A general framework for Dynamic Emulation Modelling in environmental problems

Castelletti, Andrea; Galelli, Stefano; Ratto, Marco; Soncini-Sessa, Rodolfo; Young, Peter C.

doi:10.1016/j.envsoft.2012.01.002

Cited by 129 publications

(97 citation statements)

References 116 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Various emulators based on statistical learning have already been developed in the climate and environmental modeling community for the last few years (e.g., [22,35,36,[39][40][41][42][43]). These emulators have in common that they were developed from adaptive, flexible nonparametric regression algorithms, typically within the family of machine learning regression algorithms (MLRAs) [44].…”

Section: Emulator Theorymentioning

confidence: 99%

Emulation of Leaf, Canopy and Atmosphere Radiative Transfer Models for Fast Global Sensitivity Analysis

et al. 2016

View full text Add to dashboard Cite

Abstract:Physically-based radiative transfer models (RTMs) help understand the interactions of radiation with vegetation and atmosphere. However, advanced RTMs can be computationally burdensome, which makes them impractical in many real applications, especially when many state conditions and model couplings need to be studied. To overcome this problem, it is proposed to substitute RTMs through surrogate meta-models also named emulators. Emulators approximate the functioning of RTMs through statistical learning regression methods, and can open many new applications because of their computational efficiency and outstanding accuracy. Emulators allow fast global sensitivity analysis (GSA) studies on advanced, computationally expensive RTMs. As a proof-of-concept, three machine learning regression algorithms (MLRAs) were tested to function as emulators for the leaf RTM PROSPECT-4, the canopy RTM PROSAIL, and the computationally expensive atmospheric RTM MODTRAN5. Selected MLRAs were: kernel ridge regression (KRR), neural networks (NN) and Gaussian processes regression (GPR). For each RTM, 500 simulations were generated for training and validation. The majority of MLRAs were excellently validated to function as emulators with relative errors well below 0.2%. The emulators were then put into a GSA scheme and compared against GSA results as generated by original PROSPECT-4 and PROSAIL runs. NN and GPR emulators delivered identical GSA results, while processing speed compared to the original RTMs doubled for PROSPECT-4 and tripled for PROSAIL. Having the emulator-GSA concept successfully tested, for six MODTRAN5 atmospheric transfer functions (outputs), i.e., direct and diffuse at-surface solar irradiance (E di f , E dir ), direct and diffuse upward transmittance (T dir , T di f ), spherical albedo (S) and path radiance (L 0 ), the most accurate MLRA's were subsequently applied as emulator into the GSA scheme. The sensitivity analysis along the 400-2500 nm spectral range took no more than a few minutes on a contemporary computer-in comparison, the same analysis in the original MODTRAN5 would have taken over a month. Key atmospheric drivers were identified, which are on the one hand aerosol optical properties, i.e., aerosol optical thickness (AOT), Angstrom coefficient (AMS) and scattering asymmetry variable (G), mostly driving diffuse atmospheric components, E di f and T di f ; and those affected by atmospheric scattering, L 0 and S. On the other hand, as expected, AOT, AMS and columnar water vapor (CWV) in the absorption regions mostly drive E dir and T dir atmospheric functions. The presented emulation schemes showed very promising results in replacing costly RTMs, and we think they can contribute to the adoption of machine learning techniques in remote sensing and environmental applications.

show abstract

Section: Emulator Theorymentioning

confidence: 99%

Emulation of Leaf, Canopy and Atmosphere Radiative Transfer Models for Fast Global Sensitivity Analysis

et al. 2016

View full text Add to dashboard Cite

show abstract

“…The application of emulators has emerged in many different fields of science and thus the theoretical background is relatively well developed (e.g. O'Hagan, 2000, 2001;Phillips, 2003;Lucia et al, 2004;van der Merwe et al, 2007;Bliznyuk et al, 2008;Conti et al, 2009;Liu and West, 2009;Castelletti et al, 2012). Two distinct approaches to emulation exist, which we refer to as dynamic emulation and statistical emulation.…”

Section: Emulator Approachesmentioning

confidence: 99%

“…However, approaches based on stochastic models can be computationally intensive and methodologically complex, and parameterizing all individual sources of BC/forcing error poses a major challenge. Rather than attempting a comprehensive treatment, current approaches tend to restrict the dynamical noise to certain key sources such as the atmospheric forcing (Natvik and Evensen, 2003;Simon and Bertino, 2009) or surface irradiance and background light attenuation (Torres et al, 2006;Ciavatta et al, 2011), and/or they model the net effect of BC/forcing errors and structural errors synthetically as additive (e.g. Losa et al, 2003Losa et al, , 2004 or multiplicative (e.g.…”

