|Dear editor and authors,|
I have received manuscript entitled “3D Inverse modeling of EM-LIN data for the exploration of coastal sinkholes in Quintana Roo Mexico” by Ochoa-Tinajero et al., submitted to NHESS journal and with reference nhess-2018-180.
Manuscript represents the result of a previous manuscript review and it includes a reviewer letter answer to reviewer comments. First of all, I have read new manuscript version, and after this I have read the authors answer to previous reviewer comments.
Attending to the new lecture, I have found some subjects that require some improvements and required to obtain a general scientific structure to the manuscript.
Abstract describes the eventual problematic in the study area about the hidrogeological characteristics of the Yucatán peninsula, however, the manuscript does not evaluate this problematic and has as objective a geophysical method that can give information about, but it is not explained how this information can solve the described problematic. This means that the main objective of the manuscript and description included in the abstract is not the own objective of the manuscript itself. The obtained information in the manuscript can collaborate, with the joint analysis of other techniques and methodologies, at the evaluation of the contamination of aquifers in the area, however this subject is not developed in the manuscript. In this sense, abstract does not describe what the manuscript evaluate, and differs from the obtained conclusions obtained in the manuscript, that it is centered in the geophysical data.
There are sentences that require rephrasing, eg. “In order to guarantee the sustainable use of this groundwater resource knowledge on the hydrogeological characteristics, such as geometry and position, of caverns and sinkholes and the depth of the freshwater/saltwater mixing zone (halocline) is needed.”; or “These authors performed EM-34 measurements but they did no further processing, like perform a geophysical inversion.”
Authors indicate that they do not show the results from the 2D inversion, something that can be of interest when a new processing approach is presented. However, later when authors explain their 3D approach, they are exhibiting 2D sections (or its integration that can be referred as a 2.5D distribution, see figs. 4 and 5 for example). In this sense, I am not sure about if the 3d analysis described represents a 2D analysis with a perspective view. In data inversion it should be of interest to include all data, it could be that the carried out inversion makes reference to the 2D inversion, because if a real 3D exists, they should present the map view of different depths, or geophysical bodies in 3D fashion.
The use of roof for the thickness of the carbonatic unit over the cavity induces to mistakes, as roof is a surface (later I will enter in this subject that was also highlighted in my previous review).
In the case of figure 5, I do not find a significative signal that can be related to the underground flow in the area or related to cavities in the underground, in this sense a higher detail and resolution of figures is required in order to evaluate the real meaning of these changes and the potential availability to identify them from the rest of the area (this can be done in a discussion chapter in the manuscript).
About other terminological issues, I am not sure if the term “underground rivers” is the most recommended due the characteristics from the area, I mean, that the use of underground rivers can mean many things. Here I suppose that this term make reference to a preferential water flow path through the carbonates by grouts, fractures and caves. This means that there could exist an underground conduit or group of conduits where water flow, more than an underground river itself.
In summary (mainly referring to the conclusions chapter) one of the main highlights from the article is to apply a previous mathematical method for the inversion of 2D profiles, in this case, with arbitrary directions. This is carried out on contrary than the usual parallel and perpendicular profiles net survey. The idea to perform the inversion by homogeneous directions is to simplify the inversion method and to evaluate, in a systematical pattern, the pixel size for the inversion (element or objet defined for the model). I am not sure if the only change in the orientation of the profiles can be performed by the orientation change of the data coordinates In this case what happens is that the data distribution for the inversion does not represent a homogeneous map view distribution, producing that this requires to be analyzed in detail in the discussion chapter.
I believe that if we want to perform a new approach for data inversion modifying previous usual approaches, it requires comparing, for example, the results of the inversion by means usual and new data distribution. In this case, it has been applied a new mathematical approach without comparing if this results can be of application for the studied case. In this sense, the most recommended approach should be to perform the survey in two different survey grids and to compare results. In the case that the study area does not permit this approach, the mathematical simulation can be an alternative, at least, to evaluate if the approach can be of application in the area. The conclusions, where a integration of data and its application should be included, as stated in the abstract, is not evaluated or developed.
There it not a discussion chapter, moreover when geophysical results are ambiguous in terms of changes of the properties of the indirect obtained results, and moreover the direct data from the area does not present a clear correspondence with the models, the lack of discussion decrease the interest of the manuscript.
In this sense, I am not sure if the geophysical model represent a good approach for the carried out analysis, at least looking at the presented results, when at the same time a change in the processing methodology is presented without to compare with previous methodologies, the results are not unambiguous, the comparison with the known data are not straightforward (or they are not interpreted and discussed in terms of resolution and accuracy).
Attending to the document of answers to previous review, there are some subjects highlighted in the previous review, that has been answered but not included in the manuscript or the recommendations have not been considered. In this sense, some of the suggestions related to previous data in the area that can permit to compare geophysics with direct data, the inclusion of a geological or geomorphological map, the case that I recommend to include in a map with the surficial information as authors indicated the presence of sinkholes but they have not been included and they have answered that there are not surficial evidences (then, what authors define as sinkhole?), nevertheless they have not been included. In terms of the evaluation of the obtained values for the different underground units in terms of resistivity, authors include in the reviewer answer a plot that I consider should be included in the manuscript.
About other subjects in relation to the use of roof for the unit over the cavity, authors indicate that they have corrected it, while in the manuscript authors still use this term for the description.