"When is the next big one?" A question that stirs the minds of nearly 24 million Southern California residents fearful of the next large earthquake. In fact, so much so that the San Andreas Fault is the primary antagonist in the blockbuster earthquake disaster film "San Andreas" starring Dwayne 'The Rock' Johnson (aptly named). While the movie is missing many geologic accuracies, the full story of the San Andreas Fault may also be missing something important. Our paper reveals that the hydrologic loads due to increases in water level of an ancient lake were large enough to possibly trigger major earthquakes on the adjacent section of the San Andreas Fault.

The Concern

The Southern San Andreas Fault (SSAF), a portion of the nearly state long San Andreas Fault, is currently locked and has been devoid of any major earthquakes (M 7+) for the past ~300 years. Yet, we also know that this portion of the fault has accumulated enough tectonic strain to produce such an event (Fialko, 2006). A major earthquake on the SSAF poses the largest seismic hazard in California as it could severely damage the Los Angeles metropolitan area. By studying the history of the fault we can better understand what might cause the current "earthquake drought".

The last 1000 years

The SSAF lies within the Salton Trough, a large pull-apart basin formed by the active tectonics in the region. This basin would periodically fill then dry up as the Colorado River changed its course from filling the basin to draining out into the Gulf of California. This cycle would create a large lake called ancient Lake Cahuilla which was ~x40 larger in volume than its present day remnant, the Salton Sea. Recent advances in the geologic record for both the lake and earthquake timings of the past reveal that over the last millennium ancient Lake Cahuilla formed on 6 separate occasions. More importantly, it coincides with nearly every major event on the SSAF during this time (Figure 2).

Why Does Lake ≈ Quake ?

In order to investigate the possible causal relation of the lake and earthquake timings we built a numerical model of the region to resolve the effects the lake had on the fault over the last 1000 years. We drew inspiration from previous modeling efforts of the region (Luttrell et al., 2007; Brothers et al., 2011). However, these former models do not consider the more realistic hydro-mechanic complexity that our model does. I think these complexities, which required the CSRC high-performance computing cluster at San Diego State University to resolve, are quite fascinating and I elaborate on them in the following paragraphs.

The water increases the pore pressure on the fault, reducing the normal stress (ie. unclamps the fault). This happens in two ways: The weight of the water on the surface and the water that diffuses into the fault itself. These two effects can constructively interfere with each other.

The resulting raise in pore pressure creates a situation much like a puck on an air hockey table as Yuri Fialko (a co-author) would say. Pushing the puck along the surface with you fingers represents the tectonic stress and movement of the plates; it’s not the easiest, but if you continue to push eventually the puck will move. However, the increasing pore pressure with fluid is like turning the air on - it pushes out on the two sides of the fault making it easier to slide, thus triggering an earthquake.

With ~2 million tetrahedrons our finite-element model captures other contributors to 'unclamping' like nonvertical fault dip, permeable damage zone, and lateral pore pressure diffusion. Through modeling lakes over the last 1000 years we determined that while the pressure from the lake on the surface has the greater impact towards destabilizing the SSAF fault both discussed effects interfere with one another simultaneously, raising the pore pressure to high points coinciding with major earthquakes. There is more fascinating info, including the of rate of pressure increase, in the paper!

What does the past reveal about the future?

We find new evidence from not the geologic record and our model results to support the claim that ancient Lake Cahuilla likely triggered major earthquakes on the SSAF. The last time ancient Lake Cahuilla was full was 300 years ago and due to modern infrastructure will likely never fill again. The lack of the lake has had a stabilizing effect on the SSAF, but potentially increases the stress to be released in future earthquakes or sequences of earthquakes. However, in 1905 the Colorado River swelled and flooded into the Salton Trough creating the much smaller Salton Sea. As the Salton Sea continues to decline it too has a minor stabilizing effect on the SSAF. Regardless of how much the lake modulated the process in the past, the tectonic loading indicates the current potential for an earthquake is still very high. the  The physical insights of our work are also useful for human created reservoirs and other natural bodies of water where sudden large hydrologic loads at the surface have contributed to seismicity