Seismic Barrier Protection Of Critical Infrastructure

Seismic Barrier Protection of CriticalInfrastructureRobert Haupt, Vladimir Liberman, andMordechai RothschildMIT Lincoln Laboratory244 Wood St. Lexington, MA 020420781-981-5128 Haupt@LL.mit.eduDISTRIBUTION STATEMENT A. Approved for public release:distribution unlimited. This material is based upon worksupported under Air Force Contract No. FA8721-05-C-0002and/or FA8702-15-D-0001. Any opinions, findings, conclusionsor recommendations expressed in this material are those of theauthor(s) and do not necessarily reflect the views of the U.S. AirForce. 2016 Massachusetts Institute of Technology.Deliveredto the U.S. Government with Unlimited Rights, as defined inDFARS Part 252.227-7013 or 7014 (Feb 2014).Notwithstanding any copyright notice, U.S. Government rights inthis work are defined by DFARS 252.227-7013 or DFARS252.227-7014 as detailed above. Use of this work other than asspecifically authorized by the U.S. Government may violate anycopyrights that exist in this work.I. IntroductionAbstract - Each year, on average a major magnitude-8earthquake strikes somewhere in the world. In addition,10,000 earthquake related deaths occur annually, wherecollapsing buildings claim by far most lives. Moreover, inrecent events, industry activity of oil extraction andwastewater reinjection are suspect to cause earthquake swarmsthat threaten high-value oil pipeline networks, U.S. oil storagereserves, and civilian homes. Earthquake engineering buildingstructural designs and materials have evolved over many yearsto minimize the destructive effects of seismic surface waves.However, even under the best engineering practices,significant damage and numbers of fatalities can still occur.In this paper, we present a novel concept and approach toredirect and attenuate the ground motion amplitudes caused byearthquakes by implementing an engineered subsurfaceseismic barrier – creating a form of metamaterial. The barrieris comprised of borehole array complexes and trench designsthat impede and divert destructive seismic surface waves froma designated ‘protection zone’. The barrier is also designed todivert not only surface waves in the aerial plane, but includesvertical ‘V’ shaped muffler structures composed of opposingboreholes to mitigate seismic waves from diffracting andtraveling in the vertical plane.Computational 2D and 3D seismic wave propagationmodels developed at MIT Lincoln Laboratory suggest that theborehole array and trench arrangements are critical to theredirection and self-interference reduction of broadbandhazardous seismic waves in the vicinity of the structure toprotect. The computational models are compared withexperimental data obtained from large bench-scale physicalmodels that contain scaled borehole arrays and trenches. Theseexperiments are carried out at high frequencies, but withsuitable material parameters and bore-hole dimensions. Theyindicate that effects of a devastating 7.0-magnitude earthquakecan be reduced to those of a minor magnitude-2 or -3eathquake within a suitable protection zone. These results arevery promising, and warrant validation in field scale tests.Keywords – seismic, metamaterials, borehole structures,earthquake mitigationDamage caused by earthquakes to critical structuressuch as nuclear power plants, regional hospitals,military installations, airport runways, pipelines, dams,and other infrastructure facilities exacerbates thedisaster and adds tremendous cost and time of recovery.Even low energy earthquakes resulting from humanactivity can cause significant damage as well. Forexample, wastewater reinjection practices used by theoil industry resulted in over 900 earthquakes in 20142015 in the state of Oklahoma, with a recent 2016 quakeof magnitude 5.8. These continual earthquake swarms,although many small, threaten extremely high valueabove and below ground pipelines that control oilsupply, storage, and transport in the U.S., and is a majoreconomic and environmental concern.Figure 1 (left) describes pictorially the differentkinds of seismic waves, including body waves andsurface waves. From the civil engineeringstandpoint, the most destructive are surface waves(Rayleigh, Love, shear). To protect against them, alarge body of earthquake engineering has beendeveloped, and effective building practices are beingroutinely applied to new construction.However,existing high value structures are unlikely to beretrofitted, thus posing significant risk to lives,economy, and the environment. Here, we propose anovel concept to redirect and attenuate the groundmotion of earthquake surface waves byimplementing an engineered below-ground seismicbarrier around such high value structures. As shownschematically in Figure 1 (right), the structuresproposed employ borehole arrays and trenchcomponents to act as broadband seismic wavebarriers that conceptually create a cloaking devicethat significantly reduces the energy that wouldotherwise reach an area we desire to protect, such asa power plant, oil pipeline complex, or other criticalstructures.

