The Aftershock Sequence of the 2011 Mineral, Virginia, Earthquake: Locations, Focal Mechanisms, Regional Stress and the Role of Coulomb Stress Transfer Qimin Wu and Martin Chapman, Department of Geosciences, Virginia Tech, Blacksburg, VA, 24061, wqimin86@vt.edu, mcc@vt.edu

The aftershock sequence of the Mw 5.8, August 23, 2011 Mineral, Virginia earthquake was well-recorded by 36 temporary stations installed by several institutions. The detailed investigation of thousands of aftershocks resolves spatial details of the aftershock hypocenter distribution. And focal mechanisms of 393 aftershocks and stress inversion results exhibit substantial variability.

Aftershocks near the mainshock define a previously recognized tabular cluster with orientation similar to a mainshock nodal plane; other aftershocks occurred 10-20 kilometers to the northeast. A large percentage of the aftershocks occurred in regions of positive Coulomb static stress change and approximately 80% of the focal mechanism nodal planes were brought closer to failure. Moreover, the aftershock distribution near the mainshock appears to have been influenced strongly by rupture directivity. Aftershocks at depths less than 4 km exhibit reverse mechanisms with N-NW trending nodal planes. Most focal mechanisms at depths greater than 6 km are similar to the mainshock, with N-NE trending nodal planes. A concentration of aftershocks in the 4-6 km depth range near the mainshock are mostly of reverse type, but display a 90-degree range of nodal plane trend. Those events appear to outline the periphery of mainshock rupture, where positive Coulomb stress transfer is largest. The focal mechanisms of aftershocks at depths less than 4 km and those at depths greater than 6, along with the mainshock, point to the possibility of a depth-dependent stress field prior to the occurrence of the mainshock. Presumably, the aftershock process of this event is representative of other moderate to large shocks that have occurred and will occur in central and eastern North America, and a better understanding of the aftershocks of this event could shed more light on the state of stress in intraplate North America. The dataset and some of the preliminary results are available to public at the webpage of Virginia Tech Seismological Observatory (VTSO, www.magma.geos.vt.edu/vtso/).

http://www.magma.geos.vt.edu/vtso/

EarthScope+National+Meeting+2015+Poster'abstract''

Need+Broader+Impacts?+How+UNAVCO+can+support+you+in+dissemination+of+your+science+'Donna%J.%Charlevoix,%Beth%Bartel,%Aisha%Morris,%Shelley%Olds,%Beth%Pratt;Sitaula%UNAVCO,%Education%and%Community%Engagement%'UNAVCO'operates'the'Geodesy'Advancing'Geosciences'and'EarthScope'(GAGE)'facility.'The'GAGE'Facility'provides'support'to'the'NSF'investigator'community'for'geodesy'related'research,'education,'and'workforce'training'with'broad'societal'benefits.'A'core'community'support'service'led'by'the'Education'and'Community'Engagement'(ECE)'program'is'Principal'Investigator'(PI)'planning'support'and'core'programs'to'advance'geoscience'education'resources'and'geodesy'community'engagement.''UNAVCO'provides'guidance'to'PIs'with'proposal'planning'and'budgeting'for'broader'impacts.'Support'is'provided'in'all'aspects'of'broader'impacts'including,'but'not'limited'to:'Research'Experiences'for'Undergraduates'(REUs,'RESESS),'field'education,'short'courses,'curriculum'design'and'instructional'support,'scientific'results'dissemination'support'including'social'media,'and'evaluation.'''This'presentation'will'provide'examples'of'how'UNAVCO'has'supported'PIs'with'the'broader'impacts'component'of'their'scientific'research'and'offer'suggestions'of'how'we'can'help'future'PIs.'''

'

Three-dimensional P- and S-wave velocity structure along the central Alpine Fault, South Island, New Zealand B. Guo1, C. H. Thurber1, S. W Roecker2, J. Townend3, C. Rawles1, C. J. Chamberlain3, C. M. Boese4 and S. C. Bannister5 1University of Wisconsin-Madison, 2Rensselaer Polytechnic Institute, 3Victoria University of Wellington, 4The University of Auckland, 5GNS Science Abstract: The Deep Fault Drilling Project (DFDP) on the central Alpine Fault, South Island, New Zealand, has motivated a broad range of geophysical and geological studies aiming to characterize the fault system in the locality of the drill site at various scales. We have been developing three-dimensional P- and S-wave velocity models of the region by double-difference tomography utilizing datasets from multiple seismic networks (WIZARD, SAMBA, ALFA, GeoNet, and others). In our previous work, the quality of the S-wave model was relatively poor due to the small number of available S-wave picks. We have utilized a new high-accuracy automatic S-wave picker to increase the number of usable S arrivals by an order of magnitude, thereby dramatically improving the S-wave velocity model. Compared to previous studies, e.g. Eberhart-Phillips and Bannister (2002) and Feenstra et al (2013), our updated P-wave model shows a clear high Vp body (Vp > 6km/s) at depths of 5 to 15 km near the drill site. With our better resolved S-wave velocity model, we can see a sharp high Vs body (Vs > 3.7 km/s) in the same region. Besides the newly added S-picks, we have done cross-correlation to calculate the differential times between event pairs in order to improve the precision of the relocations of the earthquakes. This in turn has highlighted the presence of earthquake swarms around an upper crustal low velocity zone in the vicinity of Mt. Cook. Together with the updated earthquake relocations, the P- and S-wave tomography results reveal the Alpine Fault to be marked by a velocity contrast throughout most of the study region. The fault dips steeply from 5 to 20 km depth with an average dip of 50-60 SE, as inferred from the velocity structure and the seismicity.

Collection, Dissemination, and Analysis of USArray Transportable Array Surface Pressure Observations within the Atmospheric Science Community

Alexander A. Jacques, John D. Horel, and Erik T. Crosman

University of Utah The addition of atmospheric pressure sensors to the USArray Transportable Array (TA) provided several avenues to expand use of this unique dataset within the atmospheric science community. This presentation highlights some of the operational and research efforts that utilize both real-time and archived surface pressure observations as part of various projects. Real-time 1 Hz sampled measurements are collected, averaged into five minute intervals, and made available by MesoWest (http://mesowest.utah.edu), an ongoing project that collects surface observations from numerous meteorological networks. These measurements are also distributed to the National Weather Service and NOAA Meteorological Automated Data Ingest System (MADIS). MADIS provides these observations to the NOAA National Centers for Environmental Prediction for potential use in numerical weather models and analyses. The 1 Hz surface pressure observations are also archived and analyzed as part of an ongoing NSF-funded study to examine pressure perturbations produced by mesoscale (minutes-hours) and synoptic scale (hours-days) phenomena. A review of recently published work is provided, detailing temporal analyses of perturbations at each TA station through February 2014 and web products used to display the data and research results (http://meso1.chpc.utah.edu/usarray). Current research efforts using the TA pressure data with gridded numerical model analyses to describe spatial characteristics of pressure perturbations are also presented.

USArray TA mesoscale (10 min - 4 h) pressure signatures per season with perturbation magnitudes exceeding 3.0 hPa during (a) winter (DJF), (b) spring (MAM), (c) summer (JJA), and (d) autumn (SON). Figure Reference: Jacques, A. A., J. D. Horel, E. T. Crosman, and F. L. Vernon, 2015: Central and Eastern United States Surface Pressure Variations Derived from the USArray Network. Monthly Weather Review, 143, 1472-1493, doi:10.1175/MWR-D-14-00274.1

http://mesowest.utah.edu/http://meso1.chpc.utah.edu/usarray

Constraining the lithosphere-asthenosphere coupling from geodynamical modeling based on tomography models over the North American continent C. Adam1, S. D. King1, and M. Caddick1

1 Department of Geosciences, Virginia Polytechnic Institute and State University, 4044 Derring Hall, 1405 Perry Street, Blacksburg, VA 24061, USA

ca3@vt.edu, sdk@vt.edu,

Teleseismic P wave Spectra from USArray and Implications for Scattering and Intrinsic Attenuation

Maps of relative variations in the upper mantle attenuation parameter t* are estimated by inversion of inter-station spectral ratios from teleseismic deep earthquakes recorded by USArray. High t* areas include much of the western Cordillera and eastern passive margin, and low t* dominates across the central U.S. Smoothed t* variations (Figure A) are moderately correlated with long-period surface wave attenuation tomography (0.6 Spearman rank) and anti-correlated with velocity tomography (-0.4). However, the two standard deviation magnitude of t* variations is a factor of ~3-10 greater than predicted by prior surface wave attenuation tomography or an anelastic olivine model. Similarly high t* in parts of the passive margin and western Cordillera suggest that the effect of thermally activated intrinsic attenuation can be overwhelmed by non-dissipative effects such as elastic scattering. Transverse component spectra are used to investigate the importance of scattering because they would receive negligible P wave energy in the absence of 3-D heterogeneity or anisotropy. Transverse-to-vertical spectral ratios (T/Z) show greater partitioning of P energy onto the transverse component and increasing T/Z with frequency for stations with high t* (Figure B). Our results indicate that scattering strongly influences spectral ratio t* estimates. Broadly similar geographic patterns of teleseismic t*, surface wave Q tomography, and velocity tomography may primarily reflect spatial covariance between intrinsic attenuation and scattering intensity.

