GENERAL COMMENT

The manuscript attempts to relate the creation and   evolution of the proterozoic Bijawar and Vindhyan sedimentary basins of India to the plume-related rifting mechanism and the associated mafic subcrustal underplate emplaced at a depth of greater than 30 km. The authors used the global gravity grid data to show the spatial and depth extent of the high density mafic underplate in the two forward models they prepared.  The topic is important and interesting and the manuscript is well written. However, there are points that need addressing in order to improve the quality of the manuscript and to make it relevant and make it suitable for publication in the form of a research paper.

SPECIFIC COMMENTS

1. The title: Basin formation and evolution is attributed to the combined effects of all the sublithospheric actions including underplating as well as the lithospheric plate actions and the supergene action of the lithosphere itself as expressed in all the processes that take place on the earth’s surface. The title, as it is and at a glance, seems to carry the idea that underplating alone can play a significant role in basin formation.
2. The authors have not shown the locations of the two basins. What they show on the map is the lithostratigraphic units bearing the names of the two basins.
3. What is the rationale for upward continuing the gravity data to a 30 km height? It is more reasonable to base the continuation height on the corresponding depth estimates obtained from the radially averaged power spectrum.
4. Gravity modeling is loosely constrained with only limited information. Hence the modeling result should be interpreted with caution.
5. To generalize that the high density anomaly has sources extending from deep to shallow is an over simplification. The authors have done upward continuation to a single height and they base their conclusion on it. Try to upward continue to a height of 60 (corresponding to sources at a depth of 30 km and below) to justify presence of underplate at depths in the order of 30 km in the models. In addition, the central part of the models (central region) do not show the extension of high anomaly sources to shallower depths as shown by the residual anomaly.
6. On Line 332 the paper asserts that there is striking similarity between the inverted Moho topography and the gravity signature. Please tone down this assertion. There is similarity but pay attention to the following remarks: a. the gravity anomaly from the inverted Moho topography shows a slightly southward shift and centered at the southern margin of the regional anomaly. b. at the southwestern corner the effect is in fact opposite. There is high anomaly in the both the regional and residual gravity but low anomaly in the gravity obtained from the Moho.
7. I have serious reservation on the contradictory results obtained from the two approaches of the gravity modeling. The authors obtained the undulation of the Moho interface using the downward continuation formula as given by the modification of Parker-Oldenburg algorithm (Fig. 5a). As is also noted by the authors the Moho depth undulations obtained by this method beneath the basins where underplate is observed do not correspond to the depths obtained by forward modeling (Fig. 6 & 7). The underplate top is considered as Moho in Fig. 5a whereas the underplate bottom is considered as Moho in both Fig. 6 & 7. In addition, there is also a clear discrepancy in areas where there is no underplate. Compare, for example, Fig. 5a with Fig. 7. In northeast where there is no underplate the Moho depth varies between 37 and 39 km in the forward model in Fig. 7 whereas in Fig. 5a the depth variation for the same area is between 33 and 44 km. These contradictions should not have occurred since there is a clear density contrast between the mantle and the underplate and the Parker-Oldenburg approach is capable of recognizing the difference between them.

TECHNICAL CORRECTIONS

1. On Line 29 “…..attributed to the presence of underplated layers, like rift basins…” Please remove the word “like”.
2. On Lines 194 and 197- 8 the phrase h(x) and z_o are redundant. Please eliminate the repetition.
3. On Line 346 it must be “crustal uplift” and not “crustal upliftment”.
4. In Fig. 5   contour values can be indicated on every other line to avoid congestion.

egusphere-2023-1389 Manuscript review

In the manuscript “Magmatic underplating with Proterozoic basin formation: insights from gravity study over the southern margin of Bundelkhand craton, India”, the authors investigate the tectonic mechanisms leading to the formation of two basins: the Bijawar and Vindhyan basins, located in proximity of the southern boundary of the Bundelkhand craton. 

The authors suggest an extensional mechanism favored by magmatic underplating below the crust at relatively shallow depth (ca. 30 km), based on the analysis and modeling of global gravity data, and Bouguer gravity anomaly in particular. 

The scientific question concerning the origin of the basins is relevant and of interest for the scientific community, and this is well explained in the manuscript Introduction. In general, the chosen methods are suitable for the proposed investigation and the manuscript is well written. 

However, I think that some methodological choices require better clarification and support, to strengthen the results interpretation. Also, the quality of some of the figures requires improvement. 

Here I provide some important points, which I believe should be addressed prior to considering the manuscript for final publication:

1. The authors use global gravity data and upward continue Bouguer gravity anomaly to investigate the crustal structure within the study area. I understand that upward continuing acts as a low pass filter to enhance the long-wavelength regional Bouguer anomaly trend.
   
   
   
   Please, can the authors explain why they chose 30 km elevation for the upward continuation? Is 30 km the elevation at which the effect of surface structures become negligible?
   
   
   
   It would be interesting to provide a few test examples of upward continuation at different elevations (possibly as supplementary material); also, see e.g. Zeng et al. (2007), Geophysics, for approaches to estimate ideal elevation for upward continuation.
   
   
   
   Also, the terrain correction seems to be small (less than 1 mGal), but topography ranges from 175 m to 617 m (line 152). Can the authors provide a topographic map of the study area, together with the grid points from the global gravity data they used?
   
   
   
   
2. The authors extract the regional Bouguer anomaly trend and separate it from the gravity effect of surface geological structures (residuals). However, they often refer to “Bouguer gravity anomaly” throughout the manuscript.
   
