Bayesian network learning for natural hazard analyses

Abstract

Abstract. Modern natural hazards research requires dealing with several uncertainties that arise from limited process knowledge, measurement errors, censored and incomplete observations, and the intrinsic randomness of the governing processes. Nevertheless, deterministic analyses are still widely used in quantitative hazard assessments despite the pitfall of misestimating the hazard and any ensuing risks.In this paper we show that Bayesian networks offer a flexible framework for capturing and expressing a broad range of uncertainties encountered in natural hazard assessments. Although Bayesian networks are well studied in theory, their application to real-world data is far from straightforward, and requires specific tailoring and adaptation of existing algorithms. We offer suggestions as how to tackle frequently arising problems in this context and mainly concentrate on the handling of continuous variables, incomplete data sets, and the interaction of both. By way of three case studies from earthquake, flood, and landslide research, we demonstrate the method of data-driven Bayesian network learning, and showcase the flexibility, applicability, and benefits of this approach.Our results offer fresh and partly counterintuitive insights into well-studied multivariate problems of earthquakeinduced ground motion prediction, accurate flood damage quantification, and spatially explicit landslide prediction at the regional scale. In particular, we highlight how Bayesian networks help to express information flow and independence assumptions between candidate predictors. Such knowledge is pivotal in providing scientists and decision makers with well-informed strategies for selecting adequate predictor variables for quantitative natural hazard assessments.

Identifying Driving Factors in Flood‐Damaging Processes Using Graphical Models

“…Due to these constraints an application of standard software is not advantageous. Instead, we make use of an advanced score‐based algorithm that was developed for incomplete, hybrid data sets (Vogel et al, ). The single challenges addressed by the algorithm are explained in the following, while Table provides a summary of the applied approaches.…”

“…They consequently often do a better job in variable selection tasks than BNs. While BNs have already successfully been applied in natural hazard assessments (Blaser et al, ; Kühn et al, ; Straub, ; Vogel et al, ), we have no knowledge about direct applications of MBs in that field. Yet MBs are successfully applied for feature selection in several other scientific disciplines, for example, in the field of bioinformatics and microarrays (Dernoncourt et al, ; Saeys et al, ; Zhu et al, ) as well as for text classification (Javed et al, ).…”

Quantifying Flood Vulnerability Reduction via Private Precaution

“…The loss estimate is then determined using the mean as the prediction of the ensemble of trees. Also, Bayesian networks are used in flood loss estimation (Schröter et al, 2014;Vogel et al, 2014;Wagenaar et al, 2018). Bayesian networks are Directed Acyclic Graphs (DAG) constructed from assertion of dependencies and principle of conditional independence (Heckerman, 1998).…”

An Experimental Research on the Use of Recurrent Neural Networks in Landslide Susceptibility Mapping

“…Natural hazards such as earthquakes, landslides, tsunamis, and volcanic activities that all have serious effects on human beings have a great number of influencing factors. The individual and joint effects of these factors are not always fully understood since each factor introduces a potentially large degree of uncertainty into any quantitative analysis [1]. Additionally, the data acquired from observations are usually sparse and lack accuracy and completeness, which is another kind of uncertainty [1].…”

“…The individual and joint effects of these factors are not always fully understood since each factor introduces a potentially large degree of uncertainty into any quantitative analysis [1]. Additionally, the data acquired from observations are usually sparse and lack accuracy and completeness, which is another kind of uncertainty [1]. To reduce these uncertainties, machine learning (ML) algorithms have been implemented, particularly for the last decade (Table 1).…”