Using B4 LiDAR and CRN age data to constrain slip rates along the San Andreas Fault System at Millard Canyon, San Gorgonio Pass Desjarlais, Ian; Yule, Doug; Heermance, Richard Fault scarps cut a series of Holocene alluvial fan surfaces in Millard Canyon, within the San Gorgonio Pass (SGP). These fault scarps are likely the result of coseismic slip along the San Andreas Fault system during potentially large magnitude (Mw7+) earthquakes. Here we provide new ages for Holocene surfaces Qf2, Qf3, and Qf4. Charcoal fragments beneath Qf2 limits the surface to 1270 80 years before present (ybp) and new 10Be exposure ages from the two older Holocene surfaces provide age constraints of 4800 1600 ybp for Qf3 and 6800 550 ybp for Qf4. These new ages provide limits on the timing of slip through the San Gorgonio Pass. Airborne LiDAR from the B4 dataset was used to identify and measure preserved scarps that cut the terrace surfaces. The northernmost fault (F1) with an observed northward dip of 45 vertically offsets units Qf2 and Qf3 by 1.4 0.7m and 3.1 0.7m respectively. The southern fault (F2), a 30 north dipping active oblique strike slip thrust fault, vertically offsets units Qf1 and Qf4 by 1.5 0.6m, and 12.7 1.4m respectively. We then mathematically resolve these vertical slip parameters onto their respective fault plane geometries to evaluate the net slip component of motion along the N45W slip vector of the San Andreas Fault. The net slip component, in conjunction with the age constraints gives the following Holocene slip rates: northern fault (F1): 1.8 0.7 mm/yr; southern fault (F2): 8.8 1.6 mm/yr. Summation of these rates across the study area yields 10.6 2.3 mm/yr for the Holocene slip rates through the San Gorgonio Pass. These faults, suspected of carrying the majority of San Andreas motion through the SGP are interpreted to release interseismic strain during large magnitude earthquakes of Mw 7 or greater (Yule and Sieh, JGR 2003).

Figure 1. Schematic interpretation showing geomorphic evolution of Millard Canyon surfaces through time with cumulative imprint of (what may be an incomplete) paleoseismic record obtained from nearby trenching activities (McBurnett, 2011; Wolf et al., in progress). Model assumes contemporaneous rupture on northern and southern faults, however this behavior has not been proven to occur. Up to 2 seismic events may have occured between EQ4 and EQ3, but were not confirmed in this study. Ruptures represented as purple lines, showing known earthquakes 1-4. Disappearance of line illustrates erosion of scarp due to fluvial processes in active channel. Dashed blue line illustrates fluvial activity in channel. Black arrows illustrate suspected direction of hydrologic forces pertinent to morphology of channel. Text in red designates the time frame in which the earthquake occurred. Qf1, Qf2, Qf3, Qf4: faulted alluvium; Q: undifferentiated. See Figure 6 of Yule and Sieh, (2003) for detailed geologic map. Slip rates are determined by dividing the cumulative kinematics by the interpreted ages of the surfaces cut.

IRIS PASSCAL has supported portable broadband seismic experiments for close to 30 years. During that time we have seen a variety of sensor vaults deployed. The vaults deployed fall into two broad categories, a PASSCAL style vault and a Flexible Array style vault. The PASSCAL vault was constructed of materials available in-county and it was the Principle Investigator (PI) who established the actual field deployed design. The Flexible Array vault was provided to PIs by the EarthScope program, offering a uniform portable vault for these deployments. Cost, logistics, and the availability of materials in-country are usually the deciding factors for PIs when choosing a vault design and frequently trade-offs are made given available resources. Recently a third type of portable broadband installation, direct burial, is being tested. In this case a sensor designed for shallow, direct burial is installed in a ~20 cm diameter by 1 m deep borehole. Direct burial installation costs are limited to the time and effort required to dig the borehole and emplace the sensor. Our initial analyses suggest that direct burial sensors have lower noise levels than vault installations on both horizontal and vertical channels across a range of periods spanning

Wave gradiometry is an array processing technique utilizing the shape of seismic wave-fields captured by USArray TA stations to determine fundamental wave propagation char-acteristics. We first explore a compatibility relation that links spatial gradients of the wave-field with the displacements and the time derivatives of displacements through two unknowncoecients ~A and ~B, which are solved through iterative, damped least-square inversion, toprovide estimates of phase velocity, back-azimuth, radiation pattern and geometrical spread-ing. We show that the ~A-coecient corresponds to the gradient of logarithmic amplitudeand the ~B-coecient corresponds approximately to the local dynamic phase velocity. Thesevector fields are interpolated to explore a second compatibility relation through solutionsto the Helmholtz equation. For most wavefields passing through the eastern U.S., we showthat the ~A-coecients are generally orthogonal to the ~B-coecients. Where they are notcompletely orthogonal, there is a strong positive correlation between the gradients of ~B-coecients and changes in geometrical spreading, which can be further linked with areas ofstrong energy focusing and defocusing. We then obtain isotropic phase velocity maps acrossthe contiguous United States for 20 - 150 s Rayleigh wave by stacking results from 700 earth-quakes. The strong velocity variations in the western U.S., correlate quite well with knowngeological features and the amplitude correction term generally improved the resolution ofsmall-scale structures for all periods we analyzed. We also observe a velocity change alongthe approximate boundary of the early Paleozoic continental margin in the eastern U.S andtwo significant low velocity anomalies within the central Appalachians, one centered whereEocene basaltic volcanism has occurred, and the other within the northeastern U.S., possiblyassociated with the Great Meteor Hotspot track.

128 120 112 104 96 88 80 72

28

32

36

40

44

48

128 120 112 104 96 88 80 72

28

32

36

40

44

48

CB

CV

A YS

NR

SR

GRB

GB

WF

CV

SN

TR

CP

SRP

128 120 112 104 96 88 80 72

28

32

36

40

44

48

Period: 60 s

3.5 3.6 3.7 3.8 3.9 4.0 4.1 4.2 4.3 4.4 4.5

km/s

Figure 1: Isotropic phase velocity map for 60 s Rayleigh wave. Geographic features asindicated: CB, Columbia Basin; CP, Colorado Plateau; CVA, Cascade Volcanic Arc; CV,Central Valley; GB, Great Basin; GRB, Green River Basin; NR, North Rocky Mountain;SN, Sierra Nevada; SRP, Snake River Plain; SR, South Rocky Mountain; TR, TransverseRanges; WF, Wasatch Front; YS, Yellowstone.

1

!Analysis!of!the!Far0Field!Motion!From!an!Underground!Chemical!Explosion!!!Arben!Pitarka,!Souheil!M.!Ezzedine,!Oleg!Yu.!Vorobiev,!Robert!J.!Mellors,!and!William!R.!Walter.!!!Three!dimensional! numerical! simulations! of! the!motions! generated!by! the! Source!Physics! Experiments! (SPE)! of! chemical! underground! explosions! conducted! at! the!Nevada!National!Security!Site!(NNSS)!have!shown!that!the!observed!near0field!shear!motion!can!be!generated!by!sliding!on!the!joints!due!to!spherical!wave!propagation.!In! this! study!we! carried! the! sensitivity! analysis! to! far0field!motion!using! a!hybrid!physics0based! ! approach! that! combines! hydro0regime! !modeling! of! the! near0field!source! with! the! far0field! elastic! modeling! of! wave! propagation.! We! analyzed! the!effect!of!the!near0source!structural!complexities!on!the!simulated!near0field!source!motions.!The!simulations!are!performed! in! the! frequency!range!of!0.1010!Hz.! !The!near0field! ground! motions! simulated! for! several! source! realizations! of! the! SPE3!explosion,!using!different!equally!probable! joint! realizations,!were!propagated!out!to! far0field!distances!using!an!elastic!wave!propagation! code,!WPP.!The! simulated!motions! were! used! to! investigate! wave! scattering! effects! due! to! structural!complexities.! !The!underlying!far0field!velocity!model!was!constrained!by!available!geological!and!geophysical!data.!In!our!model!the!wave!scattering!is!a!consequence!of! combined! large0scale! and! random! small0scale! structural! features,! and! surface!topography.!Our!numerical! investigations!suggest! that!depending!on!the!degree!of!structural! complexities! in! the! near0source! region! the! wave! scattering! acts! as! an!additional! cause! for! shear! motion! generation.! ! Wave! conversions! at! geological!model!discontinuities!create!distinguish!waveform!that!are!seen!in!the!SPE!data.!!!!!

!!This%work%was%performed%under%the%auspices%of%the%U.S.%Department%of%Energy%by%Lawrence%Livermore%National%Laboratory%under%Contract%DEAAC52A07NA27344%%Release%Number:%LLNL0ABS06659840DRAFT !!

3D Model Near-Field Simulation Far-Field Simulation

A"localized"shallow"magma"source"of"Shinmoe5dake,"Kirishima"

revealed"by"L5band"multi5temporal"InSAR"

"

Yunjun"Zhang1,"Falk"Amelung

1,"Yosuke"Aoki

2,"Heresh"Fattahi

1"

"

1."Rosenstiel"School"of"Marine"and"Atmospheric"Science,"University"of"Miami,"USA"

2."Earthquake"Research"Institute,"University"of"Tokyo,"Japan"

"

"

Ryukyu"volcanic"arc"is"Japans"triple"junction"formed"by"the"subduction"of"the"

Philippine"Sea"Plate"beneath"the"Eurasian"Plate."Lying"on"the"north"of"this"arc,"

Kyushu"Island"volcanoes"could"severely"disrupt"over"110"million"peoples"everyday"

life"with"potential"catastrophic"caldera5forming"eruption."2011"Shinmoe5dake"

eruption"is"the"latest"magmatic"eruption"on"Kyushu"Island."GPS"based"modeling"has"

been"conducted,"but"no"InSAR"yet.""

"

We"processed"three"tracks"of"ALOS"L5band"SAR"data"covering"Shinmoe5dake"crater"

using"time"series"InSAR"technique."All"show"deflation"on"and"around"the"crater."A"

shallow"magma"chamber"of"about"2.7"km"under"the"summit"is"estimated"using"half5

space"Mogi"model."This"confirms"that"shallow"magma"source"is"preferential"on"

strike5slip"tectonic"settings.""

