands, lossless and lossy compression, and values from −∞ to +∞ (including negative numbers, critical for elevation data). The Cloud Optimized GeoTIFF (COG) variant adds internal tiling for efficient HTTP range-request access from cloud storage.
Satellite imagery distribution, DEM/elevation data, aerial photography, climate data grids, remote sensing analysis, any georeferenced raster data intended for analysis or archive.
Open OGC standard. TIFF specification maintained by Adobe. No licensing restrictions. DEFLATE and LZW compression are free; some older implementations used LZW under patent (now expired).
ERDAS IMAGINE's native raster format, widely used in remote sensing and satellite image processing workflows. ESRI recommended .img as the preferred raster format for ArcGIS for many years (over the older GRID format). Supports very large rasters, hierarchical pyramid levels for fast display, and all data types. Despite being proprietary to Hexagon, it is so widely supported by GDAL that it has become semi-standard in remote sensing.
Remote sensing image processing, multispectral and hyperspectral analysis, large satellite image mosaics, ArcGIS raster analysis workflows.
Proprietary to Hexagon Geospatial (ERDAS). No published open specification. GDAL provides read/write support. Not recommended for data exchange — use GeoTIFF instead.
MrSID uses wavelet compression to produce extremely compact raster files while preserving visual quality at multiple zoom levels — think of it as JPEG for very large geospatial images. For many years it was the preferred format for distributing large aerial and satellite images because it compressed an entire state's aerial photography to manageable size. USGS distributed historical aerial imagery as .sid files. Now largely replaced by COG (Cloud Optimized GeoTIFF) in modern workflows.
Reading legacy aerial photography archives, USGS historical imagery downloads, large image mosaics from legacy projects. New projects should use COG GeoTIFF.
Reading .sid files is free (GDAL supports it). Encoding/creating MrSID files requires a commercial license from Extensis. Decode SDK is royalty-free.
ECW is a wavelet-based compressed raster format for very large aerial and satellite imagery. Similar in purpose to MrSID, it was the dominant compressed imagery format in Australia and Europe, where many national mapping agencies distribute aerial imagery as ECW. Hexagon's SDK enables reading in most GIS software, but encoding requires a commercial license. Like MrSID, being supplanted by COG in modern cloud workflows.
Legacy national imagery archives (Australia, New Zealand, UK), very large aerial mosaics already in ECW format. Convert to COG GeoTIFF for new projects.
Free decode SDK from Hexagon. Commercial license required to create ECW files above 500 MB. GDAL reads ECW with the Hexagon SDK installed.
JPEG 2000 is an open ISO wavelet compression standard that supports both lossy and lossless compression, multiple bands, and georeferencing metadata. Used by the European Space Agency for Sentinel satellite data distribution, US national archives, and as the preferred format for digital cinema. Unlike standard JPEG, it avoids block artifacts and supports transparency and alpha channels. Not to be confused with standard .jpg files.
ESA Sentinel satellite imagery, archival imagery, government imagery repositories, situations requiring lossless compression with efficient random access.
Open ISO/IEC 15444 standard. No licensing fees for the format itself. Some encoder implementations may have patents but free implementations exist via OpenJPEG.
NetCDF is the dominant format for scientific climate and atmospheric data. It stores multi-dimensional arrays — think a temperature grid for every day for 50 years stacked into a single file. Used by NOAA, NASA, ECMWF, and climate researchers worldwide. Supports the CF (Climate and Forecast) metadata conventions which standardize variable names, units, and coordinate systems. HDF5 is a closely related format used for similar purposes.
Climate model output, weather forecast grids, ocean temperature time series, multi-temporal satellite data analysis, any multi-dimensional scientific raster data with time components.
Open standard maintained by Unidata. No licensing fees. Free libraries available in C, Python, Java, Fortran, and R.
Formats specifically designed for Digital Elevation Models (DEMs), terrain surfaces, and point clouds.
The USGS DEM (.dem) is a raster grid of elevation values produced by the US Geological Survey, historically distributed at 30-meter (1 arc-second) and 10-meter (1/3 arc-second) resolutions. The ASCII grid (.asc) is a simple text format where each line is a row of elevation values — easy to inspect in a text editor and import into any GIS. Both formats have largely been replaced by the more modern 3DEP (3D Elevation Program) products distributed as GeoTIFF.
Legacy USGS elevation data, terrain analysis, slope and aspect mapping, viewshed analysis. New projects should download 3DEP data as GeoTIFF from the USGS National Map.
USGS National Map Downloader: apps.nationalmap.gov. Download modern 3DEP elevation as GeoTIFF from the same source.
LAS is the standard binary format for lidar (Light Detection and Ranging) point cloud data. Each point stores X, Y, Z coordinates, intensity, return number, classification, and RGB color. LAZ is a losslessly compressed version (using LASzip) typically 5–10× smaller. COPC (Cloud Optimized Point Cloud) is a new LAZ variant optimized for cloud streaming via HTTP range requests. USGS 3DEP program now distributes national lidar coverage as LAZ/COPC.
Lidar-derived DEMs, 3D building models, forest canopy height, powerline inspection, flood inundation modeling, archaeological survey, autonomous vehicle mapping.
Open ASPRS standard. LASzip compression is open source (LGPL). USGS distributes national lidar data as LAZ for free at nationalmap.gov.
ESRI's proprietary raster format stored as a folder of .adf files rather than a single file. It was the standard raster analysis format in ArcGIS for decades before being supplanted by GeoTIFF and ERDAS .img. Still encountered in legacy ArcGIS projects and older government data archives. QGIS and GDAL can read ArcInfo GRID folders. Not recommended for new projects.
Reading legacy ESRI raster data and ArcGIS geoprocessing outputs from older workflows. Convert to GeoTIFF for modern use.
ESRI proprietary format. No public specification. Writing requires ArcGIS. GDAL can read but not write. Superseded by GeoTIFF and GPKG raster tiles.
Store multiple layers, relationships, and rules in a single container. The future of GIS data storage.
GeoPackage is a modern, open replacement for the shapefile. Based on SQLite — a self-contained database engine — a single .gpkg file can contain multiple vector layers of different geometry types, raster tile pyramids, extensions, attribute relationships, and metadata. It is now the default vector format in QGIS. Supports field names longer than 10 characters, NULL values, multiple geometry types per file, and no 2 GB file size limit. A single .gpkg file replaces an entire folder of shapefiles.
All data — geometry, attributes, projections, metadata, and raster tiles — lives inside the single .gpkg file. Copy or share one file only.
