S51B-1739 INVITED
The 21 February 2008 Wells, Nevada, USA Earthquake-Impacts of a Major Background Earthquake on a Rural Community
The 2008 Wells, Nevada Earthquake (Mw 6) was a background event that did not rupture the surface and occurred on a fault that was previously unrecognized as a seismic hazard. The earthquake occurred just north of the town of Wells and residents generally reported about 20 to 40 seconds of shaking. Ground motion caused severe structural and nonstructural damage to several older buildings, especially two-story buildings. About 60% of the unreinforced masonry buildings were damaged, causing bricks and concrete crowns from walls and parapets to fall onto sidewalks, alleys, and adjacent buildings. Newer construction generally faired well, but commonly had cosmetic interior cracks. Over 60 masonry chimneys (approximately 10% to 15% of the total chimneys) were broken or thrown down and most homes and businesses suffered the some content loss. There were no deaths and only a few minor injuries associated with the event, partly because it occurred in the morning when many people were still home. Damage to the town's infrastructure included water-main breaks, two home propane-line leaks, a few electric-line breaks, and a couple of sewer- line breaks. One large propane tank rolled over, sheared off its valve, and leaked liquid propane, creating a critical-response situation. Several objects slid, fell, or were shaken in different dominant directions. The people of Wells, Elko County, and neighboring Utah and Idaho used an effective pioneering spirit to help the community respond and recover.
S51B-1740
Implementation of IP Telemetry in Support of Portable Deployments for Earthquake Response
IP spread spectrum radios have revolutionized the operation of remote seismic networks. In two separate deployments this year, the Nevada Seismological Laboratory implemented 900 MHz point-to-multipoint IP radio systems for portable seismographs in response to two important Nevada earthquake sequences: the Mw 6.0 event that struck Wells on February 21, 2008; and an energetic earthquake swarm in urban Reno that began in mid-February (mainshock Mw 5.0, April 26, 2008, 06:40 UTC). In cooperation with the USGS, ten portable stations were deployed in the Wells area response. Also, 10 IRIS RAMP instruments were included in the urban Reno deployment. These instruments were outfitted with Motorola Canopy radios and integrated with the regional telemetry infrastructure. As configured, these radios will support a large deployment, high sample rate dataloggers, and a flexible network topology with a working range of at least 30 miles. Real time IP telemetry can improve portable network performance in the following areas: 1. Simplified data flow- Real-time data from portable deployments is integrated with the regional and national networks. Portable instrument data does not have to be retrieved from the field, extracted from mass storage, and separately incorporated into data archives. The need to reanalyze events as locally-recorded portable data becomes available is eliminated. 2. Improved real-time products- Real-time data from portable stations can be used to improve the precision and timeliness of data products (e.g., ShakeMap) for the public, the local and national media, and emergency managers. 3. State-of-health monitoring- Systems (power, memory, etc.) can be monitored, allowing for less frequent and better targeted maintenance visits. The monitoring of these parameters can then be assumed by software packages such as Nagios or SeisNetWatch. 4. Remote management- Datalogger parameters can be managed remotely. The radios can also be remotely managed, allowing for change of center frequency, power level, etc. Field management of radios and dataloggers was conducted with mobile web applications developed at NSL (Slater and others, 2008). 5. Improved data quality- Quality of data can be improved with real- time feedback regarding sensor performance and clock status. Completeness of records is improved by decreased response time to station problems and redundancy of data collection (telemetry and local recording).
