User:Rriesr/creepmeter

A creepmeter is an instrument that monitors the slow surface displacement of an active geologic fault in the Earth. Its function is to record the slow aseismic creep between or immediately following earthquakes. Measuring aseismic creep on faults is important for understanding the earthquake cycle and the way faults respond to each other.

A curb displaced by surface creep on the Hayward Fault in California. The crack in the curb is a result of surface creep over time. Once the curb is cracked, the two sides can begin to shift with respect to each other, creating the offset shown above.

Construction

Creepmeters consist of a wire or rod between two pillars that are located on either side of the fault. The wire or rod is anchored to one of the piers, allowing motion on the fault to pull on the wire or rod as the fault slips, changing the distance between the two pillars. There are various length standards that have been used, including invar, glass fiber composite, carbon fiber, or quartz fiber.

This change in distance is measured by an electronic component. In older creepmeters, this was a low-friction potentiometer, but in more recent instruments, the potentiometer has been replaced by zero-friction linear variable differential transformers (LVDTs). By combining the measurement of change in length with the known angle of installation of the creepmeter with respect to the fault, the amount of fault parallel creep can be determined.

Most creepmeters have a power source or data logger that is above ground, however, there are micro-power versions of creepmeters that have been developed for deployment in region where there are restrictions about surface instrumentation or where there is a significant concern of the station being vandalized.

Challenges with creepmeter measurements

Most creepmeters have rods that are buried only a couple meters deep, which leaves them sensitive to the swelling and contraction of clay-rich fault gauge in response to seasonal rainfall. There has been research into attaching creepmeters to deep anchors down to tens of meters to buffer the measurement from noise from the swelling and shrinking associated with seasonal rainfall. Creepmeters that are deployed with more carefully designed anchors can have a reduction in this seasonal rainfall noise down to 0.2 mm, even in a region with a high amount of rainfall.

Creepmeters are designed to be used to measure interseismic creep (creep occurring between earthquakes), rather than coseismic displacements (fault motion that occurs during an earthquake). Coseismic displacements can be a meter or more, while the interseismic creep events have displacements of a few millimeters. Because of this large difference in the magnitude of displacement, creepmeters can be damaged when they are located too close to a large earthquake, since the large coseismic displacement brings the creepmeter out of its useable range.

Creepmeters are operated under the assumption that the two sides of the fault deform as rigid blocks, with the displacement concentrated at a thin plane between them instead of over a broader area as is believed to be the case for many faults. Despite violations of the rigid-block assumption being common, the measurement accuracy is usually still high, due to the small amount of slip (on the order of a few millimeters) in comparison to the width spanned by the creepmeter (2-30 meters). It is estimated that the error associated with the violation in this assumption is ~1% or less. Faults are usually viewed as associated with diffuse shear zones at depth, which have a width that is usually wider than the length spanned by a creepmeter. This leads creepmeters to underestimate slip on a fault.

Use in Scientific Research

Creepmeters have been used to study a number of types of aseismic displacement on faults, including steady fault creep (aseismic creep that is continuous at a nearly steady rate), transient aseismic slip (aseismic creep occurring in discrete periods of time with intervals without creep in between), triggering of slip following large earthquakes (slip triggered by the passage of seismic waves from a regional or distant earthquake), and afterslip (slip on a fault near the location of an earthquake following the occurrence of that earthquake).

There has also been some suggestion that the creep rate may change in the period of time a large earthquake occurs on nearby on the same fault. This is sometimes called "pre-slip" or "precursory slip". If true, this mechanism could allow for the prediction of earthquakes and better constraints on seismic hazard. The noise level of most creepmeters is currently too high to be able to fully investigate this idea thoroughly. As a result, the proposal of the existence of precursory slip has helped drive the development of better creepmeters and the continued deployment of creepmeters in regions with active faults.

While GPS can measure displacements, creepmeters specifically measure the surface displacement across a fault. Additionally, many GPS stations don’t have a high enough resolution to determine fault displacements with only a couple millimeters of displacement. A GPS station would also need to be right on top of a fault to get an equivalent measurement to a creepmeter.

There have been deployments of creepmeters in a number of countries including the Untied States, Chile, and Turkey studying the creep behavior in a number of fault zones.

California

Locations of creepmeters operated by the USGS in California. Colored by date of first data recording.

There is a network of creepmeters in California run by the United States Geological Survey (USGS) and University of Colorado-Boulder across major fault zones including the San Andreas, Hayward, and Calaveras fault zones. This deployment includes some instruments that have been in operation since the 1970s with some instruments having 10 minute or 1 day sampling rates.

Additional creepmeters operated by UNAVCO are deployed in southern California on major faults in that region including the southern San Andreas Fault, Garlock Fault, and Superstition Hills Fault.

Following the 2019 Ridgecrest earthquake, investigations of creepmeter data found triggered slip on distant faults up to nearly 400 km distance from the epicenter of the earthquake, despite negligible slip being found on creepmeters near the location of the earthquake. It was also observed that the triggered slip had a velocity that increased over time.  More distant stations were observed to have longer triggered slip durations than stations that were closer to the location of the Ridgecrest earthquakes. The observations of the triggered slip was also associated with the occurrence of swarms of small earthquakes on the faults with triggered slip, making the study authors believe that triggered slip on a fault can be the driving force behind delayed triggering of earthquakes following the passage of body waves or surface waves from a large earthquake.

In California, data from the network of creepmeters has also been used to constrain friction models that would be able to reproduce all of these behaviors, which had previously required multiple different models. One study suggested that there could be an unstable layer near the surface, the properties of which determine whether that portion of the fault undergoes constant creep or transient creeping events.

Chile

11 creepmeters were installed across 4 different fault zones in Northern Chile as part of the Integrated Plate Boundary Observatory (IPCC). The goal of this deployment is to constrain the fault behavior of faults in the Chilean forearc.

One study of the data from this array suggests that the majority of recorded slip events in this region are triggered slip events from either global earthquakes or regional earthquakes on the nearby subduction zone. The same study also observes that the slip recorded is mostly in discrete events instead of continuous creep.

Turkey

Creepmeters are deployed along the North Anatolian fault zone, which hosted the 1999 Izmit Earthquake. The studies from these creepmeters and other techniques to determine aseismic slip have led to multiple different estimates of the slip rate on the fault.