Copernicus DEM download and formats: GLO-30, GLO-90, files, access, and GIS workflows
Copernicus DEM is a global and European digital surface model produced within the Copernicus Earth-observation ecosystem. Its main public global instances are GLO-30 and GLO-90, while EEA-10 provides 10-metre coverage for eligible users across Europe. This guide explains what the dataset represents, how its formats differ, where to download it, how to use it in GIS, and what its surface-model design means for terrain, hydrology, visibility, infrastructure, and location-based analysis.
Copernicus DEM is designed for analyzing the elevation of the Earth’s observed surface, not simply the bare ground. The model includes buildings, infrastructure, and vegetation, making it a digital surface model (DSM). For most general users, the important choices are between GLO-30 and GLO-90, between product downloads and processed API outputs, and between formats such as GeoTIFF and DTED. The right choice depends on geographic coverage, required detail, processing environment, and whether the analysis needs surface height or ground elevation.
How EDK fits
For: GIS analysts, remote-sensing specialists, cartographers, civil and environmental researchers, software developers, infrastructure planners, and anyone building spatial workflows that require elevation or surface-height information.
Copernicus DEM is a strong example of how a physical-world dataset becomes useful when connected to other geographic objects. Elevation is rarely the final answer by itself: it helps explain roads, buildings, drainage networks, rivers, infrastructure, vegetation, flood exposure, visibility, and access across real places. In an Earth-scale world model such as E2, a DEM can serve as a terrain and surface-height layer that gives surrounding features physical context. The connection is analytical rather than a claim that Copernicus DEM is part of E2: the dataset can support questions such as how steep a road corridor is, which buildings sit below a drainage divide, which infrastructure lies on a slope, or how surface elevation varies around streams and developed land.
- •Combine Copernicus DEM with roads, buildings, waterways, and infrastructure to rank or filter physical features by slope, elevation, terrain position, or proximity to drainage pathways.
- •Use the DEM as an elevation context layer when comparing places, aggregating terrain statistics by neighborhood or watershed, modeling viewsheds, or identifying geographic areas where built assets and natural systems interact.
- •Use GLO-90 for broad regional screening and GLO-30 where the available coverage and computational budget support more detailed terrain and surface analysis.
Dataset profile
Formats
DGED 32-bit floating-point GeoTIFF with XML metadata and quality layers.
DTED 16-bit signed-integer DT1 for GLO-90 and DT2 for GLO-30, with XML metadata.
INSPIRE GeoTIFF for EEA-10, using projected coordinate systems for the relevant European areas.
Processed TIFF, PNG, or other API response formats for bounded requests and visualization, depending on the service and request configuration.
Cloud Optimized GeoTIFF holdings are available through a public AWS mirror for GLO-30 Public and GLO-90.
Access methods
Copernicus Browser for interactive discovery and download.
Copernicus Data Space OData API for catalog search, filtering, and automated product downloads.
S3-compatible object storage through Copernicus Data Space for bulk access with authenticated temporary credentials.
Sentinel Hub DEM processing services for API-based, bounded-area DEM requests and derived outputs.
AWS Registry of Open Data mirror for public Cloud Optimized GeoTIFF access to specified GLO-30 Public and GLO-90 holdings.
Quality and limits
DSM semantics mean heights may reflect vegetation, buildings, infrastructure, bridges, and other surface objects.
GLO-30 public access may not expose every idealized global 30-metre tile, and some API products may use GLO-90 infilling where 30-metre tiles are not publicly released.
NoData values, including the documented -32,767 value in relevant product contexts, must be masked before calculations.
The product’s mixed-source construction means quality and characteristics can vary geographically.
The dataset is not automatically suitable for parcel-scale engineering, safety-critical decisions, cadastral surveying, or detailed flood design.
Vertical and horizontal reference systems must be reconciled before combining values with local survey data or other DEMs.
Analysis workflows
Select GLO-30, GLO-90, or EEA-10 based on geographic coverage, access eligibility, required detail, and processing budget.
Download DGED GeoTIFF tiles with XML metadata and quality layers; use DTED only when the target workflow requires it.
Build mosaics or virtual rasters, mask NoData values, and preserve the source release and product identifiers.
Reproject only when required by the analysis, documenting the target CRS and resampling method.
Derive slope, aspect, hillshade, contours, local relief, or zonal elevation statistics at a resolution appropriate to the question.
Join elevation and terrain summaries to roads, buildings, infrastructure, streams, watersheds, service areas, or administrative regions.
