ML13210A397

G20130548/LTR-13-0637 - John Pearson, MD, Oregon Physicians for Social Responsibility, Et Al. Ltr. Earthquakes and the Safety of the Columbia Generating Station

ML13210A397
Person / Time
Site:	Columbia
Issue date:	07/19/2013
From:	Gilbert S, Pearson J Oregon Physicians for Social Responsibility, Washington Physicians for Social Responsibility
To:	Macfarlane A NRC/Chairman
Shared Package
ML13210A398	List: ML13210A397 ML13224A360
References
CORR-13-0100, G20130548, LTR-13-0637
Download: ML13210A397 (27)

Text

'ETS" U.S.NRC Ticket No: G20130548 U-'d ýt- N ,d.£,

Aroti'cingPe'ople andihe'Environntrit 3013 Assigned Office: NRR OEDO Due Date: 08/15/2013 Other Assignees: SECY Due Date: 08/19/2013 Date Response Requested by Originator:

Other Parties:

Subject:

Earthquakes and the Safety of the Columbia Generating Station

Description:

CC Routing: NRO ADAMS Accession Numbers - Incoming: ML13210A397 Response I Package: ML13210A398 0 I Cross Reference No: LTR-1 3-0637 SRMOOther: No Action Type: Letter OEDO Concurrence: Yes Signature Level: Chairman Macfarlane 0CM Concurrence: No Special Instructions: OCA Concurrence: No NRR to coordinate with NRO.

Originator Name: John Pearson Date of Incoming: 07/19/2013 Originator Org: Oregon Physicians for Social Document Received by OEDO Date: 07/30/2013 Responsibility Addressee: Chairman Macfarlane Incoming Task: Letter OEDO POC: Dan Merzke

OREGON July 19, 2013

,,,

Dr. Allison M. Macfarlane PSR PHYSICIANS Chairwoman Nucleai Re'gulatory Commission FOR SOCIAL Commission Mail Stop 6-16G4 RESPONSIBILITY W~hington, DC 20555-0001 .

.Dear Dr. .Macfarlane; On behalf of the Task Fqrce on Nuclear Power for the Oregon and W~hiiigton chapters of PhysiCians for SoCial Responsibility, we invite you to come to Seattl~ to*

'discuss our concerns about earthquik:es and the safety of the Columbia Generating WASHINGTON Station on the Hanford Nuclear Reservation in Washington State. The CGS nuclear PHYSICIANS

- piant is aGE Boiling Water Reactor with a Mark II cont3inment. It was i~sued it~

FOR SOCIAL RE,SPONSIBILITY construction permit in 1973 and a license to operate in 1984. _

Our group is.one of the signatories to the April 2013 Beyond Nuclear petition to the

. Oregon PSR Board of Directors NRC requesting that all Boiling Water Reacto~s in the U~ted States be closed because*they violate G~neral Design Criteria*16. We believe that, with the Michele Bernai-Graves , MS Treasurer installation of vents without filters, they inherep.tly lack a secure containment' in a Charles Grossman , MD worst case accident. -

Emeritus Member Jennifer James-Long Susan Katz, MD *Many of our concerns about potential ~arthquake hazards at the.CGS nuclear plant President Elect - liave been documented in the locally best-selling book"Full Rip.9.0" by Seattle _

Chris Lowe, PhD Patricia Murphy, ND. Times science reporter Sandi Doughten, in which the WPPSS reactors (the CGS was Secretary formeriy known as WPPSS #2) figure into the story of the mvestigation of the Joan Nugent, RN , MN John Pearson, MD North~est's regional plate tectonics from page 1 forward. We enclose a copy for President . you and your stafftQ review.

Joy Spalding , PhD

, I ' I Washington PSR Of particular interest is the story discussed* in chapter 7, .entitled ' ~The Earthquake Board of Directors That *w ouldn't .Stay Put," which reco,unts the overlay of politics upon science in Tom Buchanan order to place the largest earthquake on record in Eastern Washington, now called _

Vice President - .the Lake Chelan earthquake of 1872, far enough away from the WPPSS site at .

Karen Bowman , RN Hanford to keep it from influencing the design criteria for construction and siting of Secretary Steve Gilbert, PhD the plant. ' ,I Treasurer

. Richard Grady, MD President Recent's_tudieshave fou.D.d evidence of7.0 magnitude earthquakes _or greater on the David C. Hall, MD Hanford site and established faulting connections under t;he Cascade Mountains with Laura Hart, MD Gerri Haynes the South Whidbey Island Fault [Brian Sherrod, USGS], a 200 mile long fault .

Howard Putter, MD leading directly to Richland. In 2009 there were swarms of sinhll magnitude shallow earthquakes on the/eastern edge'ofHamord, near.the ColumbiaGenera~g Station nuclear plant, with a peak motion force of .15 g. The CGS nuclear plant was built to .

a withstand .25 g peak motion force . .

Oregon/Washington RSR Joint Task Force on Nuclear Power

• 812 SW Washington Street, Suite 1050, Portland, OR 97205 Phone: 503-777-2794 Email: washpsr@gmail.com _

Augmenting our case for a thorough reconsideration of the current seismic design requirements for the .

CGS nuclear plant is a study completed in 2007- for the US Department of Energy's Waste Vitrification Facility, located a few miles from the CGS, which indicates that their original estimates of needing to withstand a peak motion force of just over .5 g is insufficient.' The study, entitled, "Technical Basis for

• Certification' of Seismic Design Criteria for the Waste Treatment ;?Ian(, Hariford, Washington- 8188 *

[Brouns, Rohay~ Youngs, Costentino, and Miller]," now recommends that the facility be able to withstand .82 g peak motion force: Plea5e note that this is significantly higher than the .25 g that the CGS was' designed to meet. We enclose a copy of that study for you as well.

We don't have to tell you the consequences of a failure.in the re~ctor containment or in the structure of the CGS nuclear. plant's *elevated spent fuel pool in an earthquake: If an earthquake or quakes strike the site beyond the *design capabilities of this plant, we run the risk of a massive catastrophe.

. . ' ' .

The NRC has discussed the need to reassess the potential damage from earthquakes at the plants it regul~tes as part of the post-I:ukushima reforms. So far, we have seen little at the CGS. beyond a walk through that checked to s,e e if the structures and*systems approved in the 1983 design were intact. In some cases, they were not. Energy Northwest and the NRC say these.deficiencies have now been addressed but have denied public access to a report of a second walk t:hiough completed last year.

There is cuin:intly zero evidence that the utility or the NRC have done anything to update their knowledge of the seismic ...threat to the. safety of the. nuclear plant. *

. ' '

For these reasons and others we can explain more fully ,w hen we meet with you, we ask that you personally

. get involved

. in starting this. .critical r~evaluation of what appears to~ a very s~ismic.ally acti_ve and hazardous site.

Sincerely, *

. .

i~J~.

John Pearson, ~D Oregon Physici~s for Social Responsibility

Steven *a. Gilbert, PhD, DABT Washington Physicians for Social Responsibility *

Encs.: "Full. Rip 9.0," Sandi Doughton, Sasquatch Books, Seattle, 2013 "Technical Basis for Certification of Seismic Design Criteria for the Waste Treatment Plant,

• Hanford, Washington- 8188," T. M. Brouns, A.C. Rohay, R.R. Youngs, C.J. Costentiiw,'and L.F.

Miller, WM2008, The 34th Annual Waste Management Conference & Exhibition February 24-28, 20,08. Phoenix Convention Center, Phoenix * ** * '

\V1vl2008 Conference. Febmaty 24-28. 2008. Phoenix. A2 Technical Basis for Certification of Seismic Design Criteria for the Waste Treatment Plant, Hanford, Washington- 8188 T.M. Brmms. A. C. Rohay Pacific Nottbwest National Laboratmy P.O. Box 999. Richland. WA 99352 R.R. Youngs Geomaui.x Consultants, Inc.

