the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Cross-scale strain analysis in the Afar rift (East Africa) from automatic fault mapping and geodesy
Abstract. The formation of continental rift systems is characterised by the interplay of magmatic and tectonic processes. Their evolution involves a wide range of time scales, from centennial scales of the seismic and diking cycles to strain localisation during millions of years of continental thinning. Our understanding of rift processes at different spatial and time scales is limited by relatively short temporal coverages of geophysical measurements and by spatially discontinuous geological datasets. Here we propose a novel method for the automatic extraction of faults and the calculation of time-averaged strains distributions using topographic information from Digital Elevation Models. We apply this method to map ~4000 individual faults within a ~70 thousand km2 area of the Afar rift (East Africa), where the Nubian, Arabian and Somalian plates diverge. By comparing our results to rock dating and recent decadal geodetic measurements we deduce the rift's deformation history since 4.5 Ma and study its relationship with the current tectonic and magmatic activity. We show that the external portions of the Central Afar rift are not the mail locus of strain and rifting processes have migrated toward the axis where magma emplacement focusses strain rates due to the mechanical and thermal weakening of the crust. Increasing strains toward north-west suggest a progressive migration of the rifting process in the same direction. Conversely, Southern Afar is characterized by two systems of cross-cutting faults that respond to different strain regimes driven by the separations of the Arabian and Somalian plates from Nubia. This study demonstrates the effectiveness of our new method in quantifying fault activity and strain distribution in extensional settings and provides new insights into the spatial and temporal evolution of rifting in Afar.
- Preprint
(3072 KB) - Metadata XML
-
Supplement
(3064 KB) - BibTeX
- EndNote
Status: closed
-
RC1: 'Comment on egusphere-2025-1215', Valentin Rime, 20 May 2025
General comments
Overall, the manuscript is of very high quality. The methodology is novel, seems robust, and could be applied to many other study cases. The results from the Afar region are globally relevant and allow to better understand the late phase of rifting. One of the main interests of the paper is that it links different scales of time, liking processes happening over millions of years with processes happening over decades or even shorter. The paper is very well written and has a very clear and coherent structure. The figures and supplementary figures look good and are clear. Except for one aspect to be discussed, the results support the conclusions.
I mainly have minor comments. Some uncertainties of the methodology could be discussed in more details, even though they won’t change the conclusions. I also wonder why you didn’t calculate more geological strain rates from existing datings. This parameter is very interesting and allows a direct comparison with Recent geodetic data. Finally, I question one point of the conclusion regarding Manda Inakir which should be discussed and argued before being presented in the conclusion, or be toned down.
I also made several suggestions that do not relate to scientific quality, but to style or clarity, which is subjective and personal. They are just suggestions, don’t hesitate to not follow them without justification.
Please find my detailed comments in the document attached hereby.
-
AC1: 'Reply on RC1', Alessandro La Rosa, 24 Jun 2025
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-1215/egusphere-2025-1215-AC1-supplement.pdf
-
AC1: 'Reply on RC1', Alessandro La Rosa, 24 Jun 2025
-
RC2: 'Comment on egusphere-2025-1215', Giacomo Corti, 28 May 2025
Review of the paper “Cross-scale strain analysis in the Afar rift (East Africa) from automatic fault mapping and geodesy” by La Rosa et al for EGUsphere
May 2025
The Authors present a method to automatically detect faults in the Afar depression, Ethiopia. The methods allow to analyze strain associated to these structures, and derive the geological strain rate based on available ages of deformed volcanic rocks. The strain rates and their spatial distribution are also compared with and discussed in terms of shorter-scale geodetic strain rates.
The topic of this manuscript is surely interesting and the new automatic method to detect faults may be of interest to a wide audience. Therefore, I support its publication after the minor revisions indicated below.
Line 21. ‘main’ instead of ‘mail’
Line 22. I think it is better to say ‘focus strain due to…’ rather than ‘focus strain rates due to…’ (strain is focused, not its rates)
Line 40. Other examples of this may be found in South Ethiopia – see Corti et al. 2019 Nat Comms
Line 45. Reference to Ebinger et al 2023 - nature reviews earth & environment may be pertinent here
Line 78. Better to cite Rime et al or Rooney’s review paper here instead of Wolfenden
Line 90. Viltres et al., 2020, 2022
Line 91. ‘two major flood basalt eruption’. I think the use of the term ‘flood basalt’ here can be somehow confusing, since this term is normally used for the 30Ma traps. Maybe this term can be replaced with something like ‘two major phases of basaltic eruptions’ or similar.
Line 125. ‘Topographic map’ – isn’t this a DEM (FABDEM)? (in case, modify this).
Line 125. Why is the MH magmatic segment shown with that strange elliptic shape, whereas the others are drawn with much more detail? Uniform.
Line 127. Please specify better what the black arrows indicate (motion with respect to a fixed Nubian plate). Also add a reference for this.
Lines 128-129. Add ‘small black circles show seismicity’ or similar
Line 139 and following. I think the Authors should explain in some more detail why they use both 30m- and 90m-resolution DEMs. The choice of a 90m-resolution DEM may sound indeed rather strange, being many 30m-resolution DEMs (e.g., Aster, SRTM) available. This is of course because the Authors want to apply their approach to different datasets with different resolution and compare the results, but I think something more could be said here to explain this. They could also explain, for instance, why the FABDEM has been compared to a 90m-resolution DEM instead of the abovementioned 30m Aster or SRTM DEMs.
Line 224. Full stop instead of comma after ‘detector’
Line 226 (and captions of following figures). Is it necessary to repeat every time that ‘The FABDEM ….is used a figure basis.’?
Line 237. Not clear why the Authors discuss this difference in terms of ‘fault segmentation’, whereas in the previous lines they refer to ‘faults’. If they plotting faults (and not fault segments) the difference may be simply related to the (obvious) fact that the 90m DEM has a lower resolution and cannot capture minor/shorter faults as the 30m DEM can do (indeed the main difference in the graph are related to faults <5km long). Or did I miss something?
Line 282. Remove full stop after ‘Fig. 3a’
Line 289. Why ‘1/40’? I think it should be ‘vertically exaggerated by a factor of 40’
Line 292. ‘Assal Lake’ should not it be ‘Lake Assal’?
Line 268-270. This sentence is not very clear to me. Maybe the Author should first state that strain rates are calculated by use of rock dating (where available), and then extrapolated by assuming the continuity of rifting?
Line 274. I think the third option could be that there is something wrong with the calculations of geological strain rates – this could be maybe stated.
Line 300. ‘of focus region’ maybe not needed.
Line 317. I would not use the term ‘spreading’ here, since it normally refers to oceanic domains. Maybe ‘divergence’ or similar could replace it.
Line 332. Again ‘of focus region’ is maybe not needed.
Lines 334-335. ‘…in Fig. 3a, while….’
Line 338. ‘vertically exaggerated by a factor of 7’
Line 365 and following. I would mention here that, besides being characterized by ongoing extension, the area is also particularly favorable for this analysis as the dominant lithology (basalts) allows the development and preservation of sharp fault scarps. In other parts of the EARS this may not be the case as sediments or volcano-clastic deposits may result in less pronounced (and preserved) fault morphologies.
Lines 374, 401, 437. Again I would replace ‘spreading’ with ‘extension’ or ‘divergence’ or similar
There is no need for me to see the revised version of the ms.
Giacomo Corti,
CNR-IGG, Firenze
Citation: https://doi.org/10.5194/egusphere-2025-1215-RC2 -
AC1: 'Reply on RC1', Alessandro La Rosa, 24 Jun 2025
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-1215/egusphere-2025-1215-AC1-supplement.pdf
-
AC1: 'Reply on RC1', Alessandro La Rosa, 24 Jun 2025
Status: closed
-
RC1: 'Comment on egusphere-2025-1215', Valentin Rime, 20 May 2025
General comments
Overall, the manuscript is of very high quality. The methodology is novel, seems robust, and could be applied to many other study cases. The results from the Afar region are globally relevant and allow to better understand the late phase of rifting. One of the main interests of the paper is that it links different scales of time, liking processes happening over millions of years with processes happening over decades or even shorter. The paper is very well written and has a very clear and coherent structure. The figures and supplementary figures look good and are clear. Except for one aspect to be discussed, the results support the conclusions.
I mainly have minor comments. Some uncertainties of the methodology could be discussed in more details, even though they won’t change the conclusions. I also wonder why you didn’t calculate more geological strain rates from existing datings. This parameter is very interesting and allows a direct comparison with Recent geodetic data. Finally, I question one point of the conclusion regarding Manda Inakir which should be discussed and argued before being presented in the conclusion, or be toned down.
I also made several suggestions that do not relate to scientific quality, but to style or clarity, which is subjective and personal. They are just suggestions, don’t hesitate to not follow them without justification.
Please find my detailed comments in the document attached hereby.
-
AC1: 'Reply on RC1', Alessandro La Rosa, 24 Jun 2025
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-1215/egusphere-2025-1215-AC1-supplement.pdf
-
AC1: 'Reply on RC1', Alessandro La Rosa, 24 Jun 2025
-
RC2: 'Comment on egusphere-2025-1215', Giacomo Corti, 28 May 2025
Review of the paper “Cross-scale strain analysis in the Afar rift (East Africa) from automatic fault mapping and geodesy” by La Rosa et al for EGUsphere
May 2025
The Authors present a method to automatically detect faults in the Afar depression, Ethiopia. The methods allow to analyze strain associated to these structures, and derive the geological strain rate based on available ages of deformed volcanic rocks. The strain rates and their spatial distribution are also compared with and discussed in terms of shorter-scale geodetic strain rates.
The topic of this manuscript is surely interesting and the new automatic method to detect faults may be of interest to a wide audience. Therefore, I support its publication after the minor revisions indicated below.
Line 21. ‘main’ instead of ‘mail’
Line 22. I think it is better to say ‘focus strain due to…’ rather than ‘focus strain rates due to…’ (strain is focused, not its rates)
Line 40. Other examples of this may be found in South Ethiopia – see Corti et al. 2019 Nat Comms
Line 45. Reference to Ebinger et al 2023 - nature reviews earth & environment may be pertinent here
Line 78. Better to cite Rime et al or Rooney’s review paper here instead of Wolfenden
Line 90. Viltres et al., 2020, 2022
Line 91. ‘two major flood basalt eruption’. I think the use of the term ‘flood basalt’ here can be somehow confusing, since this term is normally used for the 30Ma traps. Maybe this term can be replaced with something like ‘two major phases of basaltic eruptions’ or similar.
Line 125. ‘Topographic map’ – isn’t this a DEM (FABDEM)? (in case, modify this).
Line 125. Why is the MH magmatic segment shown with that strange elliptic shape, whereas the others are drawn with much more detail? Uniform.
Line 127. Please specify better what the black arrows indicate (motion with respect to a fixed Nubian plate). Also add a reference for this.
Lines 128-129. Add ‘small black circles show seismicity’ or similar
Line 139 and following. I think the Authors should explain in some more detail why they use both 30m- and 90m-resolution DEMs. The choice of a 90m-resolution DEM may sound indeed rather strange, being many 30m-resolution DEMs (e.g., Aster, SRTM) available. This is of course because the Authors want to apply their approach to different datasets with different resolution and compare the results, but I think something more could be said here to explain this. They could also explain, for instance, why the FABDEM has been compared to a 90m-resolution DEM instead of the abovementioned 30m Aster or SRTM DEMs.
Line 224. Full stop instead of comma after ‘detector’
Line 226 (and captions of following figures). Is it necessary to repeat every time that ‘The FABDEM ….is used a figure basis.’?
Line 237. Not clear why the Authors discuss this difference in terms of ‘fault segmentation’, whereas in the previous lines they refer to ‘faults’. If they plotting faults (and not fault segments) the difference may be simply related to the (obvious) fact that the 90m DEM has a lower resolution and cannot capture minor/shorter faults as the 30m DEM can do (indeed the main difference in the graph are related to faults <5km long). Or did I miss something?
Line 282. Remove full stop after ‘Fig. 3a’
Line 289. Why ‘1/40’? I think it should be ‘vertically exaggerated by a factor of 40’
Line 292. ‘Assal Lake’ should not it be ‘Lake Assal’?
Line 268-270. This sentence is not very clear to me. Maybe the Author should first state that strain rates are calculated by use of rock dating (where available), and then extrapolated by assuming the continuity of rifting?
Line 274. I think the third option could be that there is something wrong with the calculations of geological strain rates – this could be maybe stated.
Line 300. ‘of focus region’ maybe not needed.
Line 317. I would not use the term ‘spreading’ here, since it normally refers to oceanic domains. Maybe ‘divergence’ or similar could replace it.
Line 332. Again ‘of focus region’ is maybe not needed.
Lines 334-335. ‘…in Fig. 3a, while….’
Line 338. ‘vertically exaggerated by a factor of 7’
Line 365 and following. I would mention here that, besides being characterized by ongoing extension, the area is also particularly favorable for this analysis as the dominant lithology (basalts) allows the development and preservation of sharp fault scarps. In other parts of the EARS this may not be the case as sediments or volcano-clastic deposits may result in less pronounced (and preserved) fault morphologies.
Lines 374, 401, 437. Again I would replace ‘spreading’ with ‘extension’ or ‘divergence’ or similar
There is no need for me to see the revised version of the ms.
Giacomo Corti,
CNR-IGG, Firenze
Citation: https://doi.org/10.5194/egusphere-2025-1215-RC2 -
AC1: 'Reply on RC1', Alessandro La Rosa, 24 Jun 2025
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-1215/egusphere-2025-1215-AC1-supplement.pdf
-
AC1: 'Reply on RC1', Alessandro La Rosa, 24 Jun 2025
Viewed
HTML | XML | Total | Supplement | BibTeX | EndNote | |
---|---|---|---|---|---|---|
449 | 89 | 20 | 558 | 30 | 29 | 40 |
- HTML: 449
- PDF: 89
- XML: 20
- Total: 558
- Supplement: 30
- BibTeX: 29
- EndNote: 40
Viewed (geographical distribution)
Country | # | Views | % |
---|
Total: | 0 |
HTML: | 0 |
PDF: | 0 |
XML: | 0 |
- 1