last modified: 10-OCT-1991 | catalog | categories | new | search |

NEA-1023 SYVAC.

SYVAC, Risk Assessment from Underground Radioactive Waste Disposal in UK

top ]
1. NAME OR DESIGNATION OF PROGRAM:  SYVAC.
The simulation program SYVAC A/C (System Variability Analysis Code for deep (A) and shallow (C) burial of radioactive waste) has been developed by the UK Department of the Environment (DOE). The acronym SYVAC comes from the name of an assessment code originally obtained from Atomic Energy of Canada (AECL) in 1982. (The AECL code was found to be inappropriate for geological conditions in the UK).
top ]
2. COMPUTERS
To submit a request, click below on the link of the version you wish to order. Only liaison officers are authorised to submit online requests. Rules for requesters are available here.
Program name Package id Status Status date
SYVAC-D/2 NEA-1023/04 Tested 10-OCT-1991

Machines used:

Package ID Orig. computer Test computer
NEA-1023/04 DEC VAX 11/780 DEC VAX 8810
top ]
3. DESCRIPTION OF PROGRAM OR FUNCTION

The SYVAC program simulates the ground water mediated movement of radionuclides from underground facilities for the disposal of low and intermediate level wastes to  the accessible environment, and provides an estimate of the subsequent radiological risk to man. The simulated timescales are usually within the range 1.0E+03 to 1.0E+07 years.
SYVAC is capable of modelling both shallow disposal facilities (located in argillaceous media and overlaying an aquifer) and deep disposal facilities (in a saturated environment). It is not suitable for considering high level wastes as it does not allow for  the heat produced by such wastes.
SYVAC has been developed for use within the DOE Radioactive Waste Management Programme, as one tool in the DOE Assessment Methodology.
top ]
4. METHOD OF SOLUTION

4.1 The Physical System
    -------------------
A simplistic description of the physical simulated by SYVAC is given below.
Low and intermediate level radioactive waste is stored in a  repository or "vault". For shallow disposals, the "vault" is an  engineered trench at depths of 20-30mm; for deep disposals a  modified mine or purpose built structure at depths of 150-300m.  Over a long period of time, groundwater penetrates the vault  structure and the canisters containing the waste corrode. The  radionuclides migrate in the groundwater flow through the vault  structure into the ground - the "geosphere" - surrounding the vault.
The processes are represented in the code by a set of modules known as the Vault submodel.
The radionuclides from the vault migrate through the geosphere  to the surface and into the "biosphere" where they become  accessible to man, through drinking water, crops etc. The  migration takes place through the groundwater flows in aquifer  structure for both the shallow and deep disposal sites; for the  shallow sites, the radionuclides can also migrate directly to  the surface and, where appropriate, this latter process is simulated in SYVAC as part of the Vault submodel.
The process of migration through the Geosphere (with the  exception noted above for the shallow site direct path) is  represented in the code by the set of modules known as the Geosphere submodel.
The Vault and Geosphere submodels are controlled within the  program by the Executive, a set of management modules.  Generally, the Executive modules are concerned with data  handling and data input/output to the submodels. The Vault and  Geosphere submodels are principally concerned with the numerical calculations needed to simulate radionuclide migration from the vault through the geosphere into the biosphere.
The final output from the Geosphere submodel is in the form of a  radionuclide flux over time. To estimate the subsequent  radiological risk to man from this flux, the SYVAC program  applies a series of dose conversion factors (these are different  for each nuclide). The dose rate is calculated by multiplying  the nuclide flux by the dose conversion factors, and applies to maximally exposed individuals.
These dose conversion factors are supplied to SYVAC as input  data, in the form of biosphere data files. The files are  currently produced from a structurally separate biosphere model called ECOS.

