The PBO Borehole Strainmeter Program: Assessing Strainmeter Performance
Between June 2005 and September 2008 UNAVCO installed 74 borehole strainmeters as part of the Plate
Boundary Observatory, the geodetic component of the Earthscope program. Borehole strainmeters, with sub-
nanostrain sensitivity, were included in the observatory with the purpose of detecting short-lived strain
transients along the Western US plate boundary. Less than a year after completion the strainmeter network
has already provided a catalogue of plate boundary strain signals such as the recording of three Cascasdia
episodic tremor and slip events, several aseismic creep events in Parkfield, California, and strain transients
captured during the December 2008 Yellowstone earthquake swarm.
PBO borehole sites are multi-instrumented stations, all collect strain, seismic and environmental data while
volcanic sites include borehole tiltmeters. Combining the strain, seismic and tilt data with the measurements
from over 800 PBO GPS sites provides an unprecedented continuous 3-dimensional record of plate boundary
deformation processes. Knowledge of each strainmeter's performance is important, however, in deciding
which strainmeters should be included when analyzing strain signals. We assess how well each PBO
strainmeter, with at least 1 year of data, performs in three different frequency bands: the seismic, tidal and the
long-term band of months to years. The metrics used are: the ability to record seismic shear signals and
microseisms, instrument self consistency in measuring areal and shear tidal signals, the difference between
observed areal and shear tides and those predicted by earth tide and ocean load models, the magnitude and
variation in barometric response with time, the absence of steps in the data, the impact of cultural noise and
state of borehole compression. Knowing how each strainmeter performs in these bands will allow a data user
to select the best set of strain measurements to work with given the signal frequency they are interested in.
Transient Strain During and Between Northern Cascadia Episodic Tremor and Slip Events From Plate Boundary Observatory Borehole Strainmeters
Plate Boundary Observatory (PBO) borehole strainmeters provide unique information about the Episodic Tremor and Slip (ETS) events that have recurred every 12-16 months in the reach of the Cascadia subduction zone from south Puget Sound to northern Vancouver Island. PBO strainmeters at seven locations have now recorded the ETS events in January, 2007 and May, 2008 that propagated the entire length of this reach. During both of these events, strain transients were observed as tremor activity migrated past each strainmeter along the strike of the subducting slab. Even when deformation was observed on only one strainmeter at a time, the ability to record two shear strain components constrains the propagation direction and the sense of slip. In particular, the time histories and net offsets of differential extension and engineering shear strain are, like the GPS time series, consistent with thrust displacement on the surface of the subducting slab between 35 and 45 km deep, although the strainmeter network by itself is too sparse to provide tight constraints on the depth. The PBO borehole strainmeters can detect smaller slip events than GPS at periods of a few weeks, which is the approximate duration of a recurrent northern Cascadia ETS event. The strainmeters have shown that slip has accompanied at least four episodes of tremor on northern Vancouver Island or in the south Puget Sound area between the larger, recurrent ETS events. Little or no surface displacement from these inter-ETS tremor bursts was evident in GPS time series. An interesting feature of the January, 2007 and May, 2008 northern Cascadia ETS events is that both followed relatively long interevent intervals of 16.4 and 15.7 months, respectively. However, tremor bursts in November, 2006 on Vancouver Island, and in March, 2008 near south Puget Sound occurred when the mean ETS recurrence time of 14 months had elapsed. PBO borehole strainmeter data show that slip accompanied both of these bursts, suggesting that tremor and slip events initiating near the ends of the recurrently slipping zone may not always grow into full northern Cascadia recurrent ETS events.
Northern Cascadia Episodic Tremor and Slip: Cycles Within Cycles
Episodic tremor and slip (ETS) events, each with geodetically determined moment magnitudes in the mid-6 range, repeat with remarkable regularity every 15 months under the Olympic Peninsula/southern Vancouver Island region. We have automatically searched for non-volcanic tremor in all 5-minute time windows both during the past 4 ETS events and during the inter-ETS period from February, 2007 through April, 2008. Inter- ETS tremor was detected in nearly 3000 windows, which overlap by 50%, so tremor was seen 2% of the time. The catalog of 5-minute tremor locations cluster in time and space into groups we call tremor swarms, revealing 35 inter-ETS tremor swarms. The number of hours of tremor per swarm ranged from about one to 50 hours, totaling 193 hours. The inter-ETS tremor swarms generally locate along the downdip side of the major ETS events, and account for approximately 45% of the time that tremor has been detected during the last entire ETS cycle, which includes the May, 2008 ETS episode. Many of the inter-ETS events are near-carbon copies in duration, spatial extent and propagation direction, as is seen for the larger 15-month-interval events. These 35 inter-ETS swarms plus one major ETS episode follow a power law relationship such that the number of swarms, N, exceeding duration τ is given by N ~ τ-0.6. If we assume that seismic moment is proportional to τ as proposed by Ide et al. [Nature, 2007], we find that the tremor swarms follow a standard Gutenberg-Richter logarithmic frequency-magnitude relation, log10N ~ 10-bMw, with b = - 0.9, which lies in the range for normal earthquake catalogs. Furthermore, the major ETS events fall on the curve defined by the inter-ETS swarms, suggesting that the inter-ETS swarms are just smaller versions of the major 15-month ETS events. Only the largest events coincide with geodetically observed slip, suggesting that current geodetic observations may be missing nearly half of the total slip. Finally, crude estimates of the spatial dimensions of tremor swarms L suggest that L ~ τ1/n where n is between 2 and 3. A value of 2 is consistent with slip propagation rates being controlled by a diffusional process. In contrast, n is observed to be about 1 for normal earthquakes because rupture generally propagates at a velocity close to the shear-wave speed.
