Lithospheric Equilibrium, Environmental Changes, and Potential Induced-Earthquake Risk around the Newly Impounded Baihetan Reservoir, China
Round 1
Reviewer 1 Report
This study proposes a method to assess the earthquake risk near the Baihetan dam, which is the second largest reservoir in the world. The authors estimate an isostatic additional force (IAF) from the dense gravity anomaly data and the Coulomb stress caused by a large-scale impoundment. The authors point out that, when the impoundment generating a downward force occurs where the IAF is downward, the superimposed forces can increase seismicity (Fig. 5 (b)).
The employed computational methods are sound and they successfully demonstrate the importance of those two factors to a certain extent, although the other factors could also affect seismicity. The interpretation based on the above isostatic force balance makes a contrast with previous studies regarding reservoir-induced earthquakes which assume fluid injection into the faults. In addition, the observation data associated with such a large-scale impoundment provide us with a rare opportunity for the study of large-scale surface loading. Considering such originality, I would like to recommend publication, after my questions and suggestions below are addressed.
Major points: (about (2) of the abstract)
A) The authors argue that the superposition of the IAF and the impoundment affects seismicity dominantly. I think that assumptions behind this idea (i.e., in which conditions this idea holds) should be described more clearly. For example, the deviatoric stress is smaller than the IAF, the rate for the tectonic stress change is constant in time etc. (these are just examples and I did not confirm whether these are true).
B) l.378-382 “Finally, the blue area around Qiaojia City is the primary area of water storage where the reservoir is relatively wide. At such a place, the IAF is not large. In other words, the area with the largest water injection in the reservoir was not particularly large in the IAF. Such a distribution indicates a lower risk of a strong induced earthquake in the Baihetan area.”
Fig. 1 shows that earthquakes with Ms>5 (red circles) occurred in Qiaojia and its southern area along the fault F1, including the blue area. So, large earthquakes have already occurred where the IAF is smaller, suggesting that the other mechanisms than the IAF are dominant. On the other hand, less large earthquakes seem to have occurred in the northern area with the negative peak of the IAF (around 80 km in Fig. 5 (c)). What causes this difference in the seismicity between these two areas? Moreover, is there an evidence showing that the stress is currently accumulating in the northern area such as GNSS displacement rate data and seismicity of smaller earthquakes?
Minor points:
The units of stress should be unified throughout the manuscript (MPa or bar).
l. 20 “improved the accuracy of the gravity anomaly field”:
“improved the spatial resolution of the gravity anomaly field” may be better.
l. 32 Is “the dam height of the reservoir” the height at the bottom of the reservoir?
l. 45 “There is a possibility that an Ms 6.8-7.1 earthquake occurred in the past (when? where?)”
l. 69 Where is the Xiludou Reservoir? I could not find in Fig. 1.
l. 125 Where is the famous Yangzi River?
l. 130 “The topography of the study area is very complex, which increases the difficulty of observation and data processing.” How about showing the contour with a color scale in Fig. 1? It will help interpretation of the free-air anomaly in Fig. 2.
l. 170 “The observation error of the GNSS data for elevation is estimated to be at the semi decimeter level”: “ellipsoidal height” may be better than “elevation”
l. 181 How about adding a figure showing the variogram inferred in the Krigging? It will be useful to understand the quality of the data.
l. 184 “The results far from the observation stations (>5 km) have lower accuracy and are presented in the contours”: This should be added in the caption of Fig. 2.
l. 199 “the accuracy of the EIGEN-6C4 gravity model was not high in the study area”: A figure showing the EIGEN-6C4 model should be added in Fig. 2 so that readers can understand the spatial resolution of the model.
Furthermore, comparing Fig. 2(a) with (c), it seems that the EIGEN-6C4 model can have a bias (or a longer-wavelength discrepancy). What can be the cause of the bias?
If the gravity anomalies (a), (b) and (c) are measured relative to a reference site, its location should be mentioned in the caption.
l. 208 “The crust is in an ideal hydrostatic equilibrium state if there is no external force on it” This is needless to say, but except for gravity.
l. 243- h0, h1, h2 kg/m3: super and subscripts should be used as in eq. (1).
l. 248 rho_t is the (average) density of the material above sea level.
Section 7: How large is the uncertainty for the densities obtained by the inversion (Fig. 6)?
l. 487 It might be better to mention that delta CFS>0 promotes earthquake for clarity.
Author Response
Please see the attachment.
Author Response File: 
Author Response.doc
Reviewer 2 Report
The paper by Fu et al. is about the changes in seismicity when a reservoir in China was filled with water. I have made comments in the pdf file. Generally, I'm completely puzzled how the authors claim that the IAF explains seismicity with water dams. The IAF is related to a long-term loading scenario (several 10s of million of years) while a water reservoir is a very short-term loading (several years to 10s of years only). In addition, the estimation of the Moho is completely wrong and is instead an estimation of an isostatic surface. The authors also completely ignored the fact the background stress regime is important to consider when analyzing load changes and seismicity.
Comments for author File: 
Comments.pdf
Author Response
Please see the attachment.
Author Response File: Author Response.doc
Reviewer 3 Report
The authors of the manuscript “Lithospheric Equilibrium, Environmental Changes, and Potential Induced-Earthquake Risk around the Newly Impounded Baihetan Reservoir, China” have collected new gravity and GNSS data to study the area of the Baihetan hydropower station. The aim of the work was the evaluation of possible induces-earthquakes. The authors invert the gravity data to obtain a density structure of the lithosphere and to use this model to determine the stress changes the crust due to water storage and the Coulomb stress changes on the main faults in the area.
The authors introduce the concept of isostatic additional force (IAF). They refer to their previous works, but I not sure to really understand all the steps in the computation. In equations (1) and (2) the authors define the depth of the Moho and of the isostatic surface, however in equation (2) the terms related to the reference point are not clear. In my opinion this part of the manuscript needs a revision to better specify how the Moho and isostatic surfaces are defined.
The heights associated at the gravity points are obtained by GPS measurements, so they are ellipsoidal heights and not orthometric heights, although in the computation the authors refer to the sea level (figure 4, for instance). What is the effects of this approximation? Which is the order of magnitude of the geoid undulation in this area?
At line 197 I don’t agree with the comment about EIGEN-6C4. I suppose that you have computed the gravity values using the maximum number of degree and order, but you should specify this information. The differences are related to the high frequencies of the gravity field, due for instance at the topography.
Minor remarks:
- In the abstract: gravity anomaly field —> gravity field
- In Figure 1 I can not see any blue triangles (continuous gravity stations), described in the figure caption
- Line 145: “to and fro”, do you mean “to and from”?
- Line 252 and 253: kg/m3 (superscript symbol)
- Line 322-323: why the sentences are in quotes “ “?
Author Response
Please see the attachment.
Author Response File: Author Response.doc
Round 2
Reviewer 2 Report
Thanks for including more details.Reviewer 3 Report
Thanks to the authors for considering the suggestions and for answering reviewers' questions. I only have to report a correction: sometimes the authors write “Ariy” instead of “Airy”
This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.