"

"

Fig."Depth"of"magma"storage"of"Shinmoe5dake,"Kirishima"from"Feb"2007"to"Aug"

2010"

0 50 100 150 200 250 300 350 400 450 500

0

50

100

150

200

250

300

350

400

450

500

AlosD Data

AlosD Model

AlosD Residual

0.020.01 0 0.01 0.02

LOS displacement [m]

AlosD Data

AlosD Model

AlosD Residual

0.020.01 0 0.01 0.02

LOS displacement [m]

AlosD Data

AlosD Model

AlosD Residual

0.020.01 0 0.01 0.02

LOS displacement [m]

2 km

0.02

0.01

0.00

0.01

0.02

LOS Velocity (cm/yr)

0.02

0.01

0.00

0.01

0.02

LOS Velocity (cm/yr)

0.02

0.01

0.00

0.01

0.02LO

S Ve

loci

ty (c

m/y

r)

2 1 2 1 0.02

0.01

0.00

0.01

0.02

LOS

Velo

city

(cm

/yr)

0

(a) Quad tree Sampling

(b) InSAR Velocity

(c) Mogi Model (d) Residual

Fig. Depth of magma storage of Shinmoe-dake (Feb 2007 - Aug 2010)

Observations of Seismic "Whistlers" & Hums in USArray

Young, B.A.1

1 CERI/University of Memphis

Unusual seismic signals in several Transportable Array (TA) stations in the Central US are observed: seismic "whistlers" and long-period harmonic hums. The observed whistlers bear a superficial resemblance to downward-sweeping VLF electromagnetic waves spawned by lightning called "whistlers." Although not likely related to their electromagnetic cousins, these "seismic whistlers" are primarily seen in two distinct categories: low frequency and high frequency. Low frequency seismic whistlers are typically seen from 3 to 7 Hz over a period of about five minutes and show a multi-harmonic downsweeping signal followed 12 to 15 minutes thereafter by an upsweeping signal with similar harmonics. These coincide with long periods (tens of minutes to over an hour) of continuous low-frequency signal with multiple harmonics. The continuity of this signal over time is suggestive of a nearby man-made source. High frequency seismic whistlers are rarer, however, and they are typically observed in a frequency band from 20 to 15 Hz over a time period of one or two minutes. Like electromagnetic whistlers, these consist of a single downsweeping harmonic. Though the source of these signals is unknown, we attempt to determine when and where these signals are observed in the seismic data and, where available, barometric and infrasound data. Observed harmonic hums come in three varietes: low-frequency (where they often coincide with low-frequency whistlers), high-frequency, and full-sprectrum. The origin of these signals remains enigmatic. Audio renditions of the data were used to help identify seismic whistlers and will be available via a YouTube link on the poster.

Opportunities for E&O include showing that there are still Wow! signals out there in sciencea lot of scientific work is difficult and mundane, but there are still weird things to look at, mysteries to unravel, and the passion for discovery lives on.

11730'

11730'

11700'

11700'

11630'

11630'

11600'

11600'

3230' 3230'

3300' 3300'

3330' 3330'

3400' 3400'

3430' 3430'

3500' 3500'

3230' 3230'

3500' 3500'

N

0 10 20 30 40 50 60 70

Tilt of the symmetry axis ()

0.4 sDelay time

Fault line Coast line

Crustal(anisotropy(from(mode2converted(body(waves(at(the(Moho(discontinuity((

Ahmet(keler(&(Miaki(Ishii((

Crust(is(the(shallowest(and(the(most(accessible(part(of(the(solid(Earth.(It( is(highly(deformed(and(contains(strong(lateral(variations.(Of(various(physical(properties(of(the(crust,(seismic(anisotropy(is(perhaps(the(most(uncertain(due(to(poor(depth(resolution(of(long(period(shear(waves(such(as(SKS(and(limited(availability(of(direct(S(waves( from( local( seismic(activity.( In( contrast,(P2to2S( converted(waves(at( the(Moho(discontinuity(offer(almost(ideal(data(coverage(and(depth(sampling(in(the(crust.(Since(these(mode2converted(waves(arrive(within(the(energetic(P2wave(coda,(it(is(almost(impossible(to(identify(them(on(original(recordings,(hence(the(need(to(calculate(receiver(functions.((Over(the(last(two(decades,(receiver(functions(have(become(a(standard(tool(for(seismologist(to(study(crustal(structure( by( removing( the( P( wave( energy( from( the( recordings( in( horizontal( directions,( which( contain(primarily( the( P2to2S( converted(waves.(Most( of( the( receiver( function( studies( targeting( crustal( anisotropy(rely( upon( forward( modeling( procedures( to( extract( anisotropic( parameters.( We( developed( an( effective(technique( based( on( the( cross2convolution( method( to( obtain( complete( set( of( anisotropic( parameters(including(fast(polarization(direction,(delay(time,(tilt(of(the(symmetry(axis,(and(percent(anisotropy(without(forward(waveform(modeling.(Our(technique(also(yields(average(P(wave(speeds(above(and(below(the(Moho(discontinuity,(average(S(wave(speed(in(the(crust,(strike(and(dip(of(the(Moho(discontinuity((in(case(of(non2horizontal(boundary)(and(the(crustal( thickness(beneath(seismic(stations.(This(approach( is(relatively(easy(and(straightforward(to(be(extended(for(multiple(anisotropic(layers,(or(to(simultaneous(analysis(of(different(shear( wave( types( such( as( PKS( and( SK{K}S.( Below( we( present( our( preliminary( results( from( receiver(functions(for(stations(around(southern(California.(In(general,(fast(polarization(direction(lies(parallel(to(the(strike(of(the(main(faults,(which(is(consistent(with(previous(estimates(from(direct(S2waves(from(local(seismic(activity.((((

(Figure( 1.( Preliminary( results(showing( the( fast( propagation(directions( (red( bars),( tilt( of( the(symmetry( axis( (color( of( the(triangles),( and( the( delay( times((length( of( the( bars)( in( southern(California.( Three( second( long(splitting( analysis( windows( are(centered(on(the(P2to2S(converted(wave( arrival( times( reported( in(EARS( (Earthscope( Automated(Receiver( Survey)( database( of(IRIS((http://ears.iris.washington.edu).(((

Mt. St. Helens seismicity observed with a 900-geophone array Brandon Schmandt1, Steven Hansen1, Eric Kiser2, Alan Levander2 1. University of New Mexico, 2. Rice University Mt. St. Helens provides a natural laboratory for studying microseismicity due to the abundance of earthquakes resulting from tectonic and magmatic processes. An excellent opportunity is provided by the MSH nodal deployment, which recorded for two weeks in July 2014 and consisted of over 900 autonomous seismometers within 15 km of the summit crater. During that time, the PNSN permanent monitoring network detected 65 earthquakes within the array footprint, 45 of which were located directly beneath the summit at 15x increase in events relative to the network catalog. Preliminary results from ongoing structural seismology studies will also be presented including Moho imaging with PmP phases and Rayleigh wave phase velocity maps extracted from two weeks of ambient noise.

Longitude

Latit

ude

122.35 122.3 122.25 122.2 122.15 122.1 122.05

46.1

46.15

46.2

46.25

46.3

seismometer

5101520Event Depth (km)

10.500.511.52Event Magnitude

Total variation regularization of geodetically and geologically constrained block models for the western United States Eileen L. Evans, Department of Earth and Planetary Sciences, Harvard University, Cambridge,

MA; now at U.S. Geological Survey, Menlo Park, CA John P. Loveless, Department of Geosciences, Smith College, Northampton, MA Brendan J. Meade, Department of Earth and Planetary Sciences, Harvard University, Cambridge,

MA Geodetic observations of interseismic deformation in the Western United States provide constraints on microplate rotations, earthquake cycle processes, and slip partitioning across the Pacific-North America plate boundary. These measurements may be interpreted using block models, in which the upper crust is divided into microplates bounded by faults that accumulate strain in a first-order approximation of earthquake cycle processes. The number and geometry of microplates are typically defined with boundaries representing a limited subset of the large number of potentially seismogenic faults. An alternative approach is to include a large number of potentially active faults bounding a dense array of microplates, and then algorithmically estimate the boundaries at which strain is localized. This approach is possible through the application of a total variation regularization (TVR) optimization algorithm, which simultaneously minimizes the L2 norm of data residuals and the L1 norm of the variation in the differential block motions. Applied to three-dimensional spherical block models, the TVR algorithm can be used to reduce the total variation between estimated rotation vectors, effectively grouping microplates that rotate together as larger blocks, and localizing fault slip on the boundaries of these larger block clusters. Here we develop a block model comprised of 137 microplates derived from published fault maps, and apply the TVR algorithm to identify the kinematically most important faults in the western United States. This approach reveals that of the 137 microplates considered, only 30 unique blocks are required to approximate deformation in the western United States at a residual level of

Model decomposition of selected block model: a) velocities due to block rotations, characterized by sharp velocity gradients at block boundaries; b) velocities due to elastic strain accumulation at faults and on the subduction zone; c) forward velocities due to full block model. Block velocity gradients have been smoothed by contribution from elastic strain accumulation. Elastic earthquake cycle elastic effects explain how the apparently smooth deformation field across much of the western United States is well described by slip on only a subset of possible active structures.

a) b) c)

128

o W

124

o W

120

o W

116

o W

112

o W

total velocity

128

o W

124

o W

120

o W

116

o W

112

o W

elastic velocity

log10(velocity magnitude) (mm/yr)0 0.5 1 1.5

128

o W

124

o W

120

o W

116

o W

112

o W

32oN

36oN

40oN

44oN

48oN

block velocity

GPS$IMAGING$OF$SOLID$EARTH$FLEX$AND$FLOW$FROM$VERTICAL$MOTIONS$USING$EARTHSCOPE$NETWORKS$

W.C.$Hammond,$G.$Blewitt,$C.$Kreemer$Nevada$Geodetic$Laboratory$Nevada$Bureau$of$Mines$and$Geology$and$Nevada$Seismological$Laboratory$University$of$Nevada,$Reno$whammond@unr.edu-

