A data-mining approach towards damage modelling for El Nino events in Peru
- Compound natural hazards likeEl Ninoevents cause high damage to society, which to manage requires reliable risk assessments. Damage modelling is a prerequisite for quantitative risk estimations, yet many procedures still rely on expert knowledge, and empirical studies investigating damage from compound natural hazards hardly exist. A nationwide building survey in Peru after theEl Ninoevent 2017 - which caused intense rainfall, ponding water, flash floods and landslides - enables us to apply data-mining methods for statistical groundwork, using explanatory features generated from remote sensing products and open data. We separate regions of different dominant characteristics through unsupervised clustering, and investigate feature importance rankings for classifying damage via supervised machine learning. Besides the expected effect of precipitation, the classification algorithms select the topographic wetness index as most important feature, especially in low elevation areas. The slope length and steepness factor ranks high forCompound natural hazards likeEl Ninoevents cause high damage to society, which to manage requires reliable risk assessments. Damage modelling is a prerequisite for quantitative risk estimations, yet many procedures still rely on expert knowledge, and empirical studies investigating damage from compound natural hazards hardly exist. A nationwide building survey in Peru after theEl Ninoevent 2017 - which caused intense rainfall, ponding water, flash floods and landslides - enables us to apply data-mining methods for statistical groundwork, using explanatory features generated from remote sensing products and open data. We separate regions of different dominant characteristics through unsupervised clustering, and investigate feature importance rankings for classifying damage via supervised machine learning. Besides the expected effect of precipitation, the classification algorithms select the topographic wetness index as most important feature, especially in low elevation areas. The slope length and steepness factor ranks high for mountains and canyons. Partial dependence plots further hint at amplified vulnerability in rural areas. An example of an empirical damage probability map, developed with a random forest model, is provided to demonstrate the technical feasibility.…
Author details: | Fabio Alexander BrillORCiDGND, Silvia Passuni Pineda, Bruno Espichan Cuya, Heidi KreibichORCiDGND |
---|---|
DOI: | https://doi.org/10.1080/19475705.2020.1818636 |
ISSN: | 1947-5705 |
ISSN: | 1947-5713 |
Title of parent work (English): | Geomatics, natural hazards and risk |
Publisher: | Routledge, Taylor & Francis Group |
Place of publishing: | Abingdon |
Publication type: | Article |
Language: | English |
Date of first publication: | 2020/10/13 |
Publication year: | 2020 |
Release date: | 2023/11/21 |
Tag: | Natural hazard; damage model; data-mining; open data; remote; residential buildings; sensing |
Volume: | 11 |
Issue: | 1 |
Number of pages: | 25 |
First page: | 1966 |
Last Page: | 1990 |
Funding institution: | German Ministry of Education and Research (BMBF)Federal Ministry of; Education & Research (BMBF) [03G0876B] |
Organizational units: | Mathematisch-Naturwissenschaftliche Fakultät / Institut für Geowissenschaften |
DDC classification: | 5 Naturwissenschaften und Mathematik / 55 Geowissenschaften, Geologie / 550 Geowissenschaften |
Peer review: | Referiert |
Publishing method: | Open Access / Gold Open-Access |
DOAJ gelistet | |
License (German): | CC-BY - Namensnennung 4.0 International |