PNNL SFA: Role of Microenvironments and Transition Zones in Subsurface Reactive Contaminant Transport

PNNL SFA: Role of Microenvironments and Transition Zones in Subsurface Reactive Contaminant Transport

Facies-based Characterization of Facies-based Characterization of Hydrogeologic Structures and Reactive

Transport PropertiesHydrogeologic Structures and Reactive

Transport PropertiesScientific Staff:

Andy Ward (PNNL)

Scientific Staff:

Andy Ward (PNNL)Andy Ward (PNNL)Chris Murray (PNNL)Charlotte Sullivan (PNNL)

Andy Ward (PNNL)Chris Murray (PNNL)Charlotte Sullivan (PNNL)Charlotte Sullivan (PNNL)Steve Yabusaki (PNNL)Roelof Versteeg (INL)

Charlotte Sullivan (PNNL)Steve Yabusaki (PNNL)Roelof Versteeg (INL)

November 4, 2008 PNNL-SA-65828

Objective

Project Objectives and ScopeProject Objectives and Scope

ObjectiveDevelop 3-D facies models that reflect variations in the depositional environment and the associated subsurface depos t o a e o e t a d t e assoc ated subsu aceheterogeneity

ScopeEstablish quantitative links between sedimentary facies and primary hydrophysical properties

requires higher level of sediment characterization whole sediments and size fractionslithocomponents of size fractions

IFRC Linkage

Field Data:Surface Geophysics

Borehole Geophysics Site-Specific Property Transfer/ SFA

IFRC

p yPetrophysical Models

a3-D Inversion

SFA

FY-09

Electrofacies DistributionsBest Estimates of

Facies Distribution

YesYes

and UncertaintyFacies Distribution

Flow/Transport Inversion with PSD Correlation

Structures

NoNo Improved Facies Model

(PSD moments) Criteria met?

Identification of LithofaciesMulti-scale Heterogeneity

4

Field Characterization and Monitoring Subsurface Temperatureg

Borehole Electrical Conductivity

5

Primary sediment properties are controlled by facies distributions, which in turn are controlled by grain size distributions resulting from the depositional environment

Project PremiseProject Premise

Hydraulic PropertiesBulk density

controlled by grain size distributions resulting from the depositional environment

Natural Isotope Abundance 40K, 238U, 232ThGross -ray responseElectrical Properties Bulk density

PorosityResidual water contentWater retentionRelative Permeability

Gross ray responsepDielectric SpectroscopySurface conductivityFormation FactorChargeability Relative Permeability

Particle Size Distribution

Chargeability

Reactive PropertiesSpecific surface areaCEC

Transport PropertiesTortuosityIntrinsic PermeabilityCECpor

SporReaction Kinetics

Thermal PropertiesHeat Capacity

Intrinsic PermeabilityDispersivityFormation Factor

Heat CapacityThermal conductivity

Sediment CharacterizationEstablishing such linkages requires

Particle Size Distribution

4

6

ume

(%)

Establishing such linkages requires higher level of characterization

whole sediments 0.01 0.1 1 10 100 1000 3000

Particle Size (µm)

0

2 Vol

u

IFC 3-30 6.5-8' <2mm (Pre Sonic) - Average, Tuesday, January 13, 2009 11:37:22 AMIFC 3-30 6.5-8' <2mm (10min Sonic) - Average, Tuesday, January 13, 2009 12:05:00 PMIFC 3-30 6.5-8' <2mm (Post Sonic) - Average, Tuesday, January 13, 2009 12:10:58 PM

size fractionsSize analysis

Dry/wet sieveH d t

Particle Size Distribution

1

2

3

4

5

Vol

ume

(%)

HydrometerLaser particle sizeMineralogy of size fractions

Continuous size distributions

0.01 0.1 1 10 100 1000 3000 Particle Size (µm)

0

C6216/2-29 39'-41' 32mm (Pre sonic) - Average, Thursday, January 08, 2009 3:19:48 PMC6216/2-29 39'-41' 32mm (10min sonic) - Average, Thursday, January 08, 2009 3:46:33 PMC6216/2-29 39'-41' 32mm (Post sonic) - Average, Thursday, January 08, 2009 3:50:32 PM

Particle Size of Coatings on 32 mm GravelContinuous size distributionsDifferences in coatings on different size gravel Differences in coatings on same size from different depths 70%

80%

90%

100%

ng

C5213 3-30 6.5'-8'C6216 2-29 39'-41'C6213 2.5'-5'C6212 2-28 47'-49'C6216 2-29 30-33

siltclay

size from different depthsCould impact variability in sorption

20%

30%

40%

50%

60%

70%

Perc

ent P

assi

n

0%

10%

20%

Size (mm)

