ML22223A116
| ML22223A116 | |
| Person / Time | |
|---|---|
| Issue date: | 04/21/2022 |
| From: | Joseph Kanney NRC/RES/DRA |
| To: | |
| Kanney, Joseph 301 4155 1920 | |
| Shared Package | |
| ML22223A112 | List: |
| References | |
| Download: ML22223A116 (18) | |
Text
Sources of Uncertainty in Probabilistic Flood Hazard Assessment (PFHA)
Joseph Kanney Fire and External Hazards Analysis Branch Division of Risk Analysis Office of Nuclear Regulatory Research
OECD/NEA/CSNI/WGEV Workshop on Uncertainties in the Assessment of Natural Hazards April 19-21, 2022
1 Outline
- Probabilistic Flood Hazard and Risk Assessment Overview
- Dimensions of Uncertainty
- Flooding Scenarios
- Site-scale Flooding
- Riverine Flooding
- Coastal Flooding
- Combined Flooding
- Summary
2 1, 2,, 1 2
Hazard Curves : Fragility Curves :
Quantitative Quantitative Reliability probabilistic of Passive and assessment Active Flood Protection of flood hazard(s) Features
Load Intensity
Magnitude
Procedures Flood Protection Reliability of Procedures and Feature and Manual Actions Manual Actions Human Reliability Assessment (HRA) and Ye s Human Factors (HF) insights Protected? Impact of Environmental No Conditions on flood protection and/or mitigation procedures
Mitigation
No Impact No Ye s Impacts Impact Success? Plant PRA Model No 4
PFHA Approaches
- Statistical Approach
- Fit probability distribution(s) to flood severity metric of interest
- Flood frequency analysis
- Precipitation frequency analysis
- (Monte Carlo) Simulation Approach
- (Mechanistic) simulation models to compute flood severity metric(s)
- Probability distributions for model parameters, BCs, etc.
5 Key Challenges
- Hazard Estimation
- Range of annual exceedance probabilities (AEPs)
- Moderately rare to extreme floods
- Multiple flooding mechanisms
- Coincident and correlated mechanisms
- Uncertainty characterization and estimation Magnitude
- Aleatory (e.g., storm recurrence rates)
- Epistemic (e.g., model structure, parameters)
- Impacts
- Cliff-edge effects, nonlinearity
- Flood duration may be important
- Sparse structure/component reliability information
- Sparse human reliability information
Load Intensity
6 Dimensions of Uncertainty
- Aleatory variability*
- Natural randomness in a process
- Irreducible uncertainty
- Epistemic Uncertainty*
- Scientific uncertainty in the model of the process
- Model structure
- Model parameters
- Reducible uncertainty
- *Characterization is a function of analytical approach
- An addressed uncertainty may be aleatory in one model while in another model it may be epistemic
- These concepts only make unambiguous sense if they are defined within the confines of a model of analysis.
7 Riverine Discharge Example
- (Statistical) Flood Frequency Model
- Aleatory variability reflected by the use of a probability distribution
- For a given set of data, sampling uncertainty is not reducible
- Epistemic uncertainty in choice of distribution
- (Simulation) Hydrologic Model
- Future is stochastic (aleatory)
- Timing and spatiotemporal distribution of precipitation
- Changes in regulation, land use/land cover
- Epistemic uncertainty in watershed model structure and parameters
- Unit hydrograph, infiltration, routing
8 Site-scale Flooding (LIP)
- Aleatory variability
- Precipitation: magnitude, duration, temporal distribution
- Precipitation frequency analysis (e.g., NOAA Atlas 14)
- Stochastic weather generators
- Numerical weather prediction models (e.g., WRF, HRRR)
- Initial/antecedent conditions: soil moisture content, surface storage/ponding, initial stormwater drainage system state
- Long-term temporal trends (e.g., climate change):
can affect precipitation, temperature, initial conditions, and boundary conditions
9 Site-scale Flooding (Cont.)
- Epistemic Uncertainty:
- Boundary conditions: upstream discharge, downstream water levels
- Process representation: runoff generationprocesses including infiltration and roof runoff; - related stormwater drainage; hydraulic routing (using one-,
two-, and three-dimensional [1-D, 2-D, and 3-D]
models); flow transitions (supercritical to subcritical and vice versa); surface roughness effects in shallow, turbulent flows
- Site configuration: aboveground features including buildings and vehicle barrier systems; status of temporary flood protection; blockage of roof drains; blockage of stormwater drains
- Model resolution: spatial and temporal
10 Riverine Flooding
- Flood Frequency Analysis Approach
- Aleatory: Sampling uncertainty (e.g., limited data)
- Short records (often only a few decades or less)
- Data gaps
- Epistemic: Uncertainty in choice of distribution
- Extreme flooding not represented in the data
- Possible mixed population
- Nonstationarity (climate, regulation, land use/land cover)
