Journal
ENVIRONMENTAL MODELLING & SOFTWARE
Volume 105, Issue -, Pages 118-131Publisher
ELSEVIER SCI LTD
DOI: 10.1016/j.envsoft.2018.03.032
Keywords
Flood risk; Flood loss modeling; Standardized data; Random forests; Vulnerability; Virtual 3D city models
Categories
Funding
- European Institute of Innovation and Technology (EIT)
Ask authors/readers for more resources
Flood loss modeling provides the basis to optimize investments for flood risk management. However, detailed object-related data are not readily available to generate spatially explicit risk information. Virtual 3D city models and numerical spatial measures derived from remote sensing data provide standardized data and hold promise to fill this gap. The suitability of these data sources to characterize the vulnerability of residential buildings to flooding is investigated using the city of Dresden as a case study, where also empirical data on relative flood loss and inundation depths are available. Random forests are used for predictive analysis of these heterogeneous data sets. Results show that variables depicting building geometric properties are suitable to explain flood vulnerability. Model validation confirms that predictive accuracy and reliability are comparable to alternative models based on detailed empirical data. Furthermore, virtual 3D city models allow embedding vulnerability information into flood risk sensitive urban planning. (C) 2018 Elsevier Ltd. All rights reserved.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available