2 µm 50 µm

G l

-ray Dependence on Grain Size-ray Dependence on Grain Size6

8

)

(a)

GoalEstablish basis for using 40K, 238U, 232Th logs to quantify grain size distributions and spatial variations 0

2

4

40K

(%)

and spatial variationsMeasurements

GEA Analysis on samples from C57086

8(b)

0.01 0.1 1 10 100 1000dg (mm)

0

strong relation between GEA response and grain size statistics

Observed on samples from Hanford and 2

4

6

238 U

(ppm

)

RiflePredictions

MCNP model of -ray spectrometer

0.01 0.1 1 10 100 1000dg (mm)

0

8(c) y p

MCNP model of borehole logNeural network to predict size moments from KUT 2

4

6

232 T

h (p

pm)

0.01 0.1 1 10 100 1000dg.mm.

0

2

Predicted Grain size (C5708)Predicted Grain size (C5708)

20

Obs.Pred.

20 20

60

40

60

40 40

80

60

Dep

th (f

t)

80

60D

epth

(ft)

80

60

Dep

th (f

t)

100 100 100

140

120

140

120

140

120

grsic s

0.0 0.5 1.0IGR

0 2 4 6 840K (%)

0.001 0.1 10dg (mm)

140

Vertical variation in lithology

Integrating Multi-scale HeterogeneitiesIntegrating Multi-scale HeterogeneitiesVertical variation in lithology

Generation of fine-scale heterogeneity patterns Borehole logs for vertical transitionBorehole logs for vertical transition probabilities and conditioning

Horizontal continuity large-scale architectural structuresurface geophysics

Merge fine and large scaleMerge fine- and large-scale heterogeneities

TPROGS algorithm of Carle et al. (1998)(1998)Coupled Markov chain (Dekking et al. 1999)Sequential indicator simulation (SISIM) f GSLIB lib

From Elfeki et al. 2002, Petroleum Geoscience, 8:159-165

(SISIM) from GSLIB library

-ray Properties of Geologic UnitsAssume global histogram of totalAssume global histogram of total is sum of n normal populations

One normal distribution for each lithofaciesO ti i fit f lith f i t l b lOptimize fit of lithofacies to global histogram of data

Three lithofacies identifiedUse spectral 40K, 238U, 232ThUse spectral K, U, Th

Extend to all logged IFRC wellsNeural network approach for identification of faciesCalibrate with grain size dataCalibrate with grain size data

Extend lithofacies simulation to three dimensions

Algorithm employed will be based Facies Percent Mean g p yon comparison of 2-D results 1 32% 151.2

2 40% 161.53 29% 177.5

11

Electrical Properties of Geologic UnitsSelected 4 cores based on EBF-Ksat

Normalized Hydraulic ConductivityProfile, Well 399-2-12

30

21-22’, 43-44’ 52-53’, 58-59’Instrumented for electrical measurements

Resistivity, chargeability (IP)Three major electrofacies identified

32

34

36

38

40

42bgs)

Three major electrofacies identifiedHigh resistivity, low chargeabilityIntermediate resistivity, low chargeabilityLow resistivity, high chargeability

42

44

46

48

50

52

54

Dep

th (f

t b

Low resistivity, high chargeabilityLow fines = low chargeability (<10 mV/V)

mineral type, grain size, Spor, pore water chemistry, physics of interaction between

54

56

58

600.0 0.1 0.2

Normalized Kiy p ysurfaces and fluids

Facies a ( m) m (mV/V)Depth (ft)21-22 1 200043-44 2 90052-53 3 1500

1.04.62.0

12

58-59 4 110 13.0

PorescaleSmooth particle hydrodynamics

Aspherical particlesAspherical particles Packing based on IFRC sedimentsReactive chemistry

Impedance spectroscopy Solve 3-D Poisson equationMultiple phases

Important factorsParticle sizeParticle sizeParticle shapeSurface area/ surface charge

13

FY 2009 Science Deliverables

Property Transfer Models for IFRC SedimentsSpectral gamma logsSpectral gamma logs

3-D electrofacies model reflecting heterogeneity at IFRC well fieldwell field3-D lithofacies model reflecting heterogeneity at IFRC well field

Backup Slides- MineralogySiO275 0

60.0

65.0

70.0

75.0

Wei

ght %

WSiLN-Lysim300-FF-5

FeO12 0

55.00 1 2 3

dg (mm)

2 04.06.08.0

10.012.0

Wei

ght %

WSiLN-Lysim

0.02.0

0 1 2 3dg (mm)

300-FF-5

MnO

0.05

0.10

0.15

0.20W

eigh

t % WSiLN-Lysim300-FF-5

0.00

0.05

0 1 2 3dg (mm)

W 300 FF 5