2101 Webstersn*eet, 12thFloor. Oakland. CA 94612-3011 C.J. Costantino C.J. Costantino and Associates 4 Rockingham Road , Spti.ng Valley, NY 1097i L.F. Miller U.S. Department of Energy, Office of River Protection P.O. Box 450, Richland, WA 99352 ABSTRACT In August 2007, Secretary of Energy Samuel W. Bodman approved the final seismic and ground motion critel.ia for the WasiTe Treatment and [rmnobilization Plant (W!P) at the Depantment of Energy's (DOE)

Hanford Site. Constmction of the vVTP began in 2002 based on seismic design criteria established in 1999 and a probabilistic seismic hazard analysis completed in 1996. The design criteria were re-evaluated in 2005 Ito address C[Ues1tions from tbe Defense Nuclear Facilities Safety Board (DNFSB),

resulting in au increase by up to 40% tn [he seismic design basis. DOE a1mounced .in 2006 rue suspension of constmction on the pretreatment and high-level waste vin*ification facilities within the WTP to validate the design with more stringem seismic c1iteria. In 2007, the U.S . Congress mandated that the Secretary of Energy cettify ithe ti.ual seismic and ground motion c1.iteria pl.ior to expendimre of funds on con:stmction offhese [WO facilities . Wid11the Secretary 's approval of the ti.nal seisrni*c criteria in the summer of 2007 ,

DOE authmized resta11 of constmction of the preu*eatment and high-level waste vitrification facilities .

The technical basis for the certification o:f seismic design criteria resulted from a two*-year Seismic Boreholes Project th'I.H planned, collected, and analyzed geological data from four new boreholes drilled to depths of approximately 1400 feet below grotmd sm*face on the WTP site . .A key uncettainty identified in the 2005 analyses was the velocity conu*asts between the basalt flows and sedimenta1.y interbeds below the WTP. TI1e absence of directly-measured seismic shear wave velocirie in the edimentruy interbeds resulted iu the use of a wider and more conservative range of velocities in the 2005 analyses. The Seismic Boreholes Project was designed to directly measme the velocities and veloci-ty contrasts in the basalts and sediments below the WTP, reanalyze the ground motion response, and assess the level of conse1vatism in the 2005 seismic design cdte1ia.

The charactetization and analysis effort included 1) downhole measurements of the velocity propetties (including uncettainties) of the basalt/interbed sequences. 2) confinnation of the geometly of the contact between the variorts basalt and interbedded sediments through examination of retlieved core from the corebole and data collected througll geophysical logging of each borehole, and 3) predi-ction of ground motion response to an earthquake using newly acquired and historic data . The data and analyses reflect a significant reduction in the uncertainty in shear wave velocities below the WTP and result in a significantly lower spe*ctral acceleration (i.e., grmmd motion). The updated ground motion response

WM2008 Conference. Febmruy 24-28. 2008. Phoenix.l\Z analyses aud conesponding design response specn*a retlect a 25% lower peak horizontal acceleration thau reflected in the 2005 design ctiteria. These results provide confidence that the WfP seismic design c1ite1ia are consenrative.

~TRODUCTION The U.S. Depat1ment of Energy (DOE) is constmcting a Waste Treatment aud Immobilization Plant (WTP) to treat and vitrify undergrm.md tank waste stored at the Hanford Site in southeastem Washington State (see Fig. I.) The W!P cornptises fom- major facilities: a pren*eatrnent facility to separate the tank waste into high level waste (HL\\') and low-activity waste (LAW) fractions, a HL\V Vitrification facility to immobilize the HL W fraction in borosilicate glass, a LAW Vitlification facility to immobilize the LAW fraction in borosilicate glass, and an Analytical Laboratmy to suppoll operations of the three treannent facilities .

.

\- ---------~--------'--

' y 1-w.fm*(l Sot ~

f lb .uul<uy

} /

1'-~rt l'lDJJ id~

"""' - - __ ....__

1 l*- *

• WTP

. - - site

'\ zoo w <t i...'ll en , -

AN: o

, I I

400 ...

'

~.

I

>

'* '

I' 0 l 4 6 8 1olc .efr.r<

I I t I l Fig. I. Location of the Waste Treatment and Immobilization Plant (WTP) site.

WM2008 Conference_Febmaty 24-28 _2008_ Phoenix. AZ The Hanford Site and WTP ru*e situated on a sequence of sedimentaxy units (Hanford and Ringold Foxmatious) that overlie the Columbia River Basalt Group (CRBG). The CRBG is a sequence of flood basalt flows that erupted between 17 and 6 million years ago from fissm*es or vent systems in Oregon.

Washington_ and Idaho, and fonns the main bedrock of the WTP_ The upper fom basalt flo'.vs (Saddle Mmmtains Basalt) were laid do\vn over a period of time which allowed sediments ofthe Ellensbmg Fotmation to accumulate between basalt layers. The general stratigraphy of geologic mlits of interest below the WTP is show11 in Fig. 2.

0 100 200 300 400 500 1-w 600 w

u_

~ 700

t

1-a.

w 800 0

900 1000 1100 1200  :!

>~

BYRON *~~

1300 INTf.R'8EO

• 5" Z <

~0) 1400 1500 Fig. 2. General stratigraplly and approximate depths below grmmd smface of geologic units of interest below the WTP.

\VM2008 Conference. Febmaty 24-28. 2008. Phoeuix. A2 The seismic design basis for the W1P was established in 1999 based on a probabilistic seismic hazard analysis completed in 1996 [1]. The Defense Nuclear Facilities Safety Board (DNSFB) subsequently initiated a review of the seismic design basis of the \VTP. In March 2002. the DNFSB staff questioned the assumptions used in developing the seismic design basis. patticularly the adequacy of the site geotechnical smveys. and subsequently raised additional questions about the probability of eatthquakes.

adequacy of the "attenuation relationships that describe how ground motion changes as it moves from its source in the eatth to the site. and lat*ge tmcertainty in the extrapolation of soil response data from Califomia to the Hanford Site. Between 2002 and 2004. the DOE Office of River Protection (ORP) responded and resolved many of the questions raised, and developed a plan to acquire additional site data and analysis to address remaining questions. The key featmes of this plan were 1) acquiring new soil data dmvn to about 500 ft (152 m), 2) reanalyzing the effects of deeper layers of sediments interbedded with basalt down to about 2,000 ft (610 m) that may affect the attenuation ofeat1hquake ground motions more than previously understood, and 3) applying new models for grmmd motions as a function of magnitude and distance at the Hanford Site.

In 2004 and 2005, the Pacific N01thwest National Laborat01y (PNNL) led eff01ts for DOE-ORP to address feanu*es I and 2 of the plan by collecting site-specific geologic and geophysical characteristics of the WTP site and condncring modeling of the "WTP site-specific ground motion response. New geophysical data were acquired. analyzed, and interpreted with respect to existing geologic infmmation gathered from other Hanford-related proj ects in the WTP area. Limited infonnation from deep boreholes was collected and intetpreted to produce a model of the deeper rock layers consisting of the interlayered basalts and sedin1enta1y interbeds. The ea1.thquake grmmd motion response was modeled, and a se1ies of seusitiviry studies was conducted tto address areas in which the geologic and geophysical infonuation has significant remaining uucenainties. This effmt culminated in 2005 with issuance of an updated seismic response a11alysis for the WTP site [2, 3]. The updated seismic response a11alysis used existing and newly acquired seislnic velocity data, statistical analysis, expett elicitation, and grmmd motion simulation to develop interim design ground motion response spectra which enveloped the remaining uncertainties.

