the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 27 Dec 2025
South East France Potentially Active Faults: A database for seismic hazard assessment
Abstract. The South-East of France has been structured by numerous tectonic episodes resulting in a complex compound of crustal deformations. This fallouts in a large network of faults which have evolved throughout subsequent tectonic events. Nevertheless, the present-day tectonic deformation is slow and of debated origin, giving place to a low to moderate seismicity. This seismotectonic context is a challenge for seismic hazard analysis, specifically for the identification and characterisation of potentially active faults. In this study we present a fault-related data compilation, named South-East France Potentially Active Faults (SEFPAF, Mowbray et al., 2025, https://doi.org/10.5281/zenodo.17235395), which integrates structural components from geological maps, previous neotectonic databases and fault-specific studies. Our objective is to provide a new, well documented, and modular fault database. Multiple structural representations per fault are presented, allowing the end-user to choose from the various parameter value options. With an end purpose of identifying the most relevant faults for seismic hazard assessment, we build a series of indices which allow fault prioritization: the Importance index (I), the Documentation index (D) and, the Seismogenic Potential index (SP). The first classifies faults into major, secondary and minor based on the source's description of the structure; the second analyses the quantity of information known for each fault; the last intends to decipher seismically active faults by integrating the age of the last documented rupture, the spatial correlation with seismic flux and, the spatial correlation with geodetic strain. SEFPAF may have various applications, amongst them we illustrate the first step towards the development of a fault model for seismic hazard analysis: we identify relevant faults to model by adjusting the index criteria and, propose idealized geometries for a set of 20 faults which can subsequently be used as composite seismogenic sources.
- Preprint
(1633 KB) - Metadata XML
- BibTeX
- EndNote
Status: open (until 04 Mar 2026)
- CC1: 'Comment on essd-2025-601', André Burnol, 12 Jan 2026 reply
-
RC1: 'Comment on essd-2025-601', Octavi Gomez-Novell, 26 Feb 2026
reply
This article presents a database of potentially active faults of southeastern France, which constitutes an excellent contribution to the scientific community working on active tectonics and seismic hazard applications in France and also in Europe. The database and accompanying article are clearly explained, decisions are well justified and exhaustive, and data appears to be traceable. The article is well-written, properly illustrated and straight to the point, which is an added value in my opinion. I would also like to express my appreciation to the authors for the significant amount of time and effort necessary to review, compile and homogenize the data into a database with this a level of detail.
I have only a few minor comments that I would appreciate if the authors could clarify:
General comments
1.- The paper mentions that the database considers a fault as potentially active if at least one of three criteria are met: i) it is located near significant observed seismicity; ii) it lies within an area undergoing above-average crustal deformation; iii) there is evidence of Quaternary activity.
First of all, I was expecting the authors to frame better the database within the landscape of existing fault databases, at least in Europe. How is SEFPAF positioned along with similar databases in other countries (e.g., QAFI in Spain, ITHACA in Italy)? Are the criteria used for the definition of a potentially active fault similar to those other databases? If not, why? This point could be particularly facilitating for researchers who might need to use several databases at the same time for a large-scale application - e.g., fault-based PSHA applications at the European scale.
Second, the Quaternary activity criterion seems to be vaguely applied in the database, because the Neotectonic sub-index allows for the inclusion of faults with a pre-Quaternary last activity. I am aware that the neotectonic evidence is complemented with seismicity and geodetic deformation evidence via the SP index, but I think the decision of including pre-Quaternary fault evidence requires a better justification in the text. This is important because causative links between specific faults and regional seismicity are challenging, and most notably when faults have no Quaternary activity evidence.
2.- The database structure and design are correct, and reflect the data contained easily. However, I think some minor improvements could be made in field standardization that would enhance user experience and facilitate automatic data extraction tasks. For instance, I generally would recommend using numerical IDs in favor (or at least accompanying) of text-based fields to describe categories because they facilitate data extraction and analyses in most algorithms. If text-based categories are necessary, I also would recommend removing the spaces, commas or dots present in string-based fields, as they can also difficult such tasks. Instead, I would choose a separator (e.g., underscore), and use it consistently throughout the whole database. Finally, values that represent a numerical range, like slip rates, would be more optimally provided as separate columns, such as minimum/maximum or mean/standard deviation. Otherwise, these values will be more likely read as strings rather than numbers, which can be somewhat inconvenient.
I am aware that these changes can be difficult to implement at this stage, but I encourage the authors to implement them in further releases. I think these can help enhance the internal consistency of the database.
Minor comments
- Line 157: The statement is unsupported. Consider adding references backing it up.
- Line 168: The 2-km buffer selection for Neotectonic evidence inclusion requires justification. What criterion was followed to reach this value?
- The importance index value ordering looks a bit counterintuitive to represent relevance of a structure. A “1” for major faults might be more intuitively related to higher position (i.e., more importance) than a “3”. Also, the paper mentions that this value is weighted depending on the hierarchical order. What does this mean exactly? Does it mean that the importance index selected corresponds to the input data types with the highest hierarchy? I suggest clarifying this point in the paper.
- Line 447: This line mentions that fid values range from 1 to 2649, but in the database, these range from 0 to 2648. I guess this has to do with Python indexing during the database compilation, but I suggest fixing it.
- In section 5, a subset of the database containing the faults with better documentation and highest potential activity is shown for PSHA purposes. This is an interesting by-product of the database and showcases its potential. I think it would be interesting and useful to provide this “PSHA-ready” dataset in the Zenodo repository, if possible.
Best regards,
Octavi Gómez Novell
Citation: https://doi.org/10.5194/essd-2025-601-RC1
Data sets
South East France Potentially Active Faults Victoria Mowbray et al. https://doi.org/10.5281/zenodo.17235395
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 265 | 162 | 21 | 448 | 22 | 20 |
- HTML: 265
- PDF: 162
- XML: 21
- Total: 448
- BibTeX: 22
- EndNote: 20
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
General comments:
This manuscript provides detailed development of a new database of potentially active faults in South-east France, called South-East France Potentially Active Faults (SEFPAF), for seismic hazard assessment.
Significance: The newly developed database has a broad potential utility and is a long-awaited improvement to the “BDFA” (Base de Données de Failles Actives) from Jomard et al. (2017), which aims to map active faults in the 50 km vicinity of nuclear facilities within metropolitan France. built with a different purpose and with a different geographical extension.
Presentation: the manuscript has sections that read more like a user manual than a journal publication. I feel that this is to some extent inevitable, as the purpose of the paper is to develop a new database, however I suggest new Appendices to facilitate the reading.
Specific comment on the La Rouvière Fault (Thomasset et al., 2024) :
As described in the supplement image, the location of the La Rouvière fault (in red) of SEFPAF is not consistent with the cited publication of Thomasset et al. (2024). This location is also not consistent with surface ruptures indices from Ritz et al. ( 2020) observed just after the Le Teil earthquake (Mw 4.9, 2019/11/11, France).
I suggest to add in SEFPAF database the secondary surface-rupturing Bayne-Rochrenard fault described in the publication of Burnol et al. (2023) in MDPI Remote sensing.