Hindcasting compound pluvial, fluvial and coastal flooding during Hurricane Harvey (2017) using Delft3D-FM

Lee, Wonhyun; Sun, Alexander Y.; Scanlon, Bridget R.; Dawson, Clint

doi:10.1007/s11069-023-06247-9

Cited by 5 publications

(4 citation statements)

References 65 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…HEC-RAS produce maximum WLs in close agreement with available high-water marks in G-Bay, and additionally, both RMSE and MAE agree well with the corresponding metrics reported in similar CF studies (Huang et al, 2021;Lee et al, 2023;Sebastian et al, 2021;Wing et al, 2019). In addition, we validate peak WLs at selected tide-gauge stations (Figure 3…”

Section: Model Calibration and Validationsupporting

confidence: 85%

Quantifying cascading uncertainty in compound flood modeling with linked process-based and machine learning models

Muñoz,

Moftakhari,

Moradkhani

2024

Preprint

View full text Add to dashboard Cite

Abstract. Compound flood (CF) modeling enables the simulation of nonlinear water level dynamics in which concurrent or successive flood drivers synergize, producing larger impacts than those from individual drivers. CF modeling is yet subject to four main sources of uncertainty including (i) initial condition, (ii) forcing (or boundary) conditions, (iii) model parameters, and (iv) model structure. These sources of uncertainty, if not quantified and effectively reduced, cascade in series throughout the modeling chain and compromise the accuracy of CF hazard assessments. Here, we characterize cascading uncertainty using linked process-based and machine learning (PB-ML) models for a well-known CF event, namely Hurricane Harvey in Galveston Bay, TX. For this, we run a set of hydrodynamic model scenarios to quantify isolated and cascading uncertainty in terms of maximum water level residuals, and additionally, track the evolution of residuals during the onset, peak, and dissipation of Hurricane Harvey. We then develop multiple-linear regression (MLR) and PB-ML models to estimate the relative and cumulative contribution of the four sources of uncertainty to total uncertainty over time. Results from this study show that the proposed PB-ML model capture “hidden” nonlinear associations and interactions among the sources of uncertainty, thereby outperforming conventional MLR models. Model structure and forcing conditions are the main sources of uncertainty in CF modeling and their corresponding model scenarios, or input features, contribute to (56 %) 49 % of variance reduction in the estimation of (maximum) water level residuals. Following these results, we conclude that PB-ML models are a feasible alternative for quantifying cascading uncertainty in CF modeling.

show abstract

Section: Model Calibration and Validationsupporting

confidence: 85%

Quantifying cascading uncertainty in compound flood modeling with linked process-based and machine learning models

Muñoz,

Moftakhari,

Moradkhani

2024

Preprint

View full text Add to dashboard Cite

show abstract

“…Further, ideally, the interaction between surge, fluvial, and pluvial drivers are simulated as well to account for compound impact of all three drivers. Such compound effects played for instance a crucial role in the case of Harvey 52 . In this sense, it would be interesting to compare a global set of observational data of TC events such as the one presented in this study with a number of more complex ocean models or compound flood models that are able to take more aspects of flooding into account than the GeoClaw setup that we used here.…”

Section: Discussionmentioning

confidence: 99%

Accounting for surge dynamics is key to modeling tropical cyclone-induced coastal flooding

Vogt,

Treu,

Mengel

et al. 2024

Preprint

View full text Add to dashboard Cite

Tropical cyclone-induced storm surge is a major coastal risk which will be further amplified by rising sea levels under global warming. The surge-induced floodplains are commonly modeled using a two-step approach by first producing coastal water levels which are, secondly, translated into inundated areas. Here, we evaluate an alternative approach where ocean dynamics and coastal inundation are modeled in a single step using the open-source solver GeoClaw that accounts for the event-specific variation over time of flood drivers. We compute the surge-induced flood extents for a global set of 71 storms from 2000-2019. We show that GeoClaw better reproduces satellite observations and high water marks than two different common modeling approaches that lack a process-based representation of inundation dynamics: a state-of-the-art two-step approach combining a full-scale ocean dynamics with a static (bathtub-type) inundation model and a lightweight approach that directly translates parametric tropical cyclone wind fields into inundation maps using statistical relationships. GeoClaw does not reach the performance of the ocean dynamics model in reproducing coastal water levels, but this deficiency is overcompensated by the dynamic modeling of inundated areas. The computational efficiency of the one-step approach opens up new perspectives for global assessments of coastal risks from tropical cyclones.

show abstract

“…In light of the global impact and growing relevance of compound floods (Eilander et al, 2023;Gu et al, 2022;Lai et al, 2021;Lee et al, 2023), this study is of continued importance in understanding the probabilistic nature of compound events, with important implications for effective flood management strategies. To further advance this field, we will focus on achieving higher resolutions in bathymetry within localized areas.…”

Section: Future Workmentioning

confidence: 99%

“…For instance, hurricanes Ike (2008) and Harvey (2017) that struck the Texas Gulf Coast represent some of the costliest storms in the U.S. history, with estimated ~$160 billion damage (combined both events, National Hurricane Center, 2018). These TCs highlighted the urgent need for improved nowcasting and forecasting techniques for compound flooding events (Lee et al, 2023;Wahl et al, 2015). Hurricane Ike has been extensively studied, with researches focusing on various aspects of its dynamics, coastal impacts, and infrastructure effects (Stearns & Padgett, 2012).…”

Section: Introductionmentioning

confidence: 99%

Probabilistic Storm Surge and Flood-Inundation Modeling for the Texas Gulf Coast Using Super-Fast INundation of CoastS (SFINCS)

Lee,

Sun,

Scanlon

2024

Preprint

View full text Add to dashboard Cite

Accurately predicting the extent of compound flooding events, including storm surge, pluvial, and fluvial flooding, is vital for protecting coastal communities. However, high computational demands associated with detailed probabilistic models highlight the need for simplified models to enable rapid forecasting. The objective of this study was to assess the accuracy and efficiency of a reduced-complexity, hydrodynamic solver – the Super-Fast INundation of CoastS (SFINCS) model – in a probabilistic ensemble simulation setting, using Hurricane Ike (2008) in the Texas Gulf Coast as a case study. Results show that the SFINCS-based framework can provide probabilistic outputs under reasonable simulation times (e.g., less than 4 hours for a 100-member ensemble on a single CPU). The model agrees well with observed data from NOAA tidal stations and USGS gage height stations. The ensemble approach significantly reduced errors (average 16%) across all stations compared to a deterministic case. The ensemble improved overall performance and revealed wider flood extents and lower depths. Sensitivity studies performed on ensemble sizes (1,000, 189, 81) and lead times (1 to 3 days before landfall) further demonstrate the reliability of flood extent predictions over varying lead times. In particular, Counties adjacent to the Trinity River Basin had ≥ 80% probability in exceeding the critical 3-m flood threshold during Hurricane Ike. Our study highlights the effectiveness of the SFINCS-based framework in providing probabilistic flood extent/depth forecasts over long lead times in a timely manner. Thus, the framework constitutes a valuable tool for effective flood preparedness and response planning during compound flooding.

show abstract

Hindcasting compound pluvial, fluvial and coastal flooding during Hurricane Harvey (2017) using Delft3D-FM

Cited by 5 publications

References 65 publications

Quantifying cascading uncertainty in compound flood modeling with linked process-based and machine learning models

Quantifying cascading uncertainty in compound flood modeling with linked process-based and machine learning models

Accounting for surge dynamics is key to modeling tropical cyclone-induced coastal flooding

Probabilistic Storm Surge and Flood-Inundation Modeling for the Texas Gulf Coast Using Super-Fast INundation of CoastS (SFINCS)

Contact Info

Product

Resources

About