NCGS Regional Hayward Fault Field Trip and Tyson’s Lagoon/Tule Pond Trench Site
Russ Graymer and Jim Lienkaemper, US Geological Survey—October 13, 2002
By Richard Cardwell, from the November 2002 newsletter, with edits and photos by Andrew Alden
What do we know about the history of earthquakes on the Hayward fault? What does the fault look like as it runs through urban areas in Fremont and Hayward? And finally, when is the Hayward fault going to rupture in the next large earthquake? These are the questions that NCGS members and friends were attempting to answer on their fall field trip on Sunday, October 13.
The answers to the first two questions were provided by field trip leaders Dr. Russ Graymer of the USGS’s Western Earth Surface Process Team and Dr. Jim Lienkaemper of the USGS’s Earthquake Group. Our day began on a warm autumn morning at the Fremont BART Station parking lot. After being fortified by coffee and donuts supplied by Phil Reed, we walked across Walnut Avenue to examine our first stop—the Tyson’s Lagoon/Tule Pond Trench site. A deflection of the curb along Walnut Avenue was our first indication of where the fault was located.
Tule Pond, October 2001. The pond is the northern remnant of Tyson’s Lagoon.
Russ and Jim began by giving us an overview of earthquake activity and faulting along the Hayward fault. The only large earthquake known to occur on the Hayward fault during historical time was on October 21 [27 in newsletter], 1868. (Is it just coincidence that so many significant Bay Area earthquakes seem to occur in October?) The magnitude of this event was estimated to be about M 6.9. There was between 45 to 60 kilometers of surface rupture from Oakland to Fremont, and the amount of right-lateral offset was about 1.9 meters. Until 1906, this event was known as the Great San Francisco Earthquake.
Seismologists thought that another earthquake may have occurred along the fault in 1838, but recent work indicates that this event probably occurred along the southern Calaveras fault or the San Andreas fault. Since 1868 the Hayward fault has not generated a large earthquake. [Still true as of 2024]
One of the best techniques to determine the history of earthquakes along a fault before historical times is via trenching. Jim explained that a trench is dug across a fault, and the occurrences of large earthquakes are recognized by mapping where the fault has disturbed successive soil layers (as seen in the trench wall). Each disturbed layer is assumed to be caused by an individual earthquake. A trench is typically about a meter wide and up to 20 meters long. The depth of a trench will determine how far back in time one can investigate, but it is often determined by more practical matters such as the depth of the groundwater table.
Jim described how an individual earthquake along the fault causes strike-slip motion as well as some vertical movement. It is this vertical component of faulting that is recorded in the disturbed soil layers. Individual events show fault-scarp colluvial deposits and liquefaction features. His trenches at Tule Pond show that there is typically 0.2-0.4 meter of uplift associated with each major earthquake.
Jim Lienkaemper tacks down burlap at the site of trench 01A south of Walnut Avenue, October 2, 2001. Results were published in USGS OF03-488.
We were joined by Gordon Seitz, a colleague of Jim’s from Lawrence Livermore Lab. He explained how the time of each event can be determined by radiocarbon dating organic matter (charcoal, plant fibers, shells, humic matter, soil fungus) and also by using pollen data from disturbed soil layers. Carbon-14 ages are converted to calendar ages by calibration with tree rings.
Tule Pond is an ideal site for trenching because it is a sag pond formed by an offset step in the Hayward fault. A right step in a right-lateral, strike-slip fault will form a pull-apart graben where abundant and hopefully continuous layers of sediment can accumulate. Tule Pond accumulated sediments throughout much of the Holocene, and the stratigraphic record preserved in it provides an excellent environment for investigating earthquake recurrence. Jim and a team from the USGS have been actively investigating this site for many years. His previous work in 2000 (for trench number 00A) found four dateable events occurring in about 1470, 1630, 1730 and ending with the historical event in 1868. Statistical analysis yields a recurrence rate of about 130 ± 40 years. Interestingly, it has now been 134 years since the last large earthquake in 1868. [See https://pubs.usgs.gov/mf/2002/2386/ for these results.]
In order to extend the record farther back in time Jim and his team recently excavated several more trenches in the area that are much deeper than previous trenches. We were fortunate to have the opportunity to examine two trenches that were dug in September and are still open. Typically trenches are only open for about a month before being filled. Trenches 02A and 02E were open for examination by hardhat-wearing NCGS members. Trench 02A is one of the deepest trenches and contains evidence for 10-12 events going back 2000 years. The radiocarbon dating work has just begun, but the later events can be correlated to events in nearby trenches. Over an interval of 2000 years 10-12 events would suggest a longer recurrence rate than the most recent record, but some events may have been lost by unconformities.
Around noon we moved to Stop 2 where we had lunch at the Mission San Jose in Fremont. The original adobe church was built from 1805-1809, destroyed by the earthquake in 1868, and then finally restored in 1985. Here Russ gave us an overview of his mapping work along the Hayward fault. He noted that the surface trace of the fault is difficult to map along its entire extent, especially at its southern end where it joins the Calaveras fault.
The section of the Hayward fault in Hayward and Fremont is part of a broad zone. This fault zone has accumulated about 100 kilometers of right-lateral offset in the past 12 million years, yielding a geologically-determined slip rate of about 9 millimeters per year. Recent geodetic measurements show that the fault in the Fremont to Hayward section is creeping at the surface at an average rate of about 4.6 millimeters per year. Presumably, the slip rate difference is made up by movement during large earthquakes.
The rest of the day we examined several sites along the fault. At Stop 3 (south of Fremont) we examined a site where the Irvington gravels are offset by the Hayward fault. These gravels are 1.5–0.5 million years old and contain clasts of sandstone, metamorphic rocks (graywacke, blueschist), volcanic rocks, and red chert. From here we moved east to Stop 4 along the northern Calaveras fault (also called the Sunol fault) to examine the Livermore gravels. The Livermore gravels have the same age, lithology, and heavy minerals as the Irvington gravels. There is no evidence for significant offset along the Sunol fault.
Russ has proposed that the Irvington and Livermore gravels were probably once a single alluvial unit. Over the last 1 million years the gravels were separated by uplift of the range between the Sunol and Calaveras faults. The range consists of the northwest-trending Pleasanton, Sunol, and Walpert ridges. The range is cut by Niles Canyon—a sinuous, deeply incised, east-west oriented canyon formed by Alameda Creek. Alameda Creek predates the uplift of the range, and the drainage was superposed on the range as it was uplifted. The uplift is best explained by oblique motion along the faults resulting in compression between the Hayward and Calaveras faults. Leveling surveys are consistent with this proposal, and they show that the creeping section of the Hayward fault has about 1 mm per year vertical uplift on its eastern side.
Stop 5 was in downtown Hayward where we examined the evidence for fault creep in an urban setting. The creep has cracked the walls in the old Hayward City Hall so badly that the building was abandoned. Other effects include offset curbs on B Street and warped building walls.
Old City Hall, Hayward, 2012.
We did not have time for Stop 6 at the Marcella Street trench site in Hayward. Here the Hayward fault occupies three different strands in a zone less than a kilometer wide.
The NCGS sincerely thanks Russ Graymer and Jim Lienkaemper for the excellent field trip. They prepared a superb field guide that includes many colored geologic maps, topographic maps, air photos, and trench logs. Further information is available at their USGS website. Jean Moran did her usual excellent job in organizing the trip, handling trip registration, arranging transportation and food. Finally, our thanks to all of the drivers for use of their cars and vans.