How do normal faults grow?

Abstract

Normal faults grow via synchronous increase in displacement and length (‘propagating fault model’, also known as the ‘isolated fault model’), or by rapid length establishment and subsequent displacement accrual (constant-length fault model). We here use time-series displacement (D) and length (L) data from natural and experimental faults to elucidate growth styles and D-L trajectories throughout fault life, and to assess the applicability of the two fault models. We show that the growth of most faults is characterized by two stages, with the first defined by fault lengthening (20-30% of fault lifespan) and the second by displacement accrual (70-80% of fault lifespan). Although broadly adhering to the constant-length model, fault growth throughout the lengthening stage, during which significant displacement (10-60% of the total end-of-life fault displacement) may also accumulate, is achieved through rapid tip propagation, relay breaching, and segment linkage, characteristics perhaps most intuitively thought to reflect growth in accordance with the propagating model. The subsequent growth stage is dominated by displacement accrual with limited lateral tip propagation, a phenomenon best described by the constant-length model. We also show that, despite being used primarily in support of the propagating model, global displacement-length (D-L) datasets are equally compatible with the constant-length model.

Structural Inheritance and Rapid Rift‐Length Establishment in a Multiphase Rift: The East Greenland Rift System and its Caledonian Orogenic Ancestry

“…230 Myr), the 10–20 Myr lengthening phase of the rift system amounts to 4–9% of its active history. Regardless of what is considered the end of rift life, rapid lengthening (10–20 Myr; 4–20% of rift life) of the East Greenland rift system appears consistent with the predictions of the “constant‐length model” for normal fault growth (see e.g., Jackson et al, ; Nicol et al, ; Rotevatn et al, ). Rapid lengthening and establishment of rifts and their border fault systems is also seen in other rift systems globally.…”

“…However, Nixon et al () also show that the establishment of border fault systems, albeit rapid, is characterized by tip propagation, relay breaching, and fault linkage, which is at variance with the “constant‐length model” for fault growth (e.g., Nicol et al, , ). This implies that hybrid growth behaviors, such as recently suggested for normal faults (Finch & Gawthorpe, ; Rotevatn et al, ) likely also apply at rift scale, whereby (i) a rapid stage of fault (or, in this case, rift) length establishment achieved through tip propagation, relay breaching, and fault linkage ( i.e., “isolated fault growth”; see Jackson et al, ; Rotevatn et al, ) is followed by (ii) a stage of displacement accumulation and subsidence without significant further tip (or rift) propagation ( i.e., “constant‐length fault growth”; see Jackson et al, ; Rotevatn et al, ).…”

“…The fact that the East Greenland rift system attained its length early in its life bears similarity to the behavior seen in many normal fault systems at a smaller (fault array to individual fault) scale, whereby faults also attain their near full length at an early stage before subsequently accruing displacement. This mode of normal fault growth is termed the “coherent fault model” (Jackson & Rotevatn, ; Walsh et al, ) or, more recently, the “constant‐length model” (Jackson et al, ; Nicol et al, , ; Rotevatn et al, ). Rotevatn et al () suggest that most normal faults spend 20–30% of their slip history attaining their near full length, before accruing displacement for the remaining 70–80% of their lifetime.…”

“…This mode of normal fault growth is termed the “coherent fault model” (Jackson & Rotevatn, ; Walsh et al, ) or, more recently, the “constant‐length model” (Jackson et al, ; Nicol et al, , ; Rotevatn et al, ). Rotevatn et al () suggest that most normal faults spend 20–30% of their slip history attaining their near full length, before accruing displacement for the remaining 70–80% of their lifetime. Assuming the southcentral part of the East Greenland rift system initiated in the latest Devonian (ca.…”

“…A secondary goal of the study is to investigate whether the “constant‐length model” for fault growth (Nicol et al, , ; Walsh et al, , ), which suggests that faults attain their near‐full lengths early in their slip histories, is valid also at rift scale. The constant‐length model has been shown to accurately describe fault growth at the scale of individual faults and fault arrays (e.g., Giba et al, ; Jackson et al, ; Jackson & Rotevatn, ; Rotevatn et al, ), but has not previously been proven to be applicable to entire rift systems.…”