Modern desktop GIS projects, distributing multi-layer datasets, mobile offline mapping (SQLite is great on devices), replacing shapefiles in all workflows, storing both vector and raster tiles together.
Open OGC standard. SQLite (public domain). No licensing restrictions whatsoever. Freely readable and writable by any software. Highly recommended for all new projects.
The ESRI File Geodatabase (.gdb) is a folder-based database format for ArcGIS that supports multiple vector layers, rasters, topology, network datasets, annotation, geometric networks, and domains. It handles files up to 1 TB per table. The standard data format for ArcGIS Desktop and Pro users. ESRI released the read API publicly in 2011 (via GDAL/OGR), enabling non-ESRI software to read .gdb files. Writing still requires ArcGIS or FME. The newer OpenFileGDB driver in GDAL 3.6+ can read all versions without the ESRI SDK.
Enterprise ArcGIS projects requiring topology, network analysis, or annotation. Government and utility data often distributed as .gdb. Convert to GeoPackage for cross-platform sharing.
Read: free via GDAL OpenFileGDB driver. Write: requires ArcGIS or FME license. No open write specification published by ESRI.
SpatiaLite adds spatial capabilities to SQLite, enabling standard SQL queries with spatial operators (ST_Contains, ST_Intersects, etc.) on geographic data. Think of it as a lightweight PostGIS for a single file. QGIS supports SpatiaLite natively. It stores geometry as WKB with a spatial reference in one SQLite database file. Similar to GeoPackage conceptually, but with a different internal structure and older lineage. GeoPackage has largely superseded SpatiaLite for new projects.
Lightweight desktop spatial databases, Python/R analysis workflows, embedded GIS in applications, rapid prototyping with SQL spatial queries. Use GeoPackage for new desktop projects.
Open source (MPL 1.1). No restrictions. SQLite is public domain.
Formats designed for the era of cloud storage and browser-based GIS — access subsets of huge datasets without downloading the whole file.
Traditional formats require downloading a complete file before reading any data. Cloud-native formats use HTTP range requests to read only the bytes you need — a bounding box query on a 100 GB dataset returns in milliseconds without downloading the full file. COG, FlatGeobuf, GeoParquet, PMTiles, and COPC are the key modern formats.
A Cloud Optimized GeoTIFF is a regular GeoTIFF file with internal tiling and overview pyramids arranged so that HTTP range requests can read only the pixels needed for a given bounding box and zoom level — no server-side processing required. It is backwards compatible: any software that reads GeoTIFF can read COG. AWS, Google Cloud, and Microsoft Azure all serve large imagery datasets as COGs. Major providers including Sentinel Hub, Maxar, and Planet distribute data as COGs. The definitive modern raster format.
Serving large raster datasets from cloud storage (S3, GCS, Azure) without a map server, streaming satellite imagery to web applications, national elevation data, Sentinel and Landsat archives.
Open OGC standard. Free. Use GDAL's -of COG option or rio-cogeo Python library to create COGs from any GeoTIFF.
FlatGeobuf is a binary vector format built on Google's FlatBuffers library, designed for fast HTTP range request streaming. It includes a packed Hilbert R-tree spatial index, enabling spatial queries against a remote file without downloading it entirely. Files are typically 40–60% smaller than equivalent GeoJSON. GDAL 3.1+ and QGIS 3.16+ support it natively. Ideal for large national-scale vector datasets served from S3 or similar cloud storage. The vector equivalent of COG.
Serving large vector datasets from cloud storage with spatial querying, OpenLayers and Mapbox web apps, pipeline input for creating PMTiles, replacing GeoJSON for large datasets.
MIT open source license. No restrictions. Convert from any format using ogr2ogr -f FlatGeobuf output.fgb input.shp
GeoParquet extends Apache Parquet — the columnar analytics format used in data lakes and warehouses (BigQuery, Snowflake, DuckDB) — with spatial data types. Geometry is stored as WKB in a regular column. Columnar storage means analytical queries (e.g., "what is the average population density of all polygons in California?") run dramatically faster than row-based formats. Rapidly becoming the standard for large-scale vector analytics. Google Open Buildings, Microsoft Building Footprints, and Overture Maps all distribute data as GeoParquet.
Large-scale vector analytics at cloud scale, data science workflows, building footprint datasets (billions of features), integrating GIS data with SQL analytics engines like DuckDB.
Open OGC standard built on Apache Parquet (Apache 2.0 license). No restrictions. Convert with GDAL: ogr2ogr -f Parquet output.parquet input.gpkg
PMTiles is a single-file archive format for map tiles (vector or raster) stored on cloud storage and served directly via HTTP range requests — no tile server required. Think of it as putting your entire tileset in one file on S3 and pointing a web map at it. It replaces MBTiles for cloud-native tile serving. The Protomaps project distributes a daily OpenStreetMap-derived planet.pmtiles (roughly 80 GB) for self-hosted base maps.
Self-hosted base maps without a tile server, distributing custom styled map tiles, replacing MBTiles in cloud deployments, lightweight web map publishing from S3/R2/GCS.
BSD open source license. Create PMTiles from FlatGeobuf using Tippecanoe (for vector tiles) or from COG imagery (raster tiles).
Engineering and design formats frequently encountered when integrating GIS with architecture, surveying, and infrastructure workflows.
DWG is AutoCAD's native binary drawing format for 2D and 3D designs. DXF (Drawing Exchange Format) is Autodesk's open text-based exchange format designed for interoperability between CAD and GIS systems. Nearly all CAD/GIS software can read DXF. In GIS workflows, DXF is typically used to import building footprints, utility networks, land parcel surveys, and infrastructure drawings. Coordinate systems must be manually specified as CAD files often have no embedded CRS.
Importing building footprints and engineering drawings into GIS, utility network mapping from surveys, land parcel boundary digitizing, converting infrastructure CAD files to vector GIS data.
DXF: open specification published by Autodesk. DWG: proprietary. The Open Design Alliance (ODA) provides libraries for reading DWG outside of AutoCAD.
SVG is an XML-based vector image format for 2D graphics. In GIS contexts, it's used primarily for cartographic output — exporting finished maps from QGIS or ArcGIS as publication-ready vector artwork that scales to any size without pixelation. D3.js renders geographic data as SVG in web browsers. Not a data format for spatial analysis, but essential for map design and publishing workflows.
Exporting finished maps for print or web, post-production map styling in Illustrator or Inkscape, D3.js web map rendering, creating scalable infographic maps.
Open W3C standard. No licensing restrictions. Freely supported in all browsers without plugins.
IDRISI, DLG, GBF-DIME, ArcInfo Coverage, WKT/WKB, TIGER, Geobuf, MIF/MID.
