| 15 Apr 2026
Movement history of faults under variable stress fields – insights derived from 3D seismics in the central northern Upper Rhine Graben, Germany
Abstract. The tectonic history of the Upper Rhine Graben (URG) is still unclear, despite decades of research. This is because of the lack of good quality 3-D seismic data to investigate the subsurface. We reconstruct the tectonic evolution of the central northern Upper Rhine Graben (NURG) by analyzing a 3D seismic survey that covers approximately 300 km2 between Worms in Rhineland-Palatinate and Darmstadt in Hesse, Germany.
We aim to understand the interactions between the faults, the sedimentation pattern near the fault areas, their evolution in relation to the major structures of the URG, and to deduce the associated stress regime. Seven of the largest normal faults were selected (five strike NNW–SSE and two strike WNW–ESE), and the sediment thickness maps in each stratigraphic zone were generated using interpreted seismic horizons. Based on the thickness maps and the syn-sedimentary fault activity, the normalized Expansion Index (EI) was calculated. The analysis indicates that NNW–SSE fault activity was dominant pre-24 Ma, while the fault activity of WNW–ESE faults became more prominent post-24 Ma. Independent structural evidence, including the geometrical refraction of fault intersection traces (attributed to the reactivation of inherited Variscan basement trends) and the unilateral southward migration of displacement maxima, physically documents the rotation from E-W extension to NW-SE transtension. We postulate that all seven faults are active to the present day, as they can be traced seismically to the near surface. Recently acquired geophysical data (including shear-wave seismics) by our research group confirms recent activity on two of these faults.
We conclude, based on the seismic interpretation of our study area and the derived results, that URG underwent through two distinct tectonic phases: 1) a pre-24 Ma rifting stage and 2) a post-24 Ma transtensional stage that has lasted to the present-day. This finding aligns with other recent studies on the tectonic evolution of the URG. Rift systems evolving under variable stress regimes can develop complex fault architectures as changing stress orientations reactivate inherited structures and redistribute strain through time. Our results show that such stress-field changes can transform initially simple extensional fault systems into more segmented and interacting networks, where older rift faults and newly activated transfer structures jointly control fault propagation, basin subsidence, and sedimentation patterns.
Review of the manuscript “Movement history of faults under variable stress fields, insights derived from 3D seismics in the central northern Upper Rhine Graben, Germany”
By Sonu Roy et al.
The authors present results from a 3D seismic survey conducted in 2012 in the Rhine Graben. The initial data were recently made public and analyzed by the authors. This paper presents, interprets, and discusses these data and analyses. The Python codes used for the analysis (including calculations of the "expansion index" parameter—see below) are provided in a GitHub repository, along with the mapped faults and seismic horizons in another GitHub repository.
On the data and methodology
It is unfortunate that the authors are very sparse in describing the processing applied to the raw data: Who? When? How? A brief expansion on this would help readers familiar with seismic-reflection interpretation understand the scope of the work.
Another criticism regarding the data aspect is the lack of detail in the README files, which is supposed to describe the content of the GitHub repositories.
On the form
The text is well-written, and the structure is sound. However, some figures are too small. For example, the map illustrating faults at the UHY horizon should occupy at least half a page. Additionally, I believe it would have been wiser to present cross-sections perpendicular to the faults (NNW and WNW) to better illustrate their geometry and relationships to surrounding layers at depth. The choice of oblique sections is debatable.
One aspect that needs revision: In Section 3.2, the authors describe the subsidence history of the horizons from the most recent to the oldest in a completely counterintuitive manner.
The reference list and citations are coherent, except that there are duplicates in the reference list (e.g., Cardozo and Behrmann; Schumacher).
On the substance
The results are interesting and deserve to be highlighted. However, some interpretations of the results are, in their current form, debatable or require clarification. For this reason, I recommend major revisions.
Here are the main comments:
Introduction
The authors focus on distant geological times, particularly the late Miocene (5 Ma) for the most recent level analyzed, compared to the period of interest for "active tectonics" (Quaternary) and current seismicity. Therefore, the claim about the importance of this work for understanding hazards needs to be better justified, either in this section or later in the discussion.
The "Expansion Index" (EI)
The index calculated to estimate the active and syn-sedimentary nature of faults (EI, Expansion Index) does not, in my opinion, bear an appropriate name. It is more of a "relative subsidence index" than an "expansion" index, which would be preferable to use in the early stages of the work. It allows for the comparison of vertical elements. Moreover, the authors are convinced that this relative subsidence is associated, post-rifting, with oblique movements (along the faults). If the authors wish to retain this term, they must justify it with bibliographic references that already use it.
