Stress and fault parameters affecting fault slip magnitude and activation time during a glacial cycle

Abstract

The growing and melting of continental ice sheets during a glacial cycle is accompanied by stress changes and reactivation of faults. To better understand the relationship between stress changes, fault activation time, fault parameters, and fault slip magnitude, a new physics‐based two‐dimensional numerical model is used. In this study, tectonic background stress magnitudes and fault parameters are tested as well as the angle of the fault and the fault locations relative to the ice sheet. Our results show that fault slip magnitude for all faults is mainly affected by the coefficient of friction within the crust and along the fault and also by the depth of the fault tip and angle of the fault. Within a compressional stress regime, we find that steeply dipping faults (∼75°) can be activated after glacial unloading, and fault activity continues thereafter. Furthermore, our results indicate that low‐angle faults (dipping at 30°) may slip up to 63m, equivalent to an earthquake with a minimum moment magnitude of 7.0. Finally, our results imply that the crust beneath formerly glaciated regions was close to a critically stressed state, in order to enable activation of faults by small changes in stress during a glacial cycle.

Late Cretaceous to recent tectonic evolution of the North German Basin and the transition zone to the Baltic Shield/southwest Baltic Sea

“…4; Piotrowski, 1997;Piotrowski and Tulaczyk, 1999). Repeated ice sheet loading and unloading during the Pleistocene caused vertical and horizontal stress variation, which led to a change in the equilibrium situation of the lithosphere and mass movements in the upper and lower mantle (e.g., Al Hseinat and Hübscher, 2014;Arvidsson, 1996;Brandes et al, 2012a, b;Dyke et al, 1991;Fenton, 1994;Johnston, 1987;Kujansuu, 1964;Lagerbäck, 1978;Lagerbäck and Sundh, 2008;Lang et al, 2014;Munier and Fenton, 2004;Muir-Wood, 2000;Olesen, 1988;Quinlan, 1984;Sandersen and Jørgensen, 2015;Sauber and Molnia, 2004;Shilts et al, 1992;Steffen et al, 2014aSteffen et al, , 2014bSteffen et al, , 2014cSteffen et al, , 2016Steffen and Wu, 2011;Stewart et al, 2000;Turpeinen et al, 2008;Wu et al, 1999).…”

Reply to comment by Hampel et al. on “Stress and fault parameters affecting fault slip magnitude and activation time during a glacial cycle”

“…Hampel et al . [, hereinafter Ha15] recently commented on our study [ Steffen et al , ], which investigates the effect of stress and fault parameters on fault slip magnitude and activation time during a glacial cycle using a newly developed two‐dimensional glacial isostatic adjustment (GIA)‐fault model [ Steffen et al , ]. In summary, Ha15 criticize the following parts of our study stating “ (1) that Steffen et al [] describe our modeling approach in a misleading way, (2) that they do not mention the specific results (e.g., regarding the amount and timing of fault slip) of our studies anywhere in their article despite the similarity of the topic of their article, and (3) that the content and layout of Steffen et al's Figure 1 closely resembles two figures previously published in our studies but they do not cite the source (they, however, introduced conceptual errors concerning the glacial‐interglacial stress evolution into their figure) .”…”

“…We appreciate the comment by Ha15 as it gives us the opportunity to discuss and describe the differences between our model and Hampel et al's model in greater detail than in Steffen et al . [] and earlier publications. We show that comparison between the results of both model approaches is not feasible.…”

“…Due to the many issues raised by Ha15, we extract specific comments by Ha15 in conjunction to those above and also rephrase their statements so that we can address them in more detail: Hampel et al's Earth model neglects the important effects of viscoelasticity in the mantle on GIA and stress migration. The applied plate‐motion velocities in Hampel et al's model control the fault slip. The load models in Hampel et al are not suitable for GIA studies. The models of Hampel et al have only limited direct observational support. According to Ha15, the density of the crust in our models is wrong. According to Ha15, we do not cite Hampel et al's studies as source of Figure in Steffen et al . []. According to Ha15, Figure 1 in Steffen et al [] has conceptual errors: the Mohr circle during glaciation has to become smaller and cannot exceed the line of failure after glaciation. According to Ha15, we apply our results to the Pärvie fault and omitted a comparison of our results with those by Turpeinen et al []. According to Ha15, we do not provide a sufficient comparison between our results and those obtained by Hampel et al We will correct several statements in Ha15. …”

“…After a certain amount of time (which is often not specified in their studies, but all load processes appear to start after one million years, see Hetzel and Hampel []), the stress starts to build up and propagate from the sides to the fault, which eventually starts to slip at a steady rate in the absence of any applied load (e.g., Figure 2b in Hampel and Hetzel []). The application or removal of glacial load alters the slip rate at the fault but the applied plate motion plays a major part in their fault slip.In contrast, no plate velocity is applied to the models of Steffen et al [, , ], whereas the tectonic stress maintains the fault close to the stability limit before the glacial cycle begins. Consequently, fault slip that is induced near the end of deglaciation is only due to the relaxation of the rebound stresses.…”