S51B-1741 INVITED
The 2008 Mw 6.0 Wells, Nevada Earthquake Sequence
The Mw 6.0 February 21, 2008 (06:16 AM PDT) Wells, Nevada normal faulting earthquake occurred in Town Creek Flat about 8 km northeast of the small community of Wells. A preliminary set of about 1000 aftershock relocations clearly defines a 55-60 degree southeast dipping fault plane. The structure projects to the surface along the southern end of the Snake Range, although no surface offsets have been identified. The earthquake occurred east of the Ruby Mountains and Snake Range west dipping range front faults, possibly on a northern extension of an east dipping normal fault system on the eastern side of the East Humbolt Range. The depth of the mainshock is estimated to be 10.5 km with the aftershock sequence extending to about 15 km. Typical of moderate sized Basin and Range earthquakes, the early aftershock period included several earthquakes of M > 4 and these were felt strongly by the residents of Wells. From the preliminary relocations, the source radius of the mainshock is estimated to be about 4 km, resulting in an estimated displacement of 55 to 83 cm and static stress drop of 72 to 86 bars, depending on the seismic moment estimate used. Aftershock relocations suggest a radial rupture mechanism. Fortunately, the EarthScope USArray network was operating in Nevada at the time of the event and provided unique controls on the mainshock and early aftershock locations. The earthquake occurred in an area of relatively low seismic hazard and the only permanent seismograph in the region was the U.S. National Network broadband station east of the Ruby Mountains south of Wells. The University of Utah and University of Nevada deployed locally recorded strong motion instruments in the Wells area. Also, an 8 station IP telemetered strong motion network, jointly deployed by the U.S. Geological Survey and University of Nevada Reno, provided real-time data for quick high-quality aftershock relocations and ground motion estimates. In addition, the University of Utah established several telemetered analog stations for improved aftershock locations. IP data communications was routed through the Nevada Department of Information Technology microwave communications site north of Wells. The aftershock deployment was not possible without the considerable support of a number of public and private agencies in the Wells area and the Wells community itself. Many unreinforced masonry structures in old-town Wells, dating to the early 1900's, experienced significant damage. There was also damage to homes and businesses within the community, including the local High School, but fortunately there were no serious injuries associated with the earthquake.
S51B-1742
Finite-Fault Source Study of the 2008 Wells, Nevada Earthquake Using Regional Broadband Empirical Green's Functions
The M~6 normal-faulting earthquake of 21 February 2008 northeast of Wells, Nevada was widely recorded by numerous seismic stations of the US Transportable Array. Aftershocks following the earthquake were also recorded by many of the USArray sites, offering an unprecedented opportunity for a detailed analysis of the source using regional empirical Green's functions. We have examined the three-component broadband waveforms recorded at 46 stations located within 300 km of the mainshock epicenter using a finite-fault methodology that prescribes subfault responses based on regional recordings for 8 different aftershocks. Mainshock records are filtered using a 2-50 sec passband and inverted using a fault plane with an area of 20-km by 20-km oriented NE-SW consistent with the trend of the observed aftershock zone. The empirical Green's function method yields excellent fits to the complex regional waveforms and identifies a single, relatively compact source (6 km by 4 km) located mostly updip of the hypocenter. The maximum inferred slip is 88 cm and the calculated seismic moment is 6.2 x 1024 dyne-cm, corresponding to an estimated magnitude of 5.8 Mw. The compact nature of the inferred source is in sharp contrast to the September 2004 M~6 strike-slip earthquake near Parkfield, California, which had a more unilateral rupture extending at least 20 km northwest from the hypocenter. Our finite-fault results yield a stress drop of approximately 200 bars for the Wells earthquake, suggesting that Basin and Range normal faults are capable of producing high stress-drop seismic events. The empirical Green's function methodology used in conjunction with a finite-fault inversion is a valuable procedure for studies of moderate-sized earthquakes with limited teleseismic data or in areas with little or no strong motion records.
S51B-1743 INVITED
Finite-Source Study of the February 21, 2008 Mw 6.0 Wells, Nevada, Earthquake
The February 21, 2008, Mw 6.0 Wells, Nevada, earthquake was well recorded by the NSF EarthScope Transportable Array (TA). This event occurred in an area with historically low seismicity, and unfortunately due to its proximity to the town of Wells, it inflicted considerable damage to the unreinforced masonry buildings of the historic district. We use broadband, three-component displacement records from the TA to invert for kinematic finite source models. The seismic moment tensor analysis shows that this event occurred on a northeast striking normal fault, and preliminary finite-source analysis shows that the causative structure is the east-dipping fault plane. The rupture appears to be bilateral however slip to the southwest, in the direction of Wells, Nevada, was appreciably higher. We investigate the sensitivity of the finite-source solutions to the fault geometry, and the distribution of available data. Unfortunately there were no near- source strong motion stations, however we demonstrate that the finite-source model obtained from regional distance records may be used to effectively simulate the level of strong ground motion (peak ground velocity) in the near-fault region (e.g. Dreger and Kaverina, 2000). Additionally, we relate simulated ground motions for the town of Wells with observations of heavy object sliding and damage patterns. Finally, we present inversions incorporating InSAR observations.
