A surrogate-based approach to nonlinear, non-Gaussian joint state-parameter data assimilation

DOI10.3934/FODS.2021019MaRDI QIDQ2072658zbMATH OpenOpenAlexFDO

Authors Elaine T. Spiller, John MacLean

Publication date 26 January 2022

Published in Foundations of Data Science (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/2012.04793

zbMATH Keywords

data assimilationdata scienceparameter estimationstatistical surrogatesuncertainty quantification

Mathematics Subject Classification ID

Empirical decision procedures; empirical Bayes procedures (62C12) Density estimation (62G07) Markov processes: estimation; hidden Markov models (62M05) Inference from stochastic processes and prediction (62M20) Statistical aspects of big data and data science (62R07) Meteorology and atmospheric physics (86A10)

Abstract: Many recent advances in sequential assimilation of data into nonlinear high-dimensional models are modifications to particle filters which employ efficient searches of a high-dimensional state space. In this work, we present a complementary strategy that combines statistical emulators and particle filters. The emulators are used to learn and offer a computationally cheap approximation to the forward dynamic mapping. This emulator-particle filter (Emu-PF) approach requires a modest number of forward-model runs, but yields well-resolved posterior distributions even in non-Gaussian cases. We explore several modifications to the Emu-PF that utilize mechanisms for dimension reduction to efficiently fit the statistical emulator, and present a series of simulation experiments on an atypical Lorenz-96 system to demonstrate their performance. We conclude with a discussion on how the Emu-PF can be paired with modern particle filtering algorithms.

Recommendations

• Efficient nonlinear data-assimilation in geophysical fluid dynamics
• Nonlinear data assimilation
• Data assimilation for models with parametric uncertainty
• A local ensemble data assimilation algorithm for nonlinear geophysical models
• Parameter sampling capabilities of sequential and simultaneous data assimilation: II. Statistical analysis of numerical results

Cites work

• Calibrate, emulate, sample
• Computer model validation with functional output
• Data Assimilation
• Design and analysis of computer experiments. With comments and a rejoinder by the authors
• Dimension Reduction for Gaussian Process Emulation: An Application to the Influence of Bathymetry on Tsunami Heights
• Gaussian processes for machine learning.
• Generalised particle filters with Gaussian mixtures
• scientific article; zbMATH DE number 1666093 (Why is no real title available?)
• Parallel partial Gaussian process emulation for computer models with massive output
• Sequential Monte Carlo Methods for Dynamic Systems
• Sequential Monte Carlo Methods in Practice
• Supervised learning from noisy observations: combining machine-learning techniques with data assimilation
• The design and analysis of computer experiments
• The ensemble Kalman filter for combined state and parameter estimation
• VISUALIZING BIFURCATIONS IN HIGH DIMENSIONAL SYSTEMS: THE SPECTRAL BIFURCATION DIAGRAM

Cited in

(2)

• Emulator-assisted reduced-rank ecological data assimilation for nonlinear multivariate dynamical spatio-temporal processes
• Blending machine learning and sequential data assimilation over latent spaces for surrogate modeling of Boussinesq systems

Describes a project that uses

Uses Software

• EnKF

This page was built for publication: A surrogate-based approach to nonlinear, non-Gaussian joint state-parameter data assimilation

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2072658)