Section: Uncertainty In Forcings and Boundary Conditions (Bcs)mentioning

confidence: 99%

Reviews and syntheses: parameter identification in marine planktonic ecosystem modelling

et al. 2017

View full text Add to dashboard Cite

Abstract. To describe the underlying processes involved in oceanic plankton dynamics is crucial for the determination of energy and mass flux through an ecosystem and for the estimation of biogeochemical element cycling. Many planktonic ecosystem models were developed to resolve major processes so that flux estimates can be derived from numerical simulations. These results depend on the type and number of parameterizations incorporated as model equations. Furthermore, the values assigned to respective parameters specify a model's solution. Representative model results are those that can explain data; therefore, data assimilation methods are utilized to yield optimal estimates of parameter values while fitting model results to match data. Central difficulties are (1) planktonic ecosystem models are imperfect and (2) data are often too sparse to constrain all model parameters. In this review we explore how problems in parameter identification are approached in marine planktonic ecosystem modelling.We provide background information about model uncertainties and estimation methods, and how these are considered for assessing misfits between observations and model results. We explain differences in evaluating uncertainties in parameter estimation, thereby also discussing issues of parameter identifiability. Aspects of model complexity are addressed and we describe how results from cross-validation studies provide much insight in this respect. Moreover, approaches are discussed that consider time-and spacedependent parameter values. We further discuss the use of dynamical/statistical emulator approaches, and we elucidate issues of parameter identification in global biogeochemical models. Our review discloses many facets of parameter identification, as we found many commonalities between the objectives of different approaches, but scientific insight differed between studies. To learn more from results of planktonic ecosystem models we recommend finding a good balance in the level of sophistication between mechanistic modelling and statistical data assimilation treatment for parameter estimation.

show abstract

“…Emulation consists of building a computationally inexpensive model and training it with data sets generated from the input-output data sets obtained from more complex models [27,28]. Hydrodynamic model emulation has proven to be a powerful tool for water level predictions while improving the calculation speed significantly [29,30].…”

Section: Introductionmentioning

confidence: 99%

Failure of Grass Covered Flood Defences with Roads on Top Due to Wave Overtopping: A Probabilistic Assessment Method

Lopez

Warmink²,

Bomers³

et al. 2018

JMSE

View full text Add to dashboard Cite

Abstract:Hard structures, i.e., roads, are commonly found over flood defences, such as dikes, in order to ensure access and connectivity between flood protected areas. Several climate change future scenario studies have concluded that flood defences will be required to withstand more severe storms than the ones used for their original design. Therefore, this paper presents a probabilistic methodology to assess the effect of a road on top of a dike: it gives the failure probability of the grass cover due to wave overtopping over a wide range of design storms. The methodology was developed by building two different dike configurations in computational fluid dynamics Navier-Stokes solution software; one with a road on top and one without a road. Both models were validated with experimental data collected from field-scale experiments. Later, both models were used to produce data sets for training simpler and faster emulators. These emulators were coupled to a simplified erosion model which allowed testing storm scenarios which resulted in local scouring conditioned statistical failure probabilities. From these results it was estimated that the dike with a road has higher probabilities (5 × 10 −5 > P f >1 × 10 −4 ) of failure than a dike without a road (P f < 1 × 10 −6 ) if realistic grass quality spatial distributions were assumed. The coupled emulator-erosion model was able to yield realistic probabilities, given all the uncertainties in the modelling process and it seems to be a promising tool for quantifying grass cover erosion failure.

show abstract

A general framework for Dynamic Emulation Modelling in environmental problems

Cited by 129 publications

References 116 publications

Emulation of Leaf, Canopy and Atmosphere Radiative Transfer Models for Fast Global Sensitivity Analysis

Emulation of Leaf, Canopy and Atmosphere Radiative Transfer Models for Fast Global Sensitivity Analysis

Reviews and syntheses: parameter identification in marine planktonic ecosystem modelling

Failure of Grass Covered Flood Defences with Roads on Top Due to Wave Overtopping: A Probabilistic Assessment Method

Contact Info

Product

Resources

About