Functional principal component analysis for global sensitivity analysis of model with spatial output

Perrin, Tran Vi-Vi Élodie; Roustant, Olivier; Rohmer, Jérémy; Alata, Olivier; Naulin, Jean-Philippe; Idier, Déborah; Pedreros, Rodrigo; Moncoulon, David; Tinard, Pierre

doi:10.1016/j.ress.2021.107522

Cited by 21 publications

(12 citation statements)

References 30 publications

(53 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Subsequently, we use the water level predictions by NOAA's Operational Oceanographic Products and Services (CO‐OPS) at this station during October 10–November 10, 2012 as the baseline water level boundary condition at the ocean boundary (Figure 9) for all scenarios. Following the approach presented in Perrin et al (2021) for storm surge analysis, we superimpose an isosceles triangular surge pulse on the baseline water level time series. However, instead of their parametrization approach, we consider the surge peak height (amplitude) as the only surge parameter and use a constant duration of two days for the rising and falling limbs of the pulse.…”

Section: Case Studymentioning

confidence: 99%

A Novel Framework for Parametric Analysis of Coastal Transition Zone Modeling

Chegini

Coelho

Ratcliff

et al. 2022

J American Water Resour Assoc

View full text Add to dashboard Cite

Vulnerability of coastal regions to extreme events motivates an operational coupled inland‐coastal modeling strategy focusing on the coastal transition zone (CTZ), an area between the coast and upland river. To tackle this challenge, we propose a top‐down framework for investigating the contribution of different processes to the hydrodynamics of CTZs with various geometrical shapes, different physical properties, and under several forcing conditions. We further propose a novel method, called tidal vanishing point (TVP), for delineating the extent of CTZs through the upland. We demonstrate the applicability of our framework over the United States East and Gulf coasts. We categorize CTZs in the region into three classes, namely, without estuary (direct river–coast connection), triangular‐, and trapezoidal‐shaped estuary. The results show that although semidiurnal tidal constituents are dominant in most cases, diurnal tidal constituents become more prevalent in the river segment as the discharge increases. Also, decreasing the bed roughness value promotes more significant changes in the results than increasing it by the same value. Additionally, the estuary promotes tidal energy attenuation and consequently decreases the reach of tidal signals through the upland. The proposed framework is generic and extensible to any coastal region.

show abstract

Section: Case Studymentioning

confidence: 99%

A Novel Framework for Parametric Analysis of Coastal Transition Zone Modeling

Chegini

Coelho

Ratcliff

et al. 2022

J American Water Resour Assoc

View full text Add to dashboard Cite

show abstract

“…In principle, many approaches can be utilised to set up a metamodel (for an overview, see [40]), including GPs, neural networks, or support vector regression. For the present work, we select GP metamodels because they have shown very high predictive capabilities in previous applications for coastal flood assessments (among others, see [44], for applications to overflow-induced marine flooding; see [19,45], for applications to hurricanes; and see [46], for an application to San Francisco Bay). In particular, high performance was shown by the extensive comparison exercise conducted by [47].…”

Section: Metamodelling Techniquementioning

confidence: 99%

A User-Oriented Local Coastal Flooding Early Warning System Using Metamodelling Techniques

Idier

Aurouet

Bachoc

et al. 2021

JMSE

Self Cite

View full text Add to dashboard Cite

Given recent scientific advances, coastal flooding events can be properly modelled. Nevertheless, such models are computationally expensive (requiring many hours), which prevents their use for forecasting and warning. In addition, there is a gap between the model outputs and information actually needed by decision makers. The present work aims to develop and test a method capable of forecasting coastal flood information adapted to users’ needs. The method must be robust and fast and must integrate the complexity of coastal flood processes. The explored solution relies on metamodels, i.e., mathematical functions that precisely and efficiently (within minutes) estimate the results that would provide the numerical model. While the principle of relying on metamodel solutions is not new, the originality of the present work is to tackle and validate the entire process from the identification of user needs to the establishment and validation of the rapid forecast and early warning system (FEWS) while relying on numerical modelling, metamodelling, the development of indicators, and information technologies. The development and validation are performed at the study site of Gâvres (France). This site is subject to wave overtopping, so the numerical phase-resolving SWASH model is used to build the learning dataset required for the metamodel setup. Gaussian process- and random forest classifier-based metamodels are used and post-processed to estimate 14 indicators of interest for FEWS users. These metamodelling and post-processing schemes are implemented in an FEWS prototype, which is employed by local users and exhibits good warning skills during the validation period. Based on this experience, we provide recommendations for the improvement and/or application of this methodology and individual steps to other sites.

show abstract

“…There exist alternative approaches that treat spatial outputs as functional data. For instance, we refer to (Marrel et al, 2011) for a framework that model the pollution produced by radioactive wastes, to (Chang and Guillas, 2019) for an approach capable to learn spatial patterns in climate experiments, and to (Perrin et al, 2020) for a surrogate model for coastal flooding risk assessment. Those approaches first project the outputs onto truncated basis representations (e.g.…”

Section: Introductionmentioning

confidence: 99%

“…Although their works scale well with the number of spatial points, they commonly require a large amount of learning simulations (e.g. over than 500 events) in order to properly capture spatial patterns (see, e.g., Perrin et al, 2020). Here, we are restricted to highly constraining situations where less than 200 flood scenarios are available.…”

Section: Introductionmentioning

confidence: 99%

“…This limitation arises from dedicated but costly-to-evaluate numerical simulators: one flood event requires almost three days of parallel computing (see, e.g., the model used by Idier et al, 2020b). We must also point out that (Marrel et al, 2011;Chang and Guillas, 2019;Perrin et al, 2020) do not account for functional inputs as our framework does, a condition that is key for properly learning hydro-meteorological forcing conditions (see the discussion in Section 2). To the best of our knowledge, our proposal is the first one that accounts for both inputs and outputs as functions in the domain of flood hazard assessment.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation