Uncertainty Reduction Through Data Management in the Development, Validation, Calibration, and Operation of a Hurricane Vulnerability Model

Abstract

Catastrophe models estimate risk at the intersection of hazard, exposure, and vulnerability. Each of these areas requires diverse sources of data, which are very often incomplete, inconsistent, or missing altogether. The poor quality of the data is a source of epistemic uncertainty, which affects the vulnerability models as well as the output of the catastrophe models. This article identifies the different sources of epistemic uncertainty in the data, and elaborates on strategies to reduce this uncertainty, in particular through identification, augmentation, and integration of the different types of data. The challenges are illustrated through the Florida Public Hurricane Loss Model (FPHLM), which estimates insured losses on residential buildings caused by hurricane events in Florida. To define the input exposure, and for model development, calibration, and validation purposes, the FPHLM teams accessed three main sources of data: county tax appraiser databases, National Flood Insurance Protection (NFIP) portfolios, and wind insurance portfolios. The data from these different sources were reformatted and processed, and the insurance databases were separately cross-referenced at the county level with tax appraiser databases. The FPHLM hazard teams assigned estimates of natural hazard intensity measure to each insurance claim. These efforts produced an integrated and more complete set of building descriptors for each policy in the NFIP and wind portfolios. The article describes the impact of these uncertainty reductions on the development and validation of the vulnerability models, and suggests avenues for data improvement. Lessons learned should be of interest to professionals involved in disaster risk assessment and management.

The value of ultra-detailed survey data for an improved flood damage modelling with explicit input data uncertainty treatment: INSYDE 2.0

“…The median estimates exhibit a satisfactory level of agreement with the observed losses, which is consistent with typical outcomes observed in validation exercises for models demonstrating overall good performances (e.g., Amadio et al 2019;Molinari et al, 2020). It should be noted that the model tends to overestimate lower entity damages across all building types (Figure 7), but this discrepancy, rather than being a consequence of any model-related issue, can be primarily attributed to the limitations in the representativeness of claim data, particularly for minor losses, as documented in the literature Molinari et al, 2020;Pinelli et al, 2020). 340 While confirming the performance of INSYDE 2.0 in accurately depicting the overall damage figures of the two events (Dottori et al, 2016;Amadio et al, 2019;Molinari et al, 2020), the results of this analysis also emphasize the advantages of incorporating the treatment of input data uncertainty when presenting the outcomes of a model validation.…”

Riverine flood risk assessment with a combined model chain in southeastern China

“…There are many uncertainties in the process of flood risk assessment. Pinelli et al (2020) found that there is inevitable uncertainty in estimating the value of assets. Koks et al (2019) showed that the economic damage caused by the failure of critical infrastructure might be underestimated.…”

Extreme Flood Disasters: Comprehensive Impact and Assessment

“…The land use data can be easily obtained through remote sensing (RS) interpretation [14], and it is effectively used to extract the spatial distribution of buildings, but it is difficult to establish the relationship between flood loss data and land use types. These problems lead to high uncertainties and disparities in flood loss assessments [19,20]. In this case, we should propose a refined method to evaluate the detailed flood loss of each affected object.…”