Urban Underground Structures and Karst Groundwater Systems Interactions: The Case of Mazzoccolo Spring in Formia, Central Italy
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsPag 1 - The title is not appropriate, actually the S.S. 7 Appia tunnel is only at the project phase, so there is no impact. Also, within the article, there is no clear evidence of urban impact assessment, only general mentions.
Pag 3 - The geographical map can be improved showing also the location in a greater area – location of the Mount Aurunci, the sea, etc. The map of the administrative boundary of Formia City is not adding any value. Also the administrative boundary of Latium Region is having sense only for readers that have very good knowledge of the geography of Italy.
Pag 3 - The geological description is better to be accompanied by a geological/hydrogeological cross section.
Pag 4 - From figure 2 it is very difficult to understand the methodology behind the delineation of the catchment area. If the delineation is based on the hydrographical basin, than a topography map is needed.
Pag. 5 - The numbering of the chapter is incorrect.
Pag 6 - From the figure 4 (correlation function for AAP calculation) the altitude values for the 3 main weather station are below 200 m a.s.l. The degree if incertitude for function extrapolation above 200 m a.s.l should be discussed and better explained as the catchment area is represented by altitudes higher than 200 m (until around 1200 m a.s.l).
Pag 6 - The Effective Infiltration calculation is not described. The quantification of the lithological outcrop is not described, and and also it is not clear how the Effective infiltration map was made.
Pag 15 - There is clear need for a piezometric map, not only a literary description of the piezometric surface.
Author Response
Pag 1 - The title is not appropriate, actually the S.S. 7 Appia tunnel is only at the project phase, so there is no impact. Also, within the article, there is no clear evidence of urban impact assessment, only general mentions.
The title has been changed to “Urban Underground Structures and Karst Groundwater Systems interactions: The Case of Mazzoccolo Spring in Formia, Central Italy”
Pag 3 - The geographical map can be improved showing also the location in a greater area – location of the Mount Aurunci, the sea, etc. The map of the administrative boundary of Formia City is not adding any value. Also the administrative boundary of Latium Region is having sense only for readers that have very good knowledge of the geography of Italy.
The geographical map (Fig. 1) has been modified, showing the location of the Aurunci Mts. and the Tyrrhenian Sea.
Pag 3 - The geological description is better to be accompanied by a geological/hydrogeological cross section.
A geological cross section has been drawn in Fig. 3.
Pag 4 - From figure 2 it is very difficult to understand the methodology behind the delineation of the catchment area. If the delineation is based on the hydrographical basin, than a topography map is needed.
In the manuscript has been explained the methodology behind the delineation of the catchment area, that came from the delineation of the catchment area proposed by Baldi [19], simplified through the analysis of digital elevation models.
Pag. 5 - The numbering of the chapter is incorrect.
Done
Pag 6 - From the figure 4 (correlation function for AAP calculation) the altitude values for the 3 main weather station are below 200 m a.s.l. The degree if incertitude for function extrapolation above 200 m a.s.l should be discussed and better explained as the catchment area is represented by altitudes higher than 200 m (until around 1200 m a.s.l).
The issue raised by Reviewer 1 is true. Thank you for highlighting that.
For the rainfall-altitude and temperature-altitude relationships, the interpolation that is normally used is the linear one, although for annual cumulative rainfall, papers from literature proved that logarithmic or exponential relationships lead to obtain much more plausible values in the case of mountain basins [Dumas D., 2011]. This is especially true where the altitudes of the stations are much lower than the basin average altitude and an extrapolation from data range is necessary, as in this case. We already published similar results in the same study area, comparing results of the hydrogeological water budget coming from the use of the linear and exponential relationship between elevation and rainfall, respectively (Sappa et al., 2018). The results indicated that the first one is more reliable in this case, as confirmed by the comparison with measured average flow rates available at springs (Mazzoccolo and Capodacqua di Spigno Springs). In the following Figure, we reported the two relationship obtained in the present study.
As it is shown in the two plots, choosing the linear relationship a clear overestimation of the rainfall can be found. Just to provide an estimate of the error: these values (points)are obtained averaging a 20 years time series data, but cumulative annual rainfall of 4000 mm (as per linear relationship) have never been recorded at any rainfall station in Italy located at similar heights (900-1000 m asl) in the Central Apennine. Hence, we have added these two references in the text, explaining the reason why we choose this relationship:
Dumas D.; The impact of forests on the evolution of water resources in the mid-altitude Alps from the middle of the 19th century (Chartreuse massif, France); Journal of Alpine Research | Revue de géographie alpine, 2011.
Sappa G., Ferranti F., Iacurto S., and Filippi F. M. D., Effects of climate change on groundwater feeding the Mazzoccolo and Capodacqua di Spigno Springs (Central Italy): first quantitative assessments, 18th International Multidisciplinary Scientific GeoConference SGEM, 2018, Albena, Bulgaria, 219–226.
Pag 6 - The Effective Infiltration calculation is not described. The quantification of the lithological outcrop is not described, and and also it is not clear how the Effective infiltration map was made.