Absolute Gravity Monitoring in the Northern Cascadia Subduction Zone: an Update
High-precision absolute gravity (AG) observations are sensitive to vertical motion of the observation site as well as mass redistribution within (and below) the underlying, slowly-deforming crust. On southern Vancouver Island, situated in the northern Cascadia forearc, long-term absolute gravity measurements have been made typically four times per year for over a decade at four sites. These AG sites are all co-located with continuous GPS stations of the Western Canada Deformation Array. For the four sites on southern Vancouver Island the comparison between the long-term observed gravity trends and vertical GPS rates indicates a linear g-dot to h- dot relationship that is appropriate for a subduction zone. However there appears to be a bias between the gravity rates and the GPS rates. The potential cause of this bias is not at present understood. In terms of shorter term phenomena, the monitoring of subduction zone Episodic Tremor and Slip (ETS) has been carried out primarily using seismic data for tremor and continuous GPS observations for transient slip. More recently, the establishment of the Plate Boundary Observatory has added borehole instrumentation consisting of borehole strainmeters (BSM) as well as pore pressure gauges. For sites on southern Vancouver Island, the regularity of ETS episodes has allowed us to schedule extended periods of continuous AG measurements to augment these other data and thereby help in understanding the fundamental physical processes involved in the generation of ETS. Our observations to date indicate that ETS in the Cascadia Subduction Zone is accompanied by gravity change that is likely caused by mass redistribution rather than height displacement. Future work is planned to determine the detailed gravity signature of a specific ETS event and to determine whether the signal is seen at other sites.
Current Tectonics of Northern Vancouver Island, Southern Queen Charlotte Islands and the Adjacent Mainland
The area south of the Queen Charlotte Islands and north of Vancouver Island is characterized by the transition from the Cascadia subduction zone to the Queen Charlotte transform fault zone. The tectonic setting involves the Pacific, North American, Juan de Fuca, and Explorer plates, and the Winona block, as well as the Queen Charlotte and Revere-Dellwood-Wilson faults, Explorer ridge, Nootka fault, and Cascadia subduction zone. On the basis of GPS campaign data from 1993 to 2008 we derive a crustal velocity field for North Vancouver Island and the adjacent mainland. This velocity data is the basis for interpretation of the tectonics of the transition from the convergent to transform boundaries. Our GPS data show significant shear velocities in the Bella Coola region, ~250 km inland from the Queen Charlotte fault, although there is no seismic activity in the area. We use geodynamic models to better understand the discrepancy between the GPS data and the seismic data. We use the GPS velocities to determine whether the measured deformation rates of northernmost Vancouver Island, related to its interaction with the Explorer Plate and possibly the Queen Charlotte transform margin, are transient or permanent. Geodynamic models are used to find out if deformation in the region including North Vancouver Island, Queen Charlotte Islands, and the adjacent mainland (Coast Shear Zone) is transient or long-term. To constrain the model, we use the rheology and structure of the region, with reasonable values for elastic thickness and viscosity. Two end-member models describing how the Pacific/North America plate convergence is accommodated off the Queen Charlotte Islands have been developed by others. They assume either internal crustal shortening or underthrusting of the Pacific plate. With the new GPS data we can further investigate which model explains the tectonic situation more appropriately. An earlier model strongly suggests an underthrusting fault fully locked down to 14 km depth, followed by a transition zone to 20 km depth. We are also looking at the contemporary transform motion between the Pacific and North American plates, and how far inland this motion can be accommodated. It is possible that this deformation zone extends up to about 400 km inland and could include the Coast Mountains and the western part of the Central Cordillera. Alternatively, the shear may represent elastic build-up to be released by large future earthquakes.