The uncettainties in these response spectra were enveloped at approximately the 84m percentile to produce consetvative design spectra , which conuibuted significantly to atl increase in the seismic design basis (see Fig. 3).

A key unceltainty identified in the 2005 analysis was the velocity contrasts between the basalt flmvs and sedimentru.y intetbeds below the WTP. Results of modeling indicated that the velocity structure of the upper four basalt flows and the interlayered sedimentaty interbeds produces strong reductions in modeled eru.thquake grmmd motions propagating through them. Uncet1ainty in the strength of velocity conu-asts between these basalts and interbeds primarily resulted fiom an absence of measmed shear 11vave velocities (Vs) in the interbeds. For the 2005 atlalysis, Vs in the interbeds was estimated from older, limited compressional wave (Vp) data using estimated ranges for the ratio of the two velocities (VpNs) based on analogues in similar materials. A range of possible Vs for the interbeds and basalts was used and produced additional tmcertai.uty in the resulting response spectra.

In late 2005, DOE-ORP initiated planning for the Seismic Boreholes Project (SBP) to emplace additional boreholes at the WTP site and obtain direct Vs measurements and other physical propetty measmements in these layers. The goal was to reduce the tmcettainty in the response spectra and seistnic design basis, and potentially recover design margin for the WTP. PNNL was selected to manage the SBP, with oversight 1iom DOE-ORP and the U.S. Anny Corps of Engineers (USACE). The p1i01ity of the SBP activities was elevated in 2006 as a result of fiscal year 2007 congressional autbmization that limited fiscal year 2007 expenditures for the WTP until ... rhe date on which the Secretary of Energy celtifies to

WM2008 Conference. Febmruy 24-28. 2008. Phoenix. A2 1.0 r---- - - - - - . . . .- - -- - - -.......- - - -- -......,.

- - Original1996

- RGM-2005 C) 0.7

-

c

.2

!

.!G)

(J 0.5 (J

c(

-..ca

-&.

(J

,,

0.25

"'

_... ,, ,

~

0.0 ' - - - - - - - - . - . .......-------~------.......

0.1 1.0 10.0 100.0 Frequency. Hz Fig. 3. Original 1996 and revised 2005 horizontal design spectra (RGM) ar 5% damping (2]

the congressional defense cmnrninees rbat the final seismic ru1d grmmd motion crite1ia have been approv~d by the Secretaty .. . 1 ~

APPROACH The approach to the SBP involved four main elements: 1) plaillling and site prepru*ation. 2) new borehole installation, 3) data collection, and 4) site seismic response analysis. A multi-contractor project Iteam was f01med to plan and implement the project, including all health and safety supe1vision and control, project management and technical direction, interface contwl, conu*acting, and environmental compliance. 11lree test boreholes were installed adjacent to the HLW Vitrification and Pretreatment facilities at the WTP to conduct downhole logging and obtain adequate data to detennine the vruiability of shear wave velocities and other physical properties across the footptint of the two facilities impacted by the revised design basis. A single wireline corehole adjacent to one of the test boreholes was also installed to provide conelation of the geology to the geophysical logging data. All fom boreholes (three "test" or **deep" boreholes and one corehole) were d!illed to a depth of approximately 1400 ft (427 m) below ground smface, so as to peneu*ate and extend past the fom sedimentat-y intet*beds and fom basalt members of interest. Locations of the fom boreholes C4993, C4996, C4997, and C4998 ru*e depicted in Fig. 4. A suite of geologic and geophysical data including in situ velocities and densities were collected from the new boreholes and ru*e summarized in Table I. The project elements of 1) Plru1ning and Site Prepru*ation,

2) New Borehole Installation, and 3) Data Collection were completed in 2006 and early 2007, and 1

John Warner National Defense Authorization Act for Fiscal Year 2007. Public Law 109-364 (H.R.5122 ENR),

Sec. 3120, Limitations on Availability of Funds for Waste Treaunent and Immobilization Plant.

WM 2008 Conference. Febm<uy 24-28. 2008. Phoenix. AZ 1\." ,.,_*, 11: .

~---------------------- ~

• Completed Borehole o 20 <O ec eo :oom 0 Comple ted Corehole tOO :co J ~O' f t S colc /~

2006/lla./Wll'SB>/QUt (II ll i) //

Fig. 4. Location of fmu* new boreholes installed adjacent to WTP Pretreatment (PTF) and HL W Villification (HL 'vV) facilities .

reported previously [4]. The approach used to analyze the new borehole data, perfonn the site response analysis, and develop final design response specn*a is desc1ibed below.

Data and interpreted results of in sim velocity and density measurements fi.*om each borehole were evaluated and analyzed to produce a set of final site-specific velocity and density models representing the W!P site. The objective was w integrate data from the new boreholes and previous site-specific studies into a set of models for use in evaluating the seismic response of the WTP.

New site response modeling and analysis was perfonned to process the new velocity and density models and detennine the overall impact of reduced Ullcenainty on the design response spectra for the WTP Site.

GeomatJ.ix Consultants of Oakland. Califomia, was selected to update the WfP site seislllic response calculations completed in 2005 by incmporating the new velocity and density models and other geophysical data collected ft*om the WTP site boreholes. A panel of expetts was convened to review the new borehole data and provide input on the approach and range of values of the input parameters to the site response models . A full probabilistic analysis was completed and generated a disl.libution of relative site response cmves for the WTP site. C.J. Costantino and Associates applied the 841h percentile results of

\VM::?008 Conference. Febmaty 24-28. 2008. Phoenix. A2 Table I. Data Collected from WTP Seismic Boreholes Property Method Sheat" (s) and compression (p) Cl Suspension (p-s) logging wave velocity 0 Downhole logging (impulsive and vibratory sources)

Density Cl Gravity-density logging Cl Compensated density (y-y logging)

Geometiy of contact 0 Geologic logs (examination of core/cuttings)

(depths/thicknesses) 0 Geophysical logging suite

- Compensated density (y-y)

- Neun*on porosity

- Dual induction resistivity

- Full wavefonn sonic Modulus reduction and damping (J Resonant column and torsional shear tests Sediment pa11icle size 0 Gradation testing Borehole condition (J Acoustic televiewer a Caliper logging 0 Gyroscope surveys the site response analysis to generate a WTP site-specific grotmd motion design response specu*a (WSGM). DOE used these results to confmn the existing seismic design ctiteria for the WTP established in 2005 was consetvative.

SITE SPECIFIC VELOCITY AND DENSITY MODEL RESULTS Shear and compression wave velocity measurements were made using two basic techniques, suspension and downhole logging. Suspension logging measures the velocities near the borehole wall using high-fi:equency signals produced and recorded on a string of instruments suspended in the boreholes.

Downhole logging measures velocities over a larger area smTotmding the borehole by using a lower-frequency surface energy source *with a geophone clamped at depth. Two different types of energy sources were used at the surface for the downhole measurements-an impulsive som-ce that produces a single, tmalllbiguous signal, and a vibratmy somce, which is more difficult to interpret but has the greater energy required to reach the depths of these boreholes. The first source was either a sledgehanm1er or small mechanical device. The second source was a large tmck-mouuted elecn*o-hydraulic vibrator. A desc1iption of the teclmiques, equipment, and detailed results of these studies m*e available elsewhere [ 5-8].

Systematic differences were found between the suspension and downhole logging measurements.

Suspension logging gives a ve1y high-resolution measmement, but the signal frequencies of the downhole method are similar to those of earthquakes impmtant in grom1d-motion response modeling. The suspen-sion logging measmements gave velocities significantly greater than the downhole measmements in the basalts for both shear and compression waves . Downhole logging shear wave velocity data fi*om the three boreholes and the core hole were combined statistically to produce an average velocity model of the WTP site. Suspension logging results were used to shape the downhole velocity profiles to address details of velocity reductions in the basalt flow tops that were not modeled previously.