We$will$present$new$results$of$imaging$vertical$motions$of$the$Earth$surface$as$measured$by$semiFcontinuous$and$continuous$GPS$networks$including$the$EarthScope$Plate$Boundary$Observatory$(PBO).$$We$show$that$by$incorporating$data$in$a$megaFnetwork$approach,$where$data$from$numerous$open$access$archives$are$obtained$and$analyzed$uniformly,$we$can$probe$Earths$deep$interior$and$better$interpret$the$data$in$terms$of$geodynamic$processes$at$work$in$the$lithosphere.$$The$method$relies$on$new$analysis$innovations$for$estimating$velocity$fields$using$the$nonFparametric$TheilFSen$medianFbased$estimator$to$get$vertical$time$series$slope$in$a$way$that$is$highly$robust$with$respect$to$unknown$steps,$outliers$and$seasonal$deviations$from$time$series$linearity.$$We$then$apply$a$weighted$median$spatial$filtering/despeckling$algorithm$and$interpolation$to$the$point$estimates$to$create$a$vertical$rate$field.$$

The$resulting$geodetic$images$reveal$geodynamically$significant$processes$in$the$Earth's$crust$and$mantle.$$Processes$seen$in$the$imagery$include$glacial$isostatic$adjustment$across$the$North$American$continent,$which$reveals$the$contribution$of$vertical$land$motion$to$relative$sea$level$rise$with$new$scope$and$detail.$$We$also$see$elastic$rebound$following$the$unburdening$of$the$lithosphere$owing$to$groundwater$pumping$in$Californias$Central$Valley,$and$viscoelastic$flow$from$postseismic$transient$response$to$the$largest$recent$earthquakes$in$Central$Nevada$and$the$Mojave$Desert.$$In$Fig.$1$the$color$indicates$vertical$rate$for$CA$and$NV$where$The$MAGNET$semiFcontinuous$network$is$essential$for$providing$the$geographic$coverage$where$PBO$is$sparse.$Integrating$InSAR,$GPS,$tideFgauge$and$leveling$data$in$Southern$California$suggest$that$including$constraints$from$InSAR$provide$even$better$control$on$vertical$rates,$making$it$possible$to$separate$contributions$from$tectonic$and$hydrological$forcing.$

EarthScope+Science+and+Discoveries+displayed+through++the+IRIS+Active+Earth+Monitor+

Patrick+J.+McQuillan,+Russ+Welti,+Danielle+F.+Sumy,+John+Taber,+Perle+Dorr++

The+IRIS+Active+Earth+Monitor+(AEM)+is+an+interactive+computer+based+educational+display+designed+for+museums,+planetariums,+libraries,+KI12+schools,+and+university+lobbies.+The+AEM+provides+a+way+to+engage+audiences+with+earth+science+information+without+spending+resources+on+a+large+exhibit,+and+has+helped+to+increase+awareness+of+USArray+and+EarthScope+within+the+USArray+footprint.+Over+80+pages+of+content+are+available+that+explain+earthquakes,+tsunamis,+plate+tectonics,+volcanoes,+and+recent+seismology+research.+Seismic+and+geodetic+data+collected+during+EarthScope+are+available+on+several+pages,+which+are+tailored+to+the+region+the+kiosk+is+located+in.+This+allows+the+public+to+view+actual+data+in+near+realItime+and+place+it+in+the+context+of+where+they+live.+++Interactive+web+based+content+allows+users+to+explore+the+EarthScope+facilities,+view+near+realItime+data+feeds+from+EarthScope+instruments,+and+understand+EarthScope+related+research+results.+AEMs+can+display+one+of+the+readyItoIrun+regional+geology+sets,+individual+pages+that+the+host+can+put+together+on+specific+topics,+or+the+host+can+create+and+display+their+own+custom+content.+Regional+content+sets+that+have+been+created+at+the+Transportable+Array+moved+across+the+US+include:+Cascadia,+Basin+and+Range,+New+Madrid,+and+Alaska.+In+addition,+Active+Earth+content+includes+flash+animations+and+short+movies+that+allow+users+to+see+how+EarthScope+instruments+are+installed,+how+they+work,+and+how+data+are+collected.+Since+2011,+IRIS+has+loaned+AEM+kiosks+to+32+museums+and+other+educational+institutions+for+one+year,+with+the+opportunity+for+the+recipient+to+purchase+the+kiosk+for+use+as+a+permanent+display,+with+another+9+kiosks+provided+under+longerIterm+arrangements.+In+2014+alone+there+were+over+1.5+million+visitors+to+institutions+hosting+a+loan+kiosk,+plus+an+additional+900,000+visitors+to+an+EarthScope+themed+visitor+center.+Many+venues+also+use+their+own+hardware+for+the+displays,+and+in+total,+over+200+users+have+applied+for+their+own+display+accounts.++

+Map+shows+the+distribution+of+Active+Earth+Monitor+loan+kiosks+across+the+central+and+eastern+United+States+since+2011.+The+symbols+represent+whether+the+AEM+was+displayed+during+a+temporary+exhibit+(circles)+or+was+adopted+for+permanent+installation+(stars).+The+color+represents+the+year+installed,+as+shown+in+the+legend.+Three+additional+displays+are+currently+on+loan+in+Alaska.+

201120122011 (Adopted)201220132012 (Adopted)201320142013 (Adopted)20132015201420152015

Characteristics*of*oceanic*strike/slip*earthquakes*along*the*Charlie/Gibbs*transform*Kasey*Aderhold*and*Rachel*E.*Abercrombie,*Boston*University**On*13*February*2015*a*MW*7.1*strike/slip*earthquake*occurred*on*the*Charlie/Gibbs*transform*in*the*north*Atlantic.*It*is*the*first*major*Mid/Atlantic*earthquake*since*the*1994*MW*7.0*Romanche*transform*fault*earthquake*over*20*years*ago.*The*2015*earthquake*is*the*seventh*M**6.25*earthquake*to*occur*on*the*northern*transform*of*the*Charlie/Gibbs*spanning*a*seismic*record*of*nearly*a*century*(See*figure).**The*most*recent*of*these*earthquakes*was*the*1998*MW*6.7*with*an*NEIC*hypocenter*~120*km*to*the*west*of*the*2015*earthquake.*We*model*the*1998*and*2015*earthquakes*to*determine*depth*and*along/strike*slip,*then*compare*these*to*thermal*models*that*are*expected*to*define*the*seismogenic*area*of*this*fault.*We*also*consider*our*results*in*the*context*of*high*resolution*but*temporary*ocean*bottom*seismometer*studies*along*the*faster*slipping*Blanco*and*Gofar*faults.*Using*observations*from*a*range*of*tectonic*settings*is*important*to*determine*what*controls*initiation*and*propagation*of*seismic*rupture*along*strike/slip*faults,*in*the*oceans*as*well*as*along*continental*faults*like*the*San*Andreas.**

*

*Earthquakes along the northern transform of the Charlie-Gibbs fracture zone.**Longitudes*are*from*the*NEIC*catalog*and*Kanamori*and*Stewart*(1976).*Filled*in*circles*correspond*to*earthquakes*in*the*quasi/periodic*sequences.*

2015044 MW 7.136 35 34 33 32 31 30 29 28

51

52

53

54

Longitude

Date

MW

EarthScopes-Plate-Boundary-Observatory-in-Alaska:-Building-on-Existing-Infrastructure-to-Provide-a-Platform-for-Integrated-Research-and-HazardDmonitoring-Efforts--Ellie%Boyce,%K.%Austin,%A.%Woolace,%K.%Feaux,%G.%Mattioli,%M.%Enders,%R.%Bierma,%R.%Busby%

%

EarthScopes%geodetic%component%in%Alaska,%the%UNAVCOHoperated%Plate%Boundary%Observatory%(PBO)%network,%

includes%139%continuous%GPS%sites%and%41%supporting%telemetry%relays.%These%are%spread%across%a%vast%area,%from%

northern%AK%to%the%Aleutians.%FortyHfive%of%these%stations%were%installed%or%have%been%upgraded%in%cooperation%

with%various%partner%agencies%and%currently%provide%data%collection%and%transmission%for%more%than%one%group.%

Leveraging%existing%infrastructure%normally%has%multiple%benefits,%such%as%easier%permitting%requirements%and%

costs%savings%through%reduced%overall%construction%and%maintenance%expenses.%%

At%some%sites,%PBOHAK%power%and%communications%systems%have%additional%capacity%beyond%that%which%is%

needed%for%reliable%acquisition%of%GPS%data.%Where%permits%allow,%such%stations%could%serve%as%platforms%for%

additional%instrumentation%or%realHtime%observing%needs.%With%the%expansion%of%the%Transportable%Array%(TA)%

into%Alaska,%there%is%increased%interest%to%leverage%existing%EarthScope%resources%for%station%coHlocation%and%

telemetry%integration.%Because%of%the%complexity%and%difficulty%of%longHterm%O&M%at%PBO%sites,%however,%actual%

integration%of%GPS%and%seismic%equipment%must%be%considered%on%a%caseHbyHcase%basis.%%

UNAVCO%currently%operates%two%integrated%GPS/seismic%stations%in%collaboration%with%the%Alaska%Earthquake%

Center,%three%with%the%Alaska%Volcano%Observatory,%and%three%with%the%TA.%%By%the%end%of%2015,%PBO%and%TA%plan%

to%install%another%three%integrated%and/or%coHlocated%geodetic%and%seismic%systems.%While%most%of%these%are%

designed%around%existing%PBO%stations,%the%2014%installation%at%Middleton%Island%was%a%completely%new%station%

for%both%groups,%providing%PBO%with%an%opportunity%to%expand%geodetic%data%collection%in%Alaska%within%the%

limited%operations%and%maintenance%phase%of%the%project.%%

We%will%present%some%of%the%design%considerations,%outcomes,%and%lessons%learned%from%past%and%ongoing%

projects%to%integrate%seismometers%and%other%instrumentation%at%PBOHAlaska%stations.%Developing%the%PBO%

network%as%a%platform%for%ongoing%research%and%hazard%monitoring%equipment%will%also%continue%to%serve%the%

needs%of%the%research%community%and%the%public%beyond%the%completion%of%EarthScope%science%plan%in%2018%

%

%

%

%

%

%

%

%

InSAR for Geothermal Reservoir Management and Sustainable Development

Mohamed Aly and Erik Bawner Department of Geosciences, University of Arkansas, Fayetteville, AR 72701

Phone: 479-575-3185, Fax: 479-575-3469, Email: aly@uark.edu

Geothermal energy is a rapidly growing source of power within the United States. Understanding the hydrothermal-geomechanical response of a geothermal reservoir to fluid production and injection is essential for integrated management and sustainable development of the reservoir. Changes in the underground water level, pressure, and temperature caused by geothermal production activities may lead to extensive ground deformation. Therefore, regular monitoring of active geothermal fields is necessary to evaluate the impact of production activities and assess local ground stability. Knowledge of the reservoir compaction, geometry, and response to production behaviors will help in defining ideal locations for new production and recharge wells that can directly improve the performance of the reservoir. This research addresses active geothermal processes and recent seismic events and investigates their impacts on the local crustal deformation at the Raft River geothermal power plant in southeastern Idaho and at the Coso geothermal site in eastern California. The study incorporates geodetic data from Interferometric Synthetic Aperture Radar (InSAR) and Global Positioning System (GPS) measurements acquired between 1992 and present. Volumetric analysis and modeling are conducted to characterize the selected geothermal reservoirs.