4.2 Probabilistic Parameter Sampling in Syvac
    -----------------------------------------
The previous section has briefly described the simulation of  radionuclide migration carried out by the Vault and Geosphere  submodels. This simulation requires data to describe the  physical structure and transport characteristics of the vault and geosphere.
The necessary data is supplied to SYVAC in the form of  parameters that represent hydraulic conductivities, diffusion coefficients, porosities etc for the vault and geosphere.  Thirteen basic vault parameters are used to model deep disposal  sites, and fifteen for shallow disposals. For both types of  site, an additional six parameters are needed for each geological layer modelled in the geosphere.
The values of some of the parameters are not precisely known,  and will vary over the timescale of the transport to the  biosphere. SYVAC allows for this with a probabilistic approach; the parameter values are sampled from distributions  chosen to cover the expected range of values that may be  encountred during the post-closure phase of a vault. For each  parameter, SYVAC allows the user to select one of five  distributions: constant, uniform, log uniform, normal and log  normal. Usually, for each parameter, the user will select the     distribution that best fits the observed parameter values.
The calculation of risk to man requires knowledge of both the     dose from a particular run, and the probability of that run.
Clearly, the dose to man calculated for a particular simulation  run will depend on the parameter values chosen for that run. The  probability of a particular run - i.e. the probability of  choosing a particular set of parameter values - also depends on  the parameter value chosen, and can be calculated from the  parameter distributions selected by the user when setting up the input data.
The probabilistic nature of these calculations means that to  obtain statistically valid estimates of risk (etc), SYVAC must be run a large number of times with different parameter  values. The user chooses the number of runs when setting up the  input data. The user also chooses the type of sampling used to  select the parameter values for each run. SYVAC stores the results of each run for later analysis.
Two different sampling methods can be chosen for SYVAC;  Random Sampling and Deterministic Generator Sampling. Random  Sampling chooses values (randomly) from the entire range of the  distribution chosen for each parameter. Deterministic Generator  sampling selects one of 11 possible (discrete) sample points for  each parameter for a run; the method of sampling is such that  all eleven points are chosen in the course of 11 consecutive  runs. The eleven points span the range of the distribution chosen for each parameter.

4.3 Run Acceptance Level
    --------------------
The calculations performed in the Vault and Geosphere submodels  use most of the CPU time needed for a run. Consequently, to  avoid unnecessary calculations, the program carries out a  screening procedure to eliminate runs (i.e. sets of parameter  values) that will produce no dose or are physically unrealistic.  After the sampled parameter values have been generated, each  SYVAC run is classified as being one of four run acceptance levels. The interpretation of run levels is as follows:
    Level 0 : No dose expected within the time of study.
Level 1 : A non-zero dose expected. This is an accepted run.               The actual dose of course may turn out to be zero.
Level 2 : Physically unreal data. An infeasible combination of               sampled parameters has been selected.
Level 3 : The data is outside the validated range of one of the  submodels and there is an expectation of a high dose.  Note that the Vault and Geosphere submodel calculations are only performed for accepted (level 1) runs.
To arrive at the classification for a particular run a series of  tests (of different levels of severity) is performed. Each test  checks whether a sampled parameter value or value calculated from several parameters is within a specified range.

4.4 Executive Control
    -----------------
The Executive is the framework that controls the simulation  program as a whole, and specifically calls the Vault and Geosphere submodels.
    The functions performed by the Executive are:
    (i) Input and validation of user defined data.
    (ii) Generation of sampled values for data defined by            probability distributions.
    (iii) Verification and classification of each complete set of            sampled data.
    (iv) Invocation and control of the submodels which represent            the stages of the nuclide transportation process.
    (v)    Generation of reports and other output.
    (vi)   Case study control.

4.5 Vault Submodel
    --------------
The Vault submodel simulates the release of radionuclides from  non heat-producing radioactive waste placed within an engineered  vault in either deep or shallow land disposal facilities. The  model includes representations of the waste matrix, waste  canister, vault liner, the vault backfill material and the  region of host rock around the vault disturbed during  construction.
This submodel consists of a hydrogeological model, representing  the groundwater flow through the different engineered barriers,  and a radionuclide migration model. In the hydrogeological  model,the different structural features of the vault are  represented by a series of compartments. The hydrogeological  model uses an electrical resistance analogy to represent flow  within each of these compartments.
The migration model includes the processes of solubility-limited  or mass-limited leaching of radionuclides from the waste matrix,  linear equilibrium sorption of radionuclides onto the barrier  materials, radioactive chain decay, dispersion and advection.  Migration of radionuclides in groundwaters is predicted using a  numerical solution to the transport equation. The leaching of  radionuclides from the waste matrix is simulated using a  diffusion model and gives the source term for the transport  equation. In solving the transport equation, the radionuclide  concentration at the vault/geosphere interface is taken to be zero.