Native vector format for Clark Labs IDRISI (TerrSet). .VCT stores feature geometry; .VDC is the companion metadata file with coordinate system, attributes, and layer info. Rarely encountered outside academic IDRISI workflows. GDAL supports read.
Software: Clark Labs TerrSet · Creator: Clark Labs
Vector format produced by USGS from paper topographic maps. Stores townships & ranges, contour lines, rivers, lakes, roads, railroads, and towns. Much of the early Census TIGER data was derived from DLG. Largely superseded by TIGER/Line shapefiles and modern topographic data.
Software: QGIS (GDAL), ArcGIS · Creator: USGS
Developed by the US Census Bureau in the late 1960s — one of the very first digital GIS formats. Stored the US road network for major urban areas and supported choropleth mapping. The direct predecessor to the TIGER system. No modern software reads it. Documented here for historical completeness.
Creator: US Census Bureau (1967) · No longer supported.
ESRI's topological vector format from the pre-shapefile era. Folders contain points, arcs, polygons, or annotations; tic points are geographic control. Attributes stored in ADF/INFO tables. Mostly unsupported in modern GIS. Files: arc, polygon, tic, info/ subfolder.
Software: Legacy ArcInfo only · Creator: ESRI
WKT is a text encoding for geometries: POINT(10 20), POLYGON(...). Used in SQL, APIs, and as the format for .PRJ projection files. WKB is the compact binary equivalent used inside spatial databases (PostGIS, SpatiaLite, GeoPackage). Not standalone file formats — encoding standards used within other systems.
Software: PostGIS, SpatiaLite, QGIS, GDAL, all spatial DBs · Creator: OGC
The US Census Bureau's authoritative vector dataset for every US geographic entity — counties, census tracts, roads, water, ZIP Code Tabulation Areas (ZCTAs), voting districts, and more. Distributed as shapefiles since 2007 and updated annually. The gold standard for US geographic boundaries. Free at census.gov.
File parts: Standard shapefile set (.shp, .dbf, .shx, .prj) · Creator: US Census Bureau
Compact binary encoding of GeoJSON using Protocol Buffers. Typically 6–8× smaller than GeoJSON while losslessly preserving all geometry and attributes. Developed by Mapbox for bandwidth-constrained transmission of geographic data. GDAL supports read/write. Less common than FlatGeobuf for new projects.
Software: GDAL, Node.js library · Creator: Mapbox / Vladimir Agafonkin
ASCII text-based exchange format for MapInfo vector data. .MIF stores geometry; .MID stores attributes as delimited text. More portable than the binary TAB format since it's editable in any text editor. GDAL/OGR supports read/write. Common in legacy GIS archives from MapInfo-heavy organisations (telecom, utilities, local government).
Software: MapInfo, QGIS (GDAL), ArcGIS · Creator: MapInfo Corp. (now Precisely)
ASCII Grid, IDRISI, ENVI BIL/BIP/BSQ, PCI PIX, HDF, GRIB, JPEG, PNG, DRG, ADRG, Zarr, VRT.
Human-readable raster: a plain text header (ncols, nrows, xllcorner, yllcorner, cellsize, NODATA_value) followed by space-delimited cell values. Editable in any text editor — invaluable for teaching or debugging raster data. QGIS, ArcGIS, GDAL, and R all read natively. Impractically large for big datasets; use GeoTIFF in production.
File parts: Single .asc file · Software: QGIS, ArcGIS, GDAL, R (terra)
Native raster for Clark Labs IDRISI (TerrSet). .RST stores grid values (integers, reals, bytes, RGB24). .RDC is the companion documentation file with columns, rows, file type, coordinate system, and positional error. GDAL supports read. Rarely encountered outside Clark Labs academic software.
Software: TerrSet/IDRISI, GDAL · Creator: Clark Labs
Raw binary raster formats for multi-band remote sensing imagery. Three band-storage layouts: BIL (Band Interleaved by Line), BIP (Band Interleaved by Pixel), BSQ (Band Sequential). Accompanied by a .HDR header file describing dimensions, bands, bit depth, and byte order. USGS used BIL for Landsat distribution. Requires .HDR to be useful.
File parts: .bil/.bip/.bsq + .hdr (required) + .prj (optional) · Software: ENVI, QGIS, ArcGIS, GDAL
Flexible self-contained raster database format from PCI Geomatics. Stores all image data and auxiliary "segments" — channels (varying bit depths), training sites, histograms, projections, attributes, metadata, and even vectors — in one file. Primarily used in Canadian government and defence remote sensing. GDAL supports read/write.
Software: PCI Geomatics, GDAL · Creator: PCI Geomatics (Canada)
Designed by NCSA to manage extremely large, complex scientific data with no limit on objects or file size. HDF4 was the NASA standard for MODIS and early Landsat products. HDF5 (also the basis for NetCDF-4) is the modern version with improved concurrency and performance. GDAL supports both. NASA distributes MODIS as HDF4; many climate models use HDF5.
File parts: Single self-describing binary · Software: ArcGIS, QGIS, GDAL, Python (h5py), HDFView · Creator: NCSA / HDF Group
The World Meteorological Organization's standard for gridded weather and climate data. GRIB2 (the current version) stores multi-dimensional grids: temperature, pressure, wind, precipitation — globally at multiple pressure levels. NOAA GFS, ECMWF ERA5, and all major forecast systems output GRIB. QGIS reads GRIB via GDAL; tools like wgrib2 and cfgrib (Python) process operationally.
Software: QGIS (GDAL), wgrib2, cfgrib, Panoply · Creator: WMO (in use since 1985)
Standard JPEG imagery with a world file (.JGW) for georeferencing. Widely used for web basemaps and aerial photography previews. Critical limitation: JPEG uses lossy compression — never use for analytical rasters (elevation, classification) as compression artifacts corrupt values. Only use JPEG for visual imagery where pixel-perfect values don't matter.
File parts: .jpg + .jgw (world file) · Software: All GIS software, all browsers
PNG uses lossless compression, supports transparency (alpha channel), and is web-native. In GIS: map tiles, scanned maps, image overlays. A world file (.PGW) provides georeferencing. Unlike JPEG, PNG preserves exact pixel values. Dominant format for web map tile caches (WMTS, TileCache, Google Maps-style grids).
File parts: .png + .pgw (optional) · Software: All GIS software, all browsers · Creator: W3C / PNG Group
A georeferenced digital scan of a paper USGS topographic map in GeoTIFF format with a specific 13-color palette. Scans of 7.5-minute and 15-minute USGS quads at 250 DPI. Excellent for historical reference — see what a location looked like before modern development. Superseded by modern vector topo maps but still used for historical comparison.
