Quantifying the Impact of Evapotranspiration at the Aquifer Scale via Groundwater Modelling and MODIS Data

Round 1

Reviewer 1 Report

Quantifying the evapotranspiration rates in aquifers characterized by coarse sediments is still a challenge, especially because of the high permeability and the heterogeneity of the subsoil as well. Therefore, evapotranspiration in low capillarity soil is generally not considered leading to an inaccuracy in the estimation of the hydrogeological balance.

The Authors quantified the role of evapotranspiration in such a type of aquifer through direct assimilation of MODIS data as model input data properly calibrated by in-situ measurements.

I enjoyed reading this paper, it is technically sound, the text is clear and well written. I greatly appreciated the combined use of numerical modeling and MODIS data for overcoming the evapotranspiration issue.

In my view, the manuscript could be accepted for publication as it is, but I suggest the Authors the few points listed below.

 

Fig. 1. I suggest to use a background map (or google earth image) including Aquasanta, Diga Talvecchia, and San Vito to better frame the study area. Also, use a km scale bar and a smaller size for North.

Fig. 2. I suggest to plot the 3D sketch with the same orientation as that used for Fig 1 and Fig. 3. Alternatively, include some references for orientation such as the North Arrow).

Line 140. Add reference for Kriging.

Fig 5. Add unit for groundwater heads (m asl).

Tab. 3 and Tab. 4. Consider if only two decimal places are enough.

 

 

Author Response

Please see the attached file.

Author Response File: Author Response.docx

Reviewer 2 Report

The topic addressed is interesting and I do thing this manuscript could be published as a technical report. 

However, before to consider it for publication, I do think that some critical points need to be clarified.

1. The model setting includes CHB condition to model the regional inflow from the west side. The use of Dirichelet condition in this case is highly questionable, and in any case no information is presented to justify this approach, nor comment/information on the imposed piezometry is presented. Authors should present and comment values of head here imposed and why they selected such values (did you have any estimate on the inflow? If yes, what kind of analysis you performed to represent the inflow as a Dirichelet condition?).
2. The point above is crucial, since the water budget (Table 3) shows that the inflow due to CHD represents around 50% of the total inflow. Therefore, all modeling results are highly influenced by this choice.
3. No information on pumping rate regimes is presented: again, from th budget, I see that water abstraction (WEL condition) is of the same order of EVT, the latter being the subject of the paper. Thus, more info on well pumping is needed, as it seems to be relevant.
4. I do not understand why Authors mention pumping rate and EVT rate as “Calibratd parameters” (Table 2)!?!. Pumping is not given as input datum? (see point 3). EVT rate is derived from MODIS data, so why do you mention it as Calibrated parameter? As in fact, at the beginning of Section 2.4 these parameters are not listed as subject of calibration. Please, clarify and correct Table 2, accordingly.
5. Once point 4 is clarified, I suggest to do and present a sensitivity analysis on EVT rate (for instance starting form the base model used as reference for calibration): since you are applying automatic calibration code (PEST) you should take the advantage of applying sensitivity analysis embedded in it. This initial sensitivity analysis will show you the importance of such parameter before to run the “manual” sensitivity using different data set. You should show the importance of such a parameter (compared to other) using classical sensitivity statics (like Composite Scaled Sensitivity, for instance).

Furthermore, I have also the following minor comments/suggestions:

1. Line 108: no transport model is here presented, the model concerns only flow. Please correct this sentence.
2. I suggest to present Figure 2 in a “planar” view, to better see the location of boundary condition. Thus, Figure 2 could present both 2D and 3D version of the model domain.
3. Table 1: units of measure are missing (I guess: mm)
4. Table 2: review this table after having clarified Point 4 above.
5. Table 3: CHD is used to represent both inflow from the west-side boundary and the sea. Despite my comments (Point 1), if you clarify it, I suggest splitting the contribute of CHD between inflow and sea-side, since these are different budget terms. You can do it easily by applying ZoneBudget capability of MODFLOW.

Author Response

Please see the attached file.

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

Authors have addressed all the critical issues pointed out in the 1st review. The minor comments have been addressed as well.

I do think that in this new version the paper is more clear and even specific issues are now better explained.

I suggest to review Figure 2 (lower panel) beacuse the quality/resolution seems not so good, but this can be addressed during the copyediting phase.

Author Response

We sincerely thank the reviewer for the positive comments, we have removed the grid in plan view that was creating the issue with the resolution.