Creep at Temescal Park: 4.1 mm/year 1995-1998.
37deg 59.35N, 122deg 21.27W
The creepmeter crosses the fault obliquely at 30 degrees near the SE entrance of Temescal Park at the intersection of Highway 24 and Highway 13 between Berkeley and Oakland.
Data from the creepmeter shows essentially linear creep 1995-1998 at a rate of 4.06 mm/year with a statistical uncertainty of 0.02 mm. This rate is applicable to the 15 m, fault-normal span of the creepmeter. (It is known from survey measurements that more than 80% of Hayward fault creep is manifest within 15 m of the fault) The creep-rate varies by 0.2 mm/year when arbitrary 6 month blocks of data are analysed. Flooding of the instrument caused temporary outages in winter months (boxes on figure below) requiring substantial editing.
The geometry of the creepmeter was modified in 9 Oct 1997. Prior to that time the creep signal was obtained by summing the outputs of two transducers, one at each end of the creepmeter. After October 1997 two independent transducers record creep at the east end of the creepmeter. The sensitivity of these two transducers differ by a factor of four. Boxes in the plot indicate times at which the instrument was flooded.
The above figure illustrates the geometry used in the Temescal creepmeter 1985-1997, and that used subsequently. The rod is buried at a depth of 1 m underground, protected from the soil by a telescopic PVC pipe, but in the above figures it is shown above ground for clarity. In the upper panel the rod/pipe assembly is free to move within the fault zone. In the lower geometry the rod is fixed to one end. The letters and double-arrows labeled E and W represent transducers at the east and west ends of the creepmeter respectively. The bold vertical lines represent attachment piers (these consist of buried 10 m tripods of helical screw piles welded at their upper end). Differential expansion between a steel pipe and the invar rod provides a measure of thermally-induced length changes in the invar rod (measured by transducer, T).
Temescal Park Creepmeter History
Current and past calibration information is embedded in the following table. The first column indicates the stored name in the Menlo Park data base, and the fifth column indicates the dextral slip calibration. If the stored number in the data base is x mV, the slip on the fault in mm is x/k where k is the negative number store in the fifth column.
| Temescal | 30 m long | 37.9892 | 122.3545 | 30° to fault |
| from 971009 | designation | volts/inch | mV/mm | dextral mV/mm |
| ctmt 7864 | low gain | 5.15 | -202.76 | -233.2 |
| ctme 22572 | temperature | 20.714 | -815.51 | -937.8 |
| ctmw 22573 | high gain | 20.485 | -806.50 | -927.5 |
| Creep= | ctmw/-927.5 | plus thermal | correction | ctme/-9378 |
| Temescal | Transducer history | volts/inch | mV/mm | dextral mV/mm |
| ctmt 22336 | 950804-980312 | 20.57 | 809.8 | -931.3 |
| ctme 22335 | 950804-971009 | 20.81 | 819.21 | -942.1 |
| ctmw 22337 | 950804-970913 | 20.53 | 808.46 | -929.7 |
| ctmw 22569 | 970913-980312 | 20.634 | 812.36 | -934.2 |
The remainder of this page contains details of transducer calibration over the period 1995-1998. The data are given as displacment of the tLVDT trasnducer. To convert displacement data to dextral creep, multiply by 1.15 ( i.e.1/cos30°). Note all the transducers are arranged such that dextral creep causes the signal to go negative. To accommodate dextral creep the transducers are set during maintenance visits near the DCP maximum input voltage (+5V) and tend toward zero as dextral creep proceeds. The piers are helical piers driven to refusal (rock at 3 m at the east end, and soil at 10 m at the west end. The transducer crosses a buried high pressure gas line near the west end.
The creepmeter is 30 m long and is installed at 30 degrees to the fault such that dextral slip causes extension. The measuring rod is buried at a depth of approximately 1 m.
March 1985-October 1997. An invar rod crosses the fault with two sensors measuring the changing distances between the ends of the rod and the piers. During the first two years of operation, transducers ctme and ctmw measured the east and west distances between the ends of the rod and the piers. Transducer ctmt measured the temperature expansion signal between a steel pipe and the invar rod. Creep in the first two years of operation is given by the following equation (after correction for calibrations given below).
Dextral creep=(-ctme-ctmw+ctmt/10)/cos30° (March 1995- October 1997)
An example of the processing involved is shown in Figure 4. The three colored traces tt, te and b are combined to yield a creep signal largely independent of thermoelastic noise. The rectangle in late February/early March indicates data lost during flooding
After Oct 1997 the rod was attached to the west pier and the transducer designations all changed such that ctme measures the differential temperature signal and ctmt is a low gain (+-2inch transducer), and ctmw is a high gain (+-0.5") transducer. Creep after October 1997 is given by either of the folowing equations. Note that the high gain creep signal is four times more sensitive than the other.