Software: All GIS software (GeoTIFF-based) · Creator: USGS
Military-standard raster format from the Defense Mapping Agency (now NGA) for mission planning, command and control systems, and aircraft cockpits. ADRG uses an ARC coordinate system. Not used outside defence/military GIS. GDAL supports read. Superseded by Compressed ADRG (CADRG) in military applications.
Software: Military GIS systems, GDAL · Creator: DMA / NGA
Open format for storing large, chunked, compressed N-dimensional arrays in cloud object storage (S3, GCS, Azure). The cloud-native equivalent of NetCDF — entire datasets can be streamed and subset without downloading. NASA's Earthdata Cloud is migrating MODIS and VIIRS to Zarr. Works natively with xarray and Dask for parallel analysis. QGIS support is emerging.
Software: Python (zarr, xarray, Dask) · Creator: Zarr community (Alistair Miles et al.)
An XML "pointer file" describing a virtual mosaic of multiple underlying raster files — no data is copied. A VRT acts like a seamless single layer over thousands of GeoTIFF tiles. Indispensable for national DEM or satellite imagery tile collections. Built in seconds; zero additional disk space. Fully supported by GDAL, QGIS, and ArcGIS.
File parts: Single .vrt XML file referencing source rasters · Software: GDAL, QGIS, ArcGIS · Creator: GDAL Project
DTED, XYZ point clouds, E57, LAS Dataset (LASD).
Military-grade terrain elevation format from the National Geospatial-Intelligence Agency (NGA). Raster grids of elevation captured from aircraft radar and satellites. Three resolution levels: Level 0 = 30 arc-second (~1 km), Level 1 = 3 arc-second (~100 m), Level 2 = 1 arc-second (~30 m). Used in military mission planning, aviation simulation, and defence. Widely supported by GDAL. Attributes via companion TAB files.
Software: ArcGIS, QGIS (GDAL), Global Mapper · Creator: NGA (Defense Mapping Agency)
Simple ASCII text: each row is one point with X, Y, Z coordinates. No standard specification — additional columns may contain RGB, intensity, or return number. Human-readable and editable in any text editor. Often used as an intermediate export step from lidar processing. Impractically slow for millions of points — convert to LAZ for serious work. Related: .txt, .asc, .pts.
Software: QGIS, ArcGIS, CloudCompare, PDAL, GRASS GIS
ASTM E2807 standard for 3D point clouds from terrestrial laser scanners and structured-light scanners. Unlike LAS (aerial lidar-focused), E57 is designed for close-range scanning: building interiors, industrial facilities, archaeological sites. Stores multiple scan positions, panoramic imagery, and full metadata. Supported by CloudCompare, Autodesk ReCap, Leica Cyclone, FARO Scene.
Software: CloudCompare, Autodesk ReCap, Leica Cyclone, PDAL · Creator: ASTM International
ESRI's reference container that points to one or more .LAS or .LAZ files without copying them. Allows ArcGIS to work with collections of lidar tiles as a single dataset — displaying triangulated surfaces and running statistical analysis across tiles. Not a data format — purely a project-management reference for ArcGIS. Stores extent, point counts, and classification statistics from referenced LAS files.
Software: ArcGIS Pro / Desktop only · Creator: ESRI
Multi-user spatial databases, enterprise geodatabases, and tile stores.
The dominant open-source spatial database. PostGIS adds spatial types, indexes (GiST, BRIN, SP-GiST), and functions to PostgreSQL. Supports points, lines, polygons, 3D, geography (sphere), raster, and topology. Powers OpenStreetMap, government agencies, and most large open GIS projects. Hundreds of spatial functions: ST_Intersects, ST_Buffer, ST_Union, ST_DWithin. QGIS, ArcGIS, GRASS, MapInfo, GeoPandas, and R (sf) connect directly.
Software: QGIS, ArcGIS, GRASS, MapInfo, GDAL · Creator: Refractions Research, now PostGIS community
ESRI's older database format based on Microsoft Access (.MDB). Could store multiple vector/raster datasets with relationship classes, but was limited to 2 GB — quickly exhausted by raster data. Superseded by the File Geodatabase (.GDB) which offers 1 TB per table. Still found in older ArcGIS projects. GDAL can read; requires Access or ArcGIS to write.
Software: ArcGIS Desktop, GDAL (read) · Creator: ESRI · Status: Deprecated
SQLite-based container for raster or vector map tile pyramids. Developed by Mapbox for offline mobile mapping. Tiles stored in spherical Mercator (EPSG:3857) only. Widely supported by Mapbox, MapLibre, QGIS, and Leaflet for offline basemaps. Being superseded by PMTiles for cloud-native deployments, but MBTiles remains the dominant format for offline mobile use cases and QGIS offline projects.
File parts: Single .mbtiles SQLite file · Software: Mapbox, QGIS, MapLibre, Leaflet · Creator: Mapbox
"Version Managed Data Store" — the native database for GE Smallworld GIS, widely used by electric, gas, water, and telecom utilities for network asset management. Stores spatial and topological utility networks with built-in version management. No open specification; requires Smallworld software. Rarely encountered outside utility-sector GIS.
Software: GE Smallworld (now Hexagon) only · Creator: GE / Smallworld Systems
ESRI's enterprise spatial layer on top of Oracle, SQL Server, IBM DB2, or PostgreSQL. Supports versioned multi-user editing, replication, backups. An .SDE connection file stores server, database, username, and version. The backbone of large-scale enterprise GIS in government and utilities. Requires ArcGIS Server license. Modern ArcGIS Pro uses direct database connections without separate SDE middleware.
Software: ArcGIS Desktop/Pro, ArcGIS Server · Creator: ESRI
OGC web services, GeoRSS, XYZ tiles — delivered by URL rather than file download.
Serves georeferenced map images (PNG/JPEG) over HTTP. Clients request a map for a bounding box; receive a rendered image — cannot access or edit underlying features. Standard for consuming remote basemaps in QGIS, ArcGIS, Leaflet. USGS, NOAA, and most government portals expose WMS. View-only; use WFS for editable feature access.
Software: QGIS, ArcGIS, Leaflet, OpenLayers · Creator: OGC
Serves actual vector features (geometry + attributes) over HTTP in GML or GeoJSON. Unlike WMS images, WFS data can be downloaded, queried, filtered, and (with WFS-T) edited. Used for live government open data feeds, GeoServer/MapServer deployments. QGIS and ArcGIS connect natively to WFS endpoints.