4.6 Geosphere Submodel
    ------------------
The Geosphere submodel is a 1-dimensional model that deals with  the migration of radionuclides from the vault to the biosphere  through the geosphere. This is through the calculation of  a response function for each geosphere layer. The response  function describes the output from each geosphere layer at a  given time for unit input to the geosphere layer at a start time  The convolution of this response function with the time-varying  input to the layer gives the total output from the geosphere  layer at any time. Up to five distinct layers can be modelled.  This submodel includes the effects of linear equilibrium  sorption (for porous or fractured media), linear dispersion and  chain decay. Up to 8 member chains can be considered. A semi-finite analytical Laplace transform solution technique is used to solve the geosphere partial differential transport equation.
4.7 Biosphere Model - ECOS
    ----------------------
The biosphere code ECOS calculates the rate of accumulation of  committed effective dose equivalent (dose) arising from the flux  of radionuclides entering the biosphere. Dose to a maximally  exposed individual and collective dose are calculated deterministically.
ECOS is an equilibrium-type compartmental model. The contents of  parent and daughter radionuclides at equilibrium in nine  activity reservoirs are computed by solving a set of linear  simultaneous equations which represent the equilibrium solution of a general differential equation description.
The transfer processes modelled are: percolation and leaching in  soils, erosion by wind and water, irrigation, cropping, water  phase and sediment phase transport, sedimentation and  resuspension in aquatic environments, equilibrium radionuclide  sorption and radioactive decay. The dose pathways to man  modelled are: drinking water, foodstuffs, external exposure and  inhalation. Consumption of aquatic foodstuffs, terrestrial crops  and animal products are included. For both drinking water and  foodstuff pathways, intakes by man are reduced if the  contaminated suply is insufficient to meet total annual needs.  The model database includes parameters (e.g. foodchain parameters and sorption coefficients) for 53 radionuclides.
ECOS can be used in two different ways. It can be used in  isolation, in order to investigate the relation between dose and  biosphere state and in order to identify biosphere scenarios  that should be considered in radiological assessment. It can  also be used, as part of the overall SYVAC simulation, to  produce dose conversion factors that are applied to the activity flux from the geosphere in order to estimate dose to man.

4.8 Nuclide Chemistry
    -----------------
Chemical processes are incorporated in the Vault and Geosphere     submodels, not in a separate chemistry submodel.
Chemical parameters are obtained from expert judgement, basic  research and site data, together with the use of more detailed  models to examine scenarios and hence estimate ranges of values.  At present, probability distributions are obtained judgementally for most cases.
The detailed models include simulation of the near-field, which  provides direct input to SYVAC in the form of limiting  solubilities, and coupled geochemical transport programs which  are intended to provide verification for SYVAC runs. The  principal programs used are based on ion-association models and  solve for mass balance and mass action iteratively using a form  of the Newton-Raphson technique and/or a continued fraction approach.
top ]
5. RESTRICTIONS ON THE COMPLEXITY OF THE PROBLEM

5.1 Restrictions in the Vault Submodel
    ----------------------------------
(a) The vault submodel allows 5 "boxes" for the internal  construction of the vault and 2 barriers (liner and clay). These cannot be changed by the user.
(b) The model includes representations of 2-dimensional diffusion,  and 1-dimensional advection for shallow sites/2-dimensional  advection for deep sites. The flow is approximated by Darcy's     Law.
(c) Time invariant chemistry; elemental (not isotopic) solubilities;  diffuse leaching; nuclide independent retardation. For the case  of shallow land disposal, retardation in the damaged region  around the Vault structure is incorporated for the direct  pathway to the biosphere but not for the pathway through the biosphere.
(d) The equations for flow and transport are uncoupled. solubility,   leaching, sorption, diffusion and permeability are all time     independent.
(e) There is no allowance for man made or natural (e.g.) tree     roots) intrusion into the Vault.
(f) The Vault structure is below the water table (saturated) after a     certain period.
(g) There is no interaction between adjacent Vault structures if     these were present.
(h) There are a number of limitations on number of canisters,     nuclides per chain, etc built into the program.

5.2 Restrictions in the Geosphere Submodel
    --------------------------------------
(a) The Geosphere submodel contains stated limitations on the  parameter values. The range of allowed values corresponds to     physically admissible values for the parameters.
(b) The timesteps in the model are of fixed duration, which limits  the minimum pore water transit times across each geosphere layer.
(c) Although the number of nuclides in each chain is theoretically  without limit, practical restrictions on CPU suggest that chains  with more than 4 members use secular equilibrium for short lived  members and chain truncationfor long lived members.
(d) SYVAC can model up to 5 geosphere layers but for practical     computations it is advisable not to exceed 3.
(e) The physical approximations incorporated in the submodel are  those normally used in one dimensional, continuum models of     transport in homogeneous porous media.

5.3 Restrictions in the Biosphere Model
    -----------------------------------
ECOS calculates estimates of activity and dose to man for a  constant radionuclide flux into the biosphere. It cannot be used  to estimate doses from rapidly changing input fluxes or "spike"  inputs. The model does not simulate the production or transport  of gaseous radionuclide species nor estimate the dose from such  species. This may lead to inaccurate estimates (usually over-  estimates) of doses from potentially gaseous radionuclides such  as H-3, C-14, I-129. The model includes 53 radionuclides and up  to 7 member chains. The allowed list has been found to be more than adequate in the studies to date.
top ]
6. TYPICAL RUNNING TIME

The running time for SYVAC depends on:
- the number of runs
- the number of nuclides in each run.
- the number of nuclides in each chain
Typical CPU times of a single run on a MicroVAX 1 are:

No. of Nuclides in Chain Time (CPU minutes)
1 0.6
2 2
3 4.4
4 11.1
Single run times on a MicroVAX 2 are approximately 1/5 of these figures. A representative example case study - say 1000 runs for the single member Iodine 129 chain - would need approximately (1000 x
0.6) x 0.2 CPU minutes (2 hours) to complete execution on the MicroVAX2.
NEA-1023/04
The system was implemented by NEADB on a VAX 8810 computer. The sample case for LAND 2 required 35 seconds, and the case for LAND 3 required 127 seconds of execution time.
top ]
7. UNUSUAL FEATURES OF THE PROGRAM:
top ]
8. RELATED AND AUXILIARY PROGRAMS

The biosphere model, ECOS, has already been described in Section 4. There are a number of other auxiliary programs concerned with graphical analysis and sensitivity analysis. They are not integral to SYVAC, and access the output  data files created by SYVAC in the course of execution. These programs are discussed briefly for completeness.
These additional programs have been written to facilitate the analysis of the results in the output files. The programs examine:
-   runs in dose or risk order
-   dose at specified times
-   high dose/risk runs
-   risk over time with confidence limits
-   the cumulative probability of dose occurrence
-   the sensitivity of dose to parameter values
-   the efficiency of sampling.
Importance sampling, a recent inclusion, uses the results of sensitivity analysis studies to significantly increase sampling efficiency. Several techniques have been investigated for quantifying expert judgement and also for incorporating measured data using a Bayesian approach.
top ]
9. STATUS
Package ID Status date Status
NEA-1023/04 10-OCT-1991 Tested at NEADB
top ]
10. REFERENCES

At present, the SYVAC documentation (prepared by SCICON Limited) totals 13 volumes, with another 2 volumes in preparation. Additional volumes describing the submodels are currently being revised by the other contractors. The documentation available is listed below:
Part 1 : SYVAC A/C Overview : Volumes 1 - 3

   Volume 1  -    Documentation Introduction
   Volume 2  -    System Overview
   Volume 3  -    Glossary

Part 2 : SYVAC A/C User Guide : Volumes 1 - 3

   Volume 1  -    Input and Execution
   Volume 2  -    Output and Error Messages
   Volume 3  -    Installation Instructions