S51B-1744
The 2008 Wells Earthquake, Nevada from Envisat ASAR data
An earthquake of magnitude 6.0 struck Wells, Nevada on February 21st, 2008 at 6:16AM PST. The epicenter was located approximately 10 km northeast of the city of Wells with a depth of 6.7 km according to the Nevada Seismological Laboratory. Over 20 buildings were reported severely damaged, over 700 buildings lightly damaged and more than three people were injured (NEIC). It was Nevada's most destructive earthquake since the magnitude 7.1 earthquake in December, 1954 near Dixie Valley. The Wells earthquake epicenter is covered by four Envisat ASAR tracks with multiple coseismic pairs of images on each. One descending pair from track 127 (baseline 210 m, spanning 140 days) and one ascending pair from track 220 (27 m, 140 days) were used in our analysis. Interferograms were processed using ROI_PAC and a digital elevation model from the Shuttle Radar Topography Mission. Our data show a four-fringe deformation signal in both pairs, which implies approximately 12 cm of peak downward displacement of the ground. We subsample our data using a quadtree decomposition and run inverse elastic dislocation models using the 'okinv' code in order to find a best-fitting set of fault parameters. We model nodal planes dipping SE and NW; however we prefer the model with a SE-dipping fault (strike 31, dip 33, rake -97, depth 3.2-9.8 km, Mw 6.0), which has a smaller misfit to data and agrees with aftershock locations, which also favor a SE dip (http://www.seismo.unr.edu/feature/2008/Preliminary_relocations1.pdf). This preliminary model has a dip of 33 degrees which is shallower than that obtained by seismic waveform inversion (e.g. global CMT, UC Berkeley).
S51B-1745
InSAR Analysis of the 2008 Wells, Nevada Earthquake
The February 21, 2008 Wells earthquake (M 6.0) was centered 10 km northeast of the Town of Wells, Nevada in Town Creek Flats, a small basin bounded on the west by the Snake Mountains and on the east by the Windermere Hills. The only mapped faults of Quaternary age in the basin are a series of north- to northeast-trending, west-dipping faults along the west flank of the Windermere Hills. Based on the geometry of these faults and the normal, northeast-oriented focal mechanism of the main event, it was initially assumed that the co-seismic slip had occurred on a northeast-striking, west-dipping fault. InSAR analysis of pre- and post-earthquake Envisat data in the WInSAR GeoEarthscope archive, however, indicates that the fault slip occurred on a northeast-striking, southeast-dipping fault bounding the eastern flank of the Snake Mountains. Both descending and ascending pairs delineate a similar oval-shaped deformation bowl about 15 km in width and 25 km in length, centered on the main event epicenter, and exhibiting more than 15 cm of subsidence. The similarity of the ascending and descending pairs indicates that there was little to no lateral displacement associated with the event. Displacement profiles through the zone of maximum surface deformation indicate that the subsidence bowl is asymmetric, with the greatest and steepest deformation occurring near the range front of the Snake Mountains and then gradually decreasing across the basin to the east. The profiles further indicate slight (1-2 cm) uplift on the footwall of the fault, an observation consistent with other normal faulting events in the Basin and Range. We applied inverse modeling to the results, in which fault parameters are selected to best estimate observed InSAR deformation patterns. The modeling shows that the deformation pattern is best fit by a normal fault striking N35E and dipping 45 degrees to the southeast. Fault slip is modeled to be 75 cm on a 4 km x 7.5 km fault patch, and the calculated moment magnitude is Mw 5.9. The InSAR results agree with independent seismologic data, including a well-defined aftershock pattern, and indicate that the earthquake occurred on a previously unknown Quaternary fault located along the eastern flank of the Snake Mountains. No surface rupture was associated with the earthquake, but field investigations found evidence for a small (15-30 cm) scarp that is likely of Holocene age. The seismogenic fault coincides with a previously mapped Paleozoic thrust fault and suggests that the earthquake may be an example of the re-activation of older, inactive bedrock structure in the present-day extension-dominated Basin and Range.