Software: QGIS, ArcGIS, OpenLayers, GeoServer, MapServer · Creator: OGC
Serves pre-rendered map tiles (PNG/JPEG) from a tile cache in a grid scheme. Much faster than WMS because tiles are cached server-side. The OGC standard equivalent to the XYZ tile scheme. QGIS and ArcGIS support WMTS natively. Most national mapping agencies expose WMTS for their topographic maps.
Software: QGIS, ArcGIS, Leaflet, OpenLayers · Creator: OGC
Extension of RSS/Atom XML feeds that adds geographic geometry (points, lines, polygons) to web feeds. USGS publishes real-time earthquake data as GeoRSS. NOAA uses it for weather alerts. Each feed item carries a location. Simple, widely supported by feed readers and web mapping libraries. Excellent for disaster notification subscriptions.
Software: QGIS, GeoServer, web browsers · Creator: OGC / community
The URL tile scheme popularised by Google Maps and OpenStreetMap: tiles addressed by zoom level (z), column (x), row (y). Not a file format — a URL pattern for raster or vector tile delivery. Every major web mapping framework (Leaflet, Mapbox GL, OpenLayers) and desktop GIS (QGIS, ArcGIS Pro) consume XYZ. Thousands of tile providers use this scheme.
Software: All web mapping libraries, QGIS, ArcGIS Pro · Creator: Google / OSM community
Formats with a time dimension — climate models, weather forecasts, time-series Earth observation.
NetCDF's defining strength is its time dimension: a single file from ECMWF ERA5 contains 40+ years of daily global temperature at 37 pressure levels — billions of values, self-described. The CF (Climate and Forecast) conventions standardise variable names and units across the atmospheric science community, enabling interoperability between NOAA, NASA, ESA, and ECMWF datasets without conversion.
Key temporal use: Climate reanalysis, NWP forecasts, oceanographic time series · Creator: Unidata/UCAR
GRIB2 is the operational weather format. NOAA's 6-hourly GFS forecast, ECMWF's 10-day model, and all hurricane track models are distributed as GRIB2. Each GRIB message contains one variable at one valid time — a full forecast run produces thousands of messages. Python tools (cfgrib, eccodes) make GRIB2 directly accessible for analysis and GIS integration.
Creator: WMO · Key tools: wgrib2, eccodes, cfgrib (Python)
Zarr's chunked array model includes a time axis, enabling efficient subset queries like "temperature for the Gulf of Mexico, 2020–2024" without downloading the full global dataset. NASA EOSDIS is migrating its archive to Zarr as part of the Earthdata Cloud initiative. Works seamlessly with xarray for multi-year time-series analysis in Python.
Creator: Zarr community · Tools: zarr, xarray, Dask
Files storing map layer references, symbology, layouts — not data, but the instructions for displaying it.
.QGZ is the default QGIS 3.2+ format — a ZIP archive of the .QGS XML file plus auxiliary data. Both store map layers (referenced by path), symbology, labelling, print layouts, variables, and QGIS settings. .qgs~ is an auto-backup. When sharing, data files must accompany the project (or use relative paths / embed in QGZ).
Software: QGIS only · Creator: QGIS Project (OSGeo)
ArcGIS Pro's project format — successor to .MXD. An APRX can contain multiple maps, 3D scenes, layouts, toolboxes, database connections, and styles — all in one ZIP-based file. Unlike MXD (one map only), APRX supports a full workspace. Not backwards-compatible with ArcGIS Desktop (ArcMap).
Software: ArcGIS Pro only · Creator: ESRI
The project file for legacy ArcGIS Desktop (ArcMap). Stores a single map's table of contents, layer symbology, labels, data frame settings, and layout. Not forward-compatible with ArcGIS Pro. Use "Save as Copy" to create an APRX. Still encountered frequently when working with pre-2019 ArcGIS projects from government and utilities.
Software: ArcGIS Desktop (ArcMap) only · Creator: ESRI · Status: Legacy
Standardized map layout templates for ArcGIS Desktop. Contain pre-built page layouts, basemap layers, scale bars, and corporate branding for reuse across an organisation. The normal.mxt is ArcGIS Desktop's default template — deleting or resetting it fixes many ArcMap startup problems.
Software: ArcGIS Desktop (ArcMap) only · Creator: ESRI · Status: Legacy
MapInfo Professional's project format. Stores open map windows, layers, symbology, and browser windows. .WOR is the legacy MapInfo workspace; .MWS is the newer MapInfo Pro format. Reopening recreates the session exactly as last saved. Data referenced by path — all data files must be accessible.
Software: MapInfo Professional (Precisely) only · Creator: Precisely / MapInfo
.3DD is for ArcGlobe (global 3D, large study areas). .SXD is for ArcScene (local 3D, smaller areas). Both are legacy from ArcGIS Desktop's 3D Analyst extension, superseded by ArcGIS Pro's integrated 3D scene capability and .APRX project files.
Software: ArcGIS Desktop 3D Analyst · Creator: ESRI · Status: Legacy
Self-contained archive packaging an MXD file with all referenced data sources into one ZIP. Designed for sharing complete map projects outside an organisation where recipients lack network drive access. Recipients extract and have immediate access to the full project. Superseded by ArcGIS Pro Package formats (.PPKX, .MPKX).
Software: ArcGIS Desktop / Pro · Creator: ESRI
Symbology, labels, and icon libraries — how layers look, not what data they contain.
Store symbology definitions without the data itself. When shared with a dataset, ensure consistent display across machines. .LYR for ArcGIS Desktop 10.x; .LYRX (XML-based) for ArcGIS Pro. Specify renderer type, color ramps, label expressions, transparency, and scale ranges. Common in government data distributions alongside shapefiles or GDBs.
Software: ArcGIS Pro/Desktop only · Creator: ESRI
.QML — XML file with full QGIS layer symbology: renderer, colors, symbol sizes, label expressions, scale visibility. Applied per-layer; shareable independently of data. .QLR — extends QML by also pointing to the referenced data source. Both are XML editable in any text editor — the open equivalent of ESRI's .LYR/.LYRX.
Software: QGIS only · Creator: QGIS Project (OSGeo)
Symbol, color, and map element libraries for ArcGIS. ESRI ships domain-specific libraries: geology, petroleum, military, transportation, and more. .STYL for ArcGIS Desktop; .STYLX (SQLite-based) for ArcGIS Pro. Organisations distribute .STYLX files to enforce cartographic standards across teams.