   
   
   Please, state explicitly which type of gravity data product is used for RAPS analysis, 3D Moho depth inversion, and 2D forward gravity modeling respectively. Is the upward-continued regional Bouguer trend used in every analysis? If yes, can it constrain the shallower (hence smaller wavelength) structures in the 2D forward modeling (e.g. basin structures at a few km depth below the surface in Figures 6 and 7)?
   
   
   
   
3. Could the authors spend a few more words on the RAPS technique? The output provides a deep density contrast interface at ca. 30.3 km depth. What is the resolution expected from this technique, which, if I understand well, provides a 1D average information across the study area?
   
   
   
   
4. The authors perform a 3D inversion for the Moho topography. The algorithm they use requires assuming a mean depth z0 and a density contrast. Why did they use 36 km for z0? And not e.g. 30 km as obtained from the previous RAPS analysis? And related to this, how do these results change as a function of z0 and density contrast?
   
   Please, provide a sensitivity test for these parameters, and a resolution test for the inversion algorithm application.
   
   
   
   
5. I believe a few important points should be clarified in the 2D modeling stage. The authors perform 2D forward modeling based on the formulas by Talwani et al. (1959). These formulas are best suited when both the data, and the target area, present a 2.5D symmetry (e.g. changing properties along the x-z plane, and no changes along y-axis); see e.g. Scarponi et al. (2021), Frontiers for one example. In particular, profile AA’ seems not to be perpendicular to visible 2.5D structures, nor in the gravity data or in the underlying, inverted Moho structure (based on Figure 2 and Figure 5a). A slight-to-moderate rotation of profile AA’ around its center (e.g. +- 20 degrees) could potentially provide a different gravity data profile, and hence lead to different results and interpretation.
   
   
   
   a) The authors could consider using a 3D inversion software (see e.g. IGMAS+ in Spooner et al. 2019, Solid Earth). If not 3D, how would the 2D gravity profiles (data and models) look like along a set of parallel profiles (e.g. at constant longitude)? Would the results, and hence interpretation, change along a different profile than AA’? This should be tested and discussed before interpretation.
   
   
   
   b) Paragraph 3.5 on the construction of the 2D profiles never mentions incorporating the results from the 3D Moho depth inversion. Were these Moho results neglected in the creation of the 2D models shown in Figure 6 and 7? If yes, why?
   
   
   
   The authors mention the RAPS estimate as reference used in the profiles, but RAPS provides an inherently 1D average information. Moreover, profile BB’ shows no interfaces around 30 km: please, can the authors explain the reason for this?
   
   According to Figure 5a, the computed Moho depths obtained along BB’ range from 44 km to 34 km depth, but this seems not to be the case when looking at Figure 7. This should be clarified (partially applies also to profile AA’ and Figure 6).
   
   
   
   c) In the definition of the structures within profiles AA’ and BB’, the authors refer to a list of previous investigations, to be used as external constraints. This is OK in principle. However, these external constraints are not explicitly indicated in Figure 6 and 7. Which geometries were imported as unmodified external information? Which ones were generated and/or modified by the authors?
   
   This information is not clear and should be made explicit. The authors could also show in Figures 6 and 7, how their new Moho estimate compares to the external information they refer to.
   
   
   
   Clarifying the points above is important to discuss the fit to the gravity data along the selected profiles. For example, was the geometry of the underplating structure in Figure 7 imported from external sources? How would the forward modeling compare to the data, with a different, or without, the underplating layer along AA’?

To address these points, I would advise starting by: 1) apply RAPS; 2) use the RAPS deepest interface estimate as z0 for the Moho inversion (provide sensitivity and resolution tests); 3) use the obtained Moho depth as starting geometrical constraint, together with those existing in the literature, either for 3D modeling, or for a set of 2D profiles (as much as possible along structures with 2.5D symmetry), providing support for the chosen 2D profiles; 4) test if the deeper underplating layers can be resolved by gravity along the chosen profiles.

Here I provide few additional specific comments:

• Figure 1b is not readable and should be larger. Figure 1a is readable, but please consider using different colors to highlight the different geological units.
  
  The perimeter box of figure 1a should appear in figure 1b to show its location;
  
  
  
  
• Figure 5 should at least contain a residual map (synthetics minus observations).
  
  It would be also beneficial to add a plot for RMS versus iteration number, to show the RMS reduction during the inversion, and a sensitivity and resolution tests (possibly in a different Figure);
  
  
  
  
• Figure 6 and 7 should show explicitly which geometries were imported as unmodified external constraints for the construction of the models. They should also show the Moho depth as obtained from the 3D inversion (Figure 5a).
  
  Also, the top banner in Figures 6 and 7 is not very clear: is it gravity or Bouguer anomaly?
  
  Please, plot the error on a separate independent scale to be more readable.
  
  
  
  
• line 175: by “Bouguer anomaly” you mean the upward-continued regional trend? Please, specify. Same for line 219, 239, 270, 286, 295 and so on.
  
  
  
  
• Line 244: Does GMSYS perform 2D forward modeling or 2D inversion?
  
  
  
  This is a crucial detail.
  
  If it performs inversion, then more information is needed here.
  
  Or, have you tested several candidate profiles? Please, explain.
  
  
  
  
• Line 375-377: To my understanding, you obtain a Moho from 3D gravity inversion.
  
  However, you do NOT obtain a Moho depth from 2D forward gravity modeling (see also question above).
  
  If you do not perform 2D inversion, then 2D forward modeling can only validate a certain profile, but not “provide” or “obtain” one from it.
  
  This is better formulated later in the conclusions, at line 442 “[...] validated by the 2D [...]”.
  
  Please, This should be clarified. And finally, why not using the 3D Moho results in the construction of the 2D models?