Accounting for covariate distributions in slope-unit-based landslide susceptibility models. A case study in the alpine environment

Gabriele Amato, Clemens Eisank, Daniela Castro-camilo, Luigi Lombardo

Research output: Contribution to journalArticleAcademicpeer-review

Abstract

Thousands or even million of pixels can be contained in a single Slope Unit. Hence, each covariate used in spatial predictive models is characterized by a distribution of values for each Slope Unit. Here, we model the whole covariates' distribution within Slope Units for landslide susceptibility purposes. This is done by finely dissecting each covariate into quantiles and then modeling the susceptibility via a LASSO penalized Binary Logistic Regression. We choose a LASSO penalization because the common Stepwise procedure is not selective enough to shrink a large number of covariates to an interpretable subset (which we also demonstrate here). LASSO mostly selects 6 covariates out of 372 to explain the spatial distribution of shallow landslides in the Upper Badia valley, Italian Alps. This allows us to verify that the selection does not include any quantile close to the median hence, nor to the mean. The latter is the common representation of the covariates' distribution within Slope Units, which we also test and report in the supplements. Overall, we suggest to always investigate the whole distribution because the mean may not be the most informative nor the most performing way to generate Slope-Unit-based susceptibility models. In this general context, we generate our landslide inventory by combining semi-automated (OBIA) and manual mapping procedures. Our inventory, quantile covariates' representation and LASSO penalization produce excellent performances and interpretable relations between covariates and landslides.

Original languageEnglish
Article number105237
JournalEngineering geology
Volume260
Early online date1 Aug 2019
DOIs
Publication statusPublished - 3 Oct 2019

Fingerprint

alpine environment
Landslides
landslide
Spatial distribution
Logistics
Pixels
logistics
pixel
distribution
spatial distribution
valley
modeling

Keywords

  • Binary logistic regression
  • Landslide susceptibility
  • Least Absolute Shrinkage Selection Operator (LASSO)
  • OBIA
  • Slope units
  • Stepwise selection
  • ITC-ISI-JOURNAL-ARTICLE

Cite this

@article{a68adc22415c4fb1bdcae539ea5d8812,
title = "Accounting for covariate distributions in slope-unit-based landslide susceptibility models. A case study in the alpine environment",
abstract = "Thousands or even million of pixels can be contained in a single Slope Unit. Hence, each covariate used in spatial predictive models is characterized by a distribution of values for each Slope Unit. Here, we model the whole covariates' distribution within Slope Units for landslide susceptibility purposes. This is done by finely dissecting each covariate into quantiles and then modeling the susceptibility via a LASSO penalized Binary Logistic Regression. We choose a LASSO penalization because the common Stepwise procedure is not selective enough to shrink a large number of covariates to an interpretable subset (which we also demonstrate here). LASSO mostly selects 6 covariates out of 372 to explain the spatial distribution of shallow landslides in the Upper Badia valley, Italian Alps. This allows us to verify that the selection does not include any quantile close to the median hence, nor to the mean. The latter is the common representation of the covariates' distribution within Slope Units, which we also test and report in the supplements. Overall, we suggest to always investigate the whole distribution because the mean may not be the most informative nor the most performing way to generate Slope-Unit-based susceptibility models. In this general context, we generate our landslide inventory by combining semi-automated (OBIA) and manual mapping procedures. Our inventory, quantile covariates' representation and LASSO penalization produce excellent performances and interpretable relations between covariates and landslides.",
keywords = "Binary logistic regression, Landslide susceptibility, Least Absolute Shrinkage Selection Operator (LASSO), OBIA, Slope units, Stepwise selection, ITC-ISI-JOURNAL-ARTICLE",
author = "Gabriele Amato and Clemens Eisank and Daniela Castro-camilo and Luigi Lombardo",
year = "2019",
month = "10",
day = "3",
doi = "10.1016/j.enggeo.2019.105237",
language = "English",
volume = "260",
journal = "Engineering geology",
issn = "0013-7952",
publisher = "Elsevier",

}

Accounting for covariate distributions in slope-unit-based landslide susceptibility models. A case study in the alpine environment. / Amato, Gabriele; Eisank, Clemens; Castro-camilo, Daniela; Lombardo, Luigi.