Software: ArcGIS Pro/Desktop only · Creator: ESRI
The OGC's open XML standard for styling vector and raster layers in WMS/WFS services. Defines fill colors, stroke widths, point symbols, label styles, and scale ranges. GeoServer, MapServer, QGIS, and ArcGIS Server all support SLD for styling published web services — the open equivalent of ESRI's .LYR for web map servers.
Software: GeoServer, MapServer, QGIS, deegree · Creator: OGC
Adobe PDF extended with geospatial capabilities. Stores points, lines, polygons, and rasters in geographic space within a PDF. Users can measure distances and view coordinates in Adobe Acrobat. USGS distributes topographic maps as Geospatial PDFs — they display at any zoom and carry the full coordinate system. QGIS and ArcGIS export and import Geospatial PDF.
Software: Adobe Acrobat, QGIS (export), ArcGIS · Creator: Adobe / ISO 32000
COLLADA, SketchUp, glTF/GLB, 3D Tiles — for 3D buildings, terrain, and digital twin city models.
XML-based 3D asset format for Google Earth 3D buildings, structures, and textured objects. A royalty-free ISO standard maintained by Khronos Group. Created in SketchUp, Maya, 3DS Max, or Blender and exported as .DAE for Google Earth placement. ArcGIS Pro also supports COLLADA in 3D scenes. Being superseded by glTF/GLB for web 3D.
Software: Google Earth, ArcGIS Pro, SketchUp, Blender · Creator: Khronos Group / Sony
Native 3D model format for Trimble SketchUp. SKP files (buildings, towers, trees, infrastructure) can be placed in Google Earth as 3D models. The 3D Warehouse hosts millions of free SKP models. Export to COLLADA (.DAE) for broader GIS compatibility. SketchUp Pro or the free web version required to create/edit.
Software: Trimble SketchUp, Google Earth · Creator: Trimble (formerly @Last Software / Google)
The modern web standard for 3D models — often called the "JPEG of 3D." Maintained by Khronos Group. .glTF is a JSON descriptor with external binary assets; .GLB is the compact single-file binary version. Used in CesiumJS, Mapbox GL, ArcGIS Maps SDK, and WebGL applications. Replacing COLLADA in 3D city models. Supported by Blender, Unity, Unreal Engine, and all major 3D tools.
Software: Cesium, Mapbox GL, ArcGIS Maps SDK, Blender, Unity · Creator: Khronos Group
OGC standard for streaming massive 3D geospatial datasets over the web. Organises 3D content (buildings, terrain, point clouds, vector data) in a hierarchical spatial tree for efficient Level-of-Detail streaming to WebGL clients. Cesium developed it; it became an OGC standard in 2019. Used for digital twin city models, photogrammetry meshes, and global terrain streaming.
Software: CesiumJS, ArcGIS Maps SDK, Cesium ion, QGIS (plugin) · Creator: Cesium / OGC
Bentley DGN, DWF — engineering formats that intersect with GIS in infrastructure workflows.
Native design format for Bentley MicroStation — widely used in infrastructure engineering (roads, rail, bridges, utilities) in Europe, Australia, and transport agencies. DGN contains annotation, polylines, polygons, multipath geometry, ColorIndex styles, and a spatial reference. ArcGIS and QGIS can read DGN; bidirectional shapefile conversion via FME or GDAL. DGN V8 (current) differs significantly from legacy DGN V7.
Software: Bentley MicroStation, ArcGIS (read), QGIS (GDAL), FME · Creator: Bentley Systems
Autodesk's "Design Web Format" — compact, secure format for sharing CAD drawings without requiring AutoCAD. DWF is to DWG what PDF is to Word: a view-only distribution format for engineering plans and utility drawings. GIS use is primarily receiving infrastructure drawings from engineering for digitising or georeferencing. Limited direct GIS analysis support.
Software: Autodesk Design Review, ArcGIS · Creator: Autodesk
Legacy exchange formats, geoprocessing tool containers, and classification files.
ESRI's legacy format for exchanging ArcInfo Coverage files. No longer supported in ArcGIS. The .E00 extension increments (.E01, .E02…) for each coverage file. ASCII text-based. To use E00 today: run "Import from E00" in ArcToolbox. Occasionally encountered in 1990s government GIS archives. No modern workflow should produce E00 files.
Software: Legacy ArcInfo only · Creator: ESRI · Status: Unsupported
Container for ESRI geoprocessing tools, toolsets, Python scripts (.PY), and ModelBuilder models. Custom toolboxes are shared across organisations to standardise spatial analysis workflows. .TBX is the ArcGIS Desktop format; .ATBX is the ArcGIS Pro Python Toolbox format. Parameters, validation, and help text are embedded in the toolbox.
Software: ArcGIS Desktop / Pro only · Creator: ESRI
Stores trained classification samples for ESRI's supervised image classification workflow. ECD files capture spectral signatures and statistical profiles learned from training polygons — used during "Classify Raster" in ArcGIS Image Analyst. Applied to classify Landsat/Sentinel imagery into land cover types. Not transferable outside the ESRI ecosystem.
Software: ArcGIS Image Analyst / Pro only · Creator: ESRI
IMDF, AVF, Revit, IFC, CityGML — where GIS meets architecture and smart buildings.
Apple's open standard for indoor venue mapping — airports, malls, transit stations. IMDF is a GeoJSON-based format (a ZIP archive of GeoJSON files) representing levels, units, amenities, openings, and relationships within a building. Apple Maps uses IMDF to display indoor navigation on iPhone and Apple Watch. Submitted via Apple Maps Connect. Progressing toward OGC standardisation.
Software: Apple Maps, FOSS4G tools · Creator: Apple
An additional Apple indoor format complementary to IMDF. Stores GeoJSON files in a structured dataset folder, each representing a feature collection as part of an indoor data collection. Primarily used by Apple's internal venue data pipeline and authorised venue partners. Less publicly documented than IMDF.
Software: Apple Maps ecosystem only · Creator: Apple
.RVT (Autodesk Revit) — dominant BIM format for architectural and structural design. .NWD (Navisworks) — coordination format for clash detection. .IFC (Industry Foundation Classes) — the open ISO 16739 standard for BIM data exchange between all AEC software. IFC is increasingly used to bring building data into GIS for digital twin and smart city applications. ArcGIS Pro and QGIS (with plugins) can import IFC.
Software: Revit (RVT/NWD), all AEC software (IFC) · Creator: Autodesk (RVT/NWD), buildingSMART (IFC)
OGC standard for 3D city models: buildings, bridges, tunnels, land use, vegetation, water bodies, and transport at multiple Levels of Detail (LoD 0–4). Used by national mapping agencies (Germany, Netherlands, Singapore) to distribute official 3D city models. ArcGIS Pro, QGIS (3DCityDB plugin), and FME support CityGML. CityJSON is a lighter JSON-based variant of CityGML with better performance for web use.