Figure 1. Left: Typical seismic wave types caused by earthquakes. Statistically, surface seismic waves (Rayleigh – groundroll, Love, Shear) can cause significant damage and destruction to man-nade structures in critical infrastructure. Right: Notionalborehole array concept that diverts and absorbs hazardous seismic waves prior to reaching critical infrastructure.II. Prior InvestigationsSeveral “earthquake cloaking” concepts havebeen proposed recently. The main limitation of theall the published approaches so far is the narrowband resonant nature of the structures, which do notaddress the typical bandwidth of a real-life tremor.Furthermore, there are limitations to scalability tolow frequencies (large structures) as well asapplicability to shear waves.(Colombi, et al,, Scientific Reports 2016, 6,19238.) suggested that tall above-ground structures,such as trees or steel towers, can be placedstrategically to absorb some of the ground vibrationsand dissipate seismic energy. However, from theprevious measurements and calculations, onlyenergy at discrete resonant frequencies can bedissipated with this approach. Furthermore, in orderto access frequencies below 1-10 Hz, the height ofthe above ground structures would have to be severalhundred meters, which is impractical.A second approach (Kim and Das, ModernPhysics Letters B 2013, 27) suggested that buryinglarge resonator tanks underground would convert theincident seismic energy into acoustic energy and,thereafter, into heat. Our analysis of this approachwith 3-dimensional finite element modeling foundthat only energy at discrete resonances can bedissipated. Furthermore, seismic shear waves cannotcouple into acoustic air filled resonator structures,thus reducing bthe utility of this approach preciselyfor the more damaging type of waves. Finally,because of great impedance mismatch between hardground and air, coupling of the resonances into theunderground tanks would be extremely inefficient, asmost of the energy would be reflected.A third approach (Krödel, S, et al. ExtremeMechanics Letters 2015, 4, 111-117)proposed burying seismic dampers into the groundfor energy dissipation. Once again, this approach isnarrowband and requires multiple resonant dampers,each working in a separate frequency band, to covera practical range of frequencies. In a scaledexperiment in a tank of sand, researchers found thatmany dozen different-size resonators would berequired to cover the relevant frequency spectrum.Scaling such approaches to full size would beimpractical.Finally, in the only controlled experimentsperformed in earth to date (Brule, et al., PhysicalReview Letters 2014, 112), a team from InstituteFresnel and Menard Construction company haveexcavated an array of several dozen holes in asedimentary basin outside of Grenoble, France. Theteam has studied elastic wave attenuation through thearray of these holes at a source frequency of 50 Hz,provided by a vibrocompactor. This experiment hasgenerated a large amount of excitement in thepopular media as it was the first and the only largescale demonstration of elastic wave diversion in soil.A modest attenuation of 3-5 dB was achieved by theborehole array. The major limitation of this

experiment is its reliance on the phononic crystalapproach, which is inherently narrowband and, often,direction-dependent. Furthermore, scaling thisphononic crystal design to frequencies below 1 Hzand to hard soil conditions that are prevalent in mostlocations would result in impractically large sizes forthe array structure.III. ‘V’ Shaped Opposing Borehole MufflerStructureBy contrast to the previously exploredapproaches, our design involves a broadbandborehole array structure, combining attenuation fromscattering theory as well as reflective muffler designsto enable not only aerial plane seismic wavediversion, but also vertical-depth plane protection.As described below, the muffler structures attenuateboth shear waves and surface waves, over thefrequency ranges relevant to earthquakes in both theaerial and vertical orientation planes.IV. Computer Modeling and Bench ScaleMeasurementsWe initially studied the phenomenology andfeasibility of using subsurface man-made structures toattenuate and divert hazardous seismic waves using 2Dand 3D computer numerical models. Our effortinvolved 1) identifying hazardous seismic waves fromearthquakes, 2) developing computational model ofseismic wave propagation in complex geological media,and 3) developing metamaterial models applicable toseismic cloaking to induce seismic wave diversion andattenuation. Modeling results show subsurface boreholearrays may reduce the seismic energy from hazardousearthquakes in the vicinity of high value structuralassets. Such seismic barrier structures may span 100s ofmeters to a kilometer in aerial range, with well depths of100 meters or less, and costs that would justifyconstruction. We conclude that such barrier arrays mayreduce seismic wave amplitudes by 30 – 50 dB thatwould otherwise reach a high value asset facility. Amodel is shown below using a recent Californiaearthquake’s parameters for input in figure 2.