Line-of-sight deformation (color scale) superimposed on the hill-shaded relief (grey scale) of the study site in southeastern Idaho

Sources: Esri, DeLorme, USGS, NPS

-11315'

-11315'

-11330'

-11330'

421

5'

421

5'

420

'

420

'

0 6.5km

24/07/2005 - 21/09/2008

LOS

dis

plac

emen

t (cm

) +2.83

0

-2.83

Km LOS

Geothermal WellsQuaternary Faults

Comparison of geodetic and geologic vertical motion rates in the Southern California

Arjun Aryal, Samuel Howell, and Bridget Smith-Konter

School of Ocean and Earth Science and Technology University of Hawaii, Honolulu, Hawaii, USA

Abstract Horizontal geologic and geodetic slip rate discrepancies are well documented for active faults in Southern California, however this discrepancy is significantly larger in the vertical direction. Therefore, rheological models used to constrain fault slip rates generally exclude vertical geodetic motions. Here, we compare geologic vertical rates from the Southern California Earthquake Center Vertical Motion Database and vertical GPS velocities from the EarthScope Plate Boundary Observatory. These two data are not necessarily co-located in space and geologic data are observed using four different markers - thermo-chronologic (TH), river terraces (RT), stratigraphic horizons (SH) and marine terraces (MT). Therefore, we extract subsets of the geologic data based on observation proximity with GPS locations, data measurement errors, as well as observation marker types and compare these subset geologic data with the geodetic data. For all subsets, geologic and geodetic data are poorly correlated, with correlation coefficient R being less than 0.3. The geologic data from SH primarily indicate subsidence, but the geologic data from other sources (TH, MT and RT) primarily indicate uplift. Furthermore, the SH and RT samples are from the same geographic location. We next compare the geologic data to a smoothed GPS velocity field that was derived from a statistically robust interpolation technique that removes high-frequency non-tectonic noise (i.e., noise related to groundwater withdrawal). Our comparison with the interpolated GPS data, as well as each subset of geologic data, shows an increased correlation (R as high as 0.38) but overall the agreement is not strong (Figure 1). Therefore, the discrepancy in geologic and geodetic vertical data along SAFS is likely to be related to the measurement bias due to different sources of the geologic vertical motion rates in addition to the noise due to non-tectonic deformations in the geodetic data. Figure 1. a) Location of geologic and geodetic data plotted in b and c. b) Scatter plot of geologic verses geodetic data and c) Scatter plot of geologic verses interpolated (modeled) geodetic data. The best-fit (BF) line is for comparison with the 1:1 line.

1.5 1 0.5 0 0.5 1 1.51.5

1

0.5

0

0.5

1

1.5

GPS Velocity (mm/yr)

Geo

logi

c ve

rtica

l vel

ocity

(mm

/yr)

b. Geologic vs nearest GPS velocities

Rvalue=0.12

1.5 1 0.5 0 0.5 1 1.51.5

1

0.5

0

0.5

1

1.5

GPS (modeled) Velocity (mm/yr)

c. Geologic vs interpolated GPS velocities

Rvalue=0.29

TCMTRTSHBF1:1

TCMTRTSHBF1:1

Geo

logi

c ve

rtica

l vel

ocity

(mm

/yr)

33

34

35

Latit

ude

(deg

ree)

120 119 118 117Longitude (degree)

TCMTRTSHGPS

a. Locatiom of geologic and GPS data

EarthScope Transportable Array Siting Outreach Activities in Alaska and Western Canada

Lea Gardine1, Perle M Dorr2, Carl Tape1, Patrick McQuillan2, Joel Cubley3, Mary Samolczyk3, John Taber2, Michael E. West1, and Robert W Busby2 1University of Alaska Fairbanks-Geophysical Institute 2IRIS Consortium 3Yukon College The EarthScope Transportable Array will be deploying about 260 stations in Alaska and western Canada. In this region, new tactics and partnerships are needed to increase outreach exposure. IRIS and EarthScope are partnering with the Alaska Earthquake Center, part of University of Alaska Geophysical Institute, and Yukon College to spread awareness of earthquakes in Alaska and western Canada and the benefits of the Transportable Array for people living in these regions. Nearly all parts of Alaska and portions of western Canada are tectonically active. The tectonic and seismic variability of Alaska, in particular, requires focused attention at the regional level, and the remoteness and inaccessibility of most Alaskan and western Canadian villages and towns often makes frequent visits difficult. For this reason, outreach most often occurs at community events. When a community is accessible, every opportunity to engage the residents is made. Booths at state fairs and large cultural gatherings, such as the annual convention of the Alaska Federation of Natives, are excellent venues to distribute earthquake information and to demonstrate a wide variety of educational products and web-based applications related to seismology and the Transportable Array that residents can use in their own communities. Region-specific publications have been developed to tie in a sense of place for residents of Alaska and the Yukon. The Alaska content for IRISs Active Earth Monitor emphasizes the widespread tectonic and seismic features and offers not just Alaska residents, but anyone interested in Alaska, a glimpse into what is going on beneath their feet. The concerted efforts of the outreach team will have lasting effects on Alaskan and Canadian understanding of the seismic hazard and tectonics of the region. Efforts to publicize the presence of the Transportable Array in Alaska, western Canada, and the Lower 48 also continue. There have been recent articles published in university, local and regional newspapers; stories appearing in national and international print and broadcast media; and documentaries produced by some of the worlds most respected scientific and educational production companies that have included a segment about EarthScope and the Transportable Array.

TowardsUsingDynamicStraininEarthquakeSourceCharacterizationAndrewBarbour 1 andBrendanCrowell 2

(1)U.S.GeologicalSurvey,MenloPark,CA (2)UniversityofWashington,Seattle,WA

Measurements of static and dynamic deformation at the Earths surface are fundamental signals used for characterizing earthquakes. In early warning systems these signals are traditionally inferred from displacements on forcebalance sensors (seismometers), and more recently with the inclusion of highrate Global Navigation Satellite System sensors however, direct strain measurements have yet to be considered.The Plate Boundary Observatory (PBO) component of EarthScope operates 78 borehole strainmeters (BSMs) and 6 longbaseline laser strainmeters (LSMs) along the western United States and British Columbia. Including these PBO stations into current characterization efforts could significantly improve station density in regions with high seismic hazard. For example, 32 (41%) of the BSMs are located along the Cascadia subduction zone, where the probability of a M 9 earthquake within 50 years might be as high as 15% and the probability of a smaller but still potentially damaging M 8 event mightbeashighas40%(Goldfinger,etal.,2012).Here we examine highfrequency (1 Hz) strains from 180 earthquakes which occurred between from 2004 through 2012, recorded by 68 PBO BSM stations these have magnitudes M ranging from 4.6 to 7.2, depths ranging from 12 to 32.7 km, and epicentral distances ranging from 13 to D 500 km. Coseismic strains seen at the BSMs may not be a reliable measure of static strain (Barbour, Agnew, and Wyatt, 2015), so we do not consider them here. But, peak dynamic strains seen at the BSMs, , can be predicted from the E magnitude of the earthquake and the logarithm of the epicentral distance between the earthquake and station, with high statistical confidence. Based on the root mean squared value of uncalibrated instrumental strains for all events, we find:

ogE .78(0.14) 2.65(0.04)logD 1.23(0.02)M , 7.7 0 .8 0l = 7 + 1 10 < Epredicted < 1 15

with standard errors of the coefficients shown in parentheses. This yields a residual standard error of

that can be reduced by either including calibration.16 0 (dof 819, R .84, p .2 0 ) 2 1 9 = 1 2 = 0 < 2 1 16 coefficients, or accounting for station effects in the regression. The figure above shows how the observed strainsvaryaccordingtothisscalingrelationshipthepointsarecoloredbystationname.Agnew and Wyatt (2014) observe a similar magnitudedistance scaling relationship based on dynamic strain from the LSMs, indicating that in general strainmeters represent a viable source of data on ground deformation that could be complementary to existing seismogeodetic techniques. With enhancements to telemetryrobustness,accurateearthquakesourceparameterscouldbeestimatedinrealtime.

