Identifying contaminant source information from limited concentration measurements downstream from the source is a crucial step in groundwater pollution investigation for accountability and remediation purposes. However, in reality, the partially known but influential hydraulic conductivity field is always a significant obstacle in this inverse problem. In this work, we apply the restart normal- score Ensemble Kalman filter (NS-EnKF) method to identify the contaminant source and a non- Gaussian conductivity field jointly in a sandbox experiment by only using concentration measurements at a few observation points. As a preliminary step, we verify the restart NS-EnKF in a synthetic case mimicking the sandbox experiment. Some simple tries demonstrate that with a small ensemble size, measurement affected by observation errors and a complex sandbox conductivity lead to filter inbreeding. Consequently, a larger ensemble size, and several inflation methods are analyzed to solve this problem. We found that using a large ensemble size or Bauser’s inflation method avoids filter inbreeding. These conclusions are used to analyse the sandbox results to identify the contaminant source and the conductivities using the restart NS-EnKF. The results show that the restart Ns-EnKF with a proper ensemble size or a suitable inflation method is capable to identify the contaminant source and the non-Gaussian conductivities in the sandbox experiment.

The use of inflation in ensemble Kalman filter for the joint identification of contaminant source parameters and hydraulic conductivities in a sandbox experiment / Chen, Zi; Gomez-Hernandez, Jaime; Xu, Teng; Zanini, Andrea. - ELETTRONICO. - (2019), p. 479. (Intervento presentato al convegno 46th IAH Congress - Malaga tenutosi a Malaga nel 22-27 settembre 2019).

The use of inflation in ensemble Kalman filter for the joint identification of contaminant source parameters and hydraulic conductivities in a sandbox experiment

Andrea Zanini
2019-01-01

Abstract

Identifying contaminant source information from limited concentration measurements downstream from the source is a crucial step in groundwater pollution investigation for accountability and remediation purposes. However, in reality, the partially known but influential hydraulic conductivity field is always a significant obstacle in this inverse problem. In this work, we apply the restart normal- score Ensemble Kalman filter (NS-EnKF) method to identify the contaminant source and a non- Gaussian conductivity field jointly in a sandbox experiment by only using concentration measurements at a few observation points. As a preliminary step, we verify the restart NS-EnKF in a synthetic case mimicking the sandbox experiment. Some simple tries demonstrate that with a small ensemble size, measurement affected by observation errors and a complex sandbox conductivity lead to filter inbreeding. Consequently, a larger ensemble size, and several inflation methods are analyzed to solve this problem. We found that using a large ensemble size or Bauser’s inflation method avoids filter inbreeding. These conclusions are used to analyse the sandbox results to identify the contaminant source and the conductivities using the restart NS-EnKF. The results show that the restart Ns-EnKF with a proper ensemble size or a suitable inflation method is capable to identify the contaminant source and the non-Gaussian conductivities in the sandbox experiment.
2019
The use of inflation in ensemble Kalman filter for the joint identification of contaminant source parameters and hydraulic conductivities in a sandbox experiment / Chen, Zi; Gomez-Hernandez, Jaime; Xu, Teng; Zanini, Andrea. - ELETTRONICO. - (2019), p. 479. (Intervento presentato al convegno 46th IAH Congress - Malaga tenutosi a Malaga nel 22-27 settembre 2019).
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11381/2888090
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact