RISK ASSESSMENT OF DISPOSAL OF HIGH-LEVEL




     RADIOACTIVE WASTES IN GEOLOGIC




              REPOSITORIES
               August 1985
  U.S. ENVIRONMENTAL PROTECTION AGENCY




      Office of Radiation Program*

-------
                         Chapter 1 -  INTRODUCTION








1 • 1  Backeround



     The  national  program for  the disposal  of  high-level  radioactive.



waste  Is  governed by  the Nuclear Waste  Policy Act  of 1982.   Three



federal   agencies   have   central   roles   In   this   program.    The




Environmental  Protection  Agency  is  responsible  for  setting  general




environmental  standards that  govern the level  of performance  to  he



expected  of  a  waste  disposal  system.    The   Nuclear   Regulatory



Commission  is  responsible for  implementing  the  standards  promulgated



by  EPA and  issuing licenses for  the construction and operation  of a



waste,  disposal  system.  The Department of Energy  is  charged with the.



responsibility  for  developing  and  implementing  a   waste  disposal



system.   These federal agencies,  as well  as other organizations and



individuals,  have been involved  in developing or  evaluating  waste



disposal  technology  for many years.   The studies  carried  out  by each



agency have  been  appropriate  to the  particular  responsibility  of that



«gency  in the  entire  waste  disposal  program.   The  purpose   of  thic



report  Is  to   summarize  the  risk  analysis  studies  that have  been



carried  out by the Environmental Protection Agency  to  provide  the



information  necessary  to   set  generally   applicable   environmental



standards.








     The  analysis  of  the  long-term  risks  f-om  the  disposal  of



high-level radioactive  waste  is an important component  in determining




the  suitability   of   any  proposed  waste   disposal   system.    Other



important  components  include  the  cost and   logistics   of   a  waste

-------
 disposal system,  the environmental  impact  of such  systems,  •»..>!  t .!,«-.

 safety  of  euch   facilities   and  system*   during  construction  and

 operation.   Th« estimation  of the risks associated with  the  disposal

 of high-level , waste  it  a particularly difficult problem,  and  one  for

 which there  exists  no   good  precedent.   The waste  disposal  system

 itself involves  the  complex  interaction  of  engineered  systems  (e.g.

 metallic  containers,  vitrified  waste  forms,   borehole  seals)  and

 natural  systems  (e.g.   geology  and hydrology).   The  time  frame  over

 which such systems must be evaluated extends far into the future.  For

 purpor «= of  the  principal  EPA  analyses,  for example, estimates  have

 been made of the  performance  of waste disposal systems   for at  least

 ten  thousand  years  into  the  future.   Predictions  of  such  complex

 systems   for  such  long  periods  of  time  necessarily  include  large
/
 uncertainties.  Therefore, an additional component of the  risk analysis

 involves estimating  or  bounding  such uncertainties;  this makes  the

 problem  even more difficult.  The approach described in this report is

 one that uses  relatively simple  mathematical models  for  the  various

 processes  and  systems   involved,  combined  with  bounding estimates

 (i.e.,  intended  to ,  overestimate  risks)  on  parameters and processes

 whenever  there  is  considerable  uncertainty associated  with  their
                < *f*»l9'»» *•* ^/^^ 4
-------
  give an overview o. c:,.» previous  literature at  the  outset.   Uh.n the




  "A published it.  draft  F.nvironm.ntal  Standard In 1982,  it  al.o




  available  a number of report, on studi.. carried  out  up to that



  These  reports included:
                                                      Rafnoaer.
-------
     models  developed by  the  EnVlrorjr.ental  Protection  Agency.

     These   models   include   the   enviroiuiitmtwl   transport  of

     radionuclides, their health effect* on  the population,  and a.

     comparative  risk analysis  between the disposal  high  level

     radioactive  waste  and  existing uranium  ore  bodies.   This

     report  presents  a reasonably  complete summary  of the risk

     analyses carried out through 1982 in support of the Agency's


     efforts  to  §et  environmental  standards  for  high   level

     radioactive waste disposal.




Environmental  Pathways  Models  for  Estimating  Population  Health

Effects from Disposal of Hieh-Level Radioactive Wasts in Geologiq


                  520/5-80-002
     This report  documents the models  developed and  applied by

     EPA  to  estimate  the  transport  of  radioactive  materials

     through  the  biosphere  (e.g.  through  food  chain or  other

     pathways  readily accessible  r   the human  population) .   In

     addition  it  summarizes other work  done  by EPA on the health


     effects  associated  with  the  exposure  of populations  to
                 x
     radioactive  materials.   It was one of  the  principal source


     documents,  in  conjunction  with  the  ADL reports,   for  the


     Integrated analysis presented in the Population Risk Report.
Population Riaks From Uranium Ore Bod^a, EPA 520/3-80-009.




     This  report was prepared  by EPA  in order to  estimate the

     risks  to a human  population from various  kinds  of uranium

     or* bodies, such as  are  found  throughout tha  United States.

                 has provided  comparative  rick information that

-------
       nas  b«.en  used  by  the  Agency in r^ ,
                       J   _   S ncy ln crying to decide how  low  the


       ri.k goal from a repertory should be sec.
              to


     out  .ub..qu.nt


              «tudle3 „.
      Th«  original risk analysis  described  above was carried out


      on che  baiis of  so-called "conceptual  models" of geological


      repositories  for nuclear waste  in varic-s  geologic media.


      These   conceptual  models  were   developed   early   in  the


      national program, when  little  site-specific information was


      available on actual  sites under evaluation.   In an effort to


      refine  the   conceptual   models  so   that  they  would  more


      realistically  describe   the   capabilities   of   geologic


      disposal,  a  review was conducted  of the  site  data collected


      by  DOE  at various  potential sites.  On the  basis  of  this


      site   specific  data,   modifications   were   made   iq  the


      conceptual models.	=	re
Individual DO a.
     A  -tudy  va.   carried  out   by  Sandi.   Laboratories   under


     contract to EPA to  use Sandia's computer  code NWTF/DVK  to


     Mtimate potential  individual  do...  at various  points  in

-------
           time,  as  a  result^ of  a  nuclear  waste  repository.  (The



           original   analyses  exclusively   „,*
-------
treat, among other topiCB, the validity of the technical basis for the


EPA  standard.    The  Vacte  Isolation  System  Panel  of  the  National


Academy of Sciences  conducted  an overall  review of the entire nuclear


waste program  and devoted particular  attention to the  approach  that


had bean adopted by  the EPA  for regulating the disposal of high level


waste.  Sandia National Laboratories, under contract to  the Nuclear

                                                                    to
Regulatory  Commission,  conducted  a  number  of  technical  analysis


intended  to   test  whether   the  EPA  standard   could   actually  be


implemented by  the  regulating  organization, i.e., the  NRC.   The EPA


itself convened  two  subcommittee,  of  its  Science  Adviaory Board.   The


first was. directed specifically at the high level waste standard and


its associated technical support material.  The  second  was addressed


towards EPA1s  regulation of airborne .radioactive  emissions under the


Clean Air Act, but it  too addressed fundamental questions relevant to


the high level waste standard.  A summary of the conclusions of  each


of these  studies with  respect  to the risk  analysis  approach  of EPA


is sunmaarized below:




     Waate Isolation System Panel (WISP)  Report




          This report  argued  that the measure of  risk used by EPA in


          its  risk analysis,  namely,  cumulative population dose  over


          ten  thousand  years,  was  inappropriate, for  two  reasons.


          First,  the  time frame  was  considered  to  be  arbitrary  and


          •hould have  been extended indefinitely  into  the future,  for


          as long  as a residual  risk  front the repository would exist.


          Second,  the  measure  used  by  EPA  did not  take  adequate


          account  of  the  risk  to  individuals,  some of  whom  might

-------
              relatively large  doses  ev«u  ti^ugh  Liie cumulative




     population  do so  might fall within acceptable  limits.  «••••*
Sandia/NRC
     An   issue   that  had  been   heavily  debated   during  the



     development  of  the  draft  Standard wa«  the  question  of



     whether it  could  actually be implemented by  the regulating



     agency,  i.e.,   the  NRC.    The  probabilistic  nature of  the



     standard and the  uncertainties  associated with  the complex



     system  being  analyzed  raised  a  question  as  to  whether



     adequate levels of  certainty could  be associated with  a



     licensee's   application  in order to  verify compliance  with



     EPA  standard.   Sandia  National  Laboratory  developed  and



     tested computer codes for generating  the  measures  of  risk



     required by the  draft  Standard,   as well  as  methods  for



     quantifying the  associated  uncertainty.   On  the  basis  of



     this  work,  Sandia  concluded that technology existed or  could



     be  further  developed  such  that   there  would  be   adequate



     technical  basis for  testing compliance with  the  standard.



     After this   work was Completed,  EPA  contracted with Sandia



     for  additional  model  development   and  testing,   as  well  as



     application  to  the  generic lithologies treated  in the  EPA



     risk  analysis.

-------
bciencc Advisory  Board Subcommttte  on ^e Htph Level Radioactive

    .g Sfandarj




     A  special   subcommittee   of  the  Science  Advisory  Board

     reviewed in detail the technical basis for the EPA Standard.
                                 *
     including  both the  risk  analysis  and consequence  models.

     There were four basic conclusions derived  from  this  review.

     First,  it  was  noted that  there were  large  uncertainties  in

     Che parameters and models used in the EPA  work.   Second,  the

     relatively restrictive level of risk imposed by  the standard

     night require  excessive  conservatism on  the part  of  the

     repository  designers  in  order  to  demonstrate,  with   an

     adequate  degree  of assurance,  that the  standard  would  be

     met.  This is essentially a corollary of the previous point.

     Third,  certain parameters  used  in the geocheraical analysis

     should  be  modified  in   order  to  reflect  the  most  recent

     information  on  the   behavior   of  appropriate,   species   in

     groundwater systems.   Fourth, certain environmental pathways

     for radlonuclide migration  and  human exposure require some

     modification in  the  models  used by  EPA  in  order to more

     accurately reflect current knowledge.




     Among these comments, the  first  and the  third are relevant

     to  the risk analysis and hence are addressed further  in this

     report.   In particular,   additional  discussion and measures

     of  uncertainty are  developed here  so as  to lend  further

     perspective  to  the  degree  of  uncertainty  that  might  be

-------
          associated   with   any   risk  estimates   generated.    The
                             *


          geochemical  values  have  also  bean  modified  taking  lnf<>



          account  the  SAB  comments,  as well  as  information  from  the



          WISP  report  and  other  sources.   The appropriate  degree of



          conservatism  in   the  standard  is  more  a  policy  than  a



          technical issue,  and it is discussed elsewhere in connection



          with  the  final  version  of  che   standard   itself.    The



          modifications Co  the models  for environmental  pathways  and



          human exposure are addressed in the companion report to this



          report.







     Science  Advisory  Board  Subcommittee   on  Risk  Assessment  for



     Radionuclides
                                 •N






     The  emphasis  of  this  review was primarily  on the  consequence



models related to human exposure.  This subject is not treated in this

                                      c*H/
-------
analytical  methods  used to  carry  it out

-------
                    CHAPTER 2  -  MODELING  APPROACH
2.1  Generic Analysis








     Th« risk  analyses  carried out in  support  of the development  of



the High Level Waste Standard  are intended  to be  "generic"  in nature.



1C  is  important  to  understand  what  this means.   In  setting  its



standards,  the Agency needs  to ensure that the performance  levels  It



requires «re realistic, that is,  that there is  a  high likelihood that



they  can  be achieved by  available  or developing  technology.   Beyond



this,  the  Agency also  wishes  to  use  its regulatory  authority  to



encourage  the  selection of waste disposal  systems that are  close  to



the   beet   that   might  reasonably  be  expected   Co  be   available.



Therefore,  an important  aspect  of the  work carried out during  the



development  of  the  standard  has  been  risk  analyses  intended  to



estimate  how  well  waste disposal   systems  might   perform  and  how



sensitive  this performance is  to various parameters.








      In  developing  this  information,  the  Agency  considered a  wide



range  of  geologic  environments  and  other  parameters.   In  the  early



stages  of this work, the  Department  of Energy, which  is  responsible



for  developing a geologic  repository, had not yet collected much data



for   its   principal   candidate   sites.    Therefore,   Che   Agency's



consideration of potential conditions at a  repository was based on the



general  literature on potential  waste disposal environments,  as well



mm the  limited data that had been obtained  by DOE in the early stages

-------
of  its  investigations.   Morfe  recently  DOE hae  developed much  more

complete d*t»  on  specific  sites that it  is ovfllviaf.tng for suitability

as  a  repository.   Vhlle   the Agency  utill  intends  that  its  risk

analysis b«  considered  as  conceptual or  generic,  in defining  the most

representative^ parameter* to  use in such analyses,  it  has  reviewed

carefully the  developments  by  DOE  at particular sites and it has based

its  conceptual models largely on  these  data.   Because  of the generic

nature  of  the analysis, the  results of the risk  calculations  do not

purport to  represent the actual risk expected at these particular DOE

sites.   They  are  still  intended  as  rough  estimates of  potential

repository  performance  in a  general  sense,  in the  full realization

that further  investigations  at particular sites  may  show that such

sites   have  a   performance   somewhat  different   from  the   generic

calculations.   At the same tine, by using the most recent information

generated  by  DOE  and   others  at  particular  sites,  the Agency has

attempted  to  ensure that   its  generic calculations are based  on the

most reasonable   parameters   and  hypothetical  scenarios that  would

control performance at  a real  site.



2.2  Waste  Disposal System Model
                            N


     The waste disposal system considered in the Agency's analyses is

based  on the  current reference plan  of  the  Department of Energy for

the  disposal  of  high-level   radioactive waste   in  mined  geologic

repositories.    Such repositories  consist  of  underground  mines  or

-------
 excavations  whose working  levels sr« between 1,000 and 3,500 ft. below



 the  surface.  Rock  formations  in which  «uch  repositories  are being



 considered include:   bedded aalt,  *alt domes,  basalt, granitic rocks,



 and   tuff.   Each   of   these  environments   has   its   own  special



 characteristic*, which are discussed in later chapters of this report.



 The  radioactive  waste*  themselves  would consist either  of  spent fuel



 from  nuclear power  reactors or  solidified reprocessing waste  in a



 relatively durable  form,  such  as  borosilicate glass.   Such  wastes



 contain  a  wide  rang*  of  radioactive  element* ranging from  highly



 active  fiaaion   products  with relatively  short lives  to  long-lived



 elements,  such  a* the  transuranic*.   The  wastes themselves would  be



 packaged  in one  or  more  containers  or canisters and these would  be



 placed  in holes in  the walls or  the  floor*  of mined vaults  in the



 repository.  After emplacement of the wastes,  the repository would  be




 backfilled to enhance  its  mechanical stability  and  Co retard  the



 movement   of fluid*.   Its  various  connection*   (e.g.   shaft*  and



 boreholes)  to the surface would also  be  severed  and  sealed.   The



 intent in  selecting  a location for  a rep  *itory would be to obtain a



 highly  stable geologic  environment and  one  in which  it  would  be.



 difficult  for groundwater  to come  in contact  with  the waste.   In



 addition,  in order  to  avoid  the preemption  of important  resources



nearby or  the attractiveness of the site  for  future excavation,  DOE




would attempt to locate the site where  other  resources  are minimal,



 and it would mark the site so  that  future generations would be aware




of the hazardous materials that had  been disposed of there.

-------
     Despite  the  care  chat DOE may exercise during the development of
                             <*


a repository,  it  olwaye remains possible that there could be A futvir«



disruption that could lead  to  the  release  of wastes.   The purpose of



the  risk analyses carried out  by  the  Agency la  to  identify the most



important  mechanism* that  could lead to  such  releases and to estimate,



on a quantitative basis, both  their likelihood and their consequences.







While  this analysis varies  somewhat from one geologic environment to



another,  the  overall common structure can  be represented  as shown in



Figure 2.1.   The  components  of the system to  the  right of the vertical



dotted  line  are  referred to  as  the  "accessible  environment".   The



Agency   has   focused  particular attention   on   tha  actual  physical



quantity of radionuclidea  that might enter  the accessible environment



over various  relatively long time periods in  the  future.







     In  order for wastes  to  reach  the  accessible  environment,  they



must be  transported through the various  components  shown  on the left



side  of the  diagram.   In  particular,  radioactive  material must  be



released from the waste form itself, which, as noted previously, might



be -either a  borosilicate  glass,  unraprocesced   spent  fuel,  or  some

               s
similar  type  material.   Having left  the  waste  form itself,  such



radionuclidea must pact through  the canister, which would of course be



designed to try to retard such movement, and  then enter the backfilled



openings of the repository.  From the  repository  there are two general



ways  in which they nay find  their  way to  accessible  environment.



There may  be  direct  pathways to the  land surface, such as might occur

-------
 if future generations penetrated the repository during an  exploratory



 drilling program and accidentally contacted the wastes, or  they  might



 migrate in slowly moving ground water to an aquafer within  which they



 could  than  b« transported some  distance  where they  might thon  have



 access to the surface.    (For  example,  many  underground   aquifers



 ultimately  discharge  into  surface  water   bodies.)   The   physical



 description of  the movement of  radionuclides  from  the  waste  form,



 through the  canister,  through  the  repository,  and  ultimately to  the




 accessible environment  depends  on the physical description  of  a number



 of possible future  scenarios  that might alter  the  conditions of  the



 underground environment.    Such processes  are called  "scenarios" or



 "release mechanisms".  They  may  affect any  of the  four components



 indicated in Figure 2.1.    For  example,  the nature of  the  waste  form



 might  be altered by crushing or  by intersection by  a drilling  tool.




 Similarly,  a canister might also be  affected by a disruptive event, or



 by  degradation as   the result  of  corrosion.   A scenario  might affect



 the  repository component  by  providing  a. flow  of  groundwater  through




 certain parts  of  the repository.   It might  affect  the availability of



 pathways  to  the surface or  to the aquifer in a number of ways,  such as



 by  the rupture of a f*ult  or by  the creation  of a  new borehole.   The



 risk  analysis  reported  here  considers a number  of  «uch  release^



nechanisms,  their probabilities,  and their  consequences.   The  results



 of the  calculations  for individual release mechanism are then combined



 into  an  integrated  representation  of  the riak from  a  hypothetical



 repository.  The  actual measures  used to characterize  auch  ri«ks are



discussed in subsequent sections.

-------
2 . 3  Time
   ,  There  is  wide variation  in  the  published literature  on  nuclear

waste  disposal  on the  subject of the time  frame  over  which disposal

alternatives should be  evaluated  or compared.   If  it  were possible to

predict  the  future with reasonable assurance,  then clearly the longer

the  time frame,  the  more complete could  ba the  comparisons  between

alternative  systems.   Some authors have attempted to compare  nuclear

waste  disposal  sites  or systems for periods  of .up to one  or even ten

million   years.    Such   results,   however,   contain   very   large

uncertainties because of  the difficulty of making such long time frame

predictions  about either the  natural or the  engineered components of

the  system.    At  the  opposite   extreme,   other  authors  have  only

considered  site  risk comparisons  for  periods  on the  order   of  300

years, arguing  that relatively reliable predictions  can  only  be  made

for  time frames of this length and that the longer term performance is

both  less  important   (due  to  the decay  of  the  radionuclides)  and

expected to be  generally proportional  to the  performance calculated

for  this period.                                   ,


x
     In  the  course  of  performing  numerous  -risk   analyses   over  a

relatively  large range of  time spans,  the Agency has  concluded  that

the  risks  identified over relatively short  time spans,  such as a few

hundred  to one  thousand  years,  do not adequately portray important

differences  among alternative  sites  or waste  disposal  systems.  This

is because  the groundwater travel times may be  sufficiently long for

-------
of
                  slt,.  that
           .  if th. .„!,...
         be  significant differences  b.t-een che  §lt..
               or  g.och.nlc.l   ch.ract.ri.clc,   of  th...  groun<.».t,r
  «y«.«,.   Th.  priory risk  ......»'.«.- ..rxi.d Me ln .upport of th.
             tlvt,l...  tUr.fore. h.v, be.n ba,.d on . el*. £,„ of «n
           y»tt.   It  t.  bellevecl th>t  tMs
             diff.r.nc.8 b«v«n on. ile. .na
          of «ne .u^,.., systen annocher
  r.strictlon to . t.n  tho.s.nd y.« clm.
  •tlU  do« noc  captur« th.
  This  U especially  the „,. when  the  dominant r.dlonuclldes «  ,oM'
  Point   in   tin,.  .„  nembera  of  th.  rain.   ^  whi^
  different  element,   In th. ch.in h.v.  different  r.t.rd«lon r.te, In
  the grounder  .y.t».   Depending on th. ri.k  »...ur. used,  It ,.y b.
  t~.port.nt to  consider .uch  process,,  m cor^.rlng .tt.._  „,.„,„._
 in . nu.b.r Of  the Agency's c.lcul.tlons.  tl.  fr.ne. Lnger th,n t.n
 thous.nd years have «lso  ,e.n considered.
      in   consldertng  the  long  t.r.  .f£.ct.  ot  .   nuel..r
"po.ttory,  the  Agency h..  con.4d.red both total cumul.tlv. popuUtlon
effects,  a,  wll „ eff.ot.  „  tnl3   ^M  ^ ^ ^^
provide 8,Mwhat different  Information.    It  1. p(1..tbi., for .XMlpl.,
">«t   a   relatively   unproductlv.   groundw.t.r   .upply  COuld   b.

-------
contaminated  at  9on,e  point  in'jrhe  future  by  nuclear waste,  but due to



the  limited  availability  of  water  from  this  supply,  only  a  few




individuals  would  actually receive  exposure   to  radioactivity as  a




result.   In  this case the  individual doses might  be relatively high,




although  the  overall  effect on the population could still be quite low




because of the small  number of people involved.  On tha other hand, it




is  also  possible that a nuclear waste repository  could at some point




in  the future lead to very low level contamination of a water supply




that  serves  a relatively large population.  In this case the dose to




any individual in that population night be small, although the overall




effect  on the population could be  substantial because  of  the large




numbers   of   people  involved.    Because   of  these  differences,   in




assessing the risk from a repository,  the  Agency  has  developed and




applied  techniques that can estimate both kinds  of risks.   Since the




ultimate  effect  of exposure of populations  or individuals within the




population to ionizing radiation  is negative health effects and since




th. hypothesis used  by  the Agency  and  others  to estimate such health




effects  is the linear mm-threshold hypothesis, the doses received by




individuals   or  by populations  can  ultimately be  reduced  to common




terms,  namely,  the total expected number of somatic or genetic health




effects.   This is  therefore the most fundamental risk measure of  all.




While of course  it  is desirable to  keep individual  doses  as low as




possible,   the   total  cumulative  release  of   radionuclides  over ten




thousand years and the  associated  population  done and health  effects




1« the most  complete measure of risk for evaluating alternative  sites




or  repository  systems.    Therefore,   the   greater  emphasis   in  the




Agency's  risk  analyses  has   been  on   total  integrated  cumulative




releases over ten  thousand years.

-------
                                                                                 /
        Even within this framework,  there are additional  alternative risk       ]/


   raensures.  The  final risk  me.Asur*  of immediate importance  is  health


   effects,   Such health effects  are  calculated by fint estimating  the


   radiological  dose to the population.   The  dose  in  turn is  estimated by


   calculating  the released quantities .of  specific radionuclides to  the


   accessible  environment.   According to the conceptual        model  for       •


.   the  risk  analyses as illustrated in  Figure  2.1,  one  would begin with


   releases  from  the waste  form and go through a sequence of models that


  would  first   calculate   the  physical  quantities  of  radionuclides


  released  to  the accessible  environment and  then,  building upon these


  numbers,  would calculate doses to the population and ultimately health


  effacts.   These last two stages,  namely,  the calculation of population


  doses and health effects, bring the calculations to their  final  point


  of  interest,   but   at   the  same  time  they   introduce   additional


  uncertainties   into  the  calculations   because  of  the  need  for the


  additional models.  Therefore,  for  some calculations  the  Agency has


  found   it  desirable   to   express  risk in   terms  of   quantities  of


 radionuclides   released   to   the  accessible  environment.    For  other


 calculations  these numbers  have  been  converted  into  dose  or health


 effects estimates^/Each  such risk measure  serves a useful  purpose as


 will be illustrated when  the  result of  the risk analyses aro discussed«

-------
               Che  .election  or a  particular  risk mea6ure  for fl  fiec  Qf
   calculations,  the'-ode  of  presentation  of eh.  ri.k  results  .iao  allow*
   —  flexibility,  .inc. the  various  relea.. MCh.ni.M  have associated
   with  them  .on,,  probabilities of  occurrence (genially qmcc small) and
   since  the  point  in  time  at which, they can  occur ha.  an  important
   influence  on their effects on th. population  (b.cause  of  the gradual
   decay  of   the  radioactivity).    The method   of  "risk  profiles"  or
   "complementary cumulative* distribution functions" (CCDF) has been used
   in  the  risk calculations.  A   hypothetical  CCDF  is   illustrated  in
  Figure  2.2.  In  general  a  point  on  such  a curve   consists  of
  conscience and a probability value.   Th. relationship between them is
  given  by the statement:  "The probability shown t. che  probability of
  obtaining a consequence at least  as  large a.  the consequence  shown.-'
  To  simplify some of the calculations, two specific probability values
  have been chosen and their corresponding consequences are calculated.
  This  is actually  the  form in which  the numerical  part  of the  high
  level  waste standard  is  presented
 2.5  Computer Codes Utilized

      A number  of  computer codec  have  been  used  a*  tools  in  the
Agency's  risk analyses.  The central  tool has been the program  REPRISK
which ha* been  under development at  the Agency  since  1978  and  for
vhich separate documentation exits.  This code will be described  late?
in this section.  REPRISK makes use of certain conversion factors that
relate the  amount of radioactive material released  to  the accessible

-------
               to  population ReaUh effects.   These  ccnver3ion  factors
  -r«  obtained by u.ing another  EPA computer  code called WESFDOSE.  This
  cod,  considers  a number  of pathways  for  th. environmental  transport of
  radionuclides   and  is  described   in   a   companion   document.   For
  calculations  involving  individual. doses and longer  time  frames than
  ten  thousand years,  the computer code  NWFT/DVM,  developed  by Sandia
  under contract  Co  NRC, has been used.   Thi. code  models  the transport
  of decay chains  whose element, have  different retardation  factors  in
  the groundwater  system.  A more complex groundwater code,  SWIFT, has
  also   been   us.d  to   support   the  EPA  risk  ..alyses,   primarily   to
  validate  so*. of the  hydrologlc c.lculatlon, carrie
-------
       After  the repository  inclosed  and Sealed,  there may  be  various
  natural  proc.s.«.  or human activity chat  could  lead to  rel.«,ses troir.
  the   repository.    Th.se   perturbations  happen  ~according  to   some
  frequency  or   probability,   eith.r   constant   or  variable.    REPRISK
  incorporate,  the  characteristics of these  perturbations  and  their
  associated  probability,  to model radioactive releases.  Four kinds  of
  "release  mechanisms'1  are addressed:

       1.   Direct impact on a waste package with associated releases  to
           th. air  and/or  the land surface  (e.g.,  volcano,  meteorite,
           drilling/dir«ct hit).

      2.   Direct impact on a wast* package.with associated releases Co
           an aquifar (e.g., faulting,  breccia pipes).

      3.   Disruption of the repository with  associated releases  to the
           land  surface (e.g.,  drilling/no hit).

      4.    Disruption of the repository with associated releases Co  an
           aquifer  (e.g., normal  groundwater flow,  faulting, breccia
           pipes, drilling/no hit).

     Afl a result of  each  release mechanism acting  at the site,  some
portioi. of the radionuclide  inventory  will begin to travel to various
components of  the  accessible  environment,  e.g.,  rivers.   Each  such
release  mode  U.ds  to several   pathways  to  human  exposure.   The

-------
consequences  of  a  radioactivity release  to  the  accessible  environment
                              4


are expressed In terms of  1) number of somatic health effects (fatal



cancers),   2)  number of genetic health effect*,   3)  ratio  of released



amount  to  the release  limit  in 40  CFR 191,  and/or  4)  curias released



of each radionuclide.








     Two  time  frames,  are  used  by the  model.   One,   called a  dose




commitment   period,   is  for  modeling  the  occurrence   of  release



mechanisms  «t  the  site.   The  other, a dose integration  period, is used



for measuring  the  consequences  of the releases.   This way consequences



may be measured  beyond the  time when a particular perturbation may be



active at a site.








     2.5.2     NWFT/EVM                      (








     NWFT/DVM is  computer  code that w«s developed  for  somewhat noro



general  purposes  than  those  for which  it  was  applied  in connection



with  the Agency's risk  analyses.   In the  paragraphs  that  follow a



general  description  of  the  capabilities  of  NWFT/DVM  is  first



presented,  followed  by a description of the way It  was applied by the

                    N

Agency.








     NWFT/DVM is essentially a pair  of two computer codes,  NWFT and



DVM, that  are compatible in the sense that  the output  from the first



can  be  fed  directly  into the second.   NWFT,  the  Network  Flow and



Transport Code,  is a simple network flow groundvater code  in which the

-------
 flow through each segmtnc of the network is modeled as one-dimenstonal



 Darcian  flow.    It  t*  thus essentially  a  steady  state  flow  code,



 although the boundary conditions can be varied  from  one  point in time



 co another so as to iimulate time varying flow in a quasi steady-state



 fashion.   The   elements  of  the ,network  might  b«  parts  of  certain



 aquifers  or pathways  connecting  one  aquifer  with another.   On  a




 smaller seal* there could be individual drifts  in  *  rained repository,



 filled  boreholes,  or  any  other  physical  pathways  for  groundwater



 within which  Darcian flow  would be  a reasonable model.   Given  the



 appropriate hydrologic boundary conditions  and parameters describing



 the  geometry, permeability,  and porosity  in  individual segments of  the



 network,  NWFT  solves  for  the  potentials  atnrodal   points  and  th«n



 calculates   the   Darcian   flow   and the   average   interstitial   fluid



 velocity in each individual segment.   DVM,  the Distributed  Velocity




 Hethod  Code, is  used to  model  the transport  of  radionuclides In  the



 flow regime  described by  th«  output of  NWFT.   In the  absence  of



 transformations  from  one  radionuclide  to another  du» to radioactive



 decay,  DVM essentially associates  with each radionuclide a transport



 velocity  which  is the fluid velocity  divided  by a retardation factor



 appropriate  for  the nuclide in  question.  Thus  in this  case  it would



 be relatively  simple,  given the output of  the flow calculations  from



NWFT,  to  track  within  DVM  the inventory  of  each   radionuclide  in




 Individual  s.egments  of the  network.   The  complication arises in  the



case  of  radioactive   decay chains,  in  which  on«  radionuclide  may



gradually   evolved   into   several   others   and  these   individual



radionuclides may have different  retardation  factors.    In  order  to



account  for this factor  (as well  as  radioactive decay  itself),  DVM

-------
 mak..  use  of  an  effici.nr. mathematical  technique  for  tracking  the



 transport  of the varlou. nuclid.fi And maint.tning a current inventory



 for  each of the  segments  of the network.  The details  of this model



 are  de.cribed   in  th.  code's  original  documentation  and  *re  noe



 important  here.   In  addition  to  the., mathematical  capabilities.



 NWFT/DVM  ha.  an  additional  useful characteristic.   The codes  arc



 relatively  simple codes,  allowing for many repetitive  calculations at




 minimal  cost.    They  are  in fact set  up in  order  to allow a large




 number of Monte  Carlo simulations  to be performed on the system, where



 the individual parameters describing flow and transport can be  modeled




 by  probability  distributions  from  which representative  valuas  are



 selected  during  each   simulation.    In  addition,  to  sharpen  the



 usefulness of the Monte  Carlo simulations, the codes are  arranged so



 that the sampling from the  set  of probability distributions for  the




 various  sample   parameters  can  be  carried  out  by a  sophisticated



 technique  called  Latin  Hypercube  Sampling   (LHS).    In   «  certain




 statistical sense, this technique maximizes the amount  of  information



 that  can  be obtained from a limited numbei  of Monte Carlo simulations.



 The  total capability of NWFT/DVM then might  be described as follows.



 The  repository and surrounding geologic  system  can  be  described by a



 relatively  large  number of parameters representing geometry, flow  and



 transport  conditions,  and  release  rates from  the waste  packages.



Uncertainties  in  each  one  of  these  parameters  can be  described  by



aligning a  probability distribution  function to the paraneter,  rather



than limply  apecifying  a  particular point value.   The computation  can



b«  run a  large   number of times  in  aMonte Carlo fashion,  each time



selecting  a  value  for  each  input   variable  from  its  probability

-------
  distribution  function.   The  output th.n.  which m.y  be  Inventories  in

  individual  compArtn,ent.,  r.l....a  to  th. .cee..ibl.  environment,  or

  even associated  health  effects (obtained by  u..r..uppll.d conversion

  factors)  can  also  be  represented  by  «  probability  distribution

  function.    The  nature   of   this .  distribution  function  can  aid  in

  understanding the uncertainties in the system's performance.




       While  the Agency  believes   that  the  techniques  contained  in

  NWFT/DVM may  prove  to  be  extremely  useful  as  the  NRG evaluates a

  propose repository system for licensing, only rather restricted use

  was  made of NWFT/DVM for  the  Agency's  risk analyses.  In' particular,

  the  Mont.  Carlo  capabilities  were not  utilized.    Individual  point

  values for  the  analyses were chosen to be  identical to those that were

 applied  with REPRISK, which  does not have  the Monte Carlo capability.

 The  resulting  flow  regimes were therefore  essentially identical since

 they are both based on sirapU networks described by Darcian flow.  The

 unique contribution of NWFT/DVM was in the DVM portion of the analysis

 where,   «or  tiraeframe.  longer  than  ten thousand   years,   certain

 r.dionuclide decay  chains were  modeled so as to obtain  groundwatcr

 concentrations  and health  effects  over  the.,  longer periods of  time.
                                                                 x
 As   described  in   the  documentation  for  REPRISK,   there  are  some

 simplified  approaches  to decay chains also contained in the EPA's  own

 codes.    However,  the availability of  DVM   enables  none  accurate

 calculations of release,  for  time   periods longer than  ten thousand

y.ars,  «;ter which the  transport of decay chains begin,  to become a

•ore  important  factor.

-------
 2.6   Site Parameters




      A number  of  parameters are  used  to described  the  geometry and

 hydrologic conditions present at  the site.   The  most  important ones
                                    *
 are  listed in Table 2.1  and can  best  be  understood in conjunction with

 the  generic cross  section shown  in Figure  2.3, with certain exceptions

 to be described shortly.   The conceptual framework of the lithology at

 the  site  is that the repository  is located between two aquifers called

 respectively  the  "upper  aquifer" and  "lower  aquifer".    For  actual

 applications of the model to simulate  conditions present at  one  or

 more  real  sites,   the  upper^  and  lower aquifer do  not  generally
                              I
 represent  single   hydrostrat/rfgraphic  units,   but  rather  represent

 "synthetic aquifers" whose properties  are defined to approximate the

 combined  properties of a number of transmissive units above and below

 the  repository horizon.   For example, if a number  of such transmissive

 units are present  above  the repository at  a particular site and if the

 application  of the  generic models  described  here  is   intended  to

 represent  conditions  similar  to  those  at  that  site,  then one can

 calculate the combined volumetric  flows in the upper units and define

Appropriate  hydrologic   parameters  so   that  the synthetic  aquifer

 represents the  sane total flow.   Similarly,  by  varying one  or more

 additional parameters it  is possible to simulate  the effective  fluid

 velocity  in any one of the actual  units.  This will be illustrated in

 subsequent sections when specific  llthologles are  discussed.




      The  general way  in which  the REPRISK model  constrains  the risk

 analyses  is that the upper aquifer is considered  to be the pathway of

 groundwater release of  any radionuclides that  leave the repository.

-------
     (An  upward gradient  is  considered  to  exist along  any  pathway  that

     might exint or  develop  between the lower .aquifer  and  upper aquifer.)


     This is why there  10  greater emphasis in Tabl* 2.1  on the  properties


     of the upper  aquifer.   It is  recognized, however,  chat  at  potential
                                     •
     repository   sites  the  hydrogaologic  environment  may   be  somewhat


    different from that assumed  in the  generic model.   For example, there


    nay be no significant  water  bearing units  balow  the repository (as in


    a number of crystalline  rock sites),  or  above  the repository  (as  in


    the case of a  repository  in  the unsaturated zone), or  there  may be  a

    prevailing  gradient  that  is  downward  from  the  upper  aquifer to  the


    lower  aquifer,  in which case  the lower aquifer would appear  to be the


   more  likely  release  pathway.   These  cases  can  all  be accommodated

   within th« modeling framework used by the Agency.





        With  respect  to the  specific  hydrologic  parameters  listed  in

  Table  2.1,  hydraulic  conductivity is  used in conjunction with  Darcy's


  Law  to  estimate  volumetric flow rates through various 'components, such


  as pathways   from  the repository  to  the upper aquifer or  along  the


  upper  aquifer itself.   For  further  elaboration on   the  mathematical


  equations referred to  here  and  below,  one may consult SPA-520/3-80-006


  and the references  cited  there.  Only  Darcian flow has been treated in


  the analyses,  and work by the  Department of Energy at specific sites


 tands to confirm  that  the flow  regimes are  such  that  this appro*;h is


 adequate.  The porosity is used  to  convert  volumetric flow rates into


 average effective fluid velocities  in  the direction of  movement.   In


 particular,  the volumeric flow  rate  ia divided by porosity to  otain


 an  effective  fluid velocity,.   It  is  important to make  this  distinction


when considering  the time of arrival  of  contaminated groundwater at

-------
                   polnt   „


                                    ., che
            b.                          .f
  other phenomena  aasum.d  to  be  present
                                   present,  ft8  described  later  in  the

  ••ctlpn  on release  mechanisms.
                 pr.viou.ly


                b.
                            tn


                    clted
                         8qul£.r,
             .1.0
             .   st.ndlrd



V.«OUB t.mp.r.tur.1 .t. built ifs th. mo
-------
analysis   modele   indicates '  chat  the   result*   of   the   Agency's



calculations   are   not   highly   lensitive   to   these   engineering



assumption*,  and therefore, Che  Agency has  adopted  a tingle  Bet  of



assumptions  to be  applied  to  repositories at all  potential  types  of



sites.   These  parameters  are  shown  in  Table  2.2.    The  repository



considered  in  the  Agency's  analyses   is  assumed  to  contain 100,000



metric  tons  of heavy  metal equivalent  (MTHM)   in the  waste,  which



assumption  can  be  used to  normalize the  results  of  the  Agency's



analyses  to  repositories  with  different  total   amounts  of  buried



radioactivity








     Because  of the  rectangular  areal dimensions  of  the repository.



for  certain  types  of   releases  it is  important  to  make a further



assumption about the  orientation  of  the repository  with respect to the



flow  direction in  the  upper aquifer.   It  is assumed that the longest



dimension of  the repository, 4  kilometers, is  in the same direction as



this flow pattern.   The mined volume of the  repository, as well as the



porosity  of  "he backfill,  enter  into calculations of  the  amount of



radionuclides that  might  desolve  in  the  water  that  would  gradually



seep into the repository after  its closure.   Such dissolution might be



United  by solubility  factors,  and therefore,  this  water  volume is



significant.   It is also important because It can be used to estimate



the  amount of  dissolved radionuclides that might  be  withdrawn by an



exploratory well that penetrates  the  repository at some point in the



future.  The  case  of a salt repository is somewhat different because



it  is  anticipated  that  such repository would gradually seal-up after



closure as  a  result of salt creep.  In this  case,  therefore, the time



to  such  closure and the amount of moisture  present at that  time, are

-------
important for  the  risk analysis.  The  other  parameters  in Table  2.2

enter  in connection  with  epecific  release  mech*ni«mu  «nd  will  be


discussed in subsequent sections.



                                      •
2.8  Waste Package Parameters




     The vast*  package consists of  essentially  two main  components:


the  waste  form and  the canister.   Each of  these may  actually  be


somewhat complex.  For example,  the waste form consists of  an initial


Inventory of  radionuclidea  contained in  some physical matrix,  which


may  be unreprocessed  spent  fuel,  borosllicate  glass, or  some  other


alternative.   Thus the waste  form might  have a  somewhat  homogeneous


physical  structure,  as in the  case  of glass, or a more  complicated


one.   The canister may  actually consist  of  a  number of  components,


such as a  more or  less concentric  set  of  containers  with  spaces,


coatings or other  materials between them.   For the conceptual analyses


reported  here,  the Agency  has  decided to  adopt a  relatively simple


model  for the  waste  package,  consisting  of a relatively  homogeneous


waste  form  and  a single  canister.   The  waste  package  parameters


entering  the analysis  are shown  in Table 2.3.




     The  assumed  total initial inventory  of radionuclides  is shown in


Table  2,4.   This  is  based on a number of detailed calculations and

                                                  s
considerations  elaborated  on  in  the  Population  Risk  Report  and


summarized  below,   The radionuclide inventory in  the  waste will vary


according to  whether  the, waste  consists  of  spent  fuel   itself or


whether  the spent fuel has been  reprocessed  to recover some of the


tranuranic  components.   Since  reprocessing  is not part of  the present

-------
 national  civilian nuclear  program,  we  have  assumed  a  radionuclide



 inventory  that  IB roughly  equivalent  to that associated, with  direct



 disposal of  spent fuel.   For a reference repository based  on  100,000



 MTHM, the total radioactivity may be characterized roughly  as shown in



 Figure  2.4.   Thia figure  is based  on  calculations  using  the  ORIGEN



 Code.   The   fuel assembly  structural  components   are   essentially



 insignificant contributors  Co the  total  radioactivity for all  periods



 of time and  have  been ignored in  the analyses.   However, it is clear




 from che  graph  that  Che  fission  products dominate the  radioactivity



 early  in  the  life of  che  repository,  and che  actinides  and  their



 daughter products  dominate  che  inventory somewhat later.    The  choice



 of radionuclides  listed  in Table  2.4 is  certainly not all  inclusive,



 but ic does include the dominant radionuclides, as well as  a range  of



 radionuclides  with  somewhat  different  properties.   Thus,  a  risk




 Analysis   based  on  Che  radionuclides  shown   provides   a   valid
»


 representation of the risk from a repository.
      The canister  is essentially a  protective container that  should




 inhibit  the leaching  or  the  disolution of  che waste  form and  Che




 consequent transport of wastes cowards Che accessible environment.   In




 the risk analyses  fcpuTLed here, che performance of  che  canister  has




 been  represented  in  an  approximate  fashion  by  a  user-specified




 caniscer lifetime.   Up until  this  time is reached,  no  radionuclides

     5


 are assumed to be released from undisturbed waste packages.

-------
     The  release  rate  of  radionuclides  from  the  waste form,  after the



beginning  of th«  period  when  the  canister can no  longer  be  relied



upon,  may  be  governed  by  one  of  two  limiting  physico-chemical



processes.   The  first  of  these  processes   la leaching,  which  is



essentially  controlled by  the  accessibility   of  the  radionuclldes



within the waste form to any groundwator with which the waste form may



come  into contact.   A leach  rate,   specified  in  terms of fraction



loached  per  year, is  the input parameter used  to characterize  this



aspect of  radionuclide  release.  The  second  limiting process is  that



of  solubility.    Even  though   the  groundwater  may  have  access  to



radionuclides within the  waste  form matrix,  the  water may  already



contain  sufficient amounts  of  the  dissolved  material so  that no  more



can enter into solution without corresponding precipitation,  resulting



in  a  constant concentration.   Such  solubility  limits  for  Individual



radionuclides are taken into account in the risk calculations reported



here.  The values are based  on  literature  values  from a  number  of



sources  and  will be  discussed in later  sections  in  connection  with



Individual geologic environments.








2.9  R»l«*flfe Mechanisms
     The release  mechanisms  by which radioactive waste  may leave the



repository and be  transported  to  the  accessible  environment have been



introduced in Section 2.2.  Both there and In Section 2.5 It was noted



that  release  mechanisms  may  lead  to  the   direct   transport  of

-------
  racilonuclides  to  the  land surface or  to  che  atmosphere,  or  they  may
  l*«d to  the  groundvat.r transport of vast, away from  the  repository.
  The purpose of this section is to review the release mechanisms  one by
  one and de.cribe  the conceptual models  that have been  incorporatsd in
  the ri.k  analyses.  A  more  detailed ' mathematical  treatment  of  the
  •quations used to  implement the conceptual models may be found  in  the
  Population  Risk  Report  and  further  background  on  the  release
  mechanisa. themselves  can be found In the Task D Report.  Six classes

 •4HHV have been  modeled by REPRISK.   The  meteorite  release
 mechanism,  however,  has  not  been included in J^j^lculations <£^H
 -^•Because  lt  ls conslder.d to be  of negligible  significance  in
 estimating the  performance  of  a  repository  or  the  selection of a
 repository «ite.


      Except for the case of a repository in salt,  normal  groundwater
 flow refers  to the movement of water through  the repository horizon,
 •ccprding  to the natural hydrologie conditions, perturbed,  perhaps,  to
 «ome  degree   by   the   presence   of   the  repository.   During  the
 construction and operation  of the repository,  it   is  expected  that
 water in  the aurrounding rock would gradually be drained so  that the
 rock will  enter an unsaturated conditon near  the  openings.    (In the
 ca«« of  tuff,   the  repository would be  located  in  a  rock mass that
would be unaaeurated at  the start,  and hence this discussion  requires
•light   modification for  the  tuff  case.)   After   the  end  of  the
operational parlod  and  sealing of the  r pository,  it is expected that
water would gradually seep back into  pores  and fractures In  the rock
and  establish  a flow regime  connected  to  the regional  groundwater

-------
system.  The  resulting flow patterns may  be  icmewhat different  from
                               •*


Choc*  prior  to  the  excav«tion  of  the  repository.   For  example,



the heat generated by  Che  waste  may modify the  hydraulic conductivity



of the  surrounding  rock and may also change  the  properties of water,



naking it less dense and less  viscous.*   The lower density can lead to



a  buoyancy  effect   that  may  cause an  increased vertical  hydraulic



gradient.   The decreased  viscosity enables  the  water  to  flow  more



easily  through the  rock and  hence allows for potential  increases in



flow rates.   In  the  case of a repository  in  tuff,  normal groundwater



flow   refers   to   the   downward   percolation   of   water   through  the



unsaturated rock  towards the water table.  This  downward movement is



not  expected to  be   influenced   greatly  by  the  presence  of  the



repository,  because  the limiting  factor is essentially  the  amount of



water   itself.   Release  mechanisms  within  the   category  of  normal



groundwater flow  lead  to potential radionuclide releases  to  an aquifer



with consequent transport  of  the radionuclides within the  aquifer.







     The faulting release mechanism is  intended  to  cover  both the



cases  of a new fault occurring  at  a repository site  (and intersecting



the  repository itself), as well  as the  case of the reactivation  of an



old  apparently  stable fault.   This  reactivation is  also assumed to



lead to an  intersection with  the  repository  itself.   This  is  treated



In Che model  as  a vertical planer structure  with increased hydraulic



conductivity  over  and above  that of  the original  rock.  Thus,  the



fault  can  lead to  an essentially new and  preferential pathway  for



groundwter aovement.   Its hydrological properties are  different from

-------
 chose of  th.  original rock and,  in  fact,  there  could be  rather  high


 volumetric  flows through  «uch featur...  In  addition to creating  a


 flow pathway, the model  for  faulting assumes that  since  thac this can



 be a  relatively viol.nt and di.ruptiv. event,  the  integrity of waste


 packages  within a certain  distance  of the  fault  la  destroyed.   The


 result is th. earlier un.et of leaching (if this  had not  already began


 by the tin,, of  fault movement).   A.  in the case  of normal  groundwater


 flow,  the  only  releases by faulting are  assured to be via  a  pathway


 connecting  th.  repository  with   an   aquifer,   thereby,   enabling


 groundwater transport of the  waste.   Faulting has  been  treated  aa  a


 random sto.cha.tic process for purposes of the Agency', risk  analyses.


 This  is not  to  say that faulting is a  random  process, but  only  that


 faulting at a real r.spository site  shold b. .bl.  to be  bounded  by  a


 similar  physical process that occurs randomly.   The likelihood of new


 or reactivated fault,  is  estimated on the basis cf  geometric  arguments


 and simple probability concepts,  as described in the next chapter for
 salt.
      The breccia pipe  release mechanism refer* to the development of a


localized  dissolution  feature  and  only  applies  to  the  case of a
  x

repository  in salt.   Breccia  pipes,  or collapse chimneys,  have been


known to develop from  the base of the salt beds and proceed vertically


in  a relatively  rapid  and  sometimes  abrupt fashion.  Such  &> breccia


pipe  is attuned  to  provide  a  relatively high permeability  zone that


can  facilitate the vertical  movement  of groundwater.   In addition,  it


ia assumed to  violate  the integrity of waste packages that are located


within  the  collapse  zone.    A probability  of  occurrence  of  such

-------
features ii based  on observed statiatical average! and  ic  assumed  to

occur randomly.   Since  good  site  ••lection procedure*  should  •n«bl«

the  selection  of  a  site where  the  likelihood of  occurrence of  a

breccia  pipe  is much  less  than  the  average  occurrence rate,   it  is
                                     •
believed that the resulting calculation will be conservative, that is,

tend to overestimate  the  risk.  This  i»  the intention with respect to

all the modeli used by the Agency.




     The  occurrence  of  future exploratory  drilling  at  a  repository

site  cannot be  ruled out,  even  though  steps will  be  taken  in  the.

decommissioning of the repository to signal to future generations that

dangerous  materials  are  buried there.   The  Agency has  considered a

wide  range of potential  purposes  for  drilling  in  different geologic

media, and ha* estimated  drilling rates which  are intended to be upper

bounds on  the  future likelihood of  drilling at a repository site.   In

estimating  these   values,   no  credit   hat  been   taken   for  the

communication  to future generations of the presence of the repository,

except  that for a limited time in  the future it is assumed chat such

communication  would  be  completely effective.   This value is specified

in  individual  risk analyses and will be discussed later.  The dominant

purposes  for future  drilling may vary  from  one  kind  of  geology to

another.   For example,  for  salt  deposits  in sedimentary basins,  the

dominant drilling  is expected to  be in search of oil and gas, whereas

in  a granitic  terrain  the  dominant purpose  might  be exploration for

water  or minerals.   The basis for  the  selection of drilling rates is

discussed in  connection  with each lithology in subsequent chapters.

The drilling  release  mechanism  contains  a  number of components.

-------
       4t  is  possible  X
release of radionuclides  to ar aquifer.   In the calculations JjBB^E

^^••IHHm^this third  release mechanism has not been explicitly

included because earlier  calculations  showed that  its contributions

were  negligible  by comparison to  other forms of release.



      Volcanism also  has the potential to release waste  from a

repository,  either by transporting  it directly to the  surface or by

-------
translocating it In an underground volcanic  structure,  such as a sill
or a  dike,  which may also  encounter an aquifer.  Calculations reported
in the  Population Risk Report and the Task D report suggest that this
latter  mode  of transport  is  overwhelmingly dominated  by  the  fault
release  mechanism, in tarns of  likelihood of  occurrence,  and that it
is roughly similar  in terms of consequences.  Therefore,  it  has not
been   included  in   the   comprehensive   risk  analysis   because  its
contribution  la essentially negligible.   The release of radionuclides
to the  surface, however,  with magma,  ash, or gases passing vertically
upward  through  the  repository  has the  potential  for  the significant
distribution   of radioactivity  to the  accessible  environment,  even
though  its  likelihood is  relatively small  in most  cases.  It  has
therefore  been  included  in the calculations  for  certain lithologles
that  have  been  investigated in relatively close  proximity  to active
volcanic regions.   This does  not  mean  that the Agency  believes that
volcanism  is  likely at any  of  the  sites  that  are  being investigated,
but only that it may be the dominant low probability high consequence
•vent  and  hence should be  included  in   the  calculations  to  give  an
adequate  perspective  on  the  risk.   In  the case  of  the  release  of
radionuclidee by  volcanism,  there are  components  in  the  model  for
waste release both to the  air  and  to  the  land surface.
       4

     The   release   mechanisms   are  further  discussed  in  subsequent
chapters,  largely with a view to  summarizing the  specific parameters
that  are  used  to  characterize  them for the risk  calculations.   As
noted earlier, further details on  the rationale and the struct re of
the models can be  found in the associated documentation.

-------
  2'10
       A companion  report  to  th. present one discusses  in  much greater


 .detail the  analysis  of  the  transport  of  radionuclide.  through  the


  accessible environment,  leading  to exposure to th. human population.
    b\t.

  T*4J  2.5  li, tg  the  pathways that  have  been  considered.   A  large


  fraction  of  this  modeling  takes  place  not within  the risk  analysis


 code  REPRISK  itself,  but   rather   in   the   form  of   radionuclide


 release/health  effects  conversion factors  that are  incorporated in


 REPRISK  on  the basis of  the moling  described  in  the  companion


 report.   Health  effects  conversion   factors  are  provided for  both


 •omatic and genetic health effects, in the form of cancer fatalities.




      Some  of  the  environmental   transport parameter,  are  included


 directly  in  the   REPRISK  code.   These  ar« described in  the  code


Documentation package, a. well as  in the Population Risk Report.   They


 are  constant  for  all  lithologies  and will  not  be  discussed  in  any

 greater detail in the present report.
      The  results  of risk analyses as discussed in this report contain

relatively  large  uncertainties   in  th.  numerical  results.   Such

uncertainties  are due  to a number cf factors, among them:


      o     the  long-time frame over which predictions are needed;


      o     the  simplified nature of the models In comparison with the

           real physical situation;

-------
      o     the  generic nature of the  modeling,  i.e.  the  lack  of

           incorporation  of   detailed  alte  «p*cific   dat*   for  *

           particular real site;

      o     the  use of simplified system models  that  do not  capture  all

           the  detail* of individual  physical,  chemical, And  geological

           processes.

The  purpose  of the risk analyst* has been  to make rough approximations

of   the   capabilities  of  geologic   disposal  of   radioactive  waste.

Therefore,  despite these uncertainties, the Agency believes that the
                JK A>  W-^» 4 ** fuAatft JUCnW W  .
estimates ger  rated ^HB*" provide an adequate technical basis  for the

associated regulations.



      In  order  to lend perspective to the uncertainties  in  the reported

calculations,   the  Agency  has  proceeded  as  follows.    First,  in

estimating parameters  or  In  choosing models  to  represent   various

processes,  an  attempt  has   been made  to  consistently   overestimate

factors  that contribute  to  risks from the repository.   That  is, we

have  attempt* '  to calculate via our  risk analyses upper bounds on the

expected performance of  the  repository.   This is  the same philosophy

that  was adopted in "the Population Risk Report and  in  the ADL Task D
   ,                             s
Report,  where  some  of  the  underlying  models  and parameters   were

developed.   Second,   extensive  use  has  been  made  of  sensitivity

analyses in order to  understand how much the  results   of  the   risk

analyses vary  with  the variation  in  certain model  components or

parameters.    For   parameters  that  are   particularly   crucial  in

determining  the  final risk results,  we have devoted special attention

to choosing  appropriate  values.  Third,   in cases  where  it  has   been

-------
difficult to model on a generic basis the characteristics of a sice or

a process, we hav« or.r.ASfiionally chosen to use more than one value for

a parameter, often referred to in the supporting documents as a "first

estimate"  and  a "second estimate",   The  first estimate  parameter is

Intended  to  represent a process  or  condition at a site that is chosen

to be  particularly favorible  with  respect to this phenomenon,  whereas

the  second estimate value  corresponds  to  a site which while it may be

favorable in an  overall sense, may  be somewhat less advantageous with

respect  to the  particular phenomenon  under discussion.   The  Agency

realizes  that  the choice  of  a  real  site  for a  geologic  repository
                      ft
will represent  a compromise  between  a  number of  desirable  goals  and

thus,  it is quite conceivable that  a  site selected using  a  full range

of  siting  criteria  responsive  to  the   intent   of  this  and  other
                                   I
regulations  could  still   include   conditions  that  would  be  better-

represented by the second estimate  values.




     It  is  important  to distinguish between the   type  of  uncertainty
.  . .,  .                           ^Jr»>v/.
included  in  the  generic analysis ^0BeV her« *nd c^» uncertainties

that would  remain with  real sites  when  they are characterized  and

modeled in connection with  the decision on where  to put a repository.

Many of the uncertainties  included  here might  better  be characterized
                   •A<*»L        A
as variabilities,  •^preal  site^ there  might  be  a wide  variation in

the property in question.   The attempt in our risk analyses to include

such variation corresponds then to  an uncertainty  in the final results

aa to how well they characterize the performance of the repository.  A

real site  will  include the additional uncertainties  associated with

data collection,  site complexity and difference  of  opinion about  a

-------
specific  sice's  character is tic.   Such uncertainties are  not  properly




within the scope  of  the work  reported here.
     It should  be  noted that in the accompanying 40 CFR 191 standard,



attention  ha* been devoted  to  the  residual  uncertainties  expected to
be present at the end  of  the  analyses of the  real  site.   Therefore,
the. results
           have  not  been  translated directly into the numerical values




of  the  standard;  rather,  an  additional margin  for  error  and/or




uncertainty has  been included  in  the numerical values in the standard.

-------
                CHAPTtR 3  •  REPOSITORIES  IN  BEDDED SALT







3.1  Introduction








     For almost  30  years  0alt deposits have  been considered to be  a



prime candidate for a nuclear waste repository.   There are a number of



reasons for this.  Salt deposits are common in  several regions  of the



United States  and  they are  found at depths  considered to be suitable



for  a repository.   By their  very presence,  they indicate  relative



geologic stability  and hydrologic  isolation,  since if g -undwater had



ready access to  them,  the salt would have  been dissolved and carried



away.   While  it  is  the  case  that almost  all  known salt  beds  are



undergoing  gradual  dissolution by  groundwater,  the  rates  of  such



dissolution  processes are  generally so  slow that  these  deposits  are



expected to  remain  substantially intact  for millions  of years.   There



is,  in  addition,   extensive  experience  in  constructing underground



mines  in salt,  and  there  is the  additional advantage  that gradual



creep of the salt will  aid  in the reseallng and  the ree»-ablishment of



total Isolation of a  repository placed in such an  environment.  On the



other   aide,   there  is  the  disadvantage   that  if  some  unforseen



circumstances  arise that  introduce groundwater  into  contact with the



aalt near  or at the repository, then the effects  might be more severe



because  they could be aggravated by relatively  rapid dissolution.  In



addition,  salt deposits  are  located in  sedimentary basins that often



contain other valuable resources such as oil,  gas,  and potash.  As  a



result,  the  adoption of a site  for  a nuclear waste repository may



either  preempt  the resources  present  at  the   site,  or may lead to



future  risks from efforts to obtain those resources.

-------
     The Department  of Energy  i«  actively  investigating sites  for  ft

repository In bedded salt deposits in the Paradox Basin in Utah «nd in

the Palo Duro  Basin in west Texas.  The  evaluation  of specific sites

is a long and  complex  process  expected to  take a decade  or  longer;

thus it  ie  not  possible  to predict* ac part of  the present study how

wall  a specific site  in  either  of these  basins might  perform  in

isolating radioactive wastes.   However,  based  on data  collected Co

date by the Department  of Energy and others,  it  is possible to define

certain idealized conceptual models of repository sites in each of the

two basins so  as to make rough  first approximations  of the potential

performance  of   such  repositories  and  to  identify  some  of  the

parameters  that are  most  critical  in  determining that  performance.
                                   Al/*-U* %V /w\*6yS  uw-
This chapter contains a summary  of ^fjjj^ff^j^ff^^y the Agency
so-called  "generic"  sites that are based on simplified  models  of the

general  geologic  and  hydrologic  conditions   reported  at  promising

locations  in  each  of  these two basins.



     In  connection  witl   the publication  of  the  Draft  High  Level

Standard in 1983,  the Agency  also reported on risk analyses associated

with the repositories in salt.   Such analyses  were based on two types

of  salt  environments.  The Agency  used  a bedded salt model  based on

data available at  that  time  for  a number of basins being investigated

for salt repositories.  (At that  time the Department of Energy had not

yet settled upon the  Paradox  Basin  and the Palo Duro Basin as the most

likely  locations  for a  salt repository.)   The  most extensive data

available  at  that  time  was from  the Delaware basin  in southeastern New

Mexico,  and this  provided important  information  to the  Agency  in the

-------
 ^uiib Liuction  ot  its  models.   I"n  order  Co compare the previous analvses

 for  bedded malt with  those  carried out in  connection with the Final

 High  Level Waste Standard, this  chapter also  contains a brief review

 of  th« parameters  and results  considered there.   Also in connection

 with  th« draft  Standard,  the Agency considered the  possibility  of a

 repository in a salt  dome.   This is  quite  different from the bedded

 salt  deposits previously considered.   Such  salt domes  are  corampn in

 the  Gulf Coastal Region  of  the United States  and are being actively

 investigated  by the Department  of  Energy.   The Agency  concluded in

 connection with  its  previous  analyses  that  there  were  sufficient

 problems  associated  with  the  disposal  of  radioactive wastes in a sale

 dome  that  it  was  unlikely  that such  a  deposit would ultimately be

 proposed  by Che  Department of Energy-for -the 'actual"construction of a

 repository.   While  it  is possible  that the  site investigations being

 carried   out  by  the  Department  of  Energy may   lead to  different

 conclusions (so  that a sale dome  repository may ultimately be found to

 be acceptable),  the Agency decided  not to expend  further  efforts in

 th«  evaluation  of  the performance  of a  repository in a  salt dome.

 Therefore,  this  geologic  formation is not discussed here in connection

 with  the  risk analyses supporting the High Level Waste  Standard.
                                                          X

                                                          Section  3.2

 cuonarizea  the  important  input parameters that have  been  used in the

Agency's  risk analyses for beddad salt.  These parameters are reviewed

 for three cases: Salt A (based on data from the Palo Duro Basin),  Salt

B (based  on data for the  Paradox Basin),  and the original bedded salt

stratigraphy  considered   in connection with  the  draft  Standard.   In

each  case  there is  a figure that  presents a  cross-section  of  the

-------
assumed stratigraphy at the site.  There is also on accompanying table



which  summarizes the  input  parameters  used to  ehftracr.nrize  the site



for the risk  analysis  models.   Following this,  there  is a sequence of



tables  summarizing  the release mechanisms  that  have  been included in



the risk analyses for  the  site.   Sirfce  most of  these  data can best be



presented  in  the form of such  tables  and figures,  there is minimal



Cextual   discussion  of   additional   details.    However,   as  noted



previously,  there  is  a  more  extensive   discussion  of  Che  actual



mathematical  models used  for  the analyses  in  the  Population  Risk



Report, as well  as  in  the model documentation packages  themselves.

-------
3 2  Inrut  Parameters



     The  Input parameters  for the  risk  analyses  carried  out  by che

Agency  fall into  several categories.  These  are  listed  below  along

with a reference  to  the  section of  this or  the companion documentation

in which more  elaborate  discussion  may be found.



            i
     o    Site Parameters.  These are discussed in this section for

          Salt A,  Salt B and  the original generic bedded salt sice.

     o    Repository Parameters.  These are the same for al'

          lithologies  and have been  presented  in  Section  2.7  with  a

          summary  table  given as Table 2.2.

     o    Waste Package  Parameters. " These have been discussed in

          Section  2.8  and  summarized  in Table  2.3. ,  Radionuclide

          inventory  is the same for all lithologies,  and leach rate is

          a  generic  parameter that  is given  the  same values and the

          sane  range of values  (for sensitivity analyses)  for  all

          lithologies. Canister lifetime  has  the  same baseline  values

          for  all  lithologies, but for certain calculations  reported

          later a  shorter canister  lifetime for a  salt repository has

          been   used.   .The  solubility    limits    for   different

          radionuclides do depend on  the  site, but are listed for all

          lithologies In Table 2.3.

     o    Release Mechanisms.   The  input parameters to characterize

          specific release  mechanisms for  the  salt  repositories  are

          discussed  in this section.

-------
Therefore.  If  water were to 
-------
effect  could  drive  groundwater  cowards the  upper  aquifer, which  is



thus modeled as the pathway along which total cumulative releases have



been   calculated.    There  are  a   number  of  differences   in  the



calculations carried  out  for  Salt A and Salt B as compared  to those



used  in connection with  the  Draft. High Level  Wait*  Standard and che



generic  salt' site  in Figure  3.3.   In this  last  case,  there  i«  a



significant natural  upward gradient that  would move  groundwater from



the repository  to  che upper aquifer and which  would  be additive with




the   additional  effect*   of   thermal   buoyancy.   In  addition,  the



temperature profile  used  for  the early calculations  with the generic



model  (Figure   3.3)  are   based  on higher  heat  loadings  than  are now



being  considered by  the Department  of  Energy.  Because of this change



in  the Department's  apparent  plans,  the  thermal buoyancy effects in



the earlier risk analyses, reported,  for  example,  in the Population




Risk Report, were much greater Chan those presently believed to be the



case.








      Let  us also  note here  certain similarities and differences in



connection  with   the release   mechanisms   Included   in   the  present



analyses.   For salt repositories, the analyses discussed  hare include



five  release  mechanisms:   Faulting,  breccia pipes,  volcanes,  drilling



without hitting  a  canister,  and  drilling  and hitting  a  canister.



Other  release  mechanisms  have  also  been  considered  and  will be



discussed  at   the   end   of  this  section.   Tables   3.4  through  3.8




summarize the  important  Input parameters associated with these release



mechanisms.  In the case of  faulting, the  probability  of occurrence



in somewhat higher  than  the original  values used in  connection with




the, draft  standard.   The new  values  are  based on  correlations  with

-------
   selsmiclty  in  a  published 'report  of  the  U.S.  Geological  Survey,
   released  since  the earlier analyses  were  carried out.   These  value*
   suggest  that  faulting  is  more  likely  to  occur  and  to  affect  the
   repository in Salt B (Paradox Basin) than in Salt A (Palo Duro  Basin).
   The likelihood  of occurrence  of  breccia pipes  has  not been  changed
   from the original generic analyses.   In both the  Palo  Duro and  Paradox
   Basins,  chere is evidence of localized dissolution  features and  in  the
   latter  basin  at   least  these  are  often  associated with  collapse
   chimneys or  breccia pipes.   One  should  not* from  the model  input
  parameters  that  no such  features are allowed to affect the  repository
  for  the  first  one  thousand  years  after   closure  because  of  the
  assumption  that  detailed site  selection and  site  confirmation  studies
  would  be almost  assured  of Identifying  such features  if  they were
  already under development.   However,  it  is  possible that either as a
  result  of  natural forces or as a  result of the  perturbation of the
  natural environment by  the  presence of the repository conditions  might
  develop  that  would  foster  such  localized  dissolution  with  the
 propagation  of a  collapse chimney vertically  towards the surface.  The
 probability  of  this is  still quite low, however.  Volcanos have been
 included  in  the  analyses  in  order   to  represent  what  may  be  the
 limiting  very  low probability but  relatively high consequence  event.
 The  parameters  used to  characterize  than  and  to  estimate   their
 probability are identical with  those  in the original ganeric analyses
 and ara  discussed in the references.


      The  possibility of   future human intrusion  leading  to   some
disruption of the  repository has been an Important concern  for the
Agency  in developing High  Level Waste Standard.   While institutional
controls nay  be  successful  In preventing such  intrusion for a  limited

-------
period 'of  time  and long-term'markers and other means of communication
                              «%


nay  preserve knowledge  of  the sire  still  further,  th« Agency  hus,




adopted  for Its  risk analyses  the underlying  assumption  that  such




measures are of limited  effectiveness  and hence  cannot be relied upon


                                     •

completely   to   eliminate   the   possibility  of   future   intrusion.




Therefore,   in   the  original   generic   analyses   described  in  the




Population  Risk Report and  in  the references cited  therein,  certain




reference  future  drilling  rates   for  each  lithology were  adopted.




These same values  have been used in  the present analyses.   Some of the




future  drillholes  might  even  intersect  actual  waste  canisters,




bringing a portion of their contents to the surface.  Other drillholes




will   not  Intersect   the  canisters   themselves,   but  may   bring




contaminated  groundwater  to  surface.   It  is also  ^ossible that  in




abandoning  such future drillholes,  a more  permeable  pathway tight be




established  between  the  repository   and  overlying  or  underlying




aquifers,  but  calculations'  the Agency  has carried  out  suggest that




this  is  of negligible importance  compared  with  the direct release to




the surface.








      In addition  to  the  release mechanisms  discussed here, a number of
                                                      X



other possible  release  mechanisms for a bedded  salt repository have




been  considered,   Of particular  concern  has  been  the   question  of




whether  there  should  be  a  so-called  "routine   release"  nu =haniem




Included  in  the  analyses.   Such  a  mechanism  is  included in  the




analyses  for hard rock repositories where a small but non-zero amount




of   groundwater  movement  is   anticipated  through  the  r«;aaitory




throughout  its  lifetime.   The Agency has  considered  flow through




gradually   degrading  seals   in  shafts  and  boreholes,   as  well  as




groundwater movement through localized areas of  the repository that

-------
may  have  higher hydraulic  conductivity  than  the  numbers  generally


assigned  to  intact ealt.   For  example,  groundwater  «cep«  frequently


occur  in  underground  salt  and potash  mines,  not throughout  the entire



aine,  but only  through  localized feature*  such  as  "dirty"  areas  of


c«lc,  dissolution/collapse  features,-or  other geologic anomalies.   On



the  basis of  the  calculations  that  have  been carried  put  for  such



features,  the  Agency  has found  no evidence  that they could  contribute



co  non-negligible  releases  from  a  salt  repqsitory.  QMBBHMBM0



                                                  sensitivity analyses



       it was  assumed that 5%  o   the area  of the  repository had  a


hydraulic  conductivity roughly  5  orders of  magnitude  higher  than  the


                                                            -16
hydraulic  conductivity generally assigned to intact salt (10   ).

-------
                                                           These are  the
                      .11
             "Port on Envlronmantal Pathways
        th.
   char.ct.rll.  th.  .tt.   „
   •°" '"Portant r.Uaaa ..chaniama.
                3.1.  ,.2. and 3.3  ,hou  th.
             u.ed to dsfine  ^  llBpUfl>d

  «.!».«.  A..Mt.t.d vtth thas, fi,uv . „. IabUs 3.1,  3.2,  and 3 3
                 th.  mo.t
              th.  r..ulta
        on ApD.ndIx A to
       be.n uaad co con.trucc th. generic, cro....ectton, tor SaU A and
       ..   Th.  cro»..«cclon  ahovn In Flgu« 3.3 u b.s.d „ th. ..rtu,
           c.rtUd «,  ln  e0nn.ctlon Blth  th«  Dr.£t
          .  It  too ,. discu.Md In Appe7 iix ^ bu£

 on. .hould conault  th. Population  ,i.k  R.port  .nd th. ,„„
                         about th.  almll.rltt.s  and «„.„„...  ,mong
               . d.pth  o£  .pprol!lMt.ly
«  th. alt. thar.  1,  .  !.rg.  productlv. aoutf.r, th. og.11.1. Aoulf.r,
»btch  .h.uld of cour..  b.  .„ lBport>nt cMc>rn ^ jtcin
in  thi. raglon.  Ho«.v.r.   ch.r.  l. . S.n.t4l  downv.r<,
hydrauuc  h..d v.iu..  «   v.rlou.  ^peh<   M  ^ r>iui
•PP.«. that any .roundw.t.r pathway, that .tght p... through th. .alt
               to   hav.   .   downward

-------
                CHAPTER  U  - REPOSITORIES IN GRANITIC ROCK







4.1   Introduction








      Granitic  rocks are  attracting, increased attention  as potential



host  rocks  for a  nuclear  waste repository.   Such rocks  are widely



distributed   throughout  the  United  Settee,  and  thus  offer  the



possibility   of  b«ing  found  in  connection  with  oeher  desirable



characteristics for a repository site.  At depth they can be extremely



"tight",  fhe naturally occurring fractures being kept almost complete



closed by the high lithostatic pressure.  Mined openings in granitic



rock  are  expected to be highly stable for well chosen sites, and there



is  considerable   experience  in  such  underground  excavations  from



various  kinds  of  hard  rock  mines end  tunnels.    Unlike  salt,  the



likelihood of associated valuable resources  Is much lower;  when they



are present  they are  often in veins at the boundaries of the granitic



bodies or plutons.  Water wells are occasionally drilled  into granitic



rock, but because of the general trend of decreasing permeability with



depth, such  wells  rarely  exceed  several  hundred feet.   An important



distinction  between granitic rocks and most of the other host rocks



being considered  for a repository is that they are not generally found



in a  layered structure as  are sedimentary rocks,  or  even basalt and



tuff, and thus  the possibility of extensive  aquifers at a depth below



the repository  Is much  less  likely.  This  decreases  the possibility of



« productive and  high pressure source of water that could cause upward



flow  and  carry  radionuclides  towards  the  surface.   On  the other hand,



the certain  presence  of fractures  and the water saturated condition



expected  at  depth virtually guarantee   that  there  would be  some

-------
 |ro«mtwatsr movement  through "R  repository  in  granite.   It  may  occur  av
 extremely low volumetric flow  rates  and  velocities;  but unlike  sale,
 which is virtually impermeable, it would b« present and must  be  taken
 into account in estimating  the  performance  of a  repository.

                                   *
      The Department  of  Energy is actively  investigating  sites for  a
 repository in  granitic  rocks   in  the North  Central  and  Northeastern
 regions  of the  United States.   The  Department had previously  carried
 out  a screening of the entire  United States  and had Identified  these
 regions  as most likely to contain suitable repository environnu "s  in
 these kinds of rocks.  The  evaluation of specific sites is a long and
 complex  process expected to  take  a   dacade or longer;  thus  it is not
 possible to predict as part of the present study how  well a specific
 site  in  either of  these regions might perform in isolating radioactive
 wastes.   However,  based on  data collected to date by the Department of
 Energy  and  others,  it  is  possible  to  define  certain  idealized
 conceptual  models  of  repository sites in each of the two regions so as
 to  make  rough first  approximations  of the  potential  performa-".e  of
 such  repositories  and to  identify  some of the parameters that are most
 critical in  determining  that  performance.   This  chapter contains  a
 summary  of Wf^fjf^^^^ by the  Agency %^so-called  "generic"
 sites  that  are based  on  simplified models of the general geologic and
 hydrologic  conditions reported at  promising locations in each of these
 two regions.


      In  connection with  the  publication  of  the  Draft  High  Level
 Standard in 1982,  the Agency also  reported on risk analyses associated
with  repositories  and granites.  At that time the Department of Energy

-------
  had  noc  yet  settled upon  the North  Central  region or  Northeastern

  region as  the most  likely granitic  terraines  for  eh«  location »[  a

  repository,  and  therefore,  the  parameters used  by  the  Agency were

  chosen from more  general  surveys  Of granitic  rocks throughout  the
                                    *
  United States.  In some  regions  the granites  extend to or outcrop at

  the surface, so that  the  only  significant water bearing zones  above  a


  granice repository  might  be   very  near  surface  deposits,   generally

  glacial in  nature,  and  in  an upper  weathered  layer  of granite.  In

  other regions the granites  might  be rather deeply burled at hundreds

  or  even thousands of feet,  so that  a granitic  repository might have

  above  it  a  sedimentary sequence  that contained aquifers  that would

  themselves  be at depth.   In the  earlier analyses carried  out  by the

 Agency, this  latter situation  was  in fact  assumed, but after  a  review

 of  recent  Department  of Energy data  of potential  granitic  sites,  the

 generic granite model has undergone some revision.  The result takes

 the form of two  new models  of  a granitic host rock, one based  on the

 North Central  region  and  one  based  on  the  Northeastern region.   It

 should be  emphasized  that these are  still  generic in nature and are

 not  intended  to   represent   performance  at   a  particular   site.

 Furthermore,  the   Department  of   Energy's   investigations  have  not

 proceeded  to the  investigation of specific sites, as they have  in the

 case of  talt,  and  therefore the selection of data  Is even nor* general

 •nd  hypothetical.    It is  the   fancy's  opinion  that  the  parameters

 represented here  present  a valid  current  estimate of the performance

 of a repository in granite and  that  they fall in  the  range values to

b« expected  at  specific real

-------
                                                    ^
                          tnput
Agency, rlak analyie,  for granite
    b.  Pr.a.nt.d ln  th.

         ««,.! dlscus.lon  of
         .   ther.  1(   .
Report, .. V.U „  in
                     th. ^j  doculMntati<]n

-------
4.2   Input Parameters
     The  input param.ters  for the  risk  analyse, carried  out  by  the
Agency  fall into  several categories.  These  are ilsCed  below along
with a reference to  th. action of  this or the companion documentation
in which more elaborate discussion  may be found.
o
          PKf PflrMnrrm.  The., are discussed ln this section for
          Granite A, Granite B and the original generic granite site.
          &eP°iUory Palmers.   These are the same for all
          lithologiea and have been presented in  Section  2.7 with  a
          •unwary table given as  Table 2.2.
                                     These  have been discussed in
          Section  2.8  and  summariz.d  in  Table  2.3.   Radionuclide
          inventory is  the same  for  all  lithologies, and leach rat.  is
          a  generic param.ter that  is  given the sane values  and the
          •am.  range of  values  (for sensitivity  analyse.)  for  all
          lithologies.  Canister  lifetime has the same baseline values
          for  all lithologies,  but  for certain  sensitivity analyses
          different   variations   on  canister   lifetime   have   been

-------
          considered,   based'  on  the  Agency's   estimation  of  the



          harshness  of  Che  environment  in which the cantstor would be



          placed.  The  solubility limits  for different radionuclides



          do depend  on  the  site, but are  listed for all lithologies in



          Table 2.3.
                                    •


     Q    Release M»$himiams.  The input parameters to characterize



          specific release  mechanisms for  the granite repositories are



          discussed  in  this section.



     o    Environmental and Health Effects Pqrfljflft^ars,.   These are the



          sane for all  lithologies and  are discussed in the companion



          report on  Environmental Pathways Analysis.



Thus  the focus  of  this  section will  be on  the  parameters used  to



characterize  the  site  as  well  as  those  used  to  represent  the



most important release mechanisms.







     Figures 4.1,  4.2,  and 4.3  show the  geologic  cross*sections that



have been used to define the  simplified  models used in  the Agency's



analyses.  Associated with  these figures are Tables 4.1, 4.2, and 4.3,



which provide  the most  important geometric  and hydrologic parameters



chat affect  the  results. These  figures and tables  of  parameters are



based on Appendix A  to  this report, which is a detailed discussion or



how  region-specific  data  from  the  North  Central  region and  the



Northeastern   region have   been  used  to  construct   the  generic



cross-sections for  Granite  A and Granite  B.   The  cross-section shown



in  Figure  4.3  Is   based   on  the  earlier  analyses  carried  out  in



connection  with  the Draft High  Level Waste  Standard.    It too  is



discussed tn Appendix A,  but for more  elaboration one  should consult



the  Population   Risk   Report   and  the   references  cited  therein.

-------
   Important  observations about  the  similarity.   ^  *<**
                                      "iwiiarities and  differences  among
   these sites are summarized below.
        Both   ,lt..   havc  che  repo$ltory
   approximat.ly 460  m.ters,  whlch la .

   offer a reasonable balance between the "tightness" of greater  depth,
   *nd the  .ngin.ering practicalities  of .halter  depths.   While th.
   repository is th.  s«e d.pth below th. surface in both instances, in
   Granite A,  the surface  deposits  and weathered rock  that constitute the
  overlying  aquifer  are  very thin  and include  ponds,  bogs,  or  other
  surface  waters.    Thus,  for  this  setting,   the  groundwater  in  the
  •quifer  t.  in  intimate  contact  with  surface  water  so  that  the
  •ce.seible environment is  at th.  sit.,  rather  than some  distance  away.
  Cranit. B, on the  other hand,  is modeled with  ,  considerably thicker
  aquifer,  but without  surface  wat.re.   For this sit.,  th. accessible
  •nvirorunent is two kilonet.rs away  and a  delay  is implicit  in the
  «roundwat.r's having to  traverse  that two kilom.ter distance.


          suggested  ..rli.r  in this  chapter,  ther.  is   no  underlying
         in  .lthtr Granlte  A o
         i. .xpected t

 -tur.ted with  water.   Under natural  conditions,  ther. would  be no
 h..d differential b.tw..n th. wat.r  at  the  r.pos-ory d.pth and  in  the
 overlying .qulftr,  .nd chufl( no  VBftlcal ^^^  o£ ^^  ^  ^

 rock.   However,  a vertical gradient may  develop  a.  . result of water
 
-------
  gradient is  dependent upon the^ temperature rise due to the HLU «nd on
  the  height of the water column that is .affected.   The  buoyancy effect
  is a factor not only  for  normal.leakage  along existing flow pathways
  but  also for n.w pathways  that might  develop or be introduced during
  the  life of  the  repository.


      The release mechanisms included in  the  present  analyses arc the
 same  for both granite  sites,  but differ  in  come  aspects  from those
 pr.viously considered for salt.   For granite,  the  release mechanisms
 examined included:   routine  releases,  faulting, volcanoes,  drilling
 without  hitting  a  canister,  and drilling  and  hitting  a  canister.
 Routine  releases,  which  involve  groundwater  migration through  bulk
 rock and engineered openings,  such as shafts,  are analyzed using  rock
 properties  cited in  the  literature  and   assumptions  about  material
 degradation which are discussed in the  Population Risk  Report.   In the
 case  of  faulting, the probability of occurrence has been changed  from
.earlier values and is based on correlation between  fault movement  and
 selsmicity in  a  recently  jablished  report   of the  U.S.   Geological
 Survey.


      Volcanoes   have   been   included  in   the   analyses  in  order  to
 represent what  may be, the limiting low probability but relatively high
 eoiu.quence event.   Since suitable granite sites may be found in many
 pares  of the country where  the probability of volcanism is vary low,
both Granite  A  and Granite  B sites are  expected to have the same very
low probability of volcanisra.   The parameters characterizing volcanoes
are identical to  those used in the original generic analyses  and  are
diacussed in  the  references.

-------
                       „,



     "" """^ " 1"                        1B  Mlt. due  «.  th.
            HO..V.,  th.r. „.


                 u.t.r
                                                        .re

                       fer
            th.t after th.t

             o£  hum.n
                                                     ^it    ^  ^





 P-rt .f . c.nut.r  on


         .u...  th«  r.t.
      u  gr.ntt..  r.th.r th.n
only P.rtuny  ...u* by ltthelt«te



           ..... "t"11' ""'""«- - «"-...ling.   Th.
           th.  fact
           throughout
            ef

         of
                                          ac

-------
                tn th.




and the original gcntric r.poaitory.

-------
                      CHAPTER 5  •' REPOSITORIES IN BASALT
                                 ^

   5.1   Introduction


        Basalt  deposits   in the Pacific  Northwest  have   been  under
   investigation for a number of years as potential host  formations for a
   nuclear waste repository.  Basalt is a dense, dark, fine-grained  rock
   formed by the solidification  of volcanic  lava.   The basalt  deposits  in
   tho  Northwest are  flood basalts.   They were extruded over extremely
   large  areas  and  formed  a layered structure of individual flows tens  to
  hundreds  of  feet thick,  separated by  relatively minor  sedimentary
  deposits  and  fractured  or highly porous  zones at the  tops  and bottoms
  of che basalt flows.  The  dense interiors of the basalt  flows  are  the
  potential repository  host rocks under consideration.   Similar  to  the
  case  of  granite,  basalt deposits are permeated  by fractures,  but at
  the depths  being considered  for  a * repository,   these fractures  are
  expected to be quite  tightly  closed,  thereby restricting  the volume
  and the velocities of any groundwater movement.   Nevertheless, there
  is  expected   to   be   some  groundwater  migration  through  a  basalt
 repository and it  is possible  this  might  be  accelerated by  repository
 induced  effects   on  the  host   rock.   Unlike  granite,  the  layered
 structure of  the  basalt deposits provides for horizontal groundwater
 movement  through  relatively  permeable  zones  between  flows.    In
 addition,   the  fracturing   in  a basalt  deposit  is  expected  to  be
 somewhat  greater   than  that  in  a   well-chosen   repository   site   in
 granite.   This  does  not mean  that such  fracturing  would  lead  to
unacceptable  repository  performance,  but  only  that  it  must  be  an
 important  consideration  in choosing  a site  and  in estimating  the
performance of  a repository  At that site.

-------
      Tha  Department  of  Energy  has  been  actively  investigating the

 possibility  of  siting   a   repository   in  basalt  at  the  Hanford

 Reservation in  southeastern Washington State.   Tha Agency  has paid

 particular  attention to tha  possibility  of  a  repository in basalt both
                                      *
 during the  development of Che Draft High Level Waste Standard, as well

 as  in  conjunction  with  the  developnent of  the  Final  Standard.  The

 relatively  advanced stags of  the  Department  of Energy investigations

 at  Hanford have provided considerable data en the characteristics of

 potential cites and repository host flows.   However,  much of the work
                 *S
 carried out at  Hanford has been  the subject of severe criticism by the

 Nuclear Regulatory  Commission and  others, and therefore the Agency has

 devoted special attention to  incorporating in  its  analyses  input not

 only  from the  Department of  Energy and its  contractors,  but also from

 technical professionals from other organizations.  Baaed on such data,

 the Agency  believes that  it  is possible  to define conceptual models of
 "»   "  •
 a  basalt  repository  that  should be adequate  to  make  rough  first

 approximations  of  the potential  performance  that  might be  expected

 from  such repositories and to  identify some of the parameters that are

 most  critical  in determining that  performance.
                                                           Section  5.2

summarizes  the important input parameters that have been  used in the

Agency's  risk  analyses  for basalt.   These  parameters  are  presented

both in  the form of a generic basalt cross -section,  which defines the

geometry  And general hydrologic structure for the modeling effort,  as

veil  as  an  accompanying  table  which  includes   the  detailed  input

parameters  that have been used  in the analyses.  Following these there

la  a  sequence  of tables characterizing  the release  mechanisms  that

-------
have  b.en included in th. riik .n*ly,ea  {ot  th.  site.   Slnce ^ of
those data  can beat  be presented  In  Che  fern,  of such  t.ble£ and
figures,  cher. i.  uinimm  textual discussion  of  additional details.
Hov,v,r,  a.  noted  previously,  there  is a more extend discussion of
the actual mathematical models used for th. analyses in th. Population
Ri.k  Report>   as   well  as   in   the   mod.l   documentation  packages
themselves.

-------
5.2  Innut Parameters
     Tho  input parameters  for  the risk  analyses  carried out  by the
Agency  fall  into  several  categories.   These  are listed  below along
with  a   reference   to  the  section ' of  this  and  the  accompanying
documentation in which wore elaborate discussion may be found.
     o     51 U P«™*«t«rs.  These  are  discussed in thi»  section for
           the current, «« veil  »* the original  generic basalt sites.
      o     Repository   Parameters.    These   are  the  same   for  all
           lithologies and are presented in  Section 2.7, with a  summary
           table given as Table 2.2.
      o    u.-t«. P*ck*«e  Poster..   The.e hav.  been  discussed  in
           Section  2.8  and  summarized  in  Table  2.3.   Radionuclide
           inventory  is  the  same  for  all lithologies  and leach rate is
           a  generic  parameter that  is given  the sane  value  (and the
           same  range  of  values  for  sensitivity analysis)   for  all
           lithologies.   Canister  lifetime  has  also  been varied for
           sensitivity analysis.   The solubility  limits for different
           radionuclides do depend on the site and are listed for all
           lithologies in Table 2.3.
       o   Pflflflj-r Maehanisma.   The  input  parameters  to  characterize
            specific  release  mechanisms  for the  basalt repository are
            discussed in this section.
       o    *nn rental and  BMUh  EffftCti Famine Wi.   These are the
            •aae   for  all  lithologies  and   are  discussed   in  the
            accompanying report on Environmental Pathways Analysis.
  Thus  the focus  of  this  section will be  on the  parameters  used  to
  characterize  the site  and  the most  important release mechanisms.

-------
      Figures  5.1  and  5.2 show the  geologic  cross-sections  that have


b»en  ue«d to  defin*  th« simplified modal*  in  the Agency's analysis.



Associated  with these  figures  are Tables 5.1  and 5.2,  which provide



the most  Important geometric  end  hydrologic parameters that affect the
                                     t

results.  The  figures  and tables  of  parameters  are based on Appendix A



to  this report, which is a  detailed  discussion  of how site-specific



data  from the  Hanford Reservation have been used to construct generic



cross-sections for a  basalt  repository.   The  cross-section  shown in



Figure  5.3  is  based on the earlier  analyses carried out in connection



with  the Draft  High Level Waste Standard.   It  too is  discussed in



Appendix  A, but  for more  elaboration one  should consult the Population


Risk  Report and  the Task  D Report.






      A  comparison  of Figures 5.1  and  5.2 indicates  that the original



model used  In  connection with the development with the draft standard



assumed a repository  at  a shallower depth than  is  now the reference



plan  of  the  Department  of  Energy.    In  addition,  the  earlier  model



grouped together  a number  of basalt  flows  into  a single unit  for



modeling  purposes, with  the  result  that  the upper  and  lower  aquifer



vere  further  removed  fron the repository.  More  recent  data  suggest
                                                                  s


that  it is important  to  consider as possible  migration  pathways  the



upper and lower boundaries  of the  actual individual basalt  flow in



which th« repository is located.  As shown in Figure 5.1 this puts the



repository much  closer to the model aquifers.   Other modifications to



the site parameters incorporated  in  the more recent modeling Include a



lowar natural  vertical  gradient from  the  repository  to the  upper



aquifer and some modifications  in other hydrologic parameters,  as can



be seen in  the tables.  In addition, the temperature profile used for

-------
 Che  earlier calculations wltfi  the generic basalt model  .re  based on



 higher heat loading,  than are  now being conalder.d  by the Department



 of Energy.  A.  . result of thi.  change in the Department', apparent



 plan., the  thermal  buoyancy  effects  in the  earlier  ri.k  analyses



 reported,  for  example in the Population Ri.k Report, are much greater



 than those presently believed to be  the  case.








      Essentially  the same  release  mechanisms  have b«tn  maintained



 throughout the Agency's  considerations  of a basalt  repository.   The



 so-called  routin. release mechanl.m described  in  Table  5.3 refers to



 th«  fact  that there  Is  expected to  be  some  amount of groundwater



 movement through a basalt repository due to the presence of fractures.



 This flow  may  be accelerated by two  factor.:  an increa.e in hydraulic



 conductivity after the thermal peak, due  to  slippage  and movement of



 the  basalt  along  fracture  planes,  and  a  reduction  in  the  water



 viscosity   due  to  higher   temperature.   Table  5.4  summarizes  the



 parameters used to characterize the faulting release  mechani.m.   The



 probability  of occurrence of faulting is somewhat  higher than used in



 che  earlier analyses  and i. based  on  a  more  recent faulting  model



developed  by  the United States  Geological  Survey.  The  effect  of



 faulting Is to  create  (either  by  the development of  a  new  fault  or



 reactivation of  the  old fault)   a  preferential  groundwater  pathway



which  would have higher potential for  flow  than the intact  rock and



would  connect  the repository with the  adjacent aquifers,  Table 5.5



summarize,  the   parameters  associated  with  the  volcanism  release



»*chani«m  which essentially  provides a limbing low  probability high



consequence  event  and suggests  the general shape of  the  CCDF  near the



lower  extremes  of  probability.  The parameters used here  are identical

-------
with thos*  used In the original generic analyse* and are discussed in

the references.



     The  possibility   of   future  human   intrusion  leading  to  some
                                      *
disruption  of  the  repository has been an important concern  for the

Agency   in   developing  the   High   Level  Waste   Standard.    While

institutional control  may be  successfully  in preventing such intrusion

for a.  limited period of time and long-term markers and other means of

communication  may preserve knowledge of the sites still further, the

Agency  has  adopted for  its  risk analysis  the  underlying assumption

that  such measures  are of limited effectiveness  and hence cannot be

relied on completely to eliminate  the possibility  of tuture intrusion.

Therefore,   in  the   origin*!   generic  analyses   described  in  the

Population Risk Report and  in  the  references  cited therein, certain

reference  future  drilling  rates  for  each  lithology  were  adopted.

These  same values  have been used in  the present  analyses.   Some  of the

future  drillholes  might  even  intersect  actual  waste  canisters,

bringing a portion of their contents to the surface.  Other drillholes

will  not   intersect   the   canisters   themselves,  but   may   bring

contaminated groundwater to  the surface.  It is also possible that in

abandoning such future drillholes,  a more permeable pathway might be

established  between   the   repository   and overlying   or  underlying

aquifers.  The calculations  the Agency has carried  out suggest  that

 this is  of negligible importance  compared with the direct release to

 tha surface.   The drilling  parameters  included  in the risk 'analyses

 are summarized in Tables 5.6 and 5.7.

-------
                    CHAPTER 6 -  REPOSITORY IN TUFF
                             <







6.1  Introduction








     Welded  tuff  has  recently  received  increased  attention  as  a



potential host  rock  for a high level waste  repository.   It is unique



among the geologic media considered for analysis by the Agency because




the  deposits of  tuff  apparently  most appropriate  for  a  repository




occur in southern Nevada, where the water table is »ome distance below



potential  repository   horizons.     .e  welded  tuff«   that  are  the



candidate host  rocks  for this analysis  consist of airborne  volcanic



d«bri,« that fused into a mass with high porosity and low permeability.



They  appear  to   have   the  necessary  engineering  properties  for



repository construction.  Because the tuff is composed of fragments of




porous volcanic rock,  the residence time of water moving through it is



relatively  long,   and  the  mineral  assemblages  can  be  expected  to



provide favorable retardation.  Tuff shares with  granite  and  basalt a



relatively low  occurrence of fuels  ^r valuable minerals that  might be



exploited by  future  drilling to  any considerable  depth.   Similarly,



the depth of the  water table is a  discouragement against the sinking



of wells or the development of underlying aquifers.








     Two  distinctive   and   important   features  emerge  from  Sandia




National Laboratory's  (SAND84-1492) and  the  Agency's  analyses of a



repository above  the water  table  in tuff.   First,  unlike any other



»«dium, upward  flow from the repository is not possible so long as the



rock remains unsaturated, and groundwater  could not accumulate in the




repository.  Second,  as long as precipitation at the  surface is  low

-------
enough  to maintain an unsaturated  condition, ws'er •*". a flowpath, such
                              <*.


as  a  faulc  zone,  will  preferentially  move  into the tuff by capillary



attraction,  rather than downward along the  fault trace (SAND84-1A92).



This helps  to  maintain long travel times.



                                    •


     The  Department  of Energy  is currently  investigating the  area



including Yucca Mountain  in southern Nevada as  a  possible candidate



site.   Other tuff sites nay  be  found, but Che  relative  abundance of



hydrogeologic  data   for  this location,  coupled  with  the very  low



preci- 
-------
* '•  ?^rut Parameters




     The  input parameters  for the  risk  analyses carried  out  by the


Agency  fall  into  several  categories.   These are  listed below along


with a reference to  the  section of this or  the companion  documentation
                                    •

in which more  elaborate  discussion may be found.




     o    Site Parameters.   These  are discussed  in this section for


          tuff.


     o    Repository Parameters.   These are the  same for  ell


          lithologies  and have been presented in Section 2.7,  with a


          summary  table  given  as Table 2.2.


     o    Waste Package  Parameters.  These have been discussed in


          Section  2.8  and  summarized  in  Table 2.3.   Radionuclide


          inventory  is the  same for all lithologies, and  leach rate is


          a .generic  parameter  that  Is  given  the same values  and the


          same range  of values  (for sensitivity  analyses) for  all


          lithologies. Canister lifetime has  the same  baseline values


          for  all  lithologies, but for certain sensitivity analyses


          different   variations   on  canister    lifetime   have   been


          considered,  based  on  the  Agency's   estimation  of  the


          harshness  of the  environment  in which  the  canister would be


          placed.  The solubility  limits for different  radionuclides


          do depend  on the  site, but are listed for all lithologies in


          Table 2.3.


     o    Raleafe  Mechanisms.   The  input parameters to characterize


          specific release  mechanisms for the fmgejs^pPHslBHssV* are


          discussed  in this  section.

-------
      Section 6.2 summarizes the importer,  input  parameters that have
                              ^


been  used in the Agency's  risk  analyses  for tuff, including a  figure



Chat  schematically  illustrates  the  cross-section of  the  site  and a



tabla that  summarizes the input parameters  used in the risk analyses.



Following this,  there is  a sequence  of tables summarizing  the release



mechanism*  that  have  been included in the site's risk analyses.  Since



most  of these data  can best be presented  in the form of  tables  and



figures,  there  is minimal discussion of additional details.  However,



the  actual   mathematical  models  used  for  analyses  are  discussed  at



length  in   the  Population  Risk Report  and in  the  Sandia  report



previously referenced.

-------
     o    Envlronmentpl and Heplfh Effects  Parameters.   These  are the



          • arao for all  litholosleo and are discussed  in the  companion



          report on Environmental Pathways  Analysis.



Thus  the  focus  of this  section will be  on the  parameters used  to
                                     «


characterize the tuff site as well as those used to represent the most



important release mechanisms.





                                                      A^                J
     Figure  6.1  shows the geologic cross-section that^PW  been used



to  define the simplified tuff model used  in the  \gency's  analyses.



Associated with  this figure  is  Tables  6.1,  which presents  geometric



and  hydrologic  parameters  that affect  the  results.  This figure and



table  are based on  Sandia  National  Laboratory's report on  the Yucca



Mountain  sice (SAND84-1492),  which  includes detailed  discussions  of



parameter values and.derivations and representative cross-sections.







      Important  observations  contrasting  and   comparing  the  tuff



repository   site  with  the   other  lithologies  may be   summarized  as



follows.   Tuff  contains  a  repository  at an  approximate  depth  of 400



meters, which is 100 meters  above  the regional groundwater table.  The



repository horizon and overlying  formations  are unsaturated.   Unlike



other lithologiea  that have been studied,  where the  host  rocks are



essentially  saturated  (though generally  highly  impermeable), flow



 through tuff is  limited by the availability of water more than by the



 impermeability  of  the rock.   In order to analyze this source-limited



 condition,  the  Agency used  an  effective  hydraulic conductivity that



 had  been back-calculated  by  Sandia  Laboratories (SAND84-1492)  from



 data on groundwater flux.  When used as the  conductivity in  darcy flow



 calculations, Sandia1s value yields a  flow  rate that approximates  the



 actual measured flow.

-------
     Another  important  difference .between  the  hydrology of  the  tuff

eito and the other iite« is the pronounced downward hydraulic gradient

found  in  tuff.   Because o£ the unsaturated  condition,  this gradient.

which  is simply  the  direct force  of gravity, ii essentially unaffected
                                   • .    '
by  temperature  changes  in the  rock due  to the  heat  output  of the

waste*.  Thu*.  the vertical  gradient is downward and constant and for

«ny  groundwater release mechanism,  and the only  flowpath  is towards

the water  table.




     The horizontal  gradient  in the  underlying aquifer  is quite  small

around the tuff site.   Although  the conductivity of the rock at the

level  of the water table is relatively  high, compared to  deep aquifers

in  other  lithologies, the very low gradient delays  the migration to

the  accessible  environment of material  in  the aquifer.




     The  release  mechanisms  included  in  the  analysis  of  tuff  are:

routine  releases,  faulting,  volcanoes, drilling and  not  hitting  a

canister,  and drilling and hitting a canister.  Tables  6.2 through 6.6

sunmariza   the   important  parameters  associated  with   these  release

mechanismsv  For faulting, the  probability,  based on U.S. Geological

Survey estimates,  is higher  than for  areas examined  for  moat  other

lithologies,  but the hydraulic  conductivity of the flowpath produced

by  A  fault is  co isiderably  smaller.   This  is  because  water movement

down the  fault trace would be effectively limited by the availability

of  the water,  much of which  would be absorbed by the rock adjacent to

the  fault.  Thus, the effective  hydraulic conductivity of a fault is

quite  low.

-------
      Volcanoes  were  analyzed  in  a-  merrier   analogous  to  the  o:her



 lithologiet,  adjusting  only  the  probability  of  a  volcanic  vent



 intersecting a repository to conform to volcanic probability estimates



 for  the  Nevada.  Teat Site  area  determined  by   the  U.S.C.S.  (U.S.



 Geological Survey, open file report 80-357).   The consequences of such



 an  event are  determined  by the  geometry of  the repoaieory  and  the
           \


 volcanic vent, and are the tame for all lithologiea analyzed.







      The potential for human Intrusion was Judged as likely in tuff as



 In granite.   While the Agency has addressed thia concern In developing



 the  High-Level Waste Standard,   it  recognize*  that  it  is  probably



 impossible to  perpetually maintain  control over access to  a  site,  or



 to rely on markers  or other  means  to  prevent drilling And exploration



 in  the  future.  Therefore,   in  accordance with  the generic  analysis



 described 'in  the  Population Risk Report,  and  references  cited there,



'future drilling rates for various litologiea  were  adopted.  Since tuff



 was not one of  the  lithologies originally  considered,  a drilling rate



 is not explicitly discussed  in  that Report.   The  Agency  subsequently'



 estimated, however,   that the rate  determined for granite would  be  an



 appropriate  surrogate for a  rate  in tuff.







      Routine releases from the  repository  are limited by the amount of



 water available.  Thermal effects  on  vertical gradient and hydraulic



 properties of  the rock are  have  been ignored  in  the analysis because



 Chay  ara negligible,  compared  to   the  restraints  imposed  by  the



 relative  lack of   water.   As  modeled,  there  is a general   flow  of



 migrating radionuclidea  toward  the  underlying aquifer,  and  thence



 horizontally to the  accessible  environment.

-------
SCENARIOS
                                                       ACCESSIBLE
                                                      ENVIRONMENT
                     DIRECT
                     PATHWAYS
                     TO SURFACE
WASTE
FORM


CANISTER



REPOSITORY
LAND
SURFACE
                                                                     HUMAN
                                                                         %
                                                                     POPULATION

PATHWAYS
TO AQUIFER
(GROUNDWATER)


AO 1 IIP F R



SURFACE
WATER
1
                         FIGURE 2.1
                    Structure of Risk Analyses
                  for High Level Waste Repository

-------
    10'2
<
CD
O
tr
10
      '6
    10'8
                             (C,P)
Probability is P  of

consequence at  least

as  large  as  C
                 I
                    I
                   0.1            1

                       CONSEQUENCE,  C
                                         10
                                            100
                          FIGURE 2.2
            Example of Complementary Cumulative Distribution
            Function  (CCDF)

-------
     Surface Deposits
     Upper Aquifer
     Upper Confining Beds
     Hose  Rock
             .  .     Repository
     Lower Confining Beds  _ ''r
     Lower Aquiter
            T_-
     Basement Rock£
                  FIGURE 2.3
General Cross-Sectional Structure Used in Risk Analysis

-------
                      Assembly Structure

                 and Components
                          10
             Decay Time  from Discharge
                    (vears)
                   FIGURE 2.4
10-
                                                              10
10
Total Radioactivity in Reference  Repository
(from 100,000 MTHM)

-------
    Upper
    Aquifer
                                                       DEPTH  (m)
                                                              0
                                                             305
r^rir_-.Upper-
    Solt
Repository
550
595
 — — Contirung
                                                     •-•."••^•••2000
                        FIGURE 3.1
            Cross-Sectional Stucture  Assumed for  Salt A

-------
                                                                  DEPTH  (m)

                                                                 	   0
—  Surfoce  Deposits
Upper  .
                                 	   	   314












rt_n_Confining Beds :j£3£H3^Lr^r£r!H?irtj£?^riJ'ir^"iri^^












~^^^^^^^^^^^^^^^^^^^^^^^^^^^^-^-^~-^~-^   945

     Solt 6                ——^—.  -»— Repository

	                              1021

















ririiC onfi m ng _B e_d£ f^—HHrlHHHHHHHKriHHHHHHHH:^

















           •"-•;-^:^-:--::.:c^:'-\;"-^:-'-',v^S :^'^V^v'>/;-v-7:v;r.'.-?;•;•;:.;.--;:;;:-^ "~'.£-V;vv'.
           Aquifer •;^>=^^^:^v..>;,;.,^.^^>v^^;^:^vx.^vV'/••.••"•/^'•>^::^•^':^vv^;'H•:.
           .y..:.^i:. ^^^•.••^>.;-^^--;-y;..;:.:;.:..:;^.-.v.:vr.••:.-.:•:•-.::••••..•V:-.-:/~x ,890
                               FIGURE  3.2
              Cross-Sectional Structure Assumed for Salt  B

-------
                                                        DEPTH
                                                         (m)
      Surface  Deposits
      Upper   \
      Aquifer!
      Salt
                                 •^•^^^^-•-ij^r: • 350
                                                            330
                                                      •—    410
Repository
^KKP: Confining Beds  gr-r-r->Jb-z->J
                                   !V"^V-^T-VV?>MriMir^  560
                    510

                    590
                  FIGURE  3.3
     Cross-Sectional Structure in Original
     Salt Analysis

-------
                                                     DEPTH
                                                      (m)
              Surface Deposits
             FIGURE 4.1.
Cross-Sectional Structure Assumed for Granite A

-------
                           Surface Deposits
Aquifer
                             FIGURE 4.2
                Cross-Sectional Structure Assumed for Granite B

-------
Aquifer
                                                                 460
                                FIGURE 4.3
              Cross-Sectional Structure Assumed in Original Granite Analysis

-------
                                                     Depth
                                                    (meters)
Surface                 	                  	
       >	 .	•  _ .  	
                                                     1S5
                  FIGURE 5.1
 Cross-Sectional Structure Assumed for Basalt Repository

-------
                                                      Depth
                                                      (meters)

                                                         0
Surface _
Deposits.
                      FIGURE 5.2


     Cross-Sectional Structure Assumed in Original Basalt Analysis

-------
                                                      Depth  (m)
                                                           0
    Welded  Tuff »ZZ"
                               !f™!^...r:.v. jlep o s ijt o r v£"S:"''"'
                                 •••••*•****  •••••••***  •••••••****  „•*•••••*"
     o a

               *
             <*   «
0  «  o



  0 4 °


     0
  »*•.-  •?,
                       ••
  "   °  *   °  0   °6
       *   «•    0-

  «' 0  *      •  .
        °  4  4     •
 0 .   Aquifer    «   »  p


 »  °   Welded  and Non-U'elded Tuff
                        M

                      <*      *


                          0
                   0
                4     *
                    0     4
                             300


                             400


                             500
                                                         1500
                      FIGURE  6.1
Cross-Sectional Structure Assumed  for  Repository in Tuff

-------
                                   TABLE  2.1

                  Site Parameters Considered  in Risk Analysis

 Distance between repository and "upper aquifer"  and/or
 "lower
      aquifer"
 Thickness of aquifers
 Hydraulic conductivity of  aquifers
 Porosity of aquifers
 Horizontal gradient in aquifers
 Retardation values for individual nuclides  in aquifer
 Average  hydraulic gradient  along a hypothetical vertical
      pathway between the upper aquifer and  lower aquifer
 Hydraulic conductivity of  the rock between  the upper aquifer
     and  the .lower aquifer
 Porosity  of the rock between the upper aquifer and the lower
     aquifer
H(  :zontal distance along aquifer to point regarded as
     release point to accessible environment

-------
                                   Table 2.2

               Repository Parameters Considered In Risk Analysis


 PARAMETER                                  VALUE              SOURCE

 Dimensions of repository                                      EPA 520/3-60-006
 Length                                     4,000 meters
 Width                                      2,000 meters
 Height                                     5 meters

Total mined-out volume                     l.OxlO7 m3         EPA 520/3-80-006

Average porosity of backfilled             0.2                EPA 520/3-80-006
repository

Time to maximum backfill compaction        200 yr             EPA 520/3-80-006
due to creep (salt only)

-------
                             Table 2.2 (continued)




               Repository Parameters Considered In Risk Analysis
 PARAMETER




 Number of canisters of HLW




 Number of waste drifts




 Canisters per drift




Length of waste drift




Canister spacing
.VALUE




 35,000




 350




 100




 500 meters




 5 meters
SOURCE




EPA 520/3-80-006




EPA 520/3-80-006




EPA 520/3-80-006




EPA 520/3-80-006




EPA 520/3-80-006

-------
                                    Table  2.3

              Waste Package Parameters Considered In Risk Analysis


     PARAMETER                                VALUE               SOURCE

 Canister  lifetime                           300  yr*           Agency Decision

 Fractional  leach rate (fraction per        1.0x10"**          EPA 520/3-80-006
 year)
*These values have been varied in the sensitivity analysis.

-------
                Table  2.3  (continued)
Waste Package Parameters Considered In Riak Analysis
                                      Solubility Limits  for Each
                                    Radionuclide by Mediua (Ci/ml)
Isotope
Anr-241
Am- 24 3
C-l'.
Cs-137
1-129
Np-237
Pu-238
Pu-240
Pu-242
Sr-90
Tc-99
Sn-126
Zr-93
Half-Life
458
7.650
5,730
30
16xl07
2.1xl07
89
6,260
380,000
29
210,000
100,000
950.000
Initial
Inventory
(per canister)
4.9xl05
4.9xl03
S.OxlO"1
2.5xlO"5
l.lxlO"1
9.4xlO~l
9.4xl02
1.4xl03
4.86
1.7xl05
4.0
1.6
5.4
Salt
3.2X10"1
1.9xlO"2
. none
none
none
7.1xlO"7
1.7xlO"2
2.2xlO"4
3.9xlO~6
none
1.7xlO~5
2.8xlO~5
4.0xlO~7
Granite
3.2x210"3
1.9xlO~4
none
none
none
7.0xlO~5
1.7xlO"2
2.2xlO~A
3.9xlO~6
none
none
2.8xlO~6
4.0xlO~9
Basalt
3.2xlO"3
1.9x10"*
none
none
• none
7.1xlO~8
1.7xlO"4
2.2xlO~6
3.9xlO~8
none
1.7xlO~5
2.8xlO"6
4.0xlO~7
Tuff
3.2xlO~3
2.0xlO"4
none
none
none
7.1xlO"5
1.7xlO"2
2,2xIO"A
3.9xlO"6
none
none
2,8xlO"6
4. 0x10" 7

-------
                                  Table 2.4

            Initial  Radionuclide  Inventory  in Reference Repository

 Radionuclide               Curies                 Half-Life

 Acr-241                     1.7 x  108                     458
 Am-243                     1.7 x  106                   7.650
 C- 14                      2.8 x  104                   5,730
 Cs-137                     8.6 x  109          --             30
 1-129                      3.8 x  103               1.6 x 107
 Np-237                     3.3 x  104                      89
 Pu-238                     3.3 x  107                  24.400
 Pu-240                    4.9 x  107                   6,260
 Pu-242                     1.7 x  105               3.8 x 105
 Sr- 90                    6.0 x 109                      29
Tc- 99                    1.4 x 106               2.1 x 105
Sn-126                    5.6 x 10*               1.0 x 105
Zr- 93                    1.9 x 105               9.5 x 105

-------
                                    Table 2.5

                  Structure of Environmental Pathways Analysis
 Release Mode


 Release to River
 Release to Ocean
 Release  Directly
  to  Land Surface
Release to Air Over Land
 Resuspended Material
Release to Air Over Ocean
                                            Pathways Analyzed
 Drinking Water Ingeition
 Freshwater Pish Ingestion
 Food Crops Ingestion
 Milk Ingestion
 Beef Ingestion
 Inhalation of Resuspended Material
 External Dose-Ground Contamination
 External Dose-Air Submersion

 Ocean Fish Ingestion
 Ocean Shellfish Ingestion

 Food Crops Ingestion
 Milk Ingestion
 Beef Ingestion
 Inhalation of Resuspended Material
 External Dose-Ground Contamination
 External Dose-Air  Submersion

 Food Crops  Ingestion
 Milk Ingestion
 Beef Ingestion
 Inhalation  of  Dispersed and
External Dose-Ground Contamination
External Dose-Air Submersion

Ocean Fish Ingestion
Ocean Shellfish Ingestion

-------
                                               Table 3.1
                              Site Parameters Considered In Risk Analysis
                                        SALT A: Palo Duro Basin
Parameter
Distance from repository to aquifer
(for Salt A, lower aquifer)

Thickness of aquifer

Hydraulic conductivity of aquifer

Porosity of aquifer

Horizontal gradient in aquifer

Retardation values for individual
nuclides
Natural hydraulic gradient between
repository and aquifer

Hydraulic conductivity of the rock
between the repository and the
aquifer, after thermal effects

Porosity of rock between the
repository and the aquifer

Horizontal distance along the aquifer
to point regarded as release point to
accessible environment
                                                   Value
NA
2,000 meters
                                                                           Source
1105 meters
300 m
1.6 m/yr
0.05
0.005
Pu - 200
Am - 1,000
Np - 50
Cs " 10
I - 1
Tc • 5
Sn - 100
Zr - 1,000
Sr " 10
C - 1
0.26
0.0 m/yr
Appendix A
Appendix A
Appendix A
Appendix A
Appendix A
Americium value: Appendix B
All others: NRC/NAS WISP
Report, 1983

/




Appendix A
Appendix A
EPA-520/3-80-006


Agency Decision

-------
                                               Table 3.2
                              Site Parameters Considered In Risk Analysis
                                         SALT B: Paradox Basin
Parameter
Distance from repoaitory to aquifer
(for Salt B, upper aquifer)

Thickness of aquifer

Hydraulic conductivity of aquifer

Porosity of aquifer

Horizontal gradient in aquifer

Retardation vt!ucs for individual
nuclidea
Natural hydraulic gradient between
repository and'aquifer

Hydraulic conductivity of the rock
between the repoaitory and the
aquifer, after thermal effects

Porosity of rock between the
repository and the aquifer

Horizontal distance along the aquifer
to point regarded as release point to
accessible environment
Value
666 meters
18 m
7.6 m/yr
0.2
0.02
Pu - 200
Am • 1,000
Np - 50
Cs • 10
I " 1
Tc - 5
Sn • 100
Zr - 1,000
Sr • 10
C - 1
0.0
0.0 m/yr
Source
Appendix A
Appendix A
Appendix A
Appendix A
Appendix A
Americium value: Appendix B
All others: KRC/NAS WISP
Report, 1983







Appendix A
Appendix A
NA
2,000 meters
EPA-520/3-80-006


Agency Decision

-------
                                               Table 3.3
                              Site Parameters Considered In Risk Analysis
                                         Original Generic Salt
Parameter
Distance from repository to aquifer

Thickness of aquifer

Hydraulic conductivity of aquifer

Porosity of aquifer

Horizontal gradient in aquifer

Retardation values for individual
nuclides
Natural hydraulic gradient between
repository and aquifer

Hydraulic conductivity of the rock
between the repository and the
aquifer, after thermal effects

Porosity of rock between the
repository and the aquifer

Horizontal diatance along the aquifer
to point regarded aa release point to
accessible en 'ironmeni
Value
100 meters
30 m
31.5 n/yr
0.15
0.0
Pu • 100
Am - 100
Np - 100
Cs - 1
I - 1
Tc - 1
Sn • 10
Zr " 100
Sr - 1
C • 1
Source
EPA 520/3-80-006
EPA 520/3-80-006
EPA 520/3-80-006
EPA 520/3-80-006
EPA 5/0/3-80-006
E A 520/3-80-006









0.1
0.0 n/yr
0.01
1,600 meters
EPA 520/3-80-006


EPA 520/3-80-006



EPA 520/3-80-006


EPA 520/3-80-006

-------
                                               Table  3.4

                Release Mechanism Parameters Considered In Salt Repository Risk Analysis

                                                Faulting

                                       Values  by  Repository Site
Parameter
Fraction of the repository
intersected by the release
mechanism

Hydraulic conductivity along
pathway created by release
mechanism

Porosity in flow path created by
release mechanism

Cross-sectional area of flow
EPA 520/3-80-006
path

Annual probability or frequency
Salt A             Salt B       Source

.003               .003         EPA 520/3-80-006



31.5 m/yr          31.5 m/yr    EPA 520/3-80-006



0.1                0.1        .EPA 520/3-80-006


4.0xl03 sq m                    4.0xl03 aq m



1.0xlO~6           l.OxlO"5     USGS Open File Report 82-972

-------
                                               Table 3.5

                Release Mechanism Parameters Considered In Salt  Repository Risk Analysis

                                             Breccia Pipes

                                       Values by Repository Site
Parameter
Hydraulic conductivity in  flow
EPA 520/3-80-006
path created by release
mechanism

Porosity in flow path created by
release mechanism

Cross-sectional area of flow
EPA 520/3-80-006
path

Probability or frequency
(after 1,000 years)
Salt A

3.2xl03 m/yr




0.2


3.0x10* sq m



      -8
                                                               Salt B
0.2
1.0x10
      ,-8
Source

3.2xl03 m/yr




EPA 520/3-80-006


3.0x10* sq m



EPA 520/3-80-006

-------
                                               Table 3.6

                Release  Mechanism Parameters  Considered  In  Salt  Repository  Risk  Analysis
                                  l«
                                               Volcanoes

                                       Values by Repository Site

Parameter                                   Salt A             Salt B       Source

Fraction of the repository                  4.0xlO~4           4.0xlO~4     EPA 520/3-80-006
intersected by the release
mechanism

Annual probability or frequency             l.OxlO""10          l.OxlO"10    EPA 520/3-80-006

-------
                                               Table 3.7

               Release  Mechanism  Parameters Considered  In Salt Repository Risk Analysis

                                  Drilling and Not Hitting a Canister

                                       Values by Repository Site

Parameter                                   Salt A             Salt B       Source

Volume of water in the repository           1.14 m3            1.14 m3      EPA 520/3-80-006
which can reach the surface.

Annual probability or frequency             2.0xlO~2           2.0xlO~2     EPA 520/3-80-006
(after control period)

-------
                                               Table 3.8

               Release Mechanism Parameters Considered In Salt Repository Risk Analysis

                                    Drilling and Hitting a Canister

                                       Values by Repository Site

Parameter                                   Salt A             Salt B       Source

Fraction of canister brought to             0.15               0.15         £p^ 520/3-80-006
surface.

Annual probability or frequency             2.5x10"'           2.0xlO~5     EPA 520/3-80-006
(after control period)

-------
                                  Table 4.1
                        Site Parameters Considered In Risk Analysis
                           GRANITE A: North Central
Parameter
Distance from repository to aquifer
Value

448 meters
                                                                            Source
Appendix A
Thickness of aquifer
10 Deters
Appendix A
Hydraulic conductivity of aquifer
5.7 m/yr
Appendix A
Porosity of aquifer

Horizontal gradient in aquifer

Retardation values for individual
nuclides
Report,
Natural hydraulic gradient between
the repository and the aquifer
0.016

0.005

Pu  • 200
An  - 3000
 Np
 Cs
 I
 Tc
 Sn
 Zr
 Sr
 C
0.0
                                                       • 100
                                                       - 1,000
                                                       • 1
                                                       - 5
                                                       " 1,000
                                                       " 5,000
                                                       • 200
                                                       • 1
Appendix A

Appendix A

Americiun value: Appendix B
All others: NRC/NAS WISP

1983
                          Appendix A
Hydraulic conductivity of the rock
between the repository and the
effects
Porosity of rock between the
repository and the aquifer
Horizontal distance along the
aquifer to point regarded aa release
environment
site)
3.2xlO~2 m/yr


.0001

0 meters
Append ix A
  aquifer, after thermal

Appendix A

Agency Decision
  point to accessible
  (aasunes surface water on

-------
Parameter
Distance from repository to aquifer
                                               Table 4.2
                               Site Parameters  Considered  In Risk Analysis
GRANITE B: North East

        Value

        370 meters
Thickness of aquifer
        80
             eters
                                  Appendix A
Hydraulic conductivity of aquifer
        315 m/yr
Appendix A
Porosity of aquifer

Horizontal gradient in aquifer

Retardation values for individual
nuclides
Report,
Natural hydraulic gradient between
the repository and the aquifer
        0.039

        0.01

        Pu  - 200
        An  • 3000
         Np
         Cs
         I
         Tc
         Sn
         Zr
         Sr
         C
        0.0
                                                       • 100
                                                       • 1.000
                                                       • 1
                                                       - 5
                                                       • 1,000
                                                       • 5,000
                                                       • 200
                                                       - 1
Appendix A

Appendix A

Americiuo value: Appendix B
All others: NRC/NAS WISP

1983
Appendix A
Hydraulic conductivity of the rock
between the repository and the
effects
Porosity of rock between the
repository and the aquifer
Horitontal distance along the
aquifer to point regarded as release
environment
        3.2xlO~2 m/yr


        .0001

        2,000 meters
Append ix A
  aquifer, after thermal

Appendix A

Agency Decision
  point to accessible

-------
Parameter
                 Table 4.3
Site Parameters Considered In Riak Analysis

    I'     Original Generic Granite

                   Value
Distance from repository to aquifer
                   230 meters
Source

EPA 520/3-80-006
Thickness of aquifer
                   30 meters
EPA 520/3-80-006
Hydraulic conductivity of aquifer
                   31.5 m/yr
EPA 520/3-80-006
Porosity of aquifer .

Horizontal gradient in aquifer

Retardation values for individual
nuclides
Natural hydraulic gradient between
the repository and the aquifer
                   0.15

                   0.01

                   Pu  • 100
                   Am  - 100
                    Np  - 100
                    Cs  - 1
                    I   - 1
                    Tc  • 1
                    Sn  - 10
                    Zr
                    Sr
                    C

                   0.1
                                                         100
                                                         1
                                                         1
EPA 520/3-80-006

EPA 520/3-80-006

EPA 520/3-80-006
EPA.520/3-80-006
Hydraulic conductivity of the rock
between the repository and the
effects
Porosity of rock between the
repository and the aquifer
Horizontal distance along the
aquifer to point regarded as release
environment
                            m/yr


                   .0001

                   1,600 meters
EPA 520/3-80-006
  aquifer, after thermal

EPA 520/3-80-006

EPA 520/3-80-006
  point to accessible

-------
                                                Table 4.A

               Release Mechanism Parameters Considered In Granite Repository Risk Analysis

                                            Routine Releases


                           Values by  Repository  Site
Parameter
Source
                                   Granite A
               Granite B
Fraction of the repository
with which groundwater can
communicate

Hydraulic conductivity of flow
path created by release
mechanism
1.0
3.2x10
      ,-2
                   1.0
                   3.2x10
      ,-2
                  EPA 520/3-80-006
                                     Append ix A
Porosity in flow path created by
release mechanism
Cross-sectional are* of flow
path (vertical flow)
0.0001
0.0001
                                     Appendix A
S.OxlO6 m2         S.OxlO6 m2         depository Area
Probability or frequency
1.0
1.0
EPA 520/3-80-006

-------
                                                Table 4.5

               Release Mechanism Parameters Considered In Granite Repository Risk Analysis

                                                Faulting


                           Values by Repository Site
Parameter
Source
                                   Granite A
                 Granite B
Fraction of the repository
intersected by release
mechanism

Hydraulic conductivity along
pathway created by release
mechanism
0.003
0.003
EPA 520/3-80-006
3.2xl03 m/yr       3.2xl03m/yr      EPA. 520/3-80-006
Porosity in flow path               0.1
created by release mechanism
Cross-sectional area of flow        4.0xl03
path
                   0.1
                    .0x10
                         3  _2
                  EPA 520/3-80-006
                  EPA 520/3-80-006
Annual probability or frequency     l.OxlO"6           l.OxlO"6          VSGS Open File Report 82-972

-------
                                                Table  4.6

               Release Mechanism Parameters Considered In Granite  Repository  Risk Analysis

                                                Volcanoes


                           Values by Repository  Site


Parameter                               Granite A
                        Source
Fraction of the repository              4.0xlO~4          '4.0xlO~^          EPA 520/3-80-006
intersected by the release
mechanism <

Annual probability or frequency         l.OxlO"10          l.OxlO"10         EPA 520/3-80-006

-------
                                 Table A. 7

Release Mechanism Parameters Considered In Granite Repository  Risk Analysis

                    Drilling and Not Hitting a Canister

            Values by Repository Site


                         Granite A        (o P(-*j\~r^   _}         J
Parameter                               Granite A        (o P(-*j\~r^  _}         Jo »'
Volume of water in the                   2.0xl02  m3          2.0xl03 m3        EPA 520/3-80-006
repository which can reach
the surface.
Annual probability or frequency          2.5xlO~3           2.5xlO~3          EPA 520/3-80-006
(after control period)

-------
                                               Table A.8

               Release  Mechanism Parameters Considered In Granite Repository Risk Analysis

                                     Drilling and Hitting a Canister


                          Values by  repository site



Parameter                            Granite  A          Granite B
Fraction of canister brought to         0.15               0.15               EPA  520/3-80-006
surface


Annual probability or frequency         2.5xlO~6           2.5xlO~6          EPA  520/3-80-006
(after control period)

-------
                                            Table 5.1
                    Site Parameters Considered In Basalt Repository Risk Aoalyei*
Parameter
Distsnce  fro* repository to aquifer
                Will-  J
Thickness  of aquifer  .'•'       ''< !

Hydraulic  conductivity iaf aquifer
Porosity of squ{
Horisontal
letsrdstioa v«
nuclides
(•port,
 Jjf:P   -ft

'W""
Natural hydraulic cradle** betveen
the repository aad the aquifer

Hydraulic  conductivity of the book
between the  repAfUary ao4 the
aquifer, after f4jif>J
Poroaity of rock ''Wf^MB' the
repository e*4 fo* «<|«4f*r
HoriconCal dis
aqvifer to poU
f«Ut to Mqaaar
                         the
                         •a ret
                                                20 meters '^

                                                30 alters
                                                0.9093
                                                Mp
                                                Cs
                                                I
                                                Tc
                                                3n
                                                Zr
                                                8r
                                                     100
                                                     1,000
                                                     1,000
                                                     5,000
                                ' f-

                             3.2««r* B/yi



                             .0001


                             2,POQ Meters
                                                                    *«*•,  '''If?;
                                                                                                                     f'V.1
                                                          ,« A    •  ''* '
                                                                                    'W$
                                                                                    :Ki
                :|fefi  ;4
                                                                                        ';'- ^jtl.iK- /
                                                                                 iff
                                                                                 -":!i >r.  ji
f|;;
 *#.
I A.**.*.'A
.  >A,P»«'^i* A ..  .,. .^..4,0,.^-9,

^^^-•^,m

  w  :   .       -.;-  -.,%  :•
                                                                              ,A


                                                                      Appendix A
                                                                      '»PA->»/3-80-006
                                                                          ,;!-.•
                                                              w;i
                    ili-
                   •: »4i .W

-------
                                               Table 5.2
                              Site Parameters Considered In Risk Analysis
                                        Original Generic Basalt
Parameter

Distance from repository to aquifer

Thickness of aquifer

Hydraulic conductivity of aquifer

Porosity of aquifer

Horizontal gradient in aquifer

Retardation values for individual
nuclides
Natural hydraulic gradient between
the repository and the aquifer

Hydraulic conductivity of the book
between the repository and the
aquifer, after thermal effects

Porosity of rock between the
repository and the aquifer

Horizontal distance along the
aquifer to point regarded as release
point to accessible environment
Value
100 meters
30 meters
31.5 a/yr
0.15
0.01
Pu - 100
Am - 100
Np - 100
Cs - 1
I - 1
Tc • 1
Sp - 10
Zr • 100
Sr - 1
C - 1
Source
EPA 520/3-80-006
EPA 570/3-80-006
EPA 520/3-80-006
EPA 520/3-80-006
EPA 520/3-80-006
EPA 520/3-80-006







0.025
3.2xlO~4 m/yr
.0001
1,600 meters
EPA 520/3-80-006


EPA 520/3-80-006



EPA 520/3-80-006


EPA 520/3-80-006

-------
                                             Table 5.3

             Release  Mechanism Parameters  Considered  In Basalt Repository Risk Analysis

                                          Routine Release


 Parameter                                 Value                Source

 Fraction of  the Repository                1.0            .      EPA 520/3-80-006
 with which groundwater can
 communicate

 Hydraulic conductivity of flow            3.2xlO~?m/yr         EPA 520/3-80-006
 path created by release
 mechanism  .

 Porosity in  flow path created by          0.001                EPA 520/3-80-006
 release mechanism

 Cross-sectional area of flow              8.0x10* sq m         Repository Area
 path (vertical flow)

Annual probability or frequency           1.0                  EPA 520/3-80-006

-------
                                             Table 5.4

            Release Mechanisn Parameters Considered In Basalt Repository Risk Analysis
                                   l»
                                             Faulting


Parameter                                 Value                Source

Fraction of the Repository                .003                 EPA 520/3-80-006
with which groundwater can
communicate

Hydraulic conductivity of flow            3.2xl03 m/yr         EPA 520/3-80-006
path created by release                                        SAND84-1492
mechanise .

Porosity in flow path created by          0.1                  EPA 520/3-80-006
release •echaniso

Cross-sectional area of flow              4.0x10-* sq o         EPA 520/3-80-006
path

Annual probability or frequency           3.0xlO~5             EPA 520/3-80-006

-------
                                             Table 5.5

            Release Mechanism Parameters Considered In Basalt Repository Risk Analye is

                                             Volcanoes


Parameter                                 Value                Source

Fraction of the repository                A.OxlO"4             EPA 520/3-80-006
intersected by the release
mechanism

Annual probability or frequency           6.0x10-1°            EPA 520/3-80-006
                                                               USGS Open File
80-357^                                                                 	J?)

-------
                                             Table  3.6

            Release Mechanism Parameters Considered In Basalt.Repository Risk Analysis

                                Drilling and Not  Hitting a Canister
Parameter
Volume of water in the
repository that can reach
the surface

Annual probability or frequency
(after control period)
Value

Z.OxlO2 a3
1.0x10
      ,-2
                                                               Source
EPA 520/3-80-006
EPA 520/3-80-006

-------
                                            Table  5.7

            Release Mechanism Parameters Considered In Basalt Repository Risk Analysis
                                   I*
                                  Drilling  and  Hitting  a  Canister


Parameter                                 Value                Source

Fraction of canister brought to           0.15                 EPA 520/3-80-006
the surface


Annual probability or frequency           l.OxlO"5             EPA 520/3-80-006
(aftei control period)

-------
                                               Table 6.1
                      Site Parameters Considered In Tuff  Repository Risk  Analysis
Parameter

Distance from repository to aquifer
                                  i •
Thickness of aquifer

Hydraulic conductivity of aquifer

Porosity of aquifer

Horizontal gradient in aquifer

Retardation value* for individual
nuclides
Report,
Natural hydrajlic gradient between
the repository and the aquifer

Hydraulic conductivity of the rock
between the repository and the
aquifer, after thermal effecta

Porosity of rock between the
repository and the aquifer

Horizontal distance along the
aquifer to point regarded as release
point to accessible environment
Value

100 meters

1000 meters

30 m/yr

0.002

0.00034

Pu - 200
An " 1,000

Np - 100
Cs " 500
I  " 1
Tc • 5
Sn " 1,000
Zr " 5,000
Sr - 200
C  - 1

1.0


l.OxlO'3 m/yr
0.1
2,000 meters
Source

SAND84-1 ,92

SAND84-1492

SAND84-1492

SAND84-1492

SAKD84-1492

Americium value: Appendix B
All others: NRC/NAS WISP

1983
SAND84-1492


SAND84-1492



SAND84-1492


Agency Decision

-------
                                             Table  6.2

             Release Mechanism Parameter* Considered In Tuff Repository Risk Analysis

                                         Routine Releases


Parameter                                 Value                Source

Hydraulic conductivity of flow            0.001 n/yr           SAND84-1492
path created by release
mechanism

Porosity in flow path created by          0.1                  EPA 520/3-80-006
release mechanism                                              SAND84-1492

Cross-sectional area of flow              8.0x10** m?           Repository Area
path

Probability or frequency                  1.0                  EPA 520/3-80-006

-------
                                            Table  6.4

             Release Mechanism Parameters Considered In Tuff Repository Risk Analysis

      <•                                     Volcanoes

Parameter                                 Value                Source

Fraction of the repository                4.0x10"*             EPA 520/3-80-006
intersected by the release
mechanism

Annual probability or frequency           2.9xlO~"             USGS Open File
                                                               Report 80-357

-------
                                            Table  6.3

             Release Mechanism Parameters Considered In Tuff Repository Risk Analysis

                                             Faulting

Parameter                                 Value                Source

Fraction of the repository                0.003                EPA 520/3-80-006
intersected by the release
mechanism

Hydraulic conductivity in-flow            0.02  m/yr           SAND84-1492
path created by release
mechanism

Porosity in flow path created by          0.1                   EPA 520/3-80-006
release mechanism

Cross-sectional area of flow              4.0X103 sq o         EPA 520/3-80-006
path

Annual probability or frequency           8.0x10"^      •       VSGS Open File
                                                               Report 82-972

-------
                                             Table  6.5

             Release Mechanism Parameters-Considered In Tuff Repository Risk Analysis

                                Drilling  and  Not  Hitting  a  Canister

Parameter                                 Value                Source

Volune of water in the                    140 m3               EPA 520/3-80-006
repository which can reach                            '         SAND84-1492
the surface

Annual probability or frequency           2.5xlO~3             EPA 520/3-80-006
(after control period)

-------
                                             Table 6.6

             Release Mechanism Parameters Considered In Tuff Repository Risk Analysis
                                  l»
                                  Drilling and Hitting a Canister

Parameter                                 Value                Source

Fraction of canister brought              0.1S                 EPA 520/3-80-006
to the surface

Annual probability or frequency           2.5xlO~6             EPA 520/3-80-006
(after control period)

-------