Strain Partitioning in the Northern Walker Lane and Western Basin and Range from GPS Measurements
The northern Walker Lane, in the western Basin and Range Province of the United States, is a complex system of dextral, normal and sinestral faults that work together to accommodate approximately 9 mm/yr of relative motion between the Sierra Nevada/Great Valley block and the more slowly extending Province. GPS measurements made using the BARGEN, EarthScope PBO and MAGNET GPS networks since 2004 are now providing improved resolution of deformation patterns and crustal fault slip rates inside the Walker Lane and western Basin and Range. We have processed all the GPS data as part of a uniform global solution, and filtered the solution on a Great Basin spatial scale to obtain rates of motion of the Walker Lane crust with respect to North America. Using these rates we have constrained slip rates on regional faults using a block model whose boundaries conform to Quaternary surface rupture geometries. These results show a very strong correlation between the geologic domains and style of strain measured with GPS. In particular, east of the Walker Lane, where the topography and crustal faulting are characteristic of classic Basin and Range tectonic extension, the GPS velocities show a highly uniform southeast to northwest uniaxial extension of 2.5 mm/yr distributed over 250 km. This uniform extension implies normal slip rates of approximately 0.1 mm/yr on average for each fault (horizontal extension). The transition between Basin and Range morphology and the Walker Lane is matched in the GPS velocities by a transition from uniaxial extension to transtension that is resolved into dextral slip on northwest trending faults, with minor contributions from left lateral slip on northeast striking faults and normal slip. Right oblique extension is well-distributed across the Walker Lane, with most faults contributing some slip to accommodate the overall slip budget. The greatest slip rates occur on the western and eastern margins, and by far the greatest amount of normal slip occurs in the westernmost fault systems near the Sierra Nevada crest, Lake Tahoe, and Carson Range faults where horizontal extension rates normal to the fault are as high as 1.8 mm/yr. Normal slip rates elsewhere in the Walker Lane are similar in rate to the rest of the Basin and Range. The Mohawk Valley fault slips faster than any fault in the entire system, about 2.8 mm/yr of dextral slip. These results suggest that contemporary extension on normal faults nearest the Sierra Nevada Range Front may drive the seismic hazard for the nearby Reno/Tahoe metropolitan areas.
Re-interpretation of Geodetic Velocity Data Near 'the Great Bend', Southern California
A new analysis of geodetic data collected and compiled by Shen et al. (1996) in the Los Angeles area of California suggests that the interseismic velocity field can be best interpreted as elastic deformation above a deep-seated transcurrent shear zone that trends about 10º clockwise from the surface expression of the San Andreas Fault within the Great Bend. The Bend would then be a transpressive jog only for the upper crust. The transpressive deformation of that upper crust appears to be mostly accomplished through discrete seismic motions and thus not seen in the interseismic velocity field. This new analysis relies on re-calculating velocities in a reference frame in which the direction of the shear zone itself is invariant and on then finding the shear zone direction that best fits a heterogeneous simple shear model.
Major and micro seismo-volcanic crises in the Asal Rift, Djibouti
The Asal-Ghoubbet Rift is located on the eastern branch of the Afar triple junction between the Arabia, Somalia, and Nubia tectonic plates. The last major seismo-volcanic crisis on this segment occurred in November 1978, involving two earthquakes of mb=5+, a basaltic fissure eruption, the development of many open fissures across the rift and up to 80 cm of vertical slip on the bordering faults. Geodetic leveling revealed ~2 m of horizontal opening of the rift accompanied by ~70 cm of subsidence of the inner-floor, consistent with models of the elastic deformation produced by the injection of magma in a system of two dykes. InSAR data acquired at 24-day intervals during the last 12 years by the Canadian Radarsat satellite over the Asal Rift show that the two main faults activated in 1978 continue to slip with periods of steady creep at rates of 0.3-1.3 mm/yr, interrupted by sudden slip events of a few millimeters, in 2000 and 2003. Slip events are coincident with bursts of micro earthquakes distributed around and over the Fieale volcanic center in the eastern part of the Asal Rift. In both cases (the 1978 crisis and micro-slip events), the observed geodetic moment released by fault slip exceeds by a few orders of magnitude the total seismic moment released by earthquakes over the same period. Aseismic fault slip is likely to be the faults response to a changing stress field associated with a volcanic process and not due to dry friction on faults. Sustained injection of magma (1978 crisis) and/or crustal fluids (micro-slip events) in dykes and fissures is a plausible mechanism to control fluid pressure in the basal parts of faults and trigger aseismic slip. In this respect, the micro-events observed by InSAR during a 12-year period of low activity in the rift and the 1978 seismo-volcanic episode are of same nature.