In: Engineering geology, Vol. 260, 105237, 03.10.2019.

Research output: Contribution to journalArticleAcademicpeer-review

TY - JOUR

T1 - Accounting for covariate distributions in slope-unit-based landslide susceptibility models. A case study in the alpine environment

AU - Amato, Gabriele

AU - Eisank, Clemens

AU - Castro-camilo, Daniela

AU - Lombardo, Luigi

PY - 2019/10/3

Y1 - 2019/10/3

N2 - Thousands or even million of pixels can be contained in a single Slope Unit. Hence, each covariate used in spatial predictive models is characterized by a distribution of values for each Slope Unit. Here, we model the whole covariates' distribution within Slope Units for landslide susceptibility purposes. This is done by finely dissecting each covariate into quantiles and then modeling the susceptibility via a LASSO penalized Binary Logistic Regression. We choose a LASSO penalization because the common Stepwise procedure is not selective enough to shrink a large number of covariates to an interpretable subset (which we also demonstrate here). LASSO mostly selects 6 covariates out of 372 to explain the spatial distribution of shallow landslides in the Upper Badia valley, Italian Alps. This allows us to verify that the selection does not include any quantile close to the median hence, nor to the mean. The latter is the common representation of the covariates' distribution within Slope Units, which we also test and report in the supplements. Overall, we suggest to always investigate the whole distribution because the mean may not be the most informative nor the most performing way to generate Slope-Unit-based susceptibility models. In this general context, we generate our landslide inventory by combining semi-automated (OBIA) and manual mapping procedures. Our inventory, quantile covariates' representation and LASSO penalization produce excellent performances and interpretable relations between covariates and landslides.

AB - Thousands or even million of pixels can be contained in a single Slope Unit. Hence, each covariate used in spatial predictive models is characterized by a distribution of values for each Slope Unit. Here, we model the whole covariates' distribution within Slope Units for landslide susceptibility purposes. This is done by finely dissecting each covariate into quantiles and then modeling the susceptibility via a LASSO penalized Binary Logistic Regression. We choose a LASSO penalization because the common Stepwise procedure is not selective enough to shrink a large number of covariates to an interpretable subset (which we also demonstrate here). LASSO mostly selects 6 covariates out of 372 to explain the spatial distribution of shallow landslides in the Upper Badia valley, Italian Alps. This allows us to verify that the selection does not include any quantile close to the median hence, nor to the mean. The latter is the common representation of the covariates' distribution within Slope Units, which we also test and report in the supplements. Overall, we suggest to always investigate the whole distribution because the mean may not be the most informative nor the most performing way to generate Slope-Unit-based susceptibility models. In this general context, we generate our landslide inventory by combining semi-automated (OBIA) and manual mapping procedures. Our inventory, quantile covariates' representation and LASSO penalization produce excellent performances and interpretable relations between covariates and landslides.

KW - Binary logistic regression

KW - Landslide susceptibility

KW - Least Absolute Shrinkage Selection Operator (LASSO)

KW - OBIA

KW - Slope units

KW - Stepwise selection

KW - ITC-ISI-JOURNAL-ARTICLE

UR - http://www.scopus.com/inward/record.url?scp=85070203672&partnerID=8YFLogxK

UR - https://ezproxy2.utwente.nl/login?url=https://library.itc.utwente.nl/login/2019/isi/lombardo_acc.pdf

U2 - 10.1016/j.enggeo.2019.105237

DO - 10.1016/j.enggeo.2019.105237

M3 - Article

VL - 260

JO - Engineering geology

JF - Engineering geology

SN - 0013-7952

M1 - 105237

ER -