Dataset Viewer
Auto-converted to Parquet Duplicate
Search is not available for this dataset
The dataset viewer is not available for this split.
Server error while post-processing the rows. Please report the issue.
Error code:   RowsPostProcessingError

Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.

chauvet-exterior

High-density aerial LiDAR over the landscape that contains the Chauvet-Pont-d'Arc cave (Ardeche, France): the Cirque d'Estre abandoned meander, the cliff that holds the cave entrance, the Pont d'Arc natural arch, and the surrounding karst plateau at the entrance of the Ardeche gorge.

Hillshade preview with landmarks

The area of interest is a 3 x 3 km block in Lambert-93 (EPSG:2154), X 811000 to 814000, Y 6365000 to 6368000. The decorated cave itself runs under this plateau. Its prehistoric entrance was sealed by successive collapses of the cliff above it roughly 21,500 years ago, which is why the art survived intact until the cave's rediscovery in 1994. The site has been a UNESCO World Heritage property since 2014. The collapse scar, the scree slope, and the geomorphology of the meander are all legible in this data.

What this is not

This dataset does not contain the cave interior. The high-resolution laser scan of the inside of the cave was commissioned by the French Ministry of Culture, is state property, and has never been publicly released. This dataset covers the exterior terrain above and around the cave, from France's national open-data LiDAR program.

Contents

Path Description
pointclouds/*.copc.laz 9 classified point-cloud tiles (1 km each), COPC LAZ, ~1.5 GB total
derived/dtm_mosaic_0m50.tif 50 cm digital terrain model mosaic (bare earth), 6000 x 6000, EPSG:2154
derived/dsm_mosaic_0m50.tif 50 cm digital surface model mosaic (with vegetation/structures)
derived/hillshade_dtm.png DTM hillshade render
derived/hillshade_dsm.png DSM hillshade render
stats.json Point counts, classification breakdown, elevation range, landmark coordinates
tiles.json Full per-tile provenance: source URLs, timestamps, IGN survey block metadata
fetch.py, process.py Scripts that produced everything above, for reproduction

Point classification follows the IGN LiDAR HD scheme: 1 unclassified, 2 ground, 3 low vegetation, 4 medium vegetation, 5 high vegetation, 6 building, 9 water, 17 bridge, 64 perennial above-ground structure, 65 artifact, 66 virtual point.

Landmarks

Landmark EPSG:2154 (X, Y) WGS84 (lon, lat)
Grotte Chauvet (entrance area) 812793, 6366429 4.4160, 44.3878
Pont d'Arc (natural arch) 813188, 6366170 4.4209, 44.3854

Tile naming: LHD_FXX_XXXX_YYYY_... encodes the tile's west edge (XXXX, km) and north edge (YYYY, km) in Lambert-93. The cave sits in tile 0812_6367; the arch spans into 0813_6367.

Stats

279,793,893 points over 9 km² (mean 31.1 pts/m²). Terrain elevation spans 71 m (river level) to 425 m (plateau). Classification mix: 62.9% high vegetation, 28.6% ground (80.0 M bare-earth returns), 4.2% medium vegetation, 3.2% low vegetation, 0.4% building, 0.4% unclassified, 0.2% water, plus marginal bridge and virtual points.

Tile Points Size (MB)
LHD_FXX_0811_6366 31,339,943 174.6
LHD_FXX_0811_6367 28,763,899 159.3
LHD_FXX_0811_6368 28,963,711 153.5
LHD_FXX_0812_6366 27,202,712 152.1
LHD_FXX_0812_6367 (cave) 28,235,737 154.6
LHD_FXX_0812_6368 35,043,895 193.7
LHD_FXX_0813_6366 28,995,193 157.7
LHD_FXX_0813_6367 (arch) 33,464,624 181.6
LHD_FXX_0813_6368 37,784,179 206.6

Quick start

# Point cloud (pip install laspy[lazrs])
import laspy
las = laspy.read("pointclouds/LHD_FXX_0812_6367_PTS_C_LAMB93_IGN69.copc.laz")
ground = las.points[las.classification == 2]

# Terrain model (pip install rasterio)
import rasterio
with rasterio.open("derived/dtm_mosaic_0m50.tif") as src:
    dtm = src.read(1)

Source and license

All data derives from the IGN LiDAR HD program (Institut national de l'information geographique et forestiere, France). Point clouds are the original IGN classified COPC tiles, unmodified. The DTM/DSM mosaics stitch IGN's official 50 cm rasters; hillshades are computed from them. Tile-level source URLs and survey metadata are preserved in tiles.json.

Licence Ouverte / Open Licence 2.0 (Etalab). Attribution: IGN, LiDAR HD. Reuse is free, including commercially, with attribution.

Downloads last month
118