\VM2008 Conference. Febmaty 24-28. 2008. Phoenix. AZ Figme 5 presents results of shear wave travel-time measurements and interpreted velocity (V s) results using the impulsive somces in borehole C4993. and includes data collected in the suprabasalt sedin1ents as well as tv.ro uppe1most basalt milts (Elephant Motmtain artd Pomona) and sediruentaty interbeds (Rattlesnake Ridge and Selah) before and after installation of stainless steel casing. Overall. ve1y similru*

results were obtained in boreholes C4996 and C4997/C4998 (not shown). with variability across the boreholes generally less than 30%. However. borehole C4993 indicated a reduction in Vs from the lower region of the Hanford foflllation (H3 unit) to the upper region of the Ringold Fmmation (Cold Creek Unit

[CCU]). whereas measm:ements in the other two boreholes indicated either a much smaller reduction or slight increase in Vs.

Figure 6 presents results of shear wave travel-time measuremenrs and interpreted *s results using the vibratmy source in boreholeC4993 , and includes data collected through all ofthe basalt and sedimentmy Vertical Travel Time - milliseconds 0 50 100 150 200 250 300 100 Hanford Fonnation 200 Ringold 300 Formation e:;.::-=:~

Elephant Mountain --~

Basalt 8000 ftlsec 500 Rattlesnake Ridge lnterbed Pomona Basalt 600

• Slingshot Soll1:e * (Jnc3sed Hole . 211 Od 06

• Slingshot Soll1:e

• stainless <:asilg - 30 Jat 111 Sledgehammer Soln:e - Staness e;s,g - 9 Feb 07 Fig. 5. Shear wave velocity measmements in borehole C4993 using an impulsive seislnic source (adapted from Redpath [5])

W112008 Conference. Febma1y 24-28. 2008. Phoenix, AZ Time (sec )

0.19 0.20 0.21 0.22 0.23 0.24 0.25 0.26 0.27 0.28 0.29 0.30 0.31 0.32 0.33 0.34 0.35 0.36 0.37 0.38 0.39 0.40 0.41 0.42 0.43 0.4-4 300 400

..... ~-* -**- -~-

1 7000fps

- -*- -- _.... ..... ***- --- -***

1-*-* 1-**

Ringold Formation Elephant Mountain Member

• • -- --* -*~

....., ____

500 --"~ -~

1 3000 fps -- - --- - **-- . -*.. ~-

I 1 Rattlesnake Ridge lnterbed h

~ I.-

..... **-- -*** -*

--* ***-* **-** **-- --* -*** *-*- -~~ ,

600 1 8440 tps 1

"

Pomona Member 700 ~ I

-- -~-1--------lil -- ~-- --

-~ 2620 fps ~* --- -*- --** --**

- Selah lntert>ed .::.

• • - --- ~-jilif!

-1 Esquatzel Member I ~

~-

g ---- --r--- -- ---

800 ,.------- 7770 fps

_.._ -- ----... ~~ --2520 fps--I -**- _ , - -- ...... --- -~ I a

..

.J:::.

Q. --l---1--~--+--4--

900 ,------ ~ Cold Creek lnterbed I

--- -~ ~ r-..., ---r-* * - -- -- --*- -- -- r---f--*** *-- ---

--* ---r-- *r--r-- -- ~. .. -~

l v~ ......

+/-

1000 I-- Umatilla Member

./ 1 8210 fps J I frequency changed 1100 1Mabton lntert>ed j I from 50 to 30 Hz v 1L 273o tps

............

- *- - -* - - -- --* -** bL I

1200

• Fixed Sine, 50 Hz; Date 1211912006 L_J ~

I l..._

1300 1--- f-1 Priest Rapids Member l-

.._----1

• Fixed Sine, 30 Hz; Date 12J20/2006 7520 fps 1400 1--*

-- --t*t=t*t*i -- ---"-:_t=i""'- -* -* -*-r-.. --t-t-t-*- *n*- l.mEl

_T_

Fig. 6 . Shear wave velocity measurements in borehole C4993 using a vibratmy seismic source.

Velocities repmted in ft/sec (.tps) (from Stokoe et al. [6]).

interbed milts. Overall, ve1y similar results were obtained in boreholes C4996 and C4997 (not shown).

Vatiability across boreholes was generally less than 20%.

Density measurements were also made using lt\¥0 different methods. A standard geophysical logging method measured density at the borehole wall. A second method using a borehole gravity meter measmed density far fi:om the borehole wall The second method is not affected by drill fluid invasion, cement, or metal casing in the borehole. Comparison of the two density measurements gave good agreement except where borehole inegularities or steel casing were present, TI1e shear wave velocity and density data from the three boreholes and the core hole were combined statistically to produce an average velocity and density model of the WfP site. The final set of profiles integrated data fi:om the new boreholes and previous studies and provided a set of updated input parameters for subsequent use in evaluating the seismic site response of the WTP site. The statistical analysis also provided bounds on the vruiability and tulcettainty of the profiles. Figure 7 depicts the 2007 velocity model for the suprabasalt sediments along with the 2005 velocity model for comparison. The 2007 model was produced by integrating and averaging the new borehole Vs data with pti or seismic cone penen*ometer and downhole data, and used geologists logs to define the range of geologic U!.lit tlllclmesses. The 2007 models for the suprabasalt sediments are comparable to the 2005 models except for a sharp Vs conn*ast fi.*om Hanford sand to gravel (H2/H3). a shrup Vs contrast fi*om Cold Creek Unit to lower Ringold Unit A, and a lllgh Vs in Ringold Unit A that is comparable to Vs in basalt flow top, Figure 8 depicts the 2007 velocity model for the basalts and interbeds along with the 2005 velocity model for cmnpru*ison. The 2007 model was produced by integrating and averaging the new borehole Vs data

\Vlv12008 Conference. Febnuuy 24-28. 2008. Phoenix. AZ 0

0+-~rr----~------~--------~---------r-------,

- ~d ZJJ1 M:rl:ls Hanford H2Unit I (sand) 100 "I

' H2/H3 Boundary

~

100 ' '

I i I

~

~

J:: 2Xl Hanford aQ)

I H3 Unit (gravel) 0 I

'

I H3/Cold;Creek 2S) ~ Boundacy I

I Cold Creek Cold Crerek!Ringold Unit Unit A Bpundary (gravel)

Ringold Unit A 300 'I 1- - I -- - - -

.. - . -- - - ~ -

:J;ieP.ha~f: : :*:

• *: Mountair\- :-;

Merni:ie:r :: :*:

0 ' ~ o A *

• 0 o ....

Fig. 7. Comparison of2007 Vs models to 2005 Vs models for suprabasalt sediments.

collected using downhole logging with the vibratmy somce for all basalts and interbeds and the impulsive somce for the upper basalts and interbeds. As with the suprabasalt sediments, geologists' logs were used to define the range of geologic unit thicknesses. Significant differences can be seen between the 2007 and 2005 models for the basalts and interbeds . The basalt Vs values for 2007 are comparable to the upper limit of the 2005 analyses, and the interbed Vs values are significantly less than the upper litnit of the 2005 analyses. The measmed results and conesponding 2007 model represents significantly greater contrast in Vs between the basalt and interbed units. The flow top gradients are also noticeably different.

The velocity profiles for these flow top gradients were estimated using density and suspension logging Vs data, which enabled unit-specific gradients to be developed for 2007 which have a gradual lise fi:om a much smaller Vs value than estimated in 2005 . Finally, the 2007 profile represents the intertlow features

WM2008 Conference. Febmaty 24-28. 2008. Phoenix. AZ She<< W:Jve 'klocity (fps) 0 400J 100)J 500 700 Selah Interned z *.

g Cold Creek

..r:: ~  ;* lnterbed

~

~

1100 Mabton lnterbed Fig. 8. Comparison of2007 Vs models to 2005 Vs models for basalts and interbeds.

that are present in the Umatilla and Priest Rapids members, introducing flow top gradients between specific basalt flows .

TI1e final velocity and density profiles are represented by a set of input parameters required for seismic site response a11alyses that include densities of all stratigraphic units, stratigraphic mlit thicknesses, basalt flow top thicknesses, Vs of all stratigraphic mtits, and basalt flow top velocity gradients (as depicted in Figures 7-8).

WM2008 Conference. Febmruy 24-28. 2008. Phoenix. AZ SITE RESPONSE Al\"ALYSIS RESULTS A.J.~ DISCUSSIO I\"

An updated site response model for the WTP site was developed by PNNL and Geomanix. This eff011 was suppmted by an expe11 panel that provided guidance on the interpretation of the data atld recommendations on fmmulating the updated site response modeL

\1odel Input Par ameters Input pru*atneters for the site response model are listed in Table II along with the relative (qualitative) lmcettainty and impact on overall site response (i.e., spectral acceleration) for both the 2005 and 2007 calculations. For model paratneters with significant uncertainty. a range of alternative parameter values and weights/probabilities was used and agreed to by expe11 panel members. Figure 9 shows the site response model logic n*ee developed to represellit the unce1tainties in the sire dyuatnic properties. A few points should be made to assist me reader .in understanding the stmcntre of the logic tree. For the sake of visual clru"ity, the logic n*ee figure does not display all of the brru1ches that exist in the actual logic n*ee used in the analysis. Those brru1ches that are repeated at multiple nodes of the n*ee ru*e not shown. For example, note that at the f1rst node for the value of"kappa" there are three branches for altemative va lues.

Tile next level (node) of the logic n*ee indicates altemarive models for H3/CCU sediment velocities. The altemative H3/CCU velocity models are only shown for a single kappa branch. Tllis is only for the sake of keeping the figme uncomplicated. In the site response calculation, all nodes of the logic n*ee are assigned the relevant branches such tba[ aU possible sets of site pru-ameters are used.

Overall damping is represented by the kappa" value in the sire response modeL Uncenaimies in kappa are represented by three alrema.tives which were desctibed previously by Rohay and Reidel [2]. The altemative models used in the 2007 site response model are consistent with those used in 2005. No new data have been collected that would wanam cbauge to the altematives used previously.

Table IT. Model Inputs to 2005 and 2007 Site Response Analyses and Relative Unce1tainties and Impact 2005 2007 Model Inputs Sediment velocities Med Med Med Basalt and interbed High High High velocities Sediment modulus reduction and damping Low Med Med curves Kappa (overall damping ) Med Med Med Geometry of contact Med Med (depths/thicknesses )

Densities Low Low

WM2008 Conference. Febmaty 24-28. 2008. Phoeuix. AZ Approac/J for H2 Unit G!Gmax H3 Unit G/Gmax Kappa H3/CCU Vs (fps) Sa::~:avs:lt/ G/Gmax anr! and Damping and Damping Damping Curves Curves Curves O.o18sec (0.3) Individual Layer Velodties (0.5)

Generic (0.25)

Meoq(2003)

Cu=5 (0.2)

Menq (2003}

Cu = 20 (0.2)

Meoq(2003) Menq (2003)

Cu= 15 Cu=30 (0.6) (0.6)

Menq (2003)

Cu= 50 (0.2)

Menq(2003) 0.031 sec Cu=25 (0.3) (0.2)

Terminology Kappa . Overo!l Damping;(#.#)

• weig/11/p..-oiJabil ity; Vs

• snear wave velocity in It/sec; G/Gmax- Modulus Reduction; Cu - Coetricient or Un/forrnlty; H3- Hanforr! fonnaUon, H3 unit; H2 - Hanford forma..'IOn, H2 un.t; CCV - Ringold Formation, Cold Creek Unit Fig. 9. Updated sire response model logic n*ee for the WTP. (Ntm1bers in parentheses below branches indicate assigned weight.)

The velocities of sediments and basalts used in the 2007 site response model were as depicted in the velocity model Figures 7 and 8. Two altemative models for the H3 and CCU sediment velocities were used to represent tmcettainty in measmed values for these specific units. For the basalt and interbed velocity model. two altemative models were used. One was based on the individual unit velocities as depicted in Figure 8. The other altemative used a single mean or common value for all basalt units and another mean value for all interbed units.

Uncettainties in the sediment modulus reduction and damping cmves were represented by two main altematives - genetic soil cu1ves fi*om the literarure. atld site-specific ctnves. Robay and Reidel [2,3]

used published genetic modulus reduction and damping relationships fi*oiD the Electlic Power Research Institute (EPRl) (9] and Rollins et al. (1998) [10] to represent the non-lineat* behavior of the suprabasalt sediments in the 2005 site response model. During the field investigation conducted in 2006, bulk samples of these sediments were obtained. Dynamic resonant column/torsional shear (RCTS) tests were perfonned on reconstituted samples by the University of Texas at Austin (UTA). The bulk samples obtained by UTA were scalped to remove large pa1ticle sizes before testing. Results of this UTA testing showed that a Menq [ 11] model provided a reasonably good match to the scalped test data, and that the model could be used to develop approptiate modulus reduction (G/Gmax) and damping relationships for the in-situ grain size distributions of the H2, H3 , and CCU sediment layers. Therefore, two modeling approaches for specification of the G/Gmax and damping relationships for the sediments were incorporated into the site response IDodel. one based on the use of genetic curves and one based on development of a range of site-specific cwves using the model developed by Menq.

\VM2008 Conference. Febmruy 24-28. 2008. Phoenix. AZ Site Response :\-'lodel and Design Response Spech*a A site response analysis was petfonned to compute the relative response of Hanford site proftles and Califomia soil site profiles to g:rmmd motions representative of the site hazard at the specified renuu petiod. These site response a.Ilalyses were pe1fonned using outcropping motions back-propagated to the cmstal depth where the Califomia and Hanford sites have similru* shear wave velocities, which is at a depth of3 km [1]. These rock motions were then propagated upward through randomized Califomia soil site profiles and randomized Hanford profiles. Geomettic mean (mean log) response specu-a for the computed smface motions were used to compute the ratio ofHanford smface motions to Califomia soil site motions. This ratio. tenned the relative runpli:fication ftmction (RAF) was used by Rohay and Reidel

[2] to adjust the miginal horizontal design response spectt1tm developed using Califomia-based empitical ground motion models to reflect the ground motions representative of the response oftbe Hanford WTP 1>ite to similar levels of shaking. The same approach was followed in this smdy. The site response model logic u*ee is used in the fttll probabilistic analysis to produce a disuibution ofRAF cmves.

Rohay ru1d Reidel developed the 2005 revised horizontal gwund motion design response specUl.llll (RGM) for the WTP site by multiplyit1g the original WTP llmizontal design response spectmm (based on the I 996 PSHA results) by the RAF detived from relative site response analyses. For consetvatism in the final design recommendation, the 84th percentile relative amplifications il"om the ftilllogic u*ee analysis were used to develop the RGM. Figure I 0 shows the oligi11al i 996 design response spectnun (1996 111 DRS): the 1996 DRS multiplied by the 2005 84 percentile RAF, and the resulting RGM.

0.9 r;::============::=;-----------,

- 199£DRS

,,

0.8 - - -

• 1996 DRS x 2005 84th RAF I I

- - RGM-2005 0.7

- 0.6 C) c:

0

~ 0.5

...

Qi 0

t. 0.4 f

t)

...

Jt 0.3 0.2 0.1 0.1 10 100 Frequency (Hz)

Fig. 10. Development of 2005 iuteritn WTP hotizontal design response spect11un (RGM-2005) compru*ed ro tbe migmal horizontal design response spectmm (1996 DRS)

WM2008 Conference. Febmruy 24-28. 2008. Phoenix. AZ The RGM was developed by smoothly enveloping and broadening the peak of the 1996 DRS x 2005 84th RAF cwve. The resulting RGM-2005 increased peak h01izoutal ground motion by up to 40% over the Oiigiuall996 design crite1ia.

Figm*e 11 shows the same three cmves as Figure 10 along with two new curves representing the 2007 site response analysis (1996 DRS x 2007 84th RAF) and updated WTP site-specific h01izontal ground motion design response spectra (WSGM-2007). The 84th perceutile RAF was again used in 2007 for conse1vatisrn, and the resulting WSGM was developed by smoothly enveloping and broadening the peak of the 1996 DRS x 2007 84th RAF cmve. The resulting WSGM-2007 decreased the peak hmizontal ground motion by approxin1ately 25% from the 2005 RGM design c1ite1ia. This significant reduction in peak ground motion is attiibuted to significantly smaller uncenainty of median sheru* wave velocities for the basalts and interbeds based on direct measmements, significantly greater conu*ast between basalts and interbeds velocities, and more non-linear and greater damping based on site-specific data.

The final results of this study. including a desc1iption of the geology, an updated velocity and density model, updated site response analysis, and updated design response spectra, were fonnally doctm1ented in May and Jtme of2007 (12-14]. These results confinned that the RGM-2005 used as the basis for design of the vVTP was conseivative. In August 2007, the Secreta1y of Energy cenified to Congress that the ground motion design criteria for the WTP were t1nal, and restart of constmction of the pretreaunent and HL W vitrit1cation facilities was authorized.

0.9

- 1996DRS

- -

• 1996 DRS x 2005 84th RAF

,,

I I 0.8

- RGM-2005

- -

• 1996 DRS x 2007 84th RAF 0.7 - WSGM-2007 0.6

§ c:

0

e G>

0.5 Gi

......

oc( 0.4

\

!>

u

\

,_

-, _ ' ' , _____ _

\

"'

Q.

0.3 \ ....

"' \

'

0.2 0.1 0.0 0 .1 10 100 Frequency (Hz)

Fig. 11 . Development of WSGM-2007 horizontal design response specU1m1. Also shown are the otiginal design response spectmm (1996 DRS), the original design response spectnun multiplied by the 2005 84th-percentile RAF, and the RGM-2005

WM2008 Conference. Februmy 24-28. 2008. Phoenix. AZ

SUMMARY

A~"D CONCLUSIO~S One of the Deparlment of Energy's top primities and toughest technical challenges has been resolving seismic issues for Hanfmd ' s Waste Treatment and Imm.obilization Plant (WTP). Constmction of the t\vo perfonnauce categmy three (PC -3) facilities was halted 1mtil the Secretaty of Energy could certify the fmal seismic and gmtmd motion critetia to Congress. The Seismic Boreholes Project was initiated by DOE to address uncel1ainties in the grmmd motion ctite1ia for WTP. In 2006. DOE-ORP assigned Pl'I'NL the responsibility of managing the effort to drill fom- deep boreholes to depths of approximately 1.400 feet directly on the \VTP const11.1ction site and collect the needed seismic data . PNNL led the team of local and national indust1y and mliversity expetts in deep borehole drilling, geologic and seismic data collection. and seismic response analysis.

TI1e project team completed all project deliverables within 15 months of the flrst milling se1vices request for proposal. The project required seventeen conu*actors, many of them working rmmd-the-clock dtuing drilling operations to install boreholes, collect geophysical data and samples, and perfmm grmmd motion response modeling. The project was completed safely, within budget, and within schedule expectations.

Oniy one lost-time injmy was expetienced dming t11e more than 170,000 work homs.

TI1e end result was the sciemifically defensible resolution of critical seisnlic safety issues that enabled the Secretruy of Energy to certify seismic design criteria and authorize WTP constmction to resume at the site. The resulting WSGM-2007 confinned that the existing design criteria are consetvative. Use ofthe updated WSGM-2007 ctitetia will be limited, but will assure substantial design ma1*gin for the WTP and may be used by DOE as needed on a case-by case basis.

ACKNOWLEDGEMENTS The authors acknowledge the U.S. Deprutment of Energy Oftke of River Protection for programmatic guidance and fmancial resources to cru1y out this work, and the U.S . A.nny Corps of Engineers fox additional technical oversight and assistance in de.fining and implementing this work scope. TI1e authors also acknowledge the many team members identified throughout this paper who conuibuted significantly to this effo11.

REFERENCES I. Tallman. A.M. 1996. Probabilistic Seismic Hn::.nrd Analvsis, DOE Hanford Site, Washington .

WHC-SD-W236A-TI-002, Rev. I. Westinghouse Hanford Company, Richland. WA.

2. Rohay, A. C. and S.P. Reidel. 2005 . Site-Specific Seismic Site Response Model for the Waste Treatment Plant, Hanford, Washington . PNNL-15089, Pacific Northwest National Laborato1y.

Richland, W A.

3. Rohay, A.C. and S.P. Reidel. 2006. "Site-Specific Seismic Site Response Model for the Waste Treatment Plaut, Hanford, Washington." In Proceedings ofWM'06 Conference, February 26-March 2, 2006, Tucson, AZ. WM-6321. Pacific Northwest National Laboratoty, Richland. W A.

4. Brouns. T.M., A.C. Rohay. S.P. Reidel, and M.G. Gardner. 2007 . "Reducing Uncettainty in the Seismic Design Basis for tl1e Waste Treatment Plant Hanford, Washington." In Proceedings of WM'07 Conference, February 25-March 1, 2007, Tucson, AZ. WM-7434/PNNL-SA-54097. Pacitic N01thwest National Laboratory, Richland, W A.

5. Redpath B.B. 2007. Doll'nhole Measurements of Shear- and Compression- Wm*e Velocities in Boreholes C4993, C4996, C4997 and C4998 at the Waste Treatment Plant DOE Hanford Site.

PNNL-16559, prepru*ed by Redpath Geophysics, Mmphys. Califomia, for Paciflc Nmtbwest National Laborarocy, Richland, Washington.

WN12008 Conference, Febmaty 24-28. 2008. Phoenix. AZ

6. Stokoe KH IT. S Li, B Cox and F-Y Menq. 2007. Deep D01111hole Seismic Testing at the Waste Treatment Plant Site, Hanford, WA . Geotechnical Engineering Rep01t GR07-10/PNNL-16678 Volmnes I-VI. prepared by University of Texas at Austill. Austin.. Texas, for Pac.ific Nmthwest National Laboratory, Richland. Washington.

7. Diehl. J. and R. Steller. 2007. Final Data Report: P- and S-Wm*e Velocity Logging Borings C4993, C4996, and C4997 Part A: Intel,'al Logs. 6303-0L VoL 1. Rev. 1/PNNL-16381. Rev. L prepared by GEOVision Geophysical Setvices. Corona. Califomia. for Pacific Northwest National Laborat01y.

Richland. Washiugton.

8. Diehl. J. and R. Steller. 2007. Final Data Report: P- and S- Wave Velocity Logging Borings C4993, C4996, and C4997 Part B: Ol*erall Logs. 6303-01, Vol. 2. Rev. 1/PNNL-16476. Rev. 1, prepru*ed by GEOVision Geophysical Setvices, Corona, Califomia. for Pacific Northwest National Laboratory.

Richland. Washiugton.

9. Elecnic Power Reseru*cb InstiUlfe (EPRI). 1993. Guidelines for detennining design basis ground motions. EPRI TR-102293, Project 3302, 5 vol. , EPRI, Palo Alto, Califomia .

10. Rollins, K.M .. M.D. Evans, N.B. Diehl, and W.D. Daily ill. 1998. "Shear modulus and datnpiug relationships for gravels,' Journal of Geotechnical and Geoenrironmental Engineering 124, 396-405.

I L Meuq, F.-Y. 2003. Dvnamic properties of sandy and grave~v soils, Ph.D. Dissertation, University of Texas, Austill, Texas, May. 364 p .

12. Bamen, D.B.. B.J. Bjomstad, K.R. Fecht, D .C. Lanigatl, S.P. Reidel, and C.F Rust 2007. Geology of the Waste Treatment Plant Seismic Boreholes . PNNL-16407, Rev 1. Pacific N01thwest National Laborat01y, Richland, Washington.

13 . Rohay A. C. and T.M. Brouns. 2007 . Site-Specific Velocity and Density Model for the Waste Treatment Plant, Hanford, Washington. PNNL-16652 , Pacific Northwest National LaboratOty, Richlatld, Washillgton.

14. Youngs R.R. 2007. Updated Site Response Ana~vses for the Was-te Treatment Plant, DOE Hanford Site, Washington. GMX-9995 .002-00l Reviston 00/PNNL- 16653, prepared by Geomanix Consultants, Inc., Oakland, Califomia, for Pacific N01thwest National Laboratory, Richland, Washington.

Notes on Full Rip 9.0 by Sandi Doughton 2013

p. xii "It wasn't until the mid-1980's that a young scientist digging in marshes along the Washington coast uncovered the first solid evidence of upheaval in the past." Note taker's comment: this was after the CGS was designed and built. All the following information was discovered after the CGS was designed and built._How can the CGS be designed to withstand earthquakes that they didn't know were possible?

"Scientists now understand that the Northwest is even more seismologically complex than California, subject to three distinct types of earthquakes: deep, shallow and 1700-style giants. California may rock more often, but it can't rock as hard or in so many ways. The 1700 megaquake was sixty times as powerful as the quake that destroyed San Francisco."

p. 2 It was economics , not seismicity that toppled WPPSS. "But WPPSS left a scientific legacy too, one that's still playing out across the region. The prospect of nuclear proliferation inspired the first hard look at the Northwest's seismic nature. Armed with insights from a new field called plate tectonics, a handful of geologists started asking questions neither the nuclear industry nor much of the scientific establishment wanted to hear."

p.3 1983 - WPPSS has assured the NRC that its reactors were designed to ride out the worst possible earthquake, but when the Satsop plant (on Grays Harbor) was being constructed, NRC decided to get a second opinion. They hired Tom Heaton from USGS.

p.4 Reviewing a "decade's worth of seismic studies on the plant site and its environs, Heaton was struck by how little was really known about earthquake risks in the Northwest."

"WPPSS reviewed the historical records, which went back 150 years, and reached the logical conclusion: What's past is prologue. The middling quakes since settlers arrived in the mid-1880s were what the region could expect in the future. The consortium added a margin of safety and for the Satsop plant set its worst-case scenario at a magnitude 7.5 quake near Olympia.

p. 11 The reason subduction zones produce the most powerful quakes is because "the interface where rocks jerk past each other in a quake, called the rupture zone, is immense. A magnitude 9 subduction zone quake can rupture an area bigger than the state of Maine.

http://www .netstate.com/states/tables/st size. htm

p. 12.

"The difference between the type of quake the Satsop plant was designed to withstand and a coast wide megathrust (that it could have been subjected to) is like the difference between twenty-five atomic bombs and twenty-five thousand. Ground shaking can last ten times longer - up to five minutes. How much more would it cost to build a nuclear plant to stand up to something that big?"

p. 13 UW geology professor Eric Cheney went up against his boss and an army of consultants and challenged Puget Sound, Power and Light's plan to build two reactors near Sedro-Wooley.(in north west Washington).

p. 14-15 The 1984 report by Heaton and Kanamori pointed out the possibility that the Cascadia Subduction Zone might be capable of producing a megathrust earthquake.

Heaton, Thomas H. and Hiroo Kanamori."Seismicpotential associated with subduction in the Northwestern United States." Bulletin of the Seismological Society of America 74, (1984): 933-41.

and Steve Malone, seismologist at UW said "the report got everyone's attention."

The fuse was lit for an explosion in seismological research. Note taker's comment:

But CGS had been designed before this- all that has been learned since 1984 about the seismology of the Northwest is NOT incorporated into the design of CGS.

p. 29 1987 Brian Atwater found that over the past 7000 years, Washington state's coastline had dropped abruptly at least six times, by as much as six feet in places.

p. 31 "By 1995, a summary report listed eighty-six studies blanketing the coast from the tip of Vancouver island to Cape Mendocino - all pointing to a long history of quakes on the 700 mile long Cascadia subduction zone.

p. 38 "The average interval between them was about five hundred years. The shortest was a scant two hundred".

p. 49 Last megaquake was at 9:00p.m. on January 26, 1700. (313 years ago).

p. 56 Chris Goldfinger discovered that the Cascadia Fault has probably unleashed quakes even more powerful than magnitude 9.0. In ocean sediments he found hints that quakes may come in clusters. And he's unearthed evidence that some parts of the subduction zone snap much more frequently than Atwater found - every 250 years or so. If Goldfinger is right, the odds are higher than one in three that a great quake will hit within the next fifty years."

p. 65 By 2012 Goldfinger had evidence of nineteen quakes that ruptured the entire Cascadia margin in the past 10,000 years = every 250 years on average.

p. 66 A magnitude 8 quake anywhere on the coast will have far-reaching effects. Based on the standard view that Cascadia uncorks every 500 years on average, there's a 10-15 percent chance the region will get clobbered in the next 5 decades. Goldfinger's interpretation raises the odds to 37 percent.

p. 71 In the early 1990's it was found out that Washington state is vulnerable to a third type of quake "which could be the most destructive of all":

p. 71-84 The Seattle Fault

-discovered by Zdenko Danes in 60's

- not until 1980's Robert Bucknam and Brian Sherrod of USGS found first physical evidence that it was real and active.

- M7.0

-right in the middle of Seattle

-thrust variety fault - a shallow fault

-slices from the Hood Canal through south Seattle

p. 84 Craig Weaver, chief of USGS earthquake contingent in Seattle

-the fault isn't a single crack but a five mile swath of as many as 8 separate fault strands extending east and west between Seattle and Vashon Island.

p. 88 Lidar mapping "is revealing a network of faults running through the sagebrush flats near the Hanford Nuclear Reservation.

Ian Madin of the Oregon Department of Geology and Mineral Industries says " we are finally starting to see the big picture".

p. 93 The Bainbridge trenches marked the beginning of an era of breakneck discovery that is still going strong more than a decade later.

-Fault that slices through Tacoma

-Fault near Olympia - The Legislature's Fault

-Saddle Mountain Fault skirts eastern edge of the Olympics

-Devil's Mountain Fault cuts path from tip of Vancouver Island to the foothills of the Cascades

-Two faults near Bellingham

-"A modern fault map shows that it's hard to find a place where an earthquake-phobe could feel cozy. The few blank spots are mostly where geologists haven't looked yet.

p. 95 NRC considers a fault active if it has ruptured with the last 10,000 years.

p. 99- 101 The South Whidbey Island Fault (SWIF)

-existence proven in the mid 1990's by USGS

-most dangerous surface fault in region - did not fault 1100 year ago .

-goes from Victoria B.C. to the Cascade foothills where it links with several other faults including the Seattle fault (The Seattle Fault is just a branch of the SWIF) -

Brian Sherrod

- it carries on across the Cascade Mountains to the town of Richland 0 .100

- 200 miles long

-it is a band of fractures up to 50 miles wide (i .e. it's not a single break in the crust)

-What we're dealing with is a system of faults that we think are linked. But if you have a fault system that's three hundred kilometers long and you rupture half or a third of it, that's a big earthquake. That's a 7.5." - Brian Sherrod

- in mid 2000's Sherrod followed SWIF's trajectory and studied east-west folds (Horse Heaven Hills, Rattlesnake Ridge, Saddle Mountain)

- "Brian Sherrod found signs " of at least seven quakes of roughly magnitude 7 "

-The faults under Central Washington's ridges aren't shallow- they originate more than 12 miles below ground and cut through massive layers of basalt. "In other words, the faults that formed the ridges are much more dangerous than anyone realized . "It's a fundamental rethinking of the seismic risk over there," Sherrod said."

-In 2012 The Department of Energy ordered new studies of earthquake risk at Hanford.

-after Fukushima the NRC ordered several safety upgrades to CGS "but decided there was no need to bolster its seismic safety".

-"The 1970s-era reactor wasn't designed for a specific earthquake but rather for a specific level of ground shaking. "Based on what they knew at the time, engineers designed the reactor to stand up to .25g."

-"So it was disconcerting in 2009 when a swarm of more than a thousand quakes shook the eastern edge of the Hanford site. None of the quakes was bigger than magnitude 3 but because they occurred so close to the surface, the peak motion force was .15 g which isn't far below the nuclear plant's design level. Blakely and Sherrod traced the swarm back to one of the ridges they've been studying - and the fault that lies beneath it."

p.102 Ray Wells' masterpiece is a laminated map of the Pacific Northwest with moveable sections. "The map represents the culmination of more than two decades of research by dozens of earth scientists - and the key to calculating an earthquake budget for the region".

"It's a train wreck on a geological scale" The main driver behind the train wreck is the giant Pacific Plate which moves northward at 2 inches a year pulling California in its wake.

California rams into Oregon which is also being shoved from the side by the Juan de Fuca Plate, which is subducting under North America Washington is caught between Oregon pushing from the south and the unyielding bedrock of inland B.C. to the north. The Evergreen State "crumples like a line of box cars slamming into a mountain - "that's why you have the Seattle Fault, you have the Tacoma Fault and you have the Whidbey Island Fault. They are all driven by this north-south compression . Ditto for the rumpled ridges and faults in Central and Eastern Washington.

"The Puget lowlands are being compressed by about a quarter of an inch a year. That adds up to more than 20 feet of crunch since the last time the Seattle Fault fired off. Central and Eastern Washington are being squeezed at a slightly lower rate. Inexorably, the pressure is accumulating, loading the Seattle Fault and its associates like springs."

"The squeeze on the Puget Sound region is enough to produce a magnitude 7 quake every 500 years"

p.110 "In order to design a nuclear power plant, utilities must identify the "maximum credible earthquake" the fac ility could face.

-But for three pages Sandi Daughton outlines the "tennis game" that went on in the attempt to place the location of the 1872 quake. . Eric Cheney University of Washington geologist said "It would have been comical if it wasn't so serious."

-Finally the NRC set up a panel to settle the debate. Howard Coombs was the man in charge of the panel "He was also a paid consultant to most of the Northwest nuclear power projects." And the panel chose a location close to the Canadian border, east of the Cascades - pleasing both the Skagit proponents and WPPSS.

Cheney said that Coombs "found a place to park it where it wouldn't be a problem and everyone was happy."

The NRC approved the analysis for the Columbia Generating Station . The study concluded that the biggest historic quake in Hanford's vicinity was not 1872, but a magnitude 5.8 that struck near the Oregon border in 1936.

It wasn't until 2002 that Bill Bakun and his colleagues assembled a picture that was unambiguous in concluding that the 1872 quake struck on a shallow fault near the southern end of Lake Chelan, just north of Entiat. He pegged the magnitude at 6.8, though with enough uncertainty that it could have fallen anywhere between 6.5 and 7.

It's all riddled with faults," Bakun said. "It wouldn't surprise me to have a magnitude 6.8 quake anywhere in that region, including near Hanford."

p. 115 In 1979 University of Washington seismologists put the 1872 atM. 7.4 Malone, Stephen D. and Sheng-Sheang Bor. "Attenuation patterns in the Pacific Northwest based on intensity data and the location of the 1872 North Cascades earthquake" Bulletin of the Seismological Society of America 69 (1979):531-46

p. 118

-The Richter scale is logarithmic instead of linear

-to get a truer mark of the destructive force, you need to multiply by 31.6 for each step up the scale

-The Richter scale is popular with the press but meaningless to a seismologist

-see p. 120 for Moment Magnitude Scale

p. 128 Wadati-Benioff zones = bands of deep seismicity "many positioned unde r volcanic arcs like the Cascades"

p. 132 The Northwest has been rattled by 18 quakes known or suspected to have deep roots since the beginning of the 20th century,

p. 134 Craig Weaver USGS Seattle says a deep quake as big as an 8 could happen in the Northwest.

p. 174 The "maximum credible earthquake" approach is still used for critical facilities like dams and nuclear power plants.

The 2500 year map for the Northwest includes a magnitude 9 Cascadia megaquake and a magnitude 6-plus shallow fault quake. But the USGS considers a massive Seattle Fault quake like the one that struck the region in 900 AD to be a 5000 year quake - such a long shot that it gets scant consideration . But therein lies the Achilles heel of probabilistic mapping: It discounts the rarest quakes, which are also the most deadly."

In 2010 research showed that the 500 year map underestimated the intensity half the time, often by more than a factor of two."

p.l75 Art Frankel :"Don't' think you've seen everything that nature can throw at us."

p. 185 John Hooper - Director of Earthquake engineering at Magnusson Klemecic Associates

- one of the country's premier structural engineering firms says "The term "earthquake-proof" is not in our lexicon - a well-designed building that meets all requirement still stands as much as a 10% chance of a collapse if it's hit by the maximum earthquake the code considers, roughly a 2000 year quake in the Northwest.

p. 198 In California the Alquist-Priolo Act restricts construction near known fault scarps.

p. 205 Perhaps the most powerful predictor of earthquake damage is whether a structure sits on solid ground or loose dirt.

p.224 On average, the Northwest moves about Y2 inch a year. The motion never ceases.

The pressure never stops building on the subduction zone, the Seattle Fault, the Tacoma Fault, the South Whidbey Island Fault. Since the 1700 megaquake, the coast has moved more than 25 feet.

p. 238 -239

-A Cascadia megaquake could disrupt supplies for weeks or months

-high-voltage transmission towers can be affected

-natural gas lines can be affected

-the states gas and diesel fuel supplies could be cut off

-electrical service in Portland could be knocked out for1-3 months

-Washington's plan estimates 1-3 months to restore internet and telephone and up to three years to rebuild damaged transmission lines.

Until roads and bridges are repaired, it will be difficult to fix downed power lines and damaged electrical stations

-without electricity it won't be possible to restore telephone and internet.

p. 240 Witt- left FEMA in 2001 and runs a consulting firm to help businesses and governments plan for disaster:

" In a major subduction zone quake, not only will the direct damages to structures and infrastructure be enormous - the long term economic impact could alter the whole economy -

note takers comment: and a meltdown could end the economy of the Columbia River basin.