A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Machine learning motivated data imputation of storm data used in coastal hazard assessments
https://orcid.org/0000-0001-6449-1812
Abstract In the Coastal Hazards System's (CHS) Probabilistic Coastal Hazard Analysis (PCHA) framework developed by the United States Army Corps of Engineers (USACE), historical records of tropical cyclone parameters have been used as data sources for statistical analysis, including fitting marginal distributions and measuring correlations between storm parameters. One limitation of the available historical databases is that observations of central pressure and radius of maximum winds are not available for a large number of storms. This may adversely affect the results of statistical analyses used to develop hazard curves. This study uses machine learning techniques to develop a data imputation method to “fill in” missing storm parameter records in historical datasets used for Joint Probability Method (JPM)-based coastal hazard analysis such as the USACE's CHS-PCHA. Specifically, Gaussian process regression (GPR) and artificial neural network (ANN) models are investigated as candidate machine learning-derived data imputation models, and the performance of different model parameterizations is assessed. Candidate imputation models are compared against existing statistical relationships. The effect of the data imputation process on statistical analyses (marginal distributions and correlation measures) is also evaluated for a series of example coastal reference locations.
Highlights Machine learning models for coastal hazard storm data imputation. Comparative assessment of imputation model performance. Assessing data imputation effect on coastal hazard statistical analysis.
Machine learning motivated data imputation of storm data used in coastal hazard assessments
https://orcid.org/0000-0001-6449-1812
Abstract In the Coastal Hazards System's (CHS) Probabilistic Coastal Hazard Analysis (PCHA) framework developed by the United States Army Corps of Engineers (USACE), historical records of tropical cyclone parameters have been used as data sources for statistical analysis, including fitting marginal distributions and measuring correlations between storm parameters. One limitation of the available historical databases is that observations of central pressure and radius of maximum winds are not available for a large number of storms. This may adversely affect the results of statistical analyses used to develop hazard curves. This study uses machine learning techniques to develop a data imputation method to “fill in” missing storm parameter records in historical datasets used for Joint Probability Method (JPM)-based coastal hazard analysis such as the USACE's CHS-PCHA. Specifically, Gaussian process regression (GPR) and artificial neural network (ANN) models are investigated as candidate machine learning-derived data imputation models, and the performance of different model parameterizations is assessed. Candidate imputation models are compared against existing statistical relationships. The effect of the data imputation process on statistical analyses (marginal distributions and correlation measures) is also evaluated for a series of example coastal reference locations.
Highlights Machine learning models for coastal hazard storm data imputation. Comparative assessment of imputation model performance. Assessing data imputation effect on coastal hazard statistical analysis.
Machine learning motivated data imputation of storm data used in coastal hazard assessments
https://orcid.org/0000-0001-6449-1812
Abstract In the Coastal Hazards System's (CHS) Probabilistic Coastal Hazard Analysis (PCHA) framework developed by the United States Army Corps of Engineers (USACE), historical records of tropical cyclone parameters have been used as data sources for statistical analysis, including fitting marginal distributions and measuring correlations between storm parameters. One limitation of the available historical databases is that observations of central pressure and radius of maximum winds are not available for a large number of storms. This may adversely affect the results of statistical analyses used to develop hazard curves. This study uses machine learning techniques to develop a data imputation method to “fill in” missing storm parameter records in historical datasets used for Joint Probability Method (JPM)-based coastal hazard analysis such as the USACE's CHS-PCHA. Specifically, Gaussian process regression (GPR) and artificial neural network (ANN) models are investigated as candidate machine learning-derived data imputation models, and the performance of different model parameterizations is assessed. Candidate imputation models are compared against existing statistical relationships. The effect of the data imputation process on statistical analyses (marginal distributions and correlation measures) is also evaluated for a series of example coastal reference locations.
Highlights Machine learning models for coastal hazard storm data imputation. Comparative assessment of imputation model performance. Assessing data imputation effect on coastal hazard statistical analysis.
Machine learning motivated data imputation of storm data used in coastal hazard assessments
Liu, Ziyue (author) / Carr, Meredith L. (author) / Nadal-Caraballo, Norberto C. (author) / Yawn, Madison C. (author) / Taflanidis, Alexandros A. (author) / Bensi, Michelle T. (author)
Coastal Engineering ; 190
2024-03-12
Article (Journal)
Electronic Resource
English
Large Scale Coastal Storm Hazard Mapping
| British Library Conference Proceedings | 2002
Quantifying the storm erosion hazard for coastal planning
| Online Contents | 2009
Quantifying the storm erosion hazard for coastal planning
| British Library Online Contents | 2009
Quantifying the storm erosion hazard for coastal planning
| Online Contents | 2009