Low gain Creep=(ctmt-ctme/10)*-1.15 (October 1997 onward) or
High gain Creep=(ctmw-ctme/10)*-1.15 (October 1997 onward)
The thermal correction factor for 1997-98 is determined empirically to be ctme/2.3.
The two equations provide identical measures of creep (in case of transducer failure). The noise level of the low gain signal is closer to the 12-bit noise level of the A-D converter.
If required the transmitted range (low gain 25.4 mm, high gain 6.3 mm) can be doubled (low gain 51 mm, high gain 12.5 mm) by including a divide-by-two resistor network at the input of the DCP. All the transducers output +-10 volts whereas the Sutron DCP accepts only 0-5V. Note also that the Sutron DCP contains clamping diodes at its input which increasingly draw current such that a negative voltage on one channel more than -0.5V will reduce the voltage on the other channels. IT IS ESSENTIAL TO MAINTAIN THE INPUT TO THE SUTRON DCP IN THE RANGE ZERO TO +5V ON ALL CHANNELS! This requires re-setting the transducer body or the position of the core.
Should the creepmeter encounter a large dextral slip signal it will go off-scale as a negative voltage, however the low-gain transducer will indicate the slip faithfully to a maximum of -10 volts = 56 mm (=51 mm rod displacement= -10Volt negative).
The site is close to Temescal Creek which floods annually. During flooding the transducers and cabling are immersed for 7-10 days. Although precautions to prevent damage to electrical cabling have been improved, in the past three years flooding has caused lost data (if power is shorted) or noisy data (if electrical connections are corroded) requiring extensive editing of the recorded data. Various 'waterproof' junction boxes were tested and abandoned after failure. The transducers (which are designed to operate underwater) in September 1998 were wired directly to the DCP without underground junction boxes.
4 Aug 1995 in situ calibration
west end 22337 808.29 mV/mm
east end 22235 824.00 mV/mm
26 April 1996
ctmt 22336 calibration -809.80 mV/mm
ctme 22335 calibration -819.21 mV/mm
ctmw 22337 calibration -808.46 mV/mm
14 Feb 1997 Wiring flooded. New junction box installed . No changes to transducers.
Voltages from notes:
1 east ctme arrive 1.062 V depart 2.15 V
2 temp ctmt arrive 2.005 V depart 2.79 V
3 west ctmw arrive -0.895 V depart3.58V
Remarks- the channels were all rewired, a new junction box was installed at the east end, and the small junction box omitted at the west end. The labeling of the channels was accidentally reversed at during the rewiring ctme is actully ctmt and ctmt is actually ctme. This was not discovered until the next visit.
12 Sept 1997
Power to creepmeter at 10V so switched to DCP power. The labeling error is clarified: the data bank at Menlo Park has the east data in ctmt, and the west data in ctme and the temperature data in ctme). Channel labeled ctmw inoperative, needs rewiring to remedy corroded cable.
13 Sept 1997
East Transducer replaced (ctmw old 22337, now 22569) and temperature transducer reset.
22569 20.634V/inch 812.36mV/mm
9 October 1997 Changed to single ended creepmeter with high and low gain channels at east end. West end of invar-rod clamped rigidly to the western helical pier.
ctmt now 7864 5.15 V/inch (202.76mV/mm) (!!!labeled ctme at the DCP)
ctme 22336 -809.80 mV/mm (!!!labeled ctmt at the DCP)
ctmw 22569 -812.36 mV/mm
14 March 1998 Ground-wire to all vault sensors found corroded due to extensive flooding of the park in the past few months. Two transducers were shorted but the low gain transducer was OK. New direct cabling was installed to remove the underground junction box.
Channel 1 7864 ctmt is the low gain channel (!!!labeled ctme at the DCP) 5.15V/inch
Channel 2 old 22569, now 22573 is ctmw, the high gain channel 20.485 V/inch
Channel 3 old 22336, now 22572 is ctme, the bimetallic rod (!!!labeled ctmt at the DCP) 20.714 V/inch
Departing voltages
ctmt -202.76 mV/mm 4.9 volts measured 4.895 volts transmitted
ctmw -806.50 mV/mm 4.45 V transmitted 4.470 V
ctme -815.51mV/mm 2.42V transmitted 2.479 V
9 December 1998 visited midday to 12:30 local. Power supply +-15.25V. Reset ctmw and ctmt and disturbed ctme (the temperature sensing LVDT on the invar/steel rod assembly).
ctme bimetal (green) 2.40 V arrive 2.30 V depart
ctmw high gain (white) 2.35 V arrive 4.72 V depart
ctmt low gain (red) 4.45 V arrive 4.85 V depart