This reference system provides a clear picture of the plant structure in an urban environment, distinguishing between the three main strata: herbaceous, shrubby and arboreal.
It is based on a combination of :
- infrared orthophotoplans (revealing the presence of vegetation),
- a digital surface model (DSM) at the height of buildings and vegetation,
- and a digital elevation model (DEM).
Using supervised classification, the vegetation zones are isolated and then their height above ground is calculated using the difference between the DTM and the DTM. The result is a georeferenced raster layer that can be interpreted continuously or using height thresholds.
This dataset is regularly used by the Apur to :
- qualify the plant structure in morphological studies (e.g. block typologies or neighbourhood shapes),
- estimate the urban canopy, in particular to monitor the evolution of trees and identify areas of heavy shading,
- provide input for climate-air-energy diagnostics, by cross-referencing plant heights with permeability, surface materials and built density,
- identify areas lacking in tall or shaded vegetation (priority neighbourhoods, dense housing areas, urban heat island zones),
- assess the potential for greening, particularly in public spaces, on private plots and on roofs.
It is also used to calculate indicators such as :
- the vegetation cover rate (vegetated area in relation to the spatial unit),
- the proportion of tall vegetation (>3 m or >10 m),
- or the density of isolated trees, after detecting the crowns or feet of trees.
This repository is therefore an essential building block in studies of resilience, green grids, urban cooling and adaptation to climate issues. It is cross-referenced with many other data sets produced by the Apur (buildings, plots, rights-of-way, waterproofing, land, etc.) to enhance multi-scale analyses.
Technical Details
The data is provided in raster format (GeoTIFF) and each pixel represents a height of vegetation in metres.
- Spatial resolution: 50 cm
- Projection: Lambert 93 (EPSG:2154)
- File type: continuous raster (GeoTIFF)
- Value: height of vegetation (metres) with a precision of less than one metre, 0 for no vegetation, values >0 for vegetation detected
- Extent: Greater Paris metropolitan area (~3,700 km²)
- Most recent production date: 2021 (based on IGN images taken in summer 2021)
- Update frequency: approximately every 4 to 5 years
Further processing enables other datasets to be derived from this raster layer, in particular :
- height strata (by threshold: <1 m, 1-10 m, >10 m),
- canopy estimates (surface area of vegetation >3 m),
- a spot base of automatically detected trees.
Sources
The data on vegetation height is produced by the Apur using remote sensing techniques applied to aerial or satellite sources, depending on the vintage.
Processing is based on a combination of orthophotoplans (PIR/RVB) and digital elevation models (DTM + DEM). The supplier of the raw data varies from year to year: InterAtlas, Aérodata, or more recently IGN, from the 2021 images.
Vegetation is identified by supervised classification on the infrared orthophotoplane, then its height is estimated by subtracting the digital terrain model (DTM) from the digital surface model (DSM).
Post-processing is carried out to smooth the surfaces, correct artefacts (cast shadows, reflections) and improve consistency with the urban context. The data is then georeferenced (Lambert 93) and cut to fit standard GIS formats.
Licence and conditions of use
Before any use, you are invited to read and accept the terms of the ODbL licence and the limits of use specified here: https://www.apur.org/open_data/resume_licence_ODbl.pdf
Reuse of data
If you have any questions or suggestions, please don't hesitate to contact us at data@apur.org.
We welcome your feedback and contributions to help us enrich these datasets.
