Inverse iterative simulation: An efficient approach for contaminant source identification

TL;DR

This paper introduces the inverse iterative simulation (iIS) method, an efficient technique for identifying contaminant source parameters and hydraulic conductivity fields in groundwater modeling, significantly reducing computational costs compared to traditional methods.

Contribution

The paper presents the iIS algorithm, combining ensemble smoothing and inverse Gaussian processes, to efficiently solve high-dimensional inverse problems in groundwater contamination analysis.

Findings

iIS achieves near-MCMC accuracy in low-dimensional cases

iIS provides accurate high-dimensional estimates with low computational cost

The method outperforms traditional approaches in efficiency and accuracy

Abstract

In groundwater contaminant remediation and risk assessment, it is important to identify parameters of the contaminant source and hydraulic conductivity field by solving an inverse problem. However, if the dimensionality of the inverse problem is high, it is usually computationally expensive to obtain accurate estimation and uncertainty assessment of these parameters. This is particularly the case when Markov Chain Monte Carlo (MCMC) sampling is used. In this paper, an efficient approach entitled inverse iterative simulation (iIS) is proposed to efficiently identify the contaminant source characteristics, together with the hydraulic conductivity field. The iIS algorithm utilizes a simple approach borrowed from Ensemble Smother (ES) to update model parameters and an inverse Gaussian process (iGP) approach to improve the accuracy of parameter updating. Two numerical experiments are tested. For the low dimensional case (with 11 parameters), the iIS algorithm can obtain parameter estimation very close to that of MCMC method. For the high dimensional case (with 108 parameters), the iIS algorithm can obtain accurate parameter estimation with very low computational cost.