S51B-1746
Coulomb Stress Analysis of the 21 February 2008 Mw= 6.0 Wells, Nevada, Earthquake
We calculated static Coulomb stress changes associated with the February 21, 2008 Wells, Nevada earthquake to assess the stress transfer to surrounding active faults. We use a rectangular source for the earthquake striking 34° and dipping 40° to the SE, centered at 41.153°N and 114.867°W, with 56 slip patches consistent with the Dreger and Ford (2008) variable-slip rupture model. The geometry of nearby faults is defined by the Frankel et al. 2002 National Seismic Hazard Map active fault database. The active faults in the study area are thought to dip about 60°, with normal slip at rates 0.15-0.5 mm/yr. We also mapped Coulomb stress changes on the rupture surface and its up-dip, down-dip and northern and southern extensions; the observed annulus of aftershocks corresponds to calcualated stress increases off the fault rupture. Both the largest stress increase (+0.3 bar) and largest decrease (-0.7 bars) are calculated on the Independence fault. Stress changes on other faults are less than 0.01 bars, which are likely insignificant. The greatest effect of Wells earthquake is to inhibit Coulomb failure on the northern half of the Independence fault, largely because of fault clamping. Modest promotion of failure is seen along the middle and southern parts of the Independence fault, the Spruce fault and the northern segment of the Ruby Mountain fault.
S51B-1747
GPS Constraints on Eastern Nevada Basin and Range Crustal Deformation Before and During the February 21, 2008 M6.0 Wells, NV Earthquake
The 21 February 2008 Wells, Nevada earthquake occurred in an area of pervasive Quaternary-age Basin and Range-style normal faulting. To characterize the far-field co-seismic displacement field and secular crustal strain rates preceding the event, we have analyzed data from continuously recording GPS sites that span 300 km of eastern Nevada. Four sites within ~90 km of the epicenter recorded significant coseismic southeast displacement southeast of the epicenter, grading into west displacement west of the epicenter. For example, the site GOSH recorded horizontal displacements of 0.7±0.2 mm south and 1.1±0.4 mm east. The signal is consistent with that predicted from seismic and InSAR data, suggesting that both far-field and near-field data provide similar estimates of the earthquake source. The occurrence of an M6.0 in eastern Nevada was somewhat surprising, because some past geodetic studies have concluded that the background secular crustal strain rates in the central Nevada Basin and Range are not significantly different than zero (to within the uncertainty of a few nanostrains/yr). Hence this part of Nevada was termed by some to be a "geodetic microplate" to recognize its apparent rigidity. However, since then several continuous sites of the EarthScope Plate Boundary Observatory have been installed, and the average length of time series on BARGEN continuous GPS sites has increased to nearly 11 years so uncertainties in rates are now substantially less than in previous studies. For the longest recording sites within 300 km west of the Nevada/Utah border (BARGEN sites ELKO, RUBY, GOSH, MINE, MONI, EGAN, and FOOT) we estimated rates of secular motion. We found that westward and northward velocities increase with distance west of the border by ~1 mm/yr and ~0.8 mm/yr respectively, in a North America reference frame. We used these rates to simultaneously solve for rigid body rotation and uniform strain rate. We found extension rates of 4.6±0.8 nanostrains per year oriented N59°W and contraction rates of 2.8±1.1 nanostrains/per year oriented N31°E (uncertainties scaled so that the normalized residual scatter in the velocities after removing the strain rate model is unity). This suggests that eastern Nevada experiences secular crustal deformation that is 1) significantly greater than zero, 2) has an orientation similar to the prevailing azimuth of shear in the Pacific/North America plate boundary zone, and 3) has direction of extension similar to the direction of coseismic extension during the earthquake.