Geodetic Observations of Human Induced Seismicity and Deformation: The 2011 Mw5.3 Trinidad, Colorado Earthquake Barnhart, W.D., Rubinstein J.L., Hayes, G.P., Benz, H. Geodetic observations are a powerful tool for identifying, quantifying, modeling, and monitoring solid earth deformation, including human induced deformation. Although induced earthquakes are a seismological phenomenon, they are driven by complex interactions between anthropogenic processes, hydrological systems, local geology, and pre-existing faults. To gain a more complete image of the dynamics of these systems, observations of ongoing surface deformation, both seismic and aseismic, can compliment and enhance ongoing seismological studies. Here, we present interferormetric synthetic aperture radar (InSAR) analysis of ground deformation in the Raton Basin of southern Colorado and northern New Mexico, including displacements from the suspected wastewater injection induced 2011 Mw5.3 Trinidad earthquake.

Using Envisat observations from the WInSAR archive spanning the Raton Basin, we image co-seismic surface displacements of the 2011 Trinidad earthquake. From these displacements, we invert for the location and geometry of the source fault and the finite distribution of slip through an iterative resampling algorithm. We find that the earthquake slipped within the crystalline basement underlying basin sedimentary rocks and in the vicinity of high-volume wastewater injection wells. The spatial and temporal separation between the location and onset of wastewater injection and the earthquake itself suggests a pore pressure migration triggering mechanism is present. The finite slip distributions, along with seismically recorded aftershocks, further highlight the location and orientation of previously unmapped, seismogenic faults. Lastly, the precise earthquake location afforded by InSAR observations provides a well-located earthquake source that can be used to calibrate other regional earthquakes locations. Additionally, we derive InSAR time series observations from ALOS imagery acquired from 2007-2011. These results highlight ongoing regions of surface subsidence within the basin, presumably from shallow withdrawal of fluids. We infer that the displacements arise from extraction of coal-bed methane and water that is later reinjected. While it is not clear if there is a causative relationship between regions of co-located surface subsidence and recorded earthquakes, the time series permits us to exclude several other hypotheses for the causes of increased seismicity in the Raton Basin, including volcanic activity related to the Rio Grande Rift. Furthermore, the InSAR time series analysis provides observations of time variable surface deformation that may be used to inform hydrological models which assess subsurface stress changes from the removal and injection of fluids. Generally, the capability of InSAR to capture sub-centimeter scale surface displacements over broad spatial areas opens many new opportunities to further assess the dynamic behavior of regions experiencing induced seismicity. While few to no InSAR observations have been available over Oklahoma and Kansas until recently, ongoing acquisitions and future InSAR missions such as NISAR will provide a valuable tool to supplement seismological observations in the quantification and monitoring of active deformation in these regions.

Understanding the Processes Driving Glacier Change with Alaskan Seismic and GPS Data Timothy C. Bartholomaus, Christopher F. Larsen, Michael E. West, Shad ONeel, Ginny Catania

Worldwide, glaciers and ice sheets are losing mass and increasing global sea level (Shepherd and others, 2012; Gardner and others, 2013). However, the processes controlling these changes are not well understood. Changes in glacier hydrology and iceberg calving can both increase rates of glacier flow, thereby hastening delivering of ice to the ocean and low elevation regions. The understanding of these two processes is not yet sufficient to reliably include them in ice flow models for the prediction of sea level rise.

The application of seismology and GPS techniques within glaciology allows insight into glacier hydrology and iceberg calving processes. At Yahtse Glacier, a tidewater glacier in Alaska, we seismically quantified calving at unprecedented tidal to seasonal timescales. Tracking of calving-generated icequakes reveals that calving of large icebergs is significantly more likely to occur during falling and low tides than during rising and high tides. We also observe that calving fluxes are greater during the late summer and fall than during winter, suggesting that, on the coast of Alaska, submarine melt of glacier termini is likely a dominant control on the calving rate (Bartholomaus and others, 2013). Background seismic noise (i.e., tremor) also offers glaciological insight. Tremor amplitude rises and falls seasonally and after storms, synchronously with subglacial discharge. Thus, subglacial discharge variations can be quantified at tidewater locations where discharge has been previously unknown.

At Yahtse Glacier and Kennicott Glacier, also in Alaska, we use GPS to observe contrasting responses in glacier motion to melt, rain, and lake-drainage events (Bartholomaus and others, 2008). At Kennicott, speedup responses are short-lived and glacier motion quickly returns to background levels. Yahtse Glaciers response to hydrologic events is long-lived and leads to progressively slower flow over the course of the summer, demonstrating that in some cases changes in subglacial water routing are not reversible on daily to weekly timescales.

Together, seismic and GPS data offer views of glacier responses to environmental change with temporal resolution that is not available through approximately weekly satellite images. These highly resolved observations allow physical insight that improves our understanding of glacier physics, eventually allowing for better inclusion of glacier dynamical processes in ice flow models. Going forward, Earthscopes Transportable Array in Alaska expands on the present opportunity to remotely track iceberg calving across coastal Alaska. New terrestrial radar interferometers offer a more complete view of ice flow variability by combining the spatial resolution of satellite imagery with the temporal resolution of GPS.

N

orm

aliz

ed c

alvi

ng ra

te

1

1.5

2

~700 ~19,000 ~88,000 ~240,000 ~510,000Iceberg Volume (m3)

High

Fallin

gLo

wRi

sing

Tidal stage

Assessing the Vulnerability of Power Grids to Space Weather the Role of EarthScope MT Data

Paul A Bedrosian and Jeffrey J Love, United States Geological Survey

The response of Earths magnetic field to solar activity is both a source field for magnetotellurics (MT) and a potential hazard to infrastructure, particularly in response to large geomagnetic storms. The storm-time induction of electric fields in Earth's conducting lithosphere can interfere with the operation of electric-power grids, damage transformers, and sometimes cause blackouts. According to some scenarios, the future occurrence of a rare but extremely intense magnetic storm, such as occurred in 1859, would cause widespread failure of electric-power grid operations, with significant deleterious impacts for society. This has motivated the US Federal Energy Regulatory Commission to require the development of reliability standards to mitigate the impact of geomagnetically induced currents (GICs) on the operation of the US-national bulk-electric power system.

Modeling storm-induced electric fields, and the GICs which they drive, requires an understanding of (1) the spatiotemporal variability in the inducing geomagnetic field, (2) the spatiotemporal variability in the induced electric field, and (3) how induced earth currents couple into a distributed power system. MT data, in the absence of any modeling and inversion, provide the linkage between 1 and 2 at each measurement location. Over an extended area, such as the mid-continent region, it is possible to use EarthScope MT data to map out variations in electric field amplitude and polarization in response to highly simplified magnetic storms (Figure 1). Electric field amplitudes, upon which GICs scale, are observed to vary by two orders of magnitude over a distance of 100 kilometers, and in comparison with 3D conductivity models are found to be driven primarily by variations in crustal conductivity. As the EarthScope MT Transportable Array continues its march across the landscape it is improving not only our understanding of lithospheric conductivity, but also providing a unique, important and timely data set needed to mitigate infrastructure hazards associated with space weather.

Figure 1. Variation in electric field amplitude and direction at EarthScope MT stations within the Midwestern US associated with a 100 sec period, north-south oriented inducing magnetic field. Background color map shows modeled conductivity variations (red is conductive, blue is resistive) at 4 km depth based upon 3D inversion of the same EarthScope MT data.

Ten Years Of Plate Boundary Observatory Borehole Strainmeter Operations and Data Products Hodgkinson, Mencin, Phillips, Fox, Gallaher, Gottlieb, Henderson, Johnson, Pyatt, Van Boskirk, Meertens and Mattioli, UNAVCO, 6350 Nautilus Drive, Boulder, CO. PBOs first borehole strainmeter, B004, was installed by UNAVCO on June 16th 2005 on the southwest shore of the Straits of Juan de Fuca on the Olympic Peninsula. It was the first installation in what would be the largest borehole strainmeter network ever built for research purposes in the US. Between June 2005 and October 2018 a further 74 strainmeters were installed as part of the EarthScope PBO project, each site forming part of a sub-array in a targeted area with specific scientific questions in mind. The purpose of the strainmeters was to record small, short period strain transients the size and duration of which would render them undetectable by GPS and seismology. Since then, PBO strainmeters have provided unprecedented temporal resolution of strain pulses that evolve over minutes, for example, nanostrain-level creep events on the central San Andreas, out to measurements of transients on the order of a 100 nanostrain over weeks during Episodic Tremor and Slip events along the Cascadia Subduction Zone. In the ten years since installation B004 has been operational 99.8% of the time with similar uptimes across the network excluding the ten volcanic installations which are unreachable in the winter months. The raw data are available to the research community in SEED format from the IRIS DMC within one to two hours of recording and processed data are available from UNAVCO within 24 hours. UNAVCO provides not just the time-series but also a rich metadata set that includes Level 2 processing information such as the Earth tide and barometric responses, drilling logs, borehole cuttings and access to station notes that capture observations made by UNAVCO field engineers about the site setting and instrument condition over the years. As of April 2015 more than 1 TB of PBO BSM raw data and products have been delivered to users and strainmeter data products have been downloaded by more than 1,400 unique users. In this presentation we will describe the PBO strain, seismic, pore and tiltmeter data sets available and show how UNAVCO monitors the data quality of each instrument type. Information on all PBO borehole datasets maybe found at http://www.unavco.org/data/strain-seismic/strain-seismic.html .

Shear Strains recorded by PBO borehole strainmeter B004. Lon-term trends removed, black traces contain the Earth tides and barometric pressure signals, red traces are the residuals after these signals have been modeled and removed. ETS events stand out clearly above the background noise.

0.10

0.05

0.00

0.05

0.10

2006 2007 2008 2009 2010 2011 2012 2013 2014 2015

2Ene Shear Strain

Eee-Enn Shear Strain

B004, Hoko Falls, Olympic Peninsula, Washington

Level2 Processed Strain Data

mic

rost

rain

GLOBAL TRAVELTIME TOMOGRAPHY WITH USARRAY TRANSPORTABLE ARRAY

SCOTT BURDICK & VEDRAN LEKIC, UNIVERSITY OF MARYLAND, COLLEGE PARK

With its images of descending slabs and rising plumes, global tomography provides us with a snapshot of the dynamic

mantle. In particular, traveltimes from the USArray Transportable Array have allowed us to create tomographic images

of large-scale structures beneath North America in ever finer detail and with less risk of artifacts due to sparse and

irregular data coverage. Tomography works by finding a model of seismic velocities that best explain the traveltime

data, and structure in the velocity model can be inferred to be variation in temperature, composition, or volatile

content.

