In this work, we apply the Ensemble Smoother with Multiple Data Assimilation (ES-MDA) method for the simultaneous identification of the source location and the release history of a pollutant in groundwater. The ES-MDA is an iterative data assimilation method, based on the knowledge of observed concentration data, that updates the unknown parameters iteratively. In this work, these parameters are the spatial coordinates of the source and the time-discretized release history. The methodology is tested on two cases. First, we apply the ES-MDA to a numerical study from the literature, with the aim to show the capabilities of the approach to solve this type of inverse problem. We try to estimate the source location and the release history of a nonreactive pollutant spreading in a 2-D homogeneous aquifer from a point source, also considering the impact of the sampling scheme of the observed concentrations. Different settings of the inverse algorithm have been tested, such as the ensemble size, the choice of ES-MDA-related coefficients, or the use of covariance localization and covariance inflation techniques, with the purpose of reaching a reliable solution with a reduced computational burden. In the second case, we use real data collected in the laboratory. The experimental device is a sandbox that reproduces an unconfined aquifer characterized by two-dimensional flow in the vertical plane. The porous media reproduces an almost homogeneous field. The pollutant is represented by fluorescein sodium salt used as tracer and injected at a variable mass rate. The observed concentrations are recorded by taking pictures with a digital camera and then converting luminosity into concentration through image processing techniques. A calibrated numerical model able to describe as accurately as possible the forward processes is required. In this case, the groundwater flow was modeled with MODFLOW 2005 and the transport process with MT3DMS. The hydraulic and transport models are preliminary calibrated by other means. The results show that the ES-MDA is able to simultaneously identify the source location and the release history of a pollutant in a groundwater system, but its reliability depends on the sampling scheme of the observations and on the different settings of the inverse procedure.

Ensemble smoother with multiple data assimilation for the simultaneous identification of the source location and the release history of a pollutant in groundwater / Todaro, V.; D'Oria, M.; Tanda, M. G.; Gómez-Hernández, J. J.. - (2020). (Intervento presentato al convegno AGU Fall Meeting 2020).

Ensemble smoother with multiple data assimilation for the simultaneous identification of the source location and the release history of a pollutant in groundwater

Todaro, V.
;
D'Oria, M.;Tanda, M. G.;
2020-01-01

Abstract

In this work, we apply the Ensemble Smoother with Multiple Data Assimilation (ES-MDA) method for the simultaneous identification of the source location and the release history of a pollutant in groundwater. The ES-MDA is an iterative data assimilation method, based on the knowledge of observed concentration data, that updates the unknown parameters iteratively. In this work, these parameters are the spatial coordinates of the source and the time-discretized release history. The methodology is tested on two cases. First, we apply the ES-MDA to a numerical study from the literature, with the aim to show the capabilities of the approach to solve this type of inverse problem. We try to estimate the source location and the release history of a nonreactive pollutant spreading in a 2-D homogeneous aquifer from a point source, also considering the impact of the sampling scheme of the observed concentrations. Different settings of the inverse algorithm have been tested, such as the ensemble size, the choice of ES-MDA-related coefficients, or the use of covariance localization and covariance inflation techniques, with the purpose of reaching a reliable solution with a reduced computational burden. In the second case, we use real data collected in the laboratory. The experimental device is a sandbox that reproduces an unconfined aquifer characterized by two-dimensional flow in the vertical plane. The porous media reproduces an almost homogeneous field. The pollutant is represented by fluorescein sodium salt used as tracer and injected at a variable mass rate. The observed concentrations are recorded by taking pictures with a digital camera and then converting luminosity into concentration through image processing techniques. A calibrated numerical model able to describe as accurately as possible the forward processes is required. In this case, the groundwater flow was modeled with MODFLOW 2005 and the transport process with MT3DMS. The hydraulic and transport models are preliminary calibrated by other means. The results show that the ES-MDA is able to simultaneously identify the source location and the release history of a pollutant in a groundwater system, but its reliability depends on the sampling scheme of the observations and on the different settings of the inverse procedure.
2020
Ensemble smoother with multiple data assimilation for the simultaneous identification of the source location and the release history of a pollutant in groundwater / Todaro, V.; D'Oria, M.; Tanda, M. G.; Gómez-Hernández, J. J.. - (2020). (Intervento presentato al convegno AGU Fall Meeting 2020).
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/11381/2911087
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact