EGU General Assembly 2020
© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.

Three-dimensional analysis of normal faults in the Horda Platform (North Sea): the possible influence of stress perturbations on fault geometries

Luca Collanega1, Donatella Mellere2, Matteo Massironi1, and Anna Breda1
Luca Collanega et al.
  • 1Università degli Studi di Padova, Padova, Italy (
  • 2DNO North Sea AS, Badehusgata 37, 4014 Stavanger, Norway

Rift-related faults often display non-rectilinear geometries, which have been interpreted as (i) the result of linkage between different fault segments -developed during a single or more tectonic phases-, (ii) as curvilinear faults due to gravitational collapse, (iii) as inherited basement trends. Disentangling these processes is generally difficult, with multi-phase rifting and reactivation of pre-existing structures being the most intuitive and commonly adopted explanations.

Here, we use 3D seismic data to reconstruct the evolution of a couple of intersecting, curvilinear faults in the Horda Platform (Northern North Sea), which is characterised by a complex history of reactivation and multi-phase extension during the Jurassic-Cretaceous rifting. By reconstructing the three-dimensional geometry of the fault planes, we highlight that one fault follows the trend of the Permian-Triassic rift along its entire length, whereas the other, strongly curvilinear, fault appears to be partially deflected from it. By using time-thickness maps and kinematic analyses, we show that the partially deflected fault initiated in the Late Jurassic, soon after the other one (which activated in the Middle Jurassic). Notably, the younger fault flexes from the inherited Permian-Triassic trend as it approaches the other, more mature, fault, getting perpendicular to -and finally crosscutting- it. Hence, the curvilinear geometry developed during the upward propagation of the fault plane during the Jurassic-Cretaceous rifting, suggesting that such change of strike was driven by the influence of the more mature fault and is not due to structural inheritance. Similar deflections can be observed also in other areas of the dataset, with incipient faults flexing towards more mature structures.

More generally, newer faults have been shown to deflect perpendicularly to pre-existing faults both in analogue and numerical models, suggesting we are facing a general process. These strike deflections suggest a stress re-orientation in the vicinity of well-developed structures, and not just simply a stress-drop as widely indicated by fault spacing and throw distribution of parallel faults. This is consistent with observations on deflected normal faults developing in correspondence to oblique basement fabrics as well as with the numerical model of the stress field by Homberg et al. (1997). Hence, our three-dimensional analysis of fault geometries suggests that the well-established concept of “fault-related stress-drop” should be broadened into the concept of “fault-related stress-reorientation”.



Homberg, C., Hu, J.C., Angelier, J., Bergerat, F., Lacombe, O., 1997. Characterization of stress perturbations near major fault zones: insights from 2-D distinct-element numerical modelling and field studies (Jura mountains). Journal of Structural Geology 19, 703–718.

How to cite: Collanega, L., Mellere, D., Massironi, M., and Breda, A.: Three-dimensional analysis of normal faults in the Horda Platform (North Sea): the possible influence of stress perturbations on fault geometries, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-21550,, 2020

Comments on the presentation

AC: Author Comment | CC: Community Comment | Report abuse

Presentation version 1 – uploaded on 30 Apr 2020
  • CC1: Comment on EGU2020-21550, Christopher Jackson, 04 May 2020

    Hi Luca et al. Very nice work, on an area that still needs lots of detailed fault analysis using the newly available CGG dataset! So, based on your work, how do you view the debate regarding temporal changes in extension direction between RP1 and RP2, and within RP2 itself? Also, the 3D geometry of the two studied faults is a little unclear to me. Given that at Jurassic levels the fault changes in strike, but does not at deeper, P-T levels, does this mean there is vertical segmentation with an increasing across-strike fault spacing in cross-section as you move northwards? Thanks! Chris Jackson (Imperial College)

    • AC2: Chris' comment, Luca Collanega, 07 May 2020

      Hi Chris!! Thanks for your comments!

      Question 1, re-orientation of the extension direction. Based on our dataset, the Permian-Triassic faults generally trend NNE-SSW, with an en-echelon segmentation. Since the imaging of the underlying crystalline basement is very poor, it is difficult to say whether these faults reflect the actual stress field of the Permian-Triassic rift or deeper basement structures. Regarding the Jurassic-Cretaceous rifting, we document two fault trends which are restricted to the Upper Triassic-Jurassic stratigraphy: NNW-SSE faults and NE-SW faults. The NNW-SSE faults of our area are laterally continuous with the similarly-oriented faults described by Duffy et al. (2015). Whereas in the northernmost part of the survey these faults appear to activate in the Cretaceous (consistently with Duffy), in the south we find clear evidence of their first activation in the Early Jurassic. Regarding the NE-SW faults, they are much less developed than the NNW-SSE and it is not possible to find indications of their time of activation. It is relevant to note that in our area these faults are present exclusively in the hanging wall of similarly orientated Permian-Triassic faults. Although the NE-SW faults are restricted to the Upper Triassic-Jurassic interval, this spatial association with deeper Permian-Triassic structures may indicate some kind of kinematic-coupling (similarly to what we observed in the Taranaki Basin). 

      Question 2, 3D geometry. You are right that a vertical segmentation of the fault plane is necessary to accommodate different fault geometries in the Permian-Triassic and in the Jurassic-Cretaceous intervals. However, such segmentation is limited to the part of the fault segment where the southernmost fault rotates perpendicularly to the northernmost fault (i.e. the central part). This leads to the interesting question of why the southernmost fault rotated towards the northermost fault during the Jurassic rifting but not during the precedent Permian-Triassic rifting. 

  • CC2: Comment on EGU2020-21550, Thomas Phillips, 04 May 2020

    HI Luca, Really cool work! In the presentation you show faults rotating to become perpendicular to the pre-existing structure. Do you document examples of the opposite occurring, where faults rotate and align with the older fault? Does this depend at all on whether the newly formed fault is in the hangingwall or footwall?

    • AC1: Tom's comment, Luca Collanega, 07 May 2020

      Hi Tom!! Thank you very much for your comment. I suspect that the orientation of the fault is somehow influenced by the fact that the new fault is in the hanging wall rather than in the footwall of the pre-existing one. All the examples of faults rotating perpendicularly that I documented are in the footwall of the pre-existing fault. Although I don't have examples of faults rotating parallel in the hangingwall, I suspect that the gravitational collapse of the hangingwall (possibly enhanced on listric fault planes) may favour the development of parallel faults. Thank you very much for your contribution!!