Software: ArcGIS Pro, QGIS (plugin), FME, 3DCityDB · Creator: OGC / SIG 3D
⭐ = most frequently encountered in practice. Sort by category using the colour bands.
| Format | Extensions | Category | Creator | Open? | Primary Use / Notes |
|---|---|---|---|---|---|
| VECTOR FORMATS | |||||
| ⭐ Shapefile | .shp .dbf .shx .prj .cpg | Vector | ESRI | Open spec | Universal exchange — 30-year de facto standard |
| ⭐ GeoJSON | .geojson .json | Vector | IETF | ✅ Open | Web mapping APIs, GitHub, REST services |
| ⭐ KML / KMZ | .kml .kmz | Vector | Google / OGC | ✅ OGC | Google Earth, animated tours, NOAA/USGS live feeds |
| GML | .gml .xml | Vector | OGC | ✅ OGC | WFS responses, European INSPIRE directive |
| ⭐ GPX | .gpx | Vector | TopoGrafix | ✅ Open | GPS tracks — Garmin, Strava, AllTrails, surveys |
| ⭐ OpenStreetMap | .osm .pbf | Vector | OSM Foundation | ODbL | Free global crowdsourced vector data |
| TopoJSON | .topojson | Vector | Mike Bostock | ✅ MIT | D3.js web maps, 80% smaller than GeoJSON |
| ⭐ MapInfo TAB | .tab .dat .id .map .ind | Vector | Precisely | ❌ Proprietary | Telecom / utility GIS on MapInfo Pro |
| MapInfo MIF/MID | .mif .mid | Vector | Precisely | Open spec | MapInfo text-based exchange format |
| IDRISI Vector | .vct .vdc | Vector | Clark Labs | ❌ Proprietary | Clark Labs TerrSet academic GIS |
| Digital Line Graph | .dlg | Vector | USGS | ✅ Open | Legacy USGS topo vector — predecessor to TIGER |
| GBF-DIME | (no ext) | Vector | US Census | Historical | 1960s–80s road network — entirely obsolete |
| ArcInfo Coverage | (folder) | Vector | ESRI | ❌ Proprietary | Pre-shapefile topological vector — obsolete |
| WKT / WKB | .wkt .wkb | Vector Encoding | OGC | ✅ OGC | Geometry encoding within databases and APIs |
| Census TIGER/Line | .shp | Vector | US Census | ✅ Public Domain | US geographic boundaries — counties, tracts, roads |
| Geobuf | .geobuf | Vector | Mapbox | ✅ Open | Compact protobuf-encoded GeoJSON |
| RASTER FORMATS | |||||
| ⭐ GeoTIFF | .tif .tiff | Raster | USGS/NASA/Adobe | ✅ OGC | Universal raster — imagery, elevation, analysis |
| ASCII Grid | .asc | Raster | Open | ✅ Open | Human-readable raster; teaching and debugging |
| ⭐ ERDAS IMG | .img .rrd .aux | Raster | Hexagon | ❌ Proprietary | Remote sensing, ArcGIS preferred raster 2000–2020 |
| IDRISI Raster | .rst .rdc | Raster | Clark Labs | ❌ Proprietary | Clark Labs TerrSet native raster |
| ENVI BIL/BIP/BSQ | .bil .bip .bsq .hdr | Raster | Open Spec | ✅ Open | Multi-band remote sensing, legacy Landsat dist. |
| PCI PCIDSK | .pix | Raster | PCI Geomatics | ❌ Proprietary | Satellite image processing — Canadian/defence GIS |
| ⭐ MrSID | .sid .sdw | Raster | Extensis | Decode free | Legacy compressed aerial imagery archives |
| ⭐ ECW | .ecw | Raster | Hexagon | Decode free | National aerial imagery — Australia, EU |
| JPEG 2000 | .jp2 .j2k | Raster | ISO/JPEG | ✅ ISO | ESA Sentinel satellite data, archival imagery |
| ⭐ NetCDF | .nc .nc4 | Raster / Temporal | Unidata / UCAR | ✅ Open | Climate model output — NOAA, NASA, ECMWF |
| HDF | .hdf .h4 .h5 | Raster / Temporal | NCSA / HDF Group | ✅ Open | NASA MODIS/VIIRS, scientific supercomputing |
| GRIB / GRIB2 | .grb .grb2 | Raster / Temporal | WMO | ✅ WMO | Operational weather forecasts — GFS, ECMWF |
| JPEG | .jpg .jgw | Raster | ISO / JPEG | ✅ Open | ⚠️ Visual imagery only — lossy, not for analysis |
| PNG | .png .pgw | Raster | W3C | ✅ Open | Web tiles, map overlays — lossless compression |
| DRG | .tif (GeoTIFF) | Raster | USGS | ✅ Open | Scanned USGS paper topographic maps |
| ADRG | .adrg | Raster | NGA / DMA | Gov | Military mission planning raster maps |
| ESRI GRID | .adf (folder) | Raster | ESRI | ❌ Proprietary | Legacy ArcGIS raster analysis output |
| GDAL VRT | .vrt | Raster (Virtual) | GDAL Project | ✅ Open | Virtual mosaic of raster tile collections |
| Zarr | .zarr | Raster / Cloud | Zarr community | ✅ Open | Cloud-native N-D arrays — NASA cloud migration |
| ELEVATION & LIDAR FORMATS | |||||
| USGS DEM | .dem .asc | Elevation | USGS | ✅ Open | Legacy elevation — superseded by 3DEP GeoTIFF |
| DTED | .dt0 .dt1 .dt2 | Elevation | NGA | Gov | Military terrain elevation — aviation, defence |
| ⭐ LAS / LAZ / COPC | .las .laz .copc.laz | LiDAR | ASPRS | ✅ Open | Lidar point clouds — USGS 3DEP national coverage |
| XYZ / ASCII Point Cloud | .xyz .txt .pts | LiDAR | Open | ✅ Open | Simple point cloud export — human readable |
| E57 | .e57 | LiDAR | ASTM | ✅ ASTM | Terrestrial laser scanning — buildings, interiors |
| LAS Dataset (LASD) | .lasd | LiDAR Reference | ESRI | ❌ Proprietary | ArcGIS lidar project file — not data |
| DATABASE & ENTERPRISE FORMATS | |||||
| ⭐ GeoPackage | .gpkg | Database | OGC | ✅ OGC | Modern shapefile replacement — single multi-layer file |
| ⭐ ESRI File Geodatabase | .gdb (folder) | Database | ESRI | Read only | ArcGIS enterprise — topology, networks, rasters |