If we wish to accurately relate these velocity variations to the physical properties of the Earth, and to estimate the

their strength and spatial extent, it is of vital importance that we have a good grasp of the uncertainty in the model.

Rigorous examination of model uncertainty has long been a thorn in the side of seismic tomography due to the typically

vast number of model parameters and the computational cost of the forward modeling problem. Standard resolution

tests (i.e. checkerboard tests) can give a qualitative picture of where the data are able to constrain velocity structure,

but they rest on questionable assumptions about uncertainty in the data, neglect forward modeling uncertainty, and

do not provide estimates of covariance between model parameters.

For these reasons, we turn to Bayesian inference. We perform a transdimensional hierarchical Bayesian inversion

on traveltimes from the USArray Transportable Array and global catalogues. In our approach, we parameterize the

structure beneath North America as a set of three dimensional Voronoi volumes. Using the reversible-jump Markov

chain Monte Carlo method, we create chains of models by allowing the location of the volumes and the velocity within

them to vary at random. New volumes can be added and old volumes removed. Each new model is accepted or rejected

according to its eect on the error function with a probability based on Bayes Theory. The end result is chains of

models, all of which satisfy the error function to some degree.

The point of this process is not only to converge to the velocity model that best fits the data, but to generate

an ensemble of models from which statistical inferences can be drawn. From the ensemble, we create a probability

distribution for the velocity at each point. By analyzing the distribution, we can determine the mean model and the

model variance, eectively allowing us to put error bars on our estimates of the velocity structure. The most likely

velocity can be determined from the peak value of the distribution, and sharp boundaries in structure can be inferred

from regions where two or more peaks are present. We can furthermore quantify the trade-os between velocity

variations inferred at dierent locations This approach gives us new insight on the major questions about the mantle

beneath North America, including the extent to which certain prominent features, like the mantle plume related to

the Yellowstone hotspot, are required by the data, the dierence in complexity of structure between the stable east

and tectonically active west, and thickness variations of the continental mantle lithosphere.

Depth 200 km

-1.5% 1.5%

0% 1.5%

-1.0% 1.0%

P wavespeed

Model Uncertainty

Depth 100 km 1.50%

Figure 1. Left: P-wave velocity model at 100 km depth given in %dV/V. Red line shows locationof cross-sections. Right: Section through model down to 1400 km. Model uncertainty calculated by

transdimensional Bayesian inversion.

In the quest of understanding the enigmatic seismicity of the New Madrid Seismic Zone (NMSZ), we study the velocity structure and anisotropy beneath this region. The study region spans longitudes 86 to 94 degrees west and latitudes 33.8 to 40 degrees north. We use data from the Northern Embayment Lithospheric Experiment (NELE), which is a lithospheric-scale passive array experiment in the northern Mississippi Embayment. The first phase of this experiment involved 6-month deployments of 6 flex array (FA) stations to fill in the Transportable Array (TA) grid stations over a period of two years (September 2011 to October 2013). The second phase is a 2 - year deployment which began in July 2013 with the installation of 51 broadband seismometers along three profiles with an average station spacing of 20 km. The Cooperative New Madrid Seismic Network (CNMSN) stations operated by the Center for Earthquake Research and Information (CERI) are also used to augment the NELE stations. In order to map the orientation and strength of mantle fabrics beneath this region, we use the SplitLab processing environment to measure shear wave splitting parameters of teleseismic SKS phases recorded from January 2011 to date. We also use arrival times from local earthquakes and travel time residuals from teleseismic earthquakes to perform a joint local and teleseismic P and S wave velocity (Vp and Vs) inversion. A comparison of the splitting measurements across different terrains within the study region indicates a complex pattern of anisotropy beneath the NMSZ (Figure 1). There is little to no agreement of the splitting patterns with the absolute plate motion directions and upon entering the embayment, we observe a change in fast axis directions from south-west to north-south and an increase in magnitude of the delay times. From the tomography study we image a prominent low velocity anomaly in both the Vp and Vs results concentrated at a depth of about 200 300 km. Combining the splitting results with these new, detailed Vp and Vs models will provide detailed knowledge of upper mantle structure, which may further our understanding of the driving mechanism of the NMSZ intraplate earthquakes and allow us to better assess the associated seismic hazard.

Figure 1: Map of average splitting parameters. Estimates were obtained by a simple average of the highest quality measurements at each station. Symbols are color coded by the magnitude of the delay time. The orientation of the bar corresponds to the fast direction. The magenta circle shows the approximate location of the low velocity anomaly from the body-wave tomography study. The dotted black line delineates the Mississippi Embayment.

Leveraging EarthScope USArray with the Central and Eastern United States Seismic Network

Danielle F. Sumy, Robert L. Woodward, Andrew M. Frassetto,

and Robert W. Busby

Recent earthquakes, such as the 2011 M5.8 Mineral, Virginia earthquake, raised awareness of the comparative lack of knowledge about seismicity, site response to ground shaking, and the basic geologic underpinnings in this densely populated region. With this in mind, the National Science Foundation, United States Geological Survey, United States Nuclear Regulatory Commission, and Department of Energy supported the creation of the Central and Eastern United States Seismic Network (CEUSN). These agencies, along with the IRIS Consortium who operates the network, recognized the unique opportunity to retain EarthScope Transportable Array (TA) seismic stations in this region beyond the standard deployment duration of two years per site. Stations were selected using multiple criteria, including proximity to known regions of seismic hazard, nuclear power plants, and other critical facilities. The CEUSN mission is to produce data that enables both researchers and federal agencies to better understand the basic geologic questions, background rates of earthquake occurrence and distribution, seismic hazard potential, and associated societal risks. This multi-agency collaboration is motivated by the opportunity to use one facility to address multiple missions and needs in a way that is rarely possible. The CEUSN will encompass 159 broadband TA stations, more than 30 with strong motion sensors added, that are scheduled to operate through 2017. Stations were prioritized in regions of elevated seismic hazard that have not been traditionally heavily monitored, such as the Charlevoix and Central Virginia Seismic Zones. The stations (network code N4) transmit data in real time, with broadband and strong motion sensors sampling at 100 samples per second. The CEUSN, together with the existing backbone coverage of permanently operating seismometers in the central and eastern United States, will form a network of over 300 broadband stations.

Map shows the 159 CEUSN stations (yellow) that will be operated and maintained by the IRIS Consortium until 2018. The CEUSN stations were chosen specifically for proximity to nuclear power plants (black squares), as well as other critical infrastructure. The distribution of seismic stations across the central and eastern United States fills in regions of seismic hazard that have not been heavily monitored in the past. !

Shear velocity structure beneath the central United States: implications for the origin of the Illinois Basin and intraplate seismicity

We investigated the lithospheric structure beneath the North American Midcontinent, including the Illinois Basin and three intraplate seismic zones. By measuring Rayleigh wave phase velocities from teleseismic earthquakes recorded at USArray Transportable Array and OIINK (Ozarks-Illinois-INdiana-Kentucky) Flexible Array stations, we obtained new estimates of lithospheric shear velocities for the Illinois Basin (IB), the New Madrid Seismic Zone (NMSZ), the Wabash Valley Seismic Zone (WVSZ) and the Ste. Genevieve Fault Zone (SGFZ). A failed rift arm, the Reelfoot Rift (RR), sits beneath the NMSZ and extends into the southern Illinois Basin. We find that the southern IB possesses high mid-crustal velocities (>4.2 km s-1) at depths between 25 km and 35 km (map left). The observed high velocities at mid-crustal depths beneath the southern basin may correspond to high-velocity mafic intrusions that were emplaced into the crust during rifting. The high-density mafic intrusions may have contributed to the subsidence of the southern IB. We also observe relatively low velocities (< 4.65 km s-1) in the mantle beneath the NMSZ at depths between 90 and 125 km (map right), compared with the average shear velocity of 4.7 km s-1 outside of the rift. The low upper mantle velocities also extend beneath the WVSZ and the SGFZ. Based on exploring the sensitivity of seismic velocities to a range of thermal and compositional variations, we infer that the low mantle velocities would likely not result from elevated temperatures alone but require a contribution from increased iron content and the presence of water. The compositional heterogeneity of the upper mantle would lead to a weak zone. The crustal seismic zones may then correspond to locations where deformation has been localized within the mantle due to their lower integrated lithospheric strength. The tectonic history of the region including rifting and interaction with a mantle plume can introduce these heterogeneities and cause velocity reduction. Similar orientations for the NE-SW low-velocity zone, and the Reelfoot Rift suggests a rifting related origin for at least the southern portion of the area of reduced velocities.

The GAGE GPS Analysis Centers process data from over 2200 GPS stations, with 1100 stations in the core PBO network, 121 stations in COCONet, and 24 stations TLALOCNet that are managed by UNAVCO. The remaining stations are operated by other agencies. The Analysis Centers and UNAVCO engineers monitor station quality, looking for abnormalities in processing such as unusually large errors, nonlinear position time series, or anomalous station velocities. Abnormalities are evaluated on a station-by-station basis with a suite of tools. Tools to analyze raw RINEX files can determine how well a station is tracking GPS satellite signals or whether there is interference near the station. Time series analysis can identify nonlinear segments, while strain rate and velocity plots highlight outliers. Abnormalities arise for many reasons. An antenna may have difficulty tracking signals if there is too much snow or vegetation growth nearby. A station position might change due to equipment replacement, site damage, co-seismic displacements, or for unknown reasons, and these changes must be accounted for when calculating the velocity of a site. The Quality Control (QC) monitoring also flags real, nonlinear ground deformation at stations. The most common and widespread is episodic tremor and slip, particularly in the Pacific Northwest. Also common is volcanic deformation at several of the active volcanoes in the network. A few stations were constructed on unstable ground, with at least one station upslope of a landslide. We present examples of QC analysis at selected stations, deducing as much about station problems as possible based on available information. Results of QC analysis are referred to the field engineers if the problem requires physical intervention and to the Analysis Centers if the problem involves metadata and maintenance-related updates. QC monitoring helps maintain the reliability of data from the UNAVCO-managed network and is an important component of efficient station maintenance and upkeep. !

