Stochastic perturbations and dimension reduction for modelling uncertainty of atmospheric dispersion simulations

Abstract Decision of emergency response to releases of hazardous material in the atmosphere increasingly rely on numerical simulations. This paper presents two contributions for accounting for the uncertainty inherent to those simulations. We first focused on one way of modelling these uncertainties, namely by applying stochastic perturbations to the inputs of the numerical dispersion model. We devised a generic mathematical formulation for time dependent perturbation of both amplitude and dynamics of the inputs. It allows a more thorough exploration of possible outcomes than simpler perturbations found in the literature. We then improved on the current state of the art on dimension reduction of atmospheric data. Indeed, most statistical methods cannot cope with high dimensional data such as the maps simulated with atmospheric dispersion models. Principal component analysis, the most widely used method for dimension reduction, relies on a linearity hypothesis that is not verified by these sets of maps. We conducted a very encouraging experiment with auto-associative models, a non-linear extension of this method.

[1]  Vivien Mallet,et al.  Automatic calibration of an ensemble for uncertainty estimation and probabilistic forecast: Application to air quality , 2011 .

[2]  Vangelis Karkaletsis,et al.  Autoencoder-Driven Weather Clustering for Source Estimation during Nuclear Events , 2017, Environ. Model. Softw..

[3]  Serge Iovleff,et al.  Auto-Associative Models, Nonlinear Principal Component Analysis, Manifolds and Projection Pursuit , 2011, 1103.6119.

[4]  J. Rougier,et al.  Parametric uncertainty in complex environmental models: a cheap emulation approach for models with high-dimensional output , 2017, 1702.03696.

[5]  Keinosuke Fukunaga,et al.  An Algorithm for Finding Intrinsic Dimensionality of Data , 1971, IEEE Transactions on Computers.

[6]  Karl Pearson F.R.S..  LIII. On lines and planes of closest fit to systems of points in space , 1901 .

[7]  Gianni Tinarelli,et al.  Review and Validation of MicroSpray, a Lagrangian Particle Model of Turbulent Dispersion , 2013 .

[8]  S. Hanna,et al.  Uncertainties in Air Toxics Calculated by the Dispersion Models AERMOD and ISCST3 in the Houston Ship Channel Area , 2007 .

[9]  Vivien Mallet,et al.  Screening sensitivity analysis of a radionuclides atmospheric dispersion model applied to the Fukushima disaster , 2014 .

[10]  Patrick Armand,et al.  Description and preliminary validation of the PMSS fast response parallel atmospheric flow and dispersion solver in complex built-up areas , 2017, Environmental Fluid Mechanics.

[11]  Jorge Cadima,et al.  Principal component analysis: a review and recent developments , 2016, Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences.

[12]  Yang Zhang,et al.  Real-time air quality forecasting, part II: State of the science, current research needs, and future prospects , 2012 .

[13]  Steven R. Hanna,et al.  Acceptance criteria for urban dispersion model evaluation , 2012, Meteorology and Atmospheric Physics.

[14]  Thomas T. Warner,et al.  Ensemble Simulations with Coupled Atmospheric Dynamic and Dispersion Models: Illustrating Uncertainties in Dosage Simulations , 2002 .

[15]  David C. Woods,et al.  Emulation of Multivariate Simulators Using Thin-Plate Splines with Application to Atmospheric Dispersion , 2015, SIAM/ASA J. Uncertain. Quantification.

[16]  Vivien Mallet,et al.  Emulation and Sobol' sensitivity analysis of an atmospheric dispersion model applied to the Fukushima nuclear accident , 2016 .

[17]  Christophe Duchenne,et al.  Validation of a Lagrangian particle dispersion model with wind tunnel and field experiments in urban environment , 2018, Atmospheric Environment.

[18]  Patrick Armand,et al.  Validation and Sensitivity Study of the PMSS Modelling System for Puff Releases in the Joint Urban 2003 Field Experiment , 2019, Boundary-Layer Meteorology.

[19]  Walter F. Dabberdt,et al.  Uncertainty, ensembles and air quality dispersion modeling: applications and challenges , 2000 .

[20]  Matthias Beekmann,et al.  Monte Carlo uncertainty analysis of a regional‐scale transport chemistry model constrained by measurements from the Atmospheric Pollution Over the Paris Area (ESQUIF) campaign , 2003 .

[21]  Marie Ekström,et al.  Combining dispersion modelling with synoptic patterns to understand the wind-borne transport into the UK of the bluetongue disease vector , 2017, International Journal of Biometeorology.

[22]  Yves Deville,et al.  DiceKriging, DiceOptim: Two R Packages for the Analysis of Computer Experiments by Kriging-Based Metamodeling and Optimization , 2012 .

[23]  G. Powers,et al.  A Description of the Advanced Research WRF Version 3 , 2008 .

[24]  M. Sofiev,et al.  Ensemble dispersion forecasting—Part I: concept, approach and indicators , 2004 .

[25]  A. Agresti,et al.  Approximate is Better than “Exact” for Interval Estimation of Binomial Proportions , 1998 .

[26]  Patrick Armand,et al.  Probabilistic safety analysis for urgent situations following the accidental release of a pollutant in the atmosphere , 2014 .

[27]  Vivien Mallet,et al.  Meta-modeling of ADMS-Urban by dimension reduction and emulation , 2018, Atmospheric Environment.

[28]  Vivien Mallet,et al.  CALIBRATION OF A SURROGATE DISPERSION MODEL APPLIED TO THE FUKUSHIMA NUCLEAR DISASTER , 2019, Proceedings of the 3rd International Conference on Uncertainty Quantification in Computational Sciences and Engineering (UNCECOMP 2019).