A glimpse into the future of exposure and vulnerabilities in cities? Modelling of residential location choice of urban population with random forest

Sebastian Scheuer,Annegret Haase,Dagmar Haase,Manuel Wolff,Thilo Wellmann

doi:10.5194/nhess-21-203-2021

Abstract

Abstract. The most common approach to assessing natural hazard risk is investigating the willingness to pay in the presence or absence of such risk. In this work, we propose a new, machine-learning-based, indirect approach to the problem, i.e. through residential-choice modelling. Especially in urban environments, exposure and vulnerability are highly dynamic risk components, both being shaped by a complex and continuous reorganization and redistribution of assets within the urban space, including the (re-)location of urban dwellers. By modelling residential-choice behaviour in the city of Leipzig, Germany, we seek to examine how exposure and vulnerabilities are shaped by the residential-location-choice process. The proposed approach reveals hot spots and cold spots of residential choice for distinct socioeconomic groups exhibiting heterogeneous preferences. We discuss the relationship between observed patterns and disaster risk through the lens of exposure and vulnerability, as well as links to urban planning, and explore how the proposed methodology may contribute to predicting future trends in exposure, vulnerability, and risk through this analytical focus. Avenues for future research include the operational strengthening of these linkages for more effective disaster risk management.

Highlights

In the human-environmental system, disaster risk arises from the interactions of different system components (Zscheischler et al, 2018)
The methodology applied in this case study embraces the following steps (Fig. 1): (i) extraction of non-spatial housing attributes, i.e. the characteristics of each actual apartment, from the scientific-use file provided by Boelmann et al (2019; see Table 1); (ii) determination of spatially homogeneous units for the geolocation of prediction targets; (iii) determination of spatial housing attributes based on ancillary data (Table 1); (iv) formulation of a set of socioeconomic profiles to account for heterogeneity of preferences (Table 2); and (v) application of the pre-trained random forest model to predict the likelihoods of positive residential-choice outcomes
We suggest increasing the spatial resolution through a mapping of apartment locations to so-called spatially homogeneous units (SHUs)

Summary

Introduction

In the human-environmental system, disaster risk arises from the interactions of different system components (Zscheischler et al, 2018). The Hyogo Framework for Action 2005– 2015 maintains that disaster risk stems from the interaction of a hazard with exposed physical, socioeconomic and environmental vulnerabilities (UNISDR, 2007), referring to the potential fatalities and losses in livelihoods, health, assets, and services. Urban processes such as the expansion into potentially hazardous areas and gentrification or densification shape exposure and vulnerabilities of services and assets within urban areas in a highly dynamic manner and are at the basis of urban disaster risk. Exposure denotes the physical aspects of disaster risk (UNISDR, 2004), referring to the socioeconomic and demographic spatiotemporal fabric, i.e. assets such as population or the built environment that are potentially affected by a Published by Copernicus Publications on behalf of the European Geosciences Union

Methods

Results

Discussion

Conclusion