Slope-Relief Threshold Landslide Susceptibility Models for the United States and Puerto Rico
Notice bibliographique
Résumé
Landslide susceptibility maps are essential tools in infrastructure planning, hazard mitigation, and risk reduction. Susceptibility maps trained in one area have been found to be unreliable when applied to different areas (Woodard et al., 2023). This limitation leads to the need for a national map that is higher resolution and rigorous, but simple enough to be applied to diverse terrains and landslide types. The susceptibility maps presented here cover the conterminous United States (CONUS), Alaska (AK), Hawaii (HI), and Puerto Rico (PR) with a resolution of 90-m. Other United States (U.S.) territories were not considered due to insufficient landslide and digital elevation data. We also provide information on the proportion of susceptible terrain as well as the density (landslides per square kilometer) of documented landslides within susceptible terrain for each U.S. county. To generate the susceptibility maps we used 1/3 arc-second digital elevation models (DEMs) (U.S. Geological Survey, 2019) to calculate slope and 100-m relief, 613,724 unique landslides from our national landslide inventory compilation (Belair et al., 2022) to train the models and compute U.S. county aggregated susceptibility information, and high-performance computing resources to train the models (Falgout and Gordon, 2023). We present two slope-relief threshold models: (1) a linear regression model weighted by landslide density of each ecoregion (Wiken et al., 2011), and (2) a quantile nonlinear regression model fitted to the 10th quantile of the data. We (1) removed extraneous landslide data, (2) averaged 50 model runs, and then (3) down-sampled the maps from 10-m to 90-m resolution to account for uncertainty in the DEM and landslide position. The nonlinear model (n10) performs better under most topographic conditions and optimally balances our priorities of capturing observed landslides (98.9%) while minimizing area covered by susceptible terrain (44.6%). The weighted linear model (lw) captures slightly fewer landslides (98.8%) and has slightly less susceptible terrain (43.1%). The values of both maps represent the number of susceptible 10-m cells within each 90-m cell after down-sampling and can range from 0 to 81. While landslides are possible within any cells containing susceptible terrain, those with the highest concentration (or cell value) capture the majority of landslides, thus representing higher susceptibility areas. The susceptibility maps were then used to determine the total area of landslide susceptible terrain (square kilometers) for each U.S. county. The national landslide inventory compilation was used to determine the number of documented landslides within susceptible terrain for each county. This information was then used to calculate the proportion of susceptible terrain and the density of documented landslides within susceptible terrain for each county in the United States. This information is provided in tabular format, with columns corresponding to the information discussed above, and each row corresponding to a U.S. county. Further information about this analysis can be found in an interpretive publication (Mirus et al., 2024). This data release includes: (1) weighted linear susceptibility maps (lw_susc.zip), (2) quantile nonlinear susceptibility maps (n10_susc.zip), (3) landslide data used to develop the models (landslides.csv), (4) county aggregated susceptibility information (county_analysis.csv), (5) readme and analysis files, and (6) metadata. References Cited Belair, G. M., Jones, E. S., Slaughter, S. L., and Mirus, B. B., 2022, Landslide Inventories across the United States version 2: U.S. Geological Survey data release, https://doi.org/10.5066/P9FZUX6N Falgout, J. T., and Gordon, J., 2023, USGS Advanced Research Computing, USGS Yeti Supercomputer: U.S. Geological Survey, https://doi.org/10.5066/F7D798MJ Mirus, B. B., Belair, G. M., Wood, N. J., Jones, J. M., and Martinez, S. M., 2024, Parsimonious high-resolution landslide susceptibility modeling at continental scales, AGU Advances, https://doi.org/10.1029/2024AV001214 U.S. Geological Survey, 2019, 3D Elevation Program (3DEP) USGS 1/3 arc-second DEM [Data set], Retrieved from https://www.usgs.gov/3d-elevation-program/about-3dep-products-services Wiken, E., Nava, F. J., and Griffith, G., 2011, North American Terrestrial Ecoregions - Level III [Data set], Montreal, Canada: Commission for Environmental Cooperation, Retrieved from https://www.epa.gov/eco-research/level-iii-and-iv-ecoregions-continental-united-states Woodard, J. B., Mirus, B. B., Crawford, M. M., Or, D., Leshchinsky, B. A., Allstadt, K. E., and Wood, N. J., 2023, Mapping Landslide Susceptibility Over Large Regions With Limited Data, Journal of Geophysical Research: Earth Surface, 128(5), e2022JF006810, https://doi.org/10.1029/2022JF006810
Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.
Comment cette classification a été obtenuedéplier
Prédiction machine sur la base complète
Imitation des enseignantsNi prévalence calibrée, ni vérité terrain. Validation humaine à venir. Le volet Gemma est une étiquette directe du modèle pour chaque travail de la base, lue sur la notice réduite au titre. Le volet Codex est un classifieur appris des 10 348 étiquettes directes de Codex et calibré sur les taux pondérés de l'échantillon; les champs sans appui suffisant ne portent aucun appel Codex. Le mode candidate est l'union des deux volets; le consensus est leur intersection. Ces sorties portent le statut machine_predicted_unvalidated et ne sont pas des étiquettes humaines.
Scores du classifieur distillé par catégorie (deux têtes)
| Catégorie | Codex | Gemma |
|---|---|---|
| Métarecherche | 0,000 | 0,001 |
| Méta-épidémiologie (sens strict) | 0,000 | 0,000 |
| Méta-épidémiologie (sens large) | 0,000 | 0,001 |
| Bibliométrie | 0,001 | 0,001 |
| Études des sciences et des technologies | 0,000 | 0,000 |
| Communication savante | 0,000 | 0,000 |
| Science ouverte | 0,001 | 0,000 |
| Intégrité de la recherche | 0,000 | 0,000 |
| Charge utile insuffisante (le modèle a refusé de juger) | 0,002 | 0,000 |
Scores machine (provisoires)
Les deux têtes enseignantes du modèle étudiant, lues sur ce travail. Un score ordonne la base pour la relecture; il n'affirme jamais une catégorie, et le statut de validation accompagne chaque rangée tel quel.
Scores de référence d'un modèle non mature (critères de maturité non atteints, 7 itérations). Un score ordonne; il n'affirme jamais une catégorie.
score_only:v0-immature-baseline · tel quel depuis la passe de notation : score_only signifie que le nombre peut ordonner les travaux, et qu'aucune étiquette de catégorie n'en découleClassification
machine, non validéePrédiction automatique; un appel candidat d’une seule source (Gemma direct ou Codex distillé), pas un consensus.
Le détail, modèle par modèle et score par score, se trouve en fin de page sous « Comment cette classification a été obtenue ».