!Quality indicators for COCONet station CN48 on the island of Dominica. The station was installed in 2014 and has a noisy time series (upper left). Error parameters from the University of Nevada QA files are plotted in the lower left. TEQC is a program that computes daily average signal-to-noise (SNR) ratio and multipath (MP) from RINEX files (upper right). The SNR is constant over time and within expected ranges, but the MP is unusually noisy. The number of cycle slips per day (lower right) are obtained from a RINEX-parsing code. There are frequent spikes in the number of slips per day that might contribute to low position solution quality.

Synthetic Testing of the G-FAST Geodetic Earthquake Early Warning System

Brendan Crowell, David Schmidt, Paul Bodin, and John Vidale

The Cascadia subduction zone poses one of the greatest risks for a megaquake in the continental United States, and because of this, the Pacific Northwest Seismic Network (PNSN) at the University of Washington is building a joint seismic and geodetic earthquake early warning system. We have taken a two-stage approach to earthquake early warning: (1) detection and initial characterization using strong-motion data from the PNSN with the ElarmS package and (2) the triggering of geodetic modeling modules using GPS data from the Pacific Northwest Geodetic Array (PANGA) and combined seismogeodetic (GPS + strong-motion) data. Because of Cascadias relatively low seismicity rate, and the paucity of data from plate boundary earthquakes there of any size, we have prioritized the development of a test system and the creation of several large simulated events. The test system permits us to: 1) replay segments of actual seismic waveform data recorded from the PNSN and contributing seismic network stations to represent both earthquakes and noise conditions, and 2) broadcast synthetic data into the system to simulate signals we anticipate from earthquakes for which we have no actual ground motion recordings. The test system lets us also simulate various error conditions (latent and/or out-of-sequence data, telemetry drop-outs, etc.) to explore how to protect the system from them. Here we report on the performance of the joint early warning system and the geodetic modeling modules in a simulated real-time mode using simulated 5-Hz displacements from Mw 6.8 Nisqually earthquake. The results show that the geodetic modeling modules are able to property characterize the event, and we discuss the limitations with respect to latency, network architecture, and earthquake location throughout the Pacific Northwest.

Figure: Flowchart of the G-FAST earthquake early warning system

Moho

Pip MohoMid-crustal reflectors

3 4 5 6 7VELOCITY (km/s)

DEPT

H (k

m)

DISTANCE (km)

2x Vertical Exaggeration

444 555

666

W E !"#$"%&'()*+,'-+"*%'.,&"

!""#$%$&'($##)*$+','-./0123)'435$#'6)+)/%+' ' ' ' ' ' ' ''7&)8.'6)%8./3%8'' '-8)//3+'5./")*3"+'' 9:%$*+3.*

ShotsEarthquakes

/0"*&10' ' ' ' ' ' ' ' ' ' ' ' ' '

EarthScope IDOR: Controlled-source seismic evidence for a Moho-penetrating steep accretionary margin, Idaho-Oregon

K. Davenport1, J.A. Hole1, S.H. Harder2, B. Tikoff3

1Virginia Tech, 2University of Texas El Paso, 3University of Wisconsin-Madison

The EarthScope IDOR project is investigating the formation and modification of a steep tectonic boundary in the U.S. Cordillera of Idaho and Oregon. The western Idaho shear zone (WISZ) juxtaposes accreted island-arc terranes against Precambrian North American craton across an unusually narrow, steep boundary adjacent to the Idaho batholith. The tectonic structure of this region has been significantly modified since accretion by large-scale transpression, the emplacement of the Idaho batholith, Challis volcanism, the Columbia River Basalts, and Basin and Range-style extension.

The IDOR controlled-source seismic survey was a 430-km refraction and wide-angle reflection line designed to image the crustal structure and velocity across these contrasting geologic features. Acquisition consisted of 2555 vertical-component seismometer stations that recorded 8 explosive shots, background seismicity, and ambient noise. Data acquisition involved a 53-person field crew composed largely of undergraduate and graduate student volunteers from 22 colleges and universities. Velocity structure across the IDOR line is well constrained by travel time inversion of direct arrivals, wide-angle reflections, and refractions from the crust and Moho. Local earthquakes recorded during the controlled-source deployment are being incorporated into the modeling process to provide additional data between shot point locations.

Results from this analysis of the IDOR controlled-source seismic data reveal significant changes in velocity and crustal structure between the accreted terranes west of the WISZ and the Idaho batholith and Precambrian craton to the east. The seismic data require a lithospheric-scale, near-vertical boundary at the WISZ. The crust west of the WISZ is characterized by faster velocities and a shallower Moho depth, consistent with oceanic-arc crust. Numerous wide-angle reflections are observed, including an arrival from the lower crust that has higher amplitude than the reflection from the Moho. This lower-crustal reflector has an underlying velocity of ~7 km/s, and the layer is interpreted to be underplating associated with the feeder dike system for the Columbia River Basalts. In contrast, the crust east of the WISZ has a much slower velocity and the Moho is 5-10km deeper, consistent with felsic-to-intermediate continental crust. Reflections from this region have lower amplitude and are less continuous than those in the western region. Complex structure underlies the Basin and Range extensional region at the eastern end of the line.

Seismic velocity model of the crust across the western Idaho shear zone based on the EarthScope IDOR controlled-source seismic data.

Acknowledgements: Data acquisition and analysis for this project were funded by NSF EarthScope Program grants EAR-0844264 and EAR-1251724. Seismographs and field support were provided by the IRIS PASSCAL facility. We would like to thank the field crew of volunteers and the landowners and managers who made the data acquisition possible.

Using&Virtual&Fieldwork,&EarthScope&and&Critical&Zone&Observatories&to&Nurture&Public&Understandings&of&the&Most&Important&Earth&System&Science&Ideas&Don$Duggan(Haas,$The$Paleontological$Research$Institution,$dad55@cornell.edu$$

Decades of science education research costing hundreds of millions of dollars have made no discernable improvement in the outcomes of school-based science education writ large. It is unlikely that this presentation will either, but it will identify possible points of departure from standard practice that may offer promise. Large science projects that are national in scope, connect a range of disciplines, and employ systems perspectives to their objects of study provide opportunity for both enhancement of, and departure from, traditional educational practice.

This presentation will begin by raising the question of what ideas and questions are most important to understand about Earth system science; address the fundamental mismatch between what research says about how people learn and the basic structures of secondary and tertiary education; and; finally describe how the in-depth study of place that employs strategies and data from projects like the Critical Zone Observatory Network and EarthScope might lead to more effective educational approaches.

Virtual Fieldwork Experiences (VFEs) are multi-media representations of actual field sites. The driving question for the work is, Why does this place look the way it does? While VFEs may stand in for actual field experiences, they are more intended to catalyze fieldwork and serve as a way to document and share work in the field. Through teacher-student collaborations, learners engage in documenting the close study of their local environment in a way that facilitates sharing with interested others, and allows for years-long interdisciplinary investigation.

Status of EarthScopes Transportable Array in Alaska

Robert Busby1, Max Enders1, Jeremy Miner1, Ryan Bierma1, Jon Meyer2 (1) IRIS Consortium, 1200 New York Avenue, NW, Suite 400, Washington, DC (2) UC San Diego, IGPP, San Diego CA

The EarthScopes Transportable Array is commencing the second year of operations in Alaska. The proposed station grid is 85 km consisting of approximately 261 locations in Alaska and Western Canada. About 71 of the grid locations will be at existing seismic stations operated by the AEC, AVO and ATWC and are being upgraded with shallow borehole installations or higher quality sensors as appropriate. 12 new stations will be collocated with PBO GPS stations. As the Transportable Array has moved to Alaska, IRIS has experimented with different portable drills and drilling techniques to create shallow holes (1-5M) in permafrost and rock outcrops. The goal of these new methods is to maintain or enhance a stations noise performance while minimizing its footprint and the equipment, materials, and overall cost required for its construction. Motivating this approach are recent developments in posthole broadband seismometer design and the unique conditions for operating in Alaska, where most areas are only accessible by small plane or helicopter, and permafrost underlies much of the region. IRIS has partnered with Genasun to produce a lightweight high capacity power systems for cold environments. Based on the proven Genasun charge controller used in PASSCAL and Polar experiments and coupled to Lithium Iron Phosphate batteries a new system allows transport of large capacity battery systems in a single sling load of a helicopter, simplifying station setup and operation.

Figure 1 Planned stations of the Alaska Transportable Array. Red symbols are currently operating, Blue are being installed this summer, while Yellow and Green are planned for FY16 and FY17. A total of 261 stations are planned for operation up to 2019 or beyond.

Simultaneous Inversion of Interpolated Receiver Functions, Surface-wave Dispersion, and Gravity Observations for Lithospheric Structure Beneath the

Western and Eastern United States

Chengping Chai1, Charles J. Ammon1, Monica Maceira2, Robert Herrmann3

1. Department of Geosciences, Pennsylvania State University, University Park, PA 2. Los Alamos National Laboratory, Los Alamos, NM 3. Department of Earth & Atmospheric Sciences, Saint Louis University, St. Louis, MO

As Earthscopes Transportable Array moves out of the east coast region, the unprecedented high-quality seismic data provide a great opportunity to investigate the subsurface structure beneath the region. Estimates of the lithospheric structure have been produced by integrating multiple observations. Though the combination of receiver functions and surface-wave dispersion is ideal for vertical resolution,