The Effective Infiltration calculation has been described, adding a table with the lithological outcrops and the potential infiltration factor (χs) values used for the calculation.
Pag 15 - There is clear need for a piezometric map, not only a literary description of the piezometric surface.
In Figure 2 has been added the piezometric surface.
Reviewer 2 Report
Comments and Suggestions for AuthorsThis manuscript presents a hydrogeological investigation into the complex interactions between urban underground infrastructure development and karst groundwater systems. Through a case study of the Mazzoccolo spring in Formia, the authors develop an innovative GIS-based methodology that integrates inverse hydrogeological water balance calculations with statistical forecasting models. This study contributes a framework for sustainable infrastructure development in sensitive geological settings. Specific comments are as follows.
(1) In Figure 4, the exponential relationship deviates from the typical linear rainfall-altitude correlation in hydrogeology. While the authors note mountainous basins may require non-linear models, no physical or statistical justification is provided for choosing this specific exponential form over alternatives. The high R² alone is insufficient without validation against independent data.
(2) The statistical discharge model uses 1974–1977 data but applies a higher amplification coefficient to 2018–2020 data without rigorous calibration.
(3) The conclusion states the tunnel will not intersect the saturated zone, but offers no hydraulic evidence to support this. Karst systems often have perched or discontinuous saturated zones.
(4) The abstract claims tunnel drainage may severely deplete the aquifer, but conclusions state impacts are avoidable with standard measures. No quantitative risk analysis bridges this gap.
(5) Table 2 lists "Depth to water table" for aquifers but leaves this column blank for springs. No explanation is given for this discrepancy.
Author Response
(1) In Figure 4, the exponential relationship deviates from the typical linear rainfall-altitude correlation in hydrogeology. While the authors note mountainous basins may require non-linear models, no physical or statistical justification is provided for choosing this specific exponential form over alternatives. The high R² alone is insufficient without validation against independent data.
For the rainfall-altitude and temperature-altitude relationships, the interpolation that is normally used is the linear one, although for annual cumulative rainfall, papers from literature proved that logarithmic or exponential relationships lead to obtain much more plausible values in the case of mountain basins [Dumas D., 2011]. This is especially true where the altitudes of the stations are much lower than the basin average altitude and an extrapolation from data range is necessary, as in this case. We already published similar results in the same study area, comparing results of the hydrogeological water budget coming from the use of the linear and exponential relationship between elevation and rainfall, respectively (Sappa et al., 2018). The results indicated that the first one is more reliable in this case, as confirmed by the comparison with measured average flow rates available at springs (Mazzoccolo and Capodacqua di Spigno Springs). In the following Figure, we reported the two relationship obtained in the present study.
As it is shown in the two plots, choosing the linear relationship a clear overestimation of the rainfall can be found. Just to provide an estimate of the error: these values (points)are obtained averaging a 20 years time series data, but cumulative annual rainfall of 4000 mm (as per linear relationship) have never been recorded at any rainfall station in Italy located at similar heights (900-1000 m asl) in the Central Apennine. Hence, we have added these two references in the text, explaining the reason why we choose this relationship:
Dumas D.; The impact of forests on the evolution of water resources in the mid-altitude Alps from the middle of the 19th century (Chartreuse massif, France); Journal of Alpine Research | Revue de géographie alpine, 2011.
Sappa G., Ferranti F., Iacurto S., and Filippi F. M. D., Effects of climate change on groundwater feeding the Mazzoccolo and Capodacqua di Spigno Springs (Central Italy): first quantitative assessments, 18th International Multidisciplinary Scientific GeoConference SGEM, 2018, Albena, Bulgaria, 219–226.
(2) The statistical discharge model uses 1974–1977 data but applies a higher amplification coefficient to 2018–2020 data without rigorous calibration.
The choice to change the amplification coefficient is mainly related to the catchment work enhancement, carried out after 1977, that that modified the flow regime and that we have cited in the manuscript. The procedure for estimating the new coefficient,
There are no flow measurements available other than the discharge exploited by the water utility (Acqualatina SpA), so we decided to proceed as follows. The new amplification coefficient was found setting an objective function that minimized the differences between the annual minimum spring flow and the exploited one, considering that, in the summer 2018-19, the shortage has led demand to exceed water availability, so that the entire flow of the spring has likely been captured.
(3) The conclusion states the tunnel will not intersect the saturated zone, but offers no hydraulic evidence to support this. Karst systems often have perched or discontinuous saturated zones.
(4) The abstract claims tunnel drainage may severely deplete the aquifer, but conclusions state impacts are avoidable with standard measures. No quantitative risk analysis bridges this gap.
The authors have revised the conclusion whit the aim of better explain the consideration about the interactions between the underground infrastructure development and Mazzoccolo karst groundwater systems.
(5) Table 2 lists "Depth to water table" for aquifers but leaves this column blank for springs. No explanation is given for this discrepancy.
In Table 3 (ex Tab. 2), “N/A” has been inserted in the column, as the "Depth to water table" parameter has no meaning in the case of springs.
Round 2
Reviewer 2 Report
Comments and Suggestions for AuthorsIt can be accepted.