11 Riverine Flooding (Cont.)
- Simulation Approach
- Aleatory: Precipitation magnitude, duration, temporal and spatial distribution
- Discharge:
- Aleatory: Initial/Antecedent conditions (e.g., soil moisture, flows)
- Epistemic: Hydrologic model structure and parameters
- e.g., infiltration, unit hydrographs, and hydrologic routing
- Epistemic: Model resolution
- Stage and velocity:
- Epistemic: Hydraulic model structure and parameters
- 1D/2D, steady/unsteady
- Epistemic: Model resolution (e.g., channel/floodplain)
12 Coastal Flooding (TCs)
- Storm recurrence rate is key aleatory variability for tropical cyclones
- Key epistemic uncertainties:
- Distributions for storm parameters
- Track location
- Heading direction
- Central pressure deficit
- Radius of maximum winds
- Translational speed
- Hydrodynamic model structure and parameters
- Bottom friction, wind stress
- Resolution of bathymetry, topography
13 Combined Flooding
- Extreme flooding due to combined processes
- Associated effects (e.g., wave action)
- Multi-mechanism flooding (e.g., rainfall w/ snowmelt, rainfall w/ dam failure)
14 Combined Flooding (Cont.)
- Statistical Approach (Joint Probability)
- Direct Estimation of Joint Distributions
- Assumed functional form for the joint distribution
- Copula-Based Approaches
- Individual variables can have different distributions
- Observations used to separately estimate
- (1) the parameters of the marginal distributions and
- (2) the parameters of the copulas, which are typically related to the correlation between the quantities, as estimated from data
- Simulation Approach
- Fully-or semi-coupled simulation models
- Coupled hydrologic, reservoir, and hydraulic models to simulate riverine flooding w/ dam failure
- Coupled atmospheric boundary layer, hydrodynamic, and wave models to simulate coastal flooding
- Sparse data for calibrating models
15 Summary
- Separating uncertainty into aleatory variability and epistemic uncertainty is a useful exercise, but these concepts are unambiguous only within a given analytical/modeling framework
- Important sources of uncertainty vary with:
- Scale and setting (e.g., site-scale, watershed-scale)
- Analysis approach (e.g., statistical vs simulation)
- Flooding metric of interest
- Interaction of multiple flooding mechanisms introduces complexity and additional uncertainty
16 References
Der Kiureghian, A. & Ditlevsen, O. (2008). Aleatory or epistemic? Does it matter? Structural Safety 31(2), 105-112, https://doi.org/10.1016/j.strusafe.2008.06.020 Asquith, W.H., Kiang, J.E., and Cohn, T.A., 2017, Application of at-site peak-streamflow frequency analyses for very low annual exceedance probabilities: U.S. Geological Survey Scientific Investigations Report 2017-5038, 93 p. https://doi.org/10.3133/sir20175038 Ryberg, K.R., Kolars, K.A., Kiang, J.E., and Carr, M.L., (2020). Flood-frequency estimation for very low annual exceedance probabilities using historical, paleoflood, and regional information with consideration of nonstationarity: U.S. Geological Survey Scientific Investigations Report 2020-5065, 89 p., https://doi.org/10.3133/sir20205065 Bensi, Michelle, Mohammadi, Somayeh, Kao, Shih-Chieh, and DeNeale, Scott T. (2020). Multi-Mechanism Flood Hazard Assessment: Critical Review of Current Practice and Approaches.
https://doi.org/10.2172/1637939 William Lehman,(2022). HEC-WAT As a Framework for PFHA, 7th Annual NRC PFHA Research Workshop, Feb 15-18, 2022, https://www.nrc.gov/docs/ML2106/ML21064A438.pdf Norberto C. Nadal-Caraballo, Jeffrey A. Melby, Victor M. Gonzalez, and Andrew T. Cox (2015).
North Atlantic Coast Comprehensive Study (NACCS), Coastal Storm Hazards from Virginia to Maine, ERDC/CHL TR-15-5, https://www.nad.usace.army.mil/Portals/40/docs/ComprehensiveStudy/Coastal_Storm_Hazard s_from_Virginia_to_Maine.pdf
17 Questions?
Contact Information Joseph.Kanney@nrc.gov
18