| SpatiaLite | .sqlite .sl3 | Database | A. Furieri | ✅ Open | Lightweight SQL spatial — Python / R workflows |
| ⭐ PostGIS + PostgreSQL | (server) | Enterprise DB | PostGIS comm. | ✅ Open | Production spatial database — powers most open GIS |
| ESRI Personal GDB | .mdb | Database | ESRI | ❌ Proprietary | Deprecated — 2 GB limit, replaced by .gdb |
| MBTiles | .mbtiles | Database / Tiles | Mapbox | ✅ Open | Offline tile storage — mobile, Mapbox, QGIS |
| GE Smallworld VMDS | .vmds | Enterprise DB | Hexagon | ❌ Proprietary | Utility network management — electric, gas, water |
| ArcSDE Enterprise GDB | .sde | Enterprise DB | ESRI | ❌ Proprietary | Multi-user GIS on Oracle / SQL Server |
| CLOUD-NATIVE & WEB FORMATS | |||||
| ⭐ COG | .tif | Cloud Raster | OGC / Community | ✅ OGC | Serverless raster — direct S3/GCS HTTP access |
| ⭐ FlatGeobuf | .fgb | Cloud Vector | B. Harrtell | ✅ MIT | Cloud vector streaming with spatial index |
| ⭐ GeoParquet | .parquet | Cloud Vector | OGC / Apache | ✅ Open | Big data analytics — DuckDB, BigQuery, Snowflake |
| PMTiles | .pmtiles | Cloud Tiles | Protomaps | ✅ BSD | Serverless tile hosting from object storage |
| OGC WMS | URL | Web Service | OGC | ✅ OGC | Map image service — view only |
| OGC WFS | URL | Web Service | OGC | ✅ OGC | Feature service — download and edit |
| OGC WMTS | URL | Web Tiles | OGC | ✅ OGC | Cached tile service — faster than WMS |
| GeoRSS | URL .xml | Web Feed | OGC / Community | ✅ Open | Location-aware feeds — earthquake / weather alerts |
| XYZ Tiles | {z}/{x}/{y} | Web Tiles | Google / OSM | ✅ Open | Slippy map tiles — Leaflet, Mapbox, QGIS basemaps |
| CAD & ENGINEERING FORMATS | |||||
| ⭐ DXF | .dxf | CAD | Autodesk | ✅ Open spec | CAD-to-GIS exchange — surveys, utilities, infrastructure |
| DWG | .dwg | CAD | Autodesk | ❌ Proprietary | AutoCAD native — requires ODA library to open |
| DWF | .dwf | CAD | Autodesk | ❌ Proprietary | View-only CAD distribution format |
| DGN | .dgn | CAD | Bentley | ❌ Proprietary | Infrastructure CAD — roads, rail, bridges |
| SVG | .svg | Graphics | W3C | ✅ Open | Cartographic output, D3.js, Illustrator finishing |
| PROJECT FILES & CARTOGRAPHIC STYLES | |||||
| QGIS Project | .qgz .qgs | Project | QGIS / OSGeo | ✅ Open | QGIS map layers, symbology, print layouts |
| ArcGIS Pro Project | .aprx | Project | ESRI | ❌ Proprietary | ArcGIS Pro maps, scenes, toolboxes, connections |
| ArcGIS MXD | .mxd | Project | ESRI | ❌ Proprietary | ArcMap project — legacy, not Pro compatible |
| MXT Template | .mxt | Project | ESRI | ❌ Proprietary | ArcGIS Desktop map layout templates |
| MapInfo Workspace | .wor .mws | Project | Precisely | ❌ Proprietary | MapInfo session workspace |
| ArcGlobe / ArcScene | .3dd .sxd | Project | ESRI | ❌ Proprietary | Legacy 3D viewer projects — superseded by Pro |
| Map Package | .mpk | Interchange | ESRI | ❌ Proprietary | MXD + data bundled for external sharing |
| ESRI Layer File | .lyr .lyrx | Cartographic | ESRI | ❌ Proprietary | Symbology template distributed with datasets |
| QGIS QML / QLR | .qml .qlr | Cartographic | QGIS / OSGeo | ✅ Open | QGIS symbology and layer definition |
| ESRI Style File | .styl .stylx | Cartographic | ESRI | ❌ Proprietary | Symbol libraries — geology, petroleum, military |
| OGC SLD | .sld | Cartographic | OGC | ✅ OGC | Open styling for WMS/WFS services (GeoServer) |
| Geospatial PDF | .pdf | Cartographic | Adobe / ISO | ✅ ISO | USGS topo maps, georeferenced PDF distribution |
| ArcInfo Interchange E00 | .e00 | Interchange | ESRI | ❌ Legacy | 1990s coverage exchange — obsolete |
| ArcGIS Toolbox | .tbx .atbx | Tool | ESRI | ❌ Proprietary | Geoprocessing tool containers — ModelBuilder / Python |
| ESRI ECD | .ecd | Classification | ESRI | ❌ Proprietary | Supervised classification training samples (ArcGIS) |
| 3D, INDOOR MAPPING & BIM FORMATS | |||||
| COLLADA | .dae | 3D | Khronos | ✅ Open ISO | 3D buildings in Google Earth, ArcGIS scenes |
| SketchUp | .skp | 3D | Trimble | ❌ Proprietary | Architectural 3D models — Google Earth placement |
| glTF / GLB | .gltf .glb | 3D | Khronos | ✅ Open | Modern web 3D — Cesium, Mapbox, ArcGIS Maps SDK |
| 3D Tiles | .3dtiles | 3D | Cesium / OGC | ✅ OGC | Streaming 3D cities, photogrammetry, point clouds |
| IMDF | .imdf (zip) | Indoor | Apple | ✅ Open | Airport / mall indoor maps for Apple Maps |
| Apple Venue Format | .avf | Indoor | Apple | ❌ Proprietary | Apple Maps indoor venue pipeline |
| Revit / Navisworks BIM | .rvt .nwd | BIM / Indoor | Autodesk | ❌ Proprietary | Architectural BIM — digital twin import to GIS |
| IFC | .ifc | BIM / Indoor | buildingSMART | ✅ ISO 16739 | Open BIM exchange — ArcGIS / QGIS for smart cities |
| CityGML / CityJSON | .gml .json | 3D City Model | OGC | ✅ OGC | 3D city models LoD 0–4 — national mapping agencies |