Part 3 : SYVAC A/C Programmers' Guide : Volumes 1 - 7

   Volume 1  -    Program Structure
   Volume 2  -    File Descriptions
   Volume 3  -    Data Dictionary
   Volume 4  -    Data Transfer
   Volume 5  -    Executive Modules
   Volume 6  -    Vault Modules
   Volume 7  -    Geosphere Modules
NEA-1023/04, included references:
- S. Carr and G.Rowling:
  Documentation Introduction.
  SYVAC-D/2 Overview, Volume 1 (April 1988)
- S. Carr, M. Fisher and G. Rowling:
  Documentation Index.
  SYVAC-D/2 Overview, Volume 2 (April 1988)
- S. Carr:
  Glossary.
  SYVAC-D/2 Overview, Volume 3 (April 1988)
- S. Carr, G. Rowling and F. Taylor:
  System Overview/
  SYVAC-D/2 Overview, Volume 4 (April 1988)
- S. Carr and G. Rowling:
  Input and Execution.
  SYVAC-D/2 User Guide, Volume 1 (February 1988)
- J. Falconer and G. Rowling:
  Output and Error Messages.
  SYVAC-D/2 User Guide, Volume 2 (February 1988)
- S. Carr, G. Rowling and F. Taylor:
  Installation Instructions.
  SYVAC-D/2 User Guide, Volume 3 (February 1988)
- J. Falconer, G. Rowling and F. Taylor:
  Analysis Programs.
  SYVAC-D/2 User Guide, Volume 4 (February 1988)
- S. Carr, G. Rowling and F. Taylor:
  Program Structure.
  SYVAC-D/2 Programmers' Guide, Volume 1 (February 1988)
- J. Falconer, and G. Rowling:
  File Descriptions.
  SYVAC-D/2 Programmers' Guide, Volume 2 (February 1988)
- S. Carr, J. Falconer and G. Rowling:
  Data Transfer.
  SYVAC-D/2 Programmers' Guide, Volume 3 (February 1988)
- J. Falconer, G. Rowling and P. Saul:
  Data Dictionary.
  SYVAC-D/2 Programmers' Guide, Volume 4 (February 1988)
- S. Carr, J. Falconer and G. Rowling:
  Executive Modules.
  SYVAC-D/2 Programmers' Guide, Volume 5 (February 1988)
- S. Carr, J. Falconer and G. Rowling:
  Vault Modules.
  SYVAC-D/2 Programmers' Guide, Volume 6 (February 1988)
- S. Carr, J. Falconer and G. Rowling:
  Geosphere Modules.
  SYVAC-D/2 Programmers' Guide, Volume 7 (February 1988)
- A. Saltelli, E. Sartori, T.H. Andres, B.W. Goodwin and
  S.G. Carlyle:
  PSACOIN Level 0 Intercomparison
  An International Code Intercomparison Exercise on a Hypothetical
  Safety Assessment Case Study for Radioactive Waste Disposal
  Systems (OECD/NEA - November 1987)
top ]
11. MACHINE REQUIREMENTS

At present, case study work is carried out by SCICON Limited (Section 15) on a DEC MicroVAX 1 and a DEC MicroVAX 2. Short term storage is via 10MByte discs for the MicroVAX 1, and a 700MByte disc for the MicroVAX 2. Tapes are used for long term storage. A printer is needed for producing physical output.
SYVAC A/C allows a case study to be restarted to increase the number of runs. This is necessary to test for convergence of results and also to work within the limitations on storage, by deleting unwanted files. Typically 18000 blocks are needed for 600 simulation runs.
A system clock is needed to continue case studies in this way. A clock is also needed for the analysis of CPU usage that is carried out by the program.
top ]
12. PROGRAMMING LANGUAGE(S) USED
Package ID Computer language
NEA-1023/04 FORTRAN-77
top ]
13. OPERATING SYSTEM UNDER WHICH PROGRAM IS EXECUTED

SYVAC is presently run under the DEC Micro VMS operating system, version 4.2. To run SYVAC A/C under version 4.2, it is necessary to have a
software modification from DEC to correct an error with double precision exponentials. The program runs under version 4.3 which does not need the modification. SYVAC uses the VAX standard mathematical library but also requires the NAG (Numerical Algorithms Group) library or an equivalent for certain numerical routines not provided in standard FORTRAN libraries.
NEA-1023/04
NEA-DB executed the test cases included in this package on a VAX 8810 computer running under VMS 5.3
top ]
14. OTHER PROGRAMMING OR OPERATING INFORMATION OR RESTRICTIONS

The MicroVax 1 and 2 are virtual memory machines and so there is no overlaying or similar restrictions due to storage. The only changes  required to run SYVAC on another machine (apart from the operating system) would be any that resulted from differences in the implementation of FORTRAN 77 on the two systems.
Converting SYVAC to run in another programming language would be a major undertaking and is unlikely to be effective; it would be preferable to start with the design documents, if it was intended to modify SYVAC in this manner.
top ]
15. NAME AND ESTABLISHMENT OF AUTHORS

The development of SYVAC has been undertaken by private contractors under DOE contracts.
Project Management
------------------
Dr. B.G.J. Thompson
Radioactive Waste (Professional) Division
Department of the Environment
Room A5.35, Romney House
43 Marsham Street
London SW1P 3PY, England
Contractors                     Areas of Responsibility
-----------                     -----------------------
Atkins Research and             Models for the migration of
Development                     radionuclides through host rock.

