Mapping landslide change and volume with repeat airborne LiDAR after the May 2023 Emilia-Romagna event: slope-dependent detection limits and lowered areas
Abstract
Landslide inventories are validated against repeat topography by comparing the ground that changes beyond a detection limit inside mapped polygons with the same fraction on stable ground. Stable ground is gentler than landslides, and LiDAR noise grows with slope. We measure the consequence for the May 2023 Emilia-Romagna event, Italy, with four pairs of 0.5 m airborne LiDAR on 919 river-corridor tiles, the only ground where pre- and post-event surveys overlap, and the 80,997-polygon RER2023 inventory. Stable ground exceeds a tile's 95th-percentile limit on 4% of flat ground and 27% of slopes above 40°, so the rate that applies to the landslides is 13.7%, not 5.9%, and the usual comparison doubles the apparent signal. A fifth pair spanning the first post-event year is consistent with the slope-matched floor in aggregate (14.5% against 14.2%, where all stable ground predicts 6.1%), although gentle and steep ground depart from it in opposite directions. With the corrected floor, 36% of mapped landslide ground changed by more than about 0.5 m. Only 20% of the mapped area of debris and earth landslides was lowered beyond the limit, mostly in the upper half of each polygon, and the same estimator finds no lowering in the post-event year. Area–volume scaling applied to the outlines overstates the measured lowering 8 to 11 times, and on the lowered area the exponent is about 1.1 to 1.3 rather than 1.45. Inventory validation needs slope-matched floors, and inventory-based volumes need lowered or source areas.
Related
- Role of Topo-Climatic Factors in Landslide Susceptibility Mapping: A Machine Learning Perspective
- Landslide susceptibility mapping using machine learning models and analytical hierarchy process along NH-513, Pasighat to Yingkiong, Arunachal Pradesh, India
- Super-resolved canopy height mapping from Sentinel-2 time series using airborne LiDAR HD reference data across metropolitan France
- Dual-stream feature fusion network for landslide detection
- Landslide Susceptibility Modelling and Infrastructure Exposure in the Zillertal, Austrian Alps
BibTeX
@misc{meena2026mapping,
title = {{Mapping landslide change and volume with repeat airborne LiDAR after the May 2023 Emilia-Romagna event: slope-dependent detection limits and lowered areas}},
author = {Sansar Raj Meena},
howpublished = {EarthArXiv},
year = {2026},
month = oct,
doi = {10.31223/x58r5c},
url = {https://eartharxiv.org/repository/view/15884/},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Mapping landslide change and volume with repeat airborne LiDAR after the May 2023 Emilia-Romagna event: slope-dependent detection limits and lowered areas
AU - Sansar Raj Meena
PY - 2026
DA - 2026/10/31
PB - EarthArXiv
DO - 10.31223/x58r5c
UR - https://eartharxiv.org/repository/view/15884/
AB - Landslide inventories are validated against repeat topography by comparing the ground that changes beyond a detection limit inside mapped polygons with the same fraction on stable ground. Stable ground is gentler than landslides, and LiDAR noise grows with slope. We measure the consequence for the May 2023 Emilia-Romagna event, Italy, with four pairs of 0.5 m airborne LiDAR on 919 river-corridor tiles, the only ground where pre- and post-event surveys overlap, and the 80,997-polygon RER2023 inventory. Stable ground exceeds a tile's 95th-percentile limit on 4% of flat ground and 27% of slopes above 40°, so the rate that applies to the landslides is 13.7%, not 5.9%, and the usual comparison doubles the apparent signal. A fifth pair spanning the first post-event year is consistent with the slope-matched floor in aggregate (14.5% against 14.2%, where all stable ground predicts 6.1%), although gentle and steep ground depart from it in opposite directions. With the corrected floor, 36% of mapped landslide ground changed by more than about 0.5 m. Only 20% of the mapped area of debris and earth landslides was lowered beyond the limit, mostly in the upper half of each polygon, and the same estimator finds no lowering in the post-event year. Area–volume scaling applied to the outlines overstates the measured lowering 8 to 11 times, and on the lowered area the exponent is about 1.1 to 1.3 rather than 1.45. Inventory validation needs slope-matched floors, and inventory-based volumes need lowered or source areas.
ER -