Section 3.3: Fault Throw Analysis
I do not share the authors' interpretation of the commented figures.
While I agree with the authors that there is a notable change in the slip pattern since the rifting phase (pre-24 Ma) and the subsequent phase (post-24 Ma), I estimate based on Figure 5 that the "southward migration of displacement maxima in post-rift sequences" is an unfounded claim. In Figures 5a to 5c, the peaks are stable in their position for the green to orange markers (CBS to JT1 top). In Figures 5f and 5g, the migrations of the two bumps on each profile change direction between pre- and post-rift, but not uniformly southward, as the authors suggest. This point needs to be clarified, discussed, and conclusions potentially revised.
However, I believe the authors overlook an interesting point: There is an evolution between the pre- and post-rift phases. The complex displacement pattern (camel-back shape for 1A, 2B, 1D) decreases over time, eventually forming only a bell-shaped pattern.
Section 4: Discussion
This section needs revision if my interpretation of Figures 5 and 6 is correct. I do not question the existence of a tectonic change around 24 Ma, which is documented in the literature and supported by the data presented in this work. However, I am not convinced by the authors' arguments (correlation between subsidence rates on the two fault families and changes in tectonic stresses).
4.1 Fault Propagation
Even if the authors' conclusion is sound, they must clearly explain why a potential southward migration of displacement (which is not demonstrated for 1A and 1B!) would indicate a shift from an extensional regime to a transtensional one.
4.2 Fault Interaction
According to the authors, the WNW transverse faults become truly active after CBS (post-rift phase). How, then, do they explain why 2A no longer impacts 1D (Figure 5f)? How do they explain Figure 5f, which shows an EI profile almost aligned with the value of 1?
The authors state that the change in tectonic regime after CBS (post-rifting phase) alters the activity of the transverse faults 2A and 2B, which then show stronger subsidence. However, in Figures 6m and 6n, this is not clearly reflected in an increase in the "normalized EI," especially for 2A.
4.3.1
The authors declare that the maximum horizontal stress field was oriented NNE-SSW during the rifting phase. What is the argument in their data, or which reference do they refer to?
4.3.2
It would be helpful to remind readers of the justification for the stress rotation proposed by Schumacher. As I understand it, this is based on a comparison between subsidence zone geometries (data provided by this author, Schumacher) and old microtectonic data (Villemin and Bergerat, 1987). While the latter data are undisputed, it should be noted that the polyphase interpretations at the time were not chronologically constrained.
The authors indicate a zig-zag structural pattern for 1E at the intersection with 2B during the CBS period as a marker of this stress state change. Cartographically, the zig-zag I see is on 1D at the intersection with 2A. Can the authors confirm their statement and explain how this is a marker of the stress state change?
Regarding the discussion of the evolution of 1E's geometry over time and depth, I suggest the authors illustrate their argument with an appropriate seismic section.
Lines 309–312: It is difficult to reconcile the "vertical displacement acceleration since the Aquitanian (UHY)" with what was described earlier. Which faults are being referred to in particular? Are these the ones described in Mair et al. (2025), which show 35 m of throw in the Quaternary?
4.4
This point should, in my opinion, be addressed first in the discussion.
4.5
I am not a specialist in "basin analysis and tectonics (at long-term scales)," so I cannot comment on whether the area is emblematic of oblique rifts. I wonder whether the bibliography is up to date.
4.6
The authors could refer to active fault systems where earthquake clusters migrate over time from one segment to another within the same fault system, such as in the Central Apennines. This could help explain such apparent discrepancies.
Additionally, regarding the recent (Quaternary) activity of the faults on the eastern border of the URG, the authors could refer to recent observations by Peña-Castellnou et al., where a horizontal slip rate was derived.
Conclusion
This section provides a good summary of the paper's findings. However, I highlight two specific points:
First, the authors begin this section with a major theme that has barely been developed earlier: the decoupling of deformation on either side of the Rupelian (RT). The paper presents results and analyses that largely focus on what happens after the Rupelian, during the Late Oligocene and Miocene. This point is not clearly addressed before Section 4.4 in the discussion, nor is it mentioned in Section 3.2, which deals with the tectonic control of sedimentation. I suggest this theme be further developed in the core of the paper and placed in its proper context, as already suggested above.
Second, the authors do not provide a clear perspective on the methods that could be deployed to obtain observables of the strike-slip component of the faults described in the basin or those in its vicinity. For example, new microtectonic studies coupled with U-Pb dating of syn-tectonic calcite?