Urban Geodesy: Monitoring Active Deformation near the Hayward fault

Matthew A. d'Alessio, David Schmidt, Roland Bürgmann, and the UC Berkeley Active Tectonics Group

If you need to take measurements in the middle of a busy intersection, or in a stadium that houses thousands of voracious football fans, or even at a freeway offramp in the middle of rush-hour, you need an urban geodesist. And the Hayward fault, which passes through one of the most densely populated strips in California, is a place where urban geodesy is king.

Geodesy, the science of surveying and measuring the earth's surface, is one of the many tools used by earthquake scientists to understand the behavior of faults. This paper describes the Berkeley Seismological Laboratory's ongoing effort to use geodesy to monitor the movement of the Hayward fault, as well as some of the challenges introduced by doing science in a densely populated urban region.

Hayward Fault GPS Campaign

The Hayward fault is unique because it not only produces large earthquakes like the estimated M7 1868 earthquake, but it also slowly slips along in a process called aseismic creep. In recent years, the Berkeley Seismological Laboratory has begun utilizing measurements with the Global Positioning System (GPS) to monitor creep along the Hayward fault. This work will provide an unprecedented spatial resolution of GPS measurements about a creeping fault and should allow us to determine where the fault is creeping, how fast it is creeping, and most importantly, how deep the creep extends. Since aseismic creep relieves stress along the fault, identifying locked patches at depth can give us clues to determine where large earthquakes might eventually nucleate and how big they could be.

In a GPS ``campaign'', we use research-grade GPS equipment to determine the location of a point on the ground to a precision of less than 5 mm. In other words, we can determine locations on earth to within an area half the size of a dime. The next year, we repeat the campaign and return to the exact same location to determine the position again. Because the Hayward fault is slowly creeping along, the spot will have moved; with our high precision positioning, we hope to determine exactly where and how fast. The precision and reliability of our measurements depend on several factors that are all more difficult in the urban setting. See the table on the following page for more details.

As part of the Berkeley Seismological Laboratory's ongoing effort to monitor active deformation along the Hayward fault, we determine positions of over fifty benchmarks within ten kilometers of the fault (Figure 25.1). Of these fifty monuments, about two thirds of them are in logistically difficult urban settings while the remainder are in the scenic rolling hills that flank the fault. The first surveys of most benchmarks were made in 1998, and our group has been returning each summer to take new measurements.

Figure 25.1: The Hayward fault and urban areas of the East Bay. Each triangle shows the location of a benchmark surveyed annually since 1998

Current Progress

In the last year, the Active Tectonics group purchased seven new, state of the art, geodetic grade GPS receivers. The Trimble 5700 receivers are smaller, lighter, and draw less power than receivers used in the past, making them ideal for both urban geodesy and more remote GPS sites that can only be accessed by hiking. Initial trials using the new receivers show that they produce good quality data that is consistent with the receivers used in past campaigns. The 2002 campaign, which is currently underway, promises to yield excellent estimates of the velocities over the last four years.

Results from our test site at UC Berkeley's Memorial Stadium appear in Figure 25.2. We should note that in order to take measurements at this site, three of the stations must be set up on staircases inside the stadium, meaning that the operator must make sure that the tripod is perfectly level even though the legs are on a completely uneven surface. Further, the tripod at station STAA must carefully straddle a railing along one stairway. Despite these difficult conditions, our calculations of the fault-parallel velocity across the stadium from these observations is $4.2\pm0.3 mm\cdot yr^{-1}$ - in very close agreement with the creep rate determined from offset curbs along nearby Dwight Way of $3.9\pm0.1$ and $4.1\pm0.1 mm\cdot yr^{-1}$. We expect results from the rest of our 2002 campaign (currently underway) to yield equally reliable velocities.

Figure 25.2: Velocities for 1998 - 2002 of four benchmark in UC Berkeley's Memorial Stadium, shown relative to station STAC. The Hayward fault cuts down the middle of the stadium. These observations show that the Hayward fault creeps at a rate of about $4.2\pm 0.2 mm\cdot yr^{-1}.$ The locations of faults are from a recent paleoseismological report by Geomatrix Consultants, 2001.

The Challenge

The more densely populated the region, the more valuable seismic hazard assessment becomes. Unfortunately, the more densely populated the region, the more difficult it is to collect reliable data. This is the challenge to the urban geodesist.

Recipe for High Precision

The following table describes conditions that contribute to high quality estimates of position during a GPS campaign, and why it is challenging to achieve these conditions in the urban environment:

Benchmark Stability

Explanation	If the location we survey is on unstable ground, the movement we record may not be related to fault motion at all.
Urban Challenges	Near-surface rocks in urban areas are often artificial fill which can be less stable.

Consistent Setup

Explanation	We must position our instruments in exactly the same location each time with less than a millimeter of tolerance - a job that requires a fair amount of patience, skill, and experience.
Urban Challenges	Even with proper safety gear, try achieving millimeter-level precision in the middle of a busy intersection!

Unobstructed Sky View

Explanation	GPS receivers determine position by recording broadcasts from satellites orbiting overhead. Overhead obstructions can therefore block or distort the signal.
Urban Challenges	Tall buildings, fences, trees, buses, etc.

Long Duration Measurements

Explanation	We must collect data for at least 6 to 72 hours continuously. During that time, the instrument must not be disturbed and cannot move even a millimeter.
Urban Challenges	An operator must guard the instrument from both theft and disturbance during the entire time. In busy settings with passing cars, the vibrations can often move the instrument slightly. When placed on asphalt, the instrument can sink into the pavement on hot days.

Berkeley Seismological Laboratory
215 McCone Hall, UC Berkeley, Berkeley, CA 94 720-4760
Questions or comments? Send e-mail: www@seismo.berkeley.edu
© 2002, The Regents of the University of California.