Three-dimensional hydrogeological parametrization using sparse piezometric data

Rambourg, Dimitri; Di Chiara, Raphaël; Ackerer, Philippe

doi:https://doi.org/10.5194/hess-26-6147-2022

Articles | Volume 26, issue 23

https://doi.org/10.5194/hess-26-6147-2022

© Author(s) 2022. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/hess-26-6147-2022

© Author(s) 2022. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 26, issue 23

Research article

|

08 Dec 2022

Research article |

| 08 Dec 2022

Three-dimensional hydrogeological parametrization using sparse piezometric data

Dimitri Rambourg, Raphaël Di Chiara, and Philippe Ackerer

Download

Final revised paper (published on 08 Dec 2022)
Preprint (discussion started on 04 Apr 2022)

Interactive discussion

Status: closed

RC1:
'Comment on egusphere-2022-81', Anonymous Referee #1, 07 May 2022

Dear authors,

I read carefully your work and I found it a very important contribution in groundwater hydrology.

The methodology is clearly presented, the mathematical background as well.

The interpolation part may require some more details but on the other hand it mainly supports the concept.

The results and discussion of the proposed methodology is well presented as well.

On the other hand, I have some overall concerns:

1) The proposed method is only digestive for those who have specialized knowledge of the entire tools presented.

2) Please mention the innovation compared to similar works .

3) Most important the presented methodology is very complex to be reproduced. I am not saying that is bad! but there also similar works in the literature that do the same work with a simpler manner. Maybe it would be good, if possible, to have a comparison with one of them. Your method is more detailed but compared to simpler approach the performance is far more efficient? Otherwise, what is the reason to have such many methodological and sometime complex steps.

4) paragraph 2.4 The optimization part needs more details. It is not clear how optimization works here.

5)The proposed 3d methodology consist of inversion, interpolation, optimization. All these steps consider parameters. Therefore, an uncertainty analysis is required to study the uncertainty propagation.

6) How realistic is the upscale of such model to a real case study. I understand the research orientation which is very strong but, this is also a matter of discussion.

7) The fixed parameters of the aquifer model regarding transport it would be good to be accompanied by a sensitivity analysis.

Citation: https://doi.org/10.5194/egusphere-2022-81-RC1
- AC1: 'Reply on RC1', Dimitri Rambourg, 23 Jun 2022
  
  Dear Referee, please find in attachment the response to your comments.
  
  We thank you for the time and attention you have dedicated to our work.
  Best regards,
  The authors.
  
  Citation: https://doi.org/10.5194/egusphere-2022-81-AC1
RC2:
'Comment on egusphere-2022-81', Anonymous Referee #2, 23 May 2022

The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2022/egusphere-2022-81/egusphere-2022-81-RC2-supplement.pdf

Citation: https://doi.org/10.5194/egusphere-2022-81-RC2
- AC2: 'Reply on RC2', Dimitri Rambourg, 23 Jun 2022
  
  Dear Referee, please find in attachment the response to your comments.
  
  We thank you for the time and attention you have dedicated to our work.
  Best regards,
  The authors.
  
  Citation: https://doi.org/10.5194/egusphere-2022-81-AC2

Peer review completion

AR: Author's response | RR: Referee report | ED: Editor decision | EF: Editorial file upload

ED: Reconsider after major revisions (further review by editor and referees) (24 Jun 2022) by Alberto Guadagnini

AR by Dimitri Rambourg on behalf of the Authors (07 Jul 2022) Author's response Author's tracked changes Manuscript

ED: Referee Nomination & Report Request started (08 Jul 2022) by Alberto Guadagnini

RR by Anonymous Referee #1 (16 Jul 2022)

Suggestions for revision or reasons for rejection

Dear authors I believe your manuscripts is a useful input but reading it I have the following concerns for clarifications and potential improvements.

The interpolation and geostatistical part needs improvement in terms of variogram calculation, stationary analysis, parameters calculation, validation and variance-uncertainty. It is a major part of your methodology and needs to be clear. You mention application of universal cokriging, of which variables, using which variogram model e.t.c

it seems that an uncertainty analysis has been carried out in line 280 but it is not clearly presented and discussed.
As long as simulations have been carried out then the overall uncertainty can be presented in the facies of the models as support figures.

Your method needs to be compared with a classical 3d inverse modelling approach (3D inverse modelling of groundwater) such the ones that have been published by J. J. Gómez-Hernández and Kitanidis.

In the abstract it is mentioned: Finally, the parameters of each facies (hydraulic conductivity and porosity)
are obtained through an optimization loop, that minimizes the difference between the 2D calibrated transmissivity and the
transmissivity computed with the estimated 3D facies parameters. According to presented methodology to achieve the target inversion, interpolation, optimization is applied. All these steps consider parameters. Therefore, an uncertainty analysis is required to study the uncertainty propagation even for synthetic data or at least show the overall uncertainty thorough the simulations, or even identify the sources of uncertainty for the readers to consider them when up-scaling.

for the B-splines method you mention that one can use GIS, for interpolation python and for the model fortan!
How all these software results are connected for your complete model? It seems that many assumptions should be made and I am still worried about uncertainty.

The figures caption require more complete descriptions.

in 3,2: Differences in the facies composition of the models are marginal. ok but are they correct/reliable to be incorporated in the model?

How the maximal discrepancy is 61 % of the initial value for the permeability and other increased discrepancies mentioned interprets a successful reconstruction of the aquifer dynamics. Obviously the model parameterization works well. Please explain clearly.

In figure 7 there are several dashed lines color, please elaborate. In addition, the aim of the work is a 3d hydrogeological characterization combining inversion, interpolation and optimization. It is not clear what fig 7 adds to the work and interpolation techniques application to water level. From the method and flowchart it seems that the interpolation applies to lithological data only.

line 400: Comparatively to joint inversion methods, the need of data acquisition and the computation efforts are lower.
I do not fully agree, here you have uncertainty that you have not discussed. Stating this you need to support it with an example or literature.

reproduction of this work by the readers is not clear, please show some guidelines.

The model results seems good in the analysis but a sensitivity analysis and uncertainty analysis in my personal opinion is needed to support the findings.

Hide

ED: Publish subject to revisions (further review by editor and referees) (16 Jul 2022) by Alberto Guadagnini

AR by Dimitri Rambourg on behalf of the Authors (08 Sep 2022) Author's response Author's tracked changes Manuscript

ED: Referee Nomination & Report Request started (09 Sep 2022) by Alberto Guadagnini

RR by Anonymous Referee #1 (16 Oct 2022)

ED: Publish subject to minor revisions (review by editor) (16 Oct 2022) by Alberto Guadagnini

AR by Dimitri Rambourg on behalf of the Authors (20 Oct 2022) Author's response Author's tracked changes Manuscript

ED: Publish as is (20 Oct 2022) by Alberto Guadagnini

AR by Dimitri Rambourg on behalf of the Authors (27 Oct 2022) Manuscript

Short summary

The reproduction of flows and contaminations underground requires a good estimation of the parameters of the geological environment (mainly permeability and porosity), in three dimensions. While most researchers rely on geophysical methods, which are costly and difficult to implement in the field, this study proposes an alternative using data that are already widely available: piezometric records (monitoring of the water table) and the lithological description of the piezometric wells.