Associated Nuclear Services     Models for the biosphere and the
                                calculation of dose to man.

CAP Scientific                  Statistical techniques used for
sensitivity and uncertainty analysis
Electrowatt Engineering         Models for the migration of
                                radionuclides from the vault.

Scicon Limited                  Software coordination, computing
                                services, system documentation.
Contractor Addresses
--------------------
1. Dr. T. Broyd, Atkins Research & Development, Woodcote Grove,
   Ashley Road, Epsom, Surrey KT18 5BW, England.
2. Dr. T. Summerling, Associated Nuclear Services, Eastleigh House,
   60 East Street, Epsom, Surrey KT17 1HB, England.
3. Dr. G. Dalrymple, CAP Scientific, 20 - 26 Lambs Conduit Street,
   London WC1N 3LF, England.
4. Dr. A. Gralewski, Electrowatt Engineering Services, Grandford
   House, 16 Carfax, Horsham, West Sussex RH12, England.
5. Dr. I.C. Rae, Scicon Limited, Wavendon Tower, Wavendon, Milton
   Kenes MK17 8LX, England.
top ]
16. MATERIAL AVAILABLE
NEA-1023/04
File name File description Records
NEA1023_04.001 Information file 275
NEA1023_04.002 Command file to compile and link SYVAC-D/2 9
NEA1023_04.003 Command file to compile FORTRAN programs 5
NEA1023_04.004 Command file to compile SYVAC-D/2 sources 157
NEA1023_04.005 Command file to link SYVAC-D/2 (nondebug) 4
NEA1023_04.006 Command file to link SYVAC-D/2 (debug) 4
NEA1023_04.007 FORTRAN source ('Executive' modules) 10873
NEA1023_04.008 FORTRAN source (Geosphere submodel modules) 3165
NEA1023_04.009 FORTRAN source (Vault submodel modules) 5617
NEA1023_04.010 FORTRAN source (included module) 1
NEA1023_04.011 FORTRAN source (included module) 474
NEA1023_04.012 FORTRAN source (included module) 271
NEA1023_04.013 FORTRAN source (included module) 19
NEA1023_04.014 FORTRAN source (included module) 164
NEA1023_04.015 FORTRAN source (included module) 79
NEA1023_04.016 FORTRAN source (included module) 34
NEA1023_04.017 FORTRAN source (included module) 43
NEA1023_04.018 FORTRAN source (NAG alternative) 2870
NEA1023_04.019 SYVAC-D/2 executable image 305
NEA1023_04.020 SYVAC-D/2 object module 634
NEA1023_04.021 LAND 2 sample input data 54
NEA1023_04.022 LAND 2 sample input data 40
NEA1023_04.023 LAND 2 sample input data 12
NEA1023_04.024 LAND 2 sample input data 45
NEA1023_04.025 LAND 2 sample input data 12
NEA1023_04.026 LAND 2 sample input data 14
NEA1023_04.027 LAND 2 sample output 62
NEA1023_04.028 LAND 2 sample output 708
NEA1023_04.029 LAND 2 sample output 810
NEA1023_04.030 LAND 2 sample output 810
NEA1023_04.031 LAND 2 sample output (binary) 7
NEA1023_04.032 LAND 3 sample input data 50
NEA1023_04.033 LAND 3 sample input data 38
NEA1023_04.034 LAND 3 sample input data 12
NEA1023_04.035 LAND 3 sample input data 54
NEA1023_04.036 LAND 3 sample input data 6
NEA1023_04.037 LAND 3 sample input data 15
NEA1023_04.038 LAND 3 sample output 47
NEA1023_04.039 LAND 3 sample output 696
NEA1023_04.040 LAND 3 sample output 1012
NEA1023_04.041 LAND 3 sample output (binary) 7
top ]
17. CATEGORIES
  • R. Environmental and Earth Sciences

Keywords: geologic strata, ground water, radioactive waste storage, risk assessment, underground storage.