How Does the Onset of Offset Influence Geologic Slip Rates?
Paper details
Citation
Abstract
Geologic slip rates are typically based on the displacement accrued by a geomorphic or stratigraphic feature and the age of the offset feature. Because slip rates are commonly calculated by dividing the displacement of a faulted marker by its age, they contain two open time intervals: the elapsed time between the age of an offset feature and the age of the earthquake that displaced the feature, and the time between the present-day and the most recent earthquake. Here, we explore the influence of including unconstrained open intervals in geologic slip rate calculations. We test the degree to which these open inter- vals affect geologic slip rates and their uncertainties, and we find that their influence depends primarily on mean earthquake recurrence intervals (RIs). Slip rates on faults with longer RIs, such as the Wasatch fault, can be greatly influenced by an increase of up to 20% when accounting for open intervals. In contrast, slip rates on faults with shorter RIs, such as the San Andreas fault, are only slightly influenced by the assumption that slip rates calculated over open intervals approximate those calculated over closed intervals. Our analyses indicate that faults with moderate slip rates (∼0.2–5 mm/yr) are sensitive to both open interval effects themselves, as well as methods to quantify and account for these effects. We re-evaluate how slip rates are calculated and defined in displacement– time space using published deformation records. We explore the utility of assigning a probability distribution to the initiation of offset of the oldest faulted feature and the timing of the most recent earthquake (MRE). We find that calculating geologic slip rates without using probability distributions that capture the timing of the MRE and the onset of offset of the oldest faulted feature, especially on slow-to-moderate slip rate faults, can lead to systematic underestimation of average geologic slip rates.
Presentation summary
Takeaway
- Geologic slip rates are systematically underestimated — potentially by 20% or more — when the time gaps before the first and after the last earthquake in a record are ignored. This bias hits hardest on slow-to-moderate faults (0.2–5 mm/yr), exactly the kind that dominate stable continental regions like Australia.
Key points
- A geologic slip rate divides displacement by time, but the time window almost never starts or ends at an actual earthquake. The gap between when a feature forms and when it first gets faulted (“onset of offset”), and the gap between the most recent earthquake and today, are both unconstrained “open intervals” that artificially stretch the denominator and push the rate down.
- Hatem et al. assigned probability distributions (triangular PDFs) to these open intervals and ran Monte Carlo simulations on synthetic data across a range of recurrence intervals (150–3,000 yr), then tested on real data from the San Andreas fault (~200 yr RI, ~25 mm/yr) and the Wasatch fault (~900 yr RI, ~1 mm/yr).
- On the San Andreas, the effect is negligible (<10% change). On the Wasatch, median slip rates increased ~20% once open intervals were accounted for. At very long RIs (3,000+ yr), the bias can reach ~50%. They also argue that single-earthquake displacement records aren’t slip rates at all — they’re just observations of paleosurface rupture.
Why it matters for ERG
- Australian intraplate faults have recurrence intervals of 10,000–100,000+ years. The paper explicitly flags that faults with RIs >10,000 yr are most affected by open intervals but notes their own method can’t fully address those cases (Fig. 3d).
- How many earthquakes are making up our slip rates of Pleistocene time markers? Another paper discussion for future (Styron, 2019).
- For PSHA in SCRs, if we’re feeding underestimated slip rates into hazard models, we’re underestimating seismicity rates on the faults we do know about — compounding the already large epistemic uncertainty from unknown faults
Session resources
Group Discussion
Discussion questions
- How should we best calculate slip rates in Australia?
- Are there easy ways which we could minimise this open interval uncertainty?
- Alternative methods for slip calculation? Expected displacement / RI?
Discussion (via GitHub)
The discussion for this session lives in the comments below.
Add thoughts, references, data links, or follow-up questions — this stays as a running record for the group.
(GitHub login required the first time.)