For hydrology, assess whether the DSM requires a bare-earth or hydro-conditioned alternative before modeling flow or flood behavior.
For application delivery, use bounded API outputs or cloud-native raster access while retaining native source data for reproducibility.
Value inside EDK
What Copernicus DEM contains
Copernicus DEM represents the Earth’s surface as a digital surface model. The Copernicus Data Space Ecosystem describes the model as including buildings, infrastructure, and vegetation, which is important when interpreting elevation values around cities, forests, industrial sites, bridges, and other above-ground features. It should not automatically be treated as a bare-earth digital terrain model. (dataspace.copernicus.eu)
The dataset is delivered in three principal instances. EEA-10 covers 39 European countries, including their islands and specified French overseas departments, at a 10-metre resolution. GLO-30 and GLO-90 provide global land coverage at approximately 30 metres and 90 metres respectively. The public-access rules are not identical across these instances: the general public can view all instances but can download GLO-30 and GLO-90, while EEA-10 access is limited to eligible categories. (dataspace.copernicus.eu)
- •GLO-30: approximately 30-metre global instance.
- •GLO-90: approximately 90-metre global instance.
- •EEA-10: 10-metre European instance with restricted download eligibility.
- •Surface model rather than an automatically bare-earth terrain model.
- •Global instances cover land surfaces; ocean areas and missing or unavailable pixels require careful handling.
Download options and file formats
The native Copernicus DEM product is available through the Copernicus Data Space Ecosystem. Registered users with the appropriate permissions can search and download GLO-30 or GLO-90 through the Copernicus Browser or the OData API. For full or bulk downloads, the documentation identifies S3-compatible object storage and OData as the main programmatic methods. S3-compatible access is suited to large-scale acquisition, while OData is useful when searching or filtering products before download. (dataspace.copernicus.eu)
The main DGED format is delivered as 32-bit floating-point GeoTIFF data, accompanied by XML metadata and quality layers. GeoTIFF is generally the most convenient choice for raster GIS, desktop analysis, cloud processing, and common geospatial libraries. The product handbook identifies DGED instances for EEA-10, GLO-30, and GLO-90. (dataspace.copernicus.eu)
- •DGED: 32-bit floating-point GeoTIFF plus XML metadata and quality layers.
- •DTED: 16-bit signed-integer DT1 or DT2 files plus XML metadata.
- •INSPIRE: an EEA-10 format using GeoTIFF and projected European or overseas coordinate systems.
- •Processed API outputs: useful for bounded requests, visualization, and application workflows.
- •For analysis archives, retain the original files, metadata, product identifier, release, and download date.
Geography, reference systems, and update cadence
GLO-30 and GLO-90 are intended for worldwide land coverage, with the Copernicus Data Space Ecosystem reporting approximately 149 million square kilometres of surface coverage for the two global instances. The EEA-10 instance covers approximately 6 million square kilometres in its European area of interest. Coverage statements should not be confused with a guarantee that every pixel has identical quality or that every public mirror exposes every tile in the same way. (dataspace.copernicus.eu)
The product handbook specifies geographic coordinates and WGS 84 for DGED and DTED global products, with EGM2008 as the vertical reference. EEA-10 INSPIRE products use projected coordinate systems appropriate to continental Europe and French overseas departments. A workflow that combines Copernicus DEM with local survey data, engineering elevations, or another DEM should explicitly reconcile horizontal and vertical reference systems rather than comparing raw values without transformation. (dataspace.copernicus.eu)
- •Horizontal reference for the global DGED and DTED products: WGS 84 geographic coordinates.
- •Vertical reference: EGM2008, with the exact handling documented in the product metadata.
- •No fixed monthly or annual refresh schedule is established for ordinary user workflows.
- •Coverage, release, permissions, and public availability can differ between GLO-30, GLO-90, EEA-10, and access channels.
Suitable GIS and spatial-analysis workflows
For desktop GIS, download the appropriate DGED GeoTIFF tiles, inspect the XML metadata and quality layers, build a virtual raster or mosaic where necessary, and reproject only when the analysis requires a projected coordinate system. Avoid resampling a 90-metre product to a fine grid and interpreting the result as newly created terrain detail; interpolation changes the grid, not the underlying information content.
For terrain analysis, derive slope, aspect, hillshade, contours, local relief, or terrain statistics after selecting a resolution appropriate to the question. GLO-90 is often suitable for continental or regional screening, broad watershed context, and exploratory visualization. GLO-30 is better suited to more detailed regional analysis when its coverage and access conditions meet the project requirements. Neither resolution should automatically be assumed adequate for parcel-scale engineering, drainage design, or safety-critical elevation decisions.
