A fault under San Diego is creeping while the city keeps building on top of it

A GPS network stretched across San Diego has measured the Rose Canyon Fault creeping at 2.4 millimeters a year, plus or minus half a millimeter, in a study published in July 2024 by researchers from the U.S. Geological Survey, San Diego State University and the Scripps Institution of Oceanography. That is a small number by itself, but it sits at or above the upper end of what geologists had previously estimated from trenching the fault alone.

The fault runs under some of the most built-up ground in the county, from La Jolla Shores south along Rose Canyon to the edge of Mission Bay and, by some evidence, on into San Diego Bay itself. The city has kept building along that corridor for decades while the debate over how dangerous the fault really is has only recently been settled.

The Fault’s Path Through the Built City

The City of San Diego’s own seismic safety planning documents trace the Rose Canyon fault zone south from the La Jolla Shores area along the general alignment of Ardath Road, through Rose Canyon itself, then along the east side of Mission Bay. Some evidence, the same planning documents note, points to a further extension south along San Diego Bay toward Tijuana. Local reporting on downtown construction sites has traced the fault more specifically still, describing it as running near the San Diego Convention Center and Mt. Soledad in La Jolla, close enough to East Village construction that heavy equipment has reportedly uncovered faulting mid-project, triggering extra trenching, redesigns and added costs. East Village’s small Fault Line Park takes its name directly from what runs beneath it.

For years, city planning documents record a genuine scientific disagreement over how dangerous that buried path actually is: one view held that motion on the fault had effectively stopped, while another argued it was locked and building strain toward a major rupture. The disagreement was not resolved by argument. It took better instruments.

GPS Data That Outran the Old Trenches

The 2024 study, published in the Bulletin of the Seismological Society of America and led by USGS research geologist Drake Moore Singleton with SDSU’s Jillian Maloney and Thomas Rockwell and Scripps geophysicist Duncan Agnew, used a combined network of campaign and continuous GPS stations to model how the crust on either side of the fault is actually moving, rather than relying only on the trenches and offset landforms earlier estimates were built on. Their preferred model puts the slip rate at 2.4 ± 0.5 millimeters a year. In the paper’s own words, “the fault may be slipping toward the higher end of geologic estimates” that had been used in prior hazard assessments.

The team also found something in the data pointing beyond Rose Canyon on its own: a possible mechanical connection to the San Miguel–Vallecitos fault system to the south, in Baja California. If the two systems are linked, a rupture would not be limited to Rose Canyon’s own length. It could run longer, which in earthquake physics generally means a bigger magnitude and stronger, more widespread shaking than a Rose Canyon-only rupture would produce.

What a Magnitude-6.9 Rupture Would Look Like

USGS has already modeled what a substantial Rose Canyon earthquake would do to the region, in a scenario animation credited to research geophysicist Robert W. Graves and released in 2019. It simulates a magnitude-6.9 earthquake rupturing a 65-kilometer stretch of the fault offshore of San Diego and Tijuana, with the rupture starting at the fault’s northern end and traveling south. Using a three-dimensional model of the crust beneath the region, the simulation shows how the direction the rupture travels concentrates shaking, an effect called rupture directivity, toward the densely populated areas in its path.

Neither Singleton’s GPS study nor Graves’s animation is a warning that a rupture is imminent. Both describe capability, not timing. But a fault modeled at magnitude 6.9 running under a convention center, a coastal bluff neighborhood and an entire construction-heavy downtown core is a different planning problem than a fault whose motion had supposedly stopped.

That scenario is not a prediction of when a rupture happens, only of what one would look like if it did. The city’s building safety element still classifies the ground near the fault trace as requiring geotechnical review before major construction, a standard that predates the 2024 GPS data and has not been revised specifically in response to it.

A Rate That Outpaced the Old Assumptions

What makes the 2024 finding matter is not the number alone but what it replaces. Hazard models used for building codes and insurance pricing across San Diego were built on slip-rate estimates drawn mostly from trenches dug across visible fault traces on land. GPS data captures the whole crustal block moving, including motion the trenches could not see or that occurs offshore, where much of the fault’s mapped length actually sits.

Neither the USGS team nor the city has proposed rewriting the seismic safety element to reflect the new number. The fault, meanwhile, keeps moving at whatever rate it moves, underneath streets, water lines and buildings the city approved under the older estimate.

This article was produced with the assistance of AI and reviewed by Morning Overview editors prior to publication.

More from Morning Overview