S51B-1748
Geophysical Setting of the February 21, 2008 M6 Wells Earthquake, Nevada, and Implications on Earthquake Hazards
The February 21, 2008 M6 Wells earthquake, centered about 10 km northeast of Wells, Nevada caused considerable damage to local buildings, especially in the historic downtown area. The earthquake occurred on a previously unmapped normal fault, and preliminary relocated events indicate a fault plane dipping about 55 degrees to the southeast (K. Smith, written commun., 2008). The epicenter lies near the intersection of north-northeast-trending Basin and Range bounding faults along the Ruby Mountains (Ruby Mountains fault system and Ruby Valley fault zone) and the northerly-trend of the Spruce Mountain Ridge and Independence fault zones. Regionally, the epicenter is approximately on trend with a geophysically defined crustal boundary on the east side of a V-shaped basement gravity high (e.g., Ponce and Glen, 2002; 2008). In addition, the Wells earthquake lies just east of a series of north-northeast-trending basins associated with Diamond, Huntington, and Lamoille Valleys along the western margin of the Ruby Mountains that extend from Elko to northwest of Wells. A regional depth to basement map, derived from the inversion of gravity data, indicates that a small oval basin north-northeast of Wells and just north of Wood Hills, may reach a depth of more than 2 km. Gravity and physical property data were collected in the vicinity of Wells during the summer of 2008 in order to improve the data coverage, which was relatively sparse throughout this part of northeast Nevada, especially on bedrock. These data will be used to better constrain the geophysical setting of the Wells earthquake and help assess the seismic hazard in this area. In particular, we will compute an updated depth to basement map in order to determine the geometry of the Wells basin which may yield information on whether or not the size and shape of the basin contributed to ground shaking. Modeling of magnetic and gravity anomaly data will help to locate subsurface structures or concealed faults that may be related to the Wells earthquake. Although large earthquakes in this part of Nevada are infrequent, the Wells earthquake indicates that normal faults are active in this region and are capable of generating large earthquakes.
S51B-1749 INVITED
What Can We Learn From the Wells, NV Earthquake Sequence About Seismic Hazard in the Intermountain West?
The February 21, 2008 Wells, NV earthquake (M 6) was felt throughout eastern Nevada, southern Idaho, and western Utah. The town of Wells sustained significant damage to unreinforced masonry buildings. The earthquake occurred in a region of low seismic hazard with little seismicity, low geodetic strain rates, and few mapped faults. The peak horizontal ground acceleration predicted by the USGS National Seismic Hazard Maps is about 0.2 g at 2% probability of exceedance in 50 years, with the contributions coming mostly from the Ruby Mountain fault and background seismicity (M5-7.0). The hazard model predicts that the probability of occurrence of an M>6 event within 50 km of Wells is about 15% in 100 years. Although the earthquake was inside the USArray Transportable Array network, the nearest on-scale recordings of ground motions from the mainshock were too distant to estimate accelerations in town. The University of Nevada Reno, the University of Utah, and the U.S. Geological Survey deployed portable instruments to capture the ground motions from aftershocks of this rare normal-faulting event. Shaking from a M 4.7 aftershock recorded on portable instruments at distances less than 10 km exceeded 0.3 g, and sustained accelerations above 0.1 g lasted for about 5 seconds. For a magnitude 5 earthquake at 10 km distance the NGA equations predict median peak ground accelerations about 0.1 g. Ground motions from normal faulting earthquakes are poorly represented in the ground motion prediction equations. We compare portable and Transportable Array ground-motion recordings with prediction equations. Advanced National Seismic System stations in Utah recorded ground motions 250 km from the mainshock of about 2% g. The maximum ground motion recorded in Salt Lake City was in the center of the basin. We analyze the spatial variability of ground motions (rock vs. soil) and the influence of the Salt Lake Basin in modifying the ground motions. We then compare this data with the September 28, 2004 Parkfield aftershocks to contrast the differences between strike-slip and normal ground motions.