- •Desktop GIS: GeoTIFF mosaics, reprojection, hillshade, slope, contours, and raster statistics.
- •Cloud or programmatic processing: S3-compatible downloads, OData filtering, or bounded DEM API requests.
- •Terrain context: join zonal elevation and slope statistics to roads, buildings, infrastructure, watersheds, or service areas.
- •Hydrology: use caution because a DSM is not automatically a conditioned bare-earth terrain model.
- •Reproducibility: preserve product IDs, releases, metadata, vertical datum, resampling method, and NoData handling.
Limitations and interpretation cautions
The most important limitation is semantic: Copernicus DEM is a DSM. In a forest, an elevation value may reflect canopy; in a city, it may reflect roofs or other structures. This can be useful for surface visibility and built-environment context but misleading for ground elevation, flood routing, excavation, or detailed engineering.
The product handbook documents a NoData value of -32,767 for pixels where elevation is not provided. Analysts should not let that value enter averages, color stretches, slope calculations, or machine-learning features as if it were a real elevation. Ocean areas may have no tiles in some public distributions, and the AWS description notes that ocean height may be assumed as zero in that mirror’s handling; this assumption should not be silently generalized to every product or workflow. (registry.opendata.aws)
- •DSM values may represent roofs, trees, bridges, and other surface objects.
- •NoData values must be masked before raster calculations.
- •Resolution is not the same as accuracy, and resampling cannot create missing detail.
- •Coverage and access conditions differ by instance and user category.
- •Redistribution and attribution require checking the applicable Copernicus license.
Questions teams ask
Is Copernicus DEM a bare-earth elevation model?
No. Copernicus DEM is described as a digital surface model that includes buildings, infrastructure, and vegetation. It can be useful for surface-height, visibility, and built-environment analysis, but workflows requiring ground elevation or detailed hydrologic modeling may need a bare-earth or conditioned terrain model.
Which Copernicus DEM format should I download for GIS?
For most modern GIS and raster-processing workflows, DGED GeoTIFF is the most practical native format because it is widely supported and comes with XML metadata and quality layers. DTED can be preferable when an existing workflow specifically requires DT1 or DT2 files. Keep the metadata and product identifiers with either format.
Can the general public download GLO-30 and GLO-90?
The Copernicus Data Space Ecosystem states that registered general-public users can download the GLO-30 and GLO-90 instances, subject to the applicable access conditions and license. EEA-10 download access is restricted to eligible user categories. Availability can also vary by access channel and product instance.
Does Copernicus DEM receive regular annual updates?
Not in the sense of a guaranteed monthly or annual refresh schedule. Copernicus DEM is distributed through releases and product versions. Public mirrors may also change tile availability or maintain a particular release. Record the exact release, product ID, and download date for reproducible work.
Can I use Copernicus DEM for flood modeling?
It can provide useful regional elevation context, but it should not be assumed to be ready for detailed flood modeling. Because it is a DSM, buildings, vegetation, bridges, and other surface objects may affect the elevation surface. Check whether a bare-earth, hydro-conditioned, locally validated DEM is more appropriate for the decision.
Keep exploring
copernicus dem analysis workflow
A companion workflow guide for preparing Copernicus DEM tiles, deriving terrain measures, and joining elevation statistics to geographic features.
copernicus dem
A broader reference page for the Copernicus DEM dataset, including its instances, geographic coverage, and role in Earth-observation analysis.
Sources
Copernicus DEM – Global and European Digital Elevation Model
Official Copernicus Data Space overview of EEA-10, GLO-30, GLO-90, access conditions, coverage, licensing, releases, and dataset specifications.
Open sourceCopernicus Contributing Missions Documentation
Official documentation describing the DSM nature of Copernicus DEM, product IDs, instances, resolutions, and download routes.
Open sourceCopernicus DEM Product Handbook
Technical product documentation for coordinate systems, vertical reference, GeoTIFF, DTED, INSPIRE, metadata, quality layers, and NoData handling.
Open sourceCopernicus Data Space FAQ
Official guidance on bulk access through S3-compatible object storage and the OData API.
Open sourceCopernicus DEM license for GLO-90
Official licensing terms describing free public use conditions and restrictions for Copernicus WorldDEM-90.
Open sourceCopernicus Digital Elevation Model on the AWS Registry of Open Data
Documentation for a public cloud mirror describing Cloud Optimized GeoTIFF access, public GLO-30 and GLO-90 holdings, and mirror-specific update behavior.
Open source