a source-responsive approach for simplified estimates …...

A SourceA Source--Responsive Responsive Approach for Simplified Approach for Simplified

Estimates of Preferential Flow Estimates of Preferential Flow at INTECat INTEC

Ben Mirus, Kim Perkins, and John NimmoUnsaturated Zone Flow ProjectU.S. Geological Survey, Menlo Park, CA

INL Monitoring and Surveillance Committee MeetingJuly 28th, 2011, Idaho Falls, ID

Key Concepts Key Concepts

Unsaturated Zone Flow ProjectUnsaturated Zone Flow Project

Source-Responsive Theory - Unsaturated preferential flow- Gravity driven films- Importance of water availability at source- Limited input data needed for modeling

1450

1455

1460

1465

4/8/2005

7/17/2005

10/25/20052/2/2006

5/13/2006

8/21/2006

11/29/20063/9/2007

Hyd

raul

ic h

ead

(m)

0

2

4

6

8

10

12

14

Dis

char

ge (c

ms)

Piezometer, 37-40 m blsTensiometer, 40 m blsBLR discharge

Sedimentary deposits

Aquifer Perched saturationFractured basalt

East Snake River Plain Aquifer

Surficial sediments

(a) Anthropogenic and natural recharge sources

B‐C Interbed

0

25

50

75

100

125

150

175

Depth below

land surface (m)

A‐B Interbed

C‐D Interbed

Groundwater flow (saturated/unsaturated)

(b) Limited recharge Approximatescale only

Conceptual Model of Vadose Zone


Monitoring• Piezometers / wells• Nested tensiometers• S. perching eliminated• N. perching remains

Wet Dry


Perched Zone Dynamics

90

-300

-200

-100

0

100

200

300

400

500

Jul-01 Jul-02 Jul-03 Jul-04 Jul-05 Jul-06 Jul-07

Pres

sure

hea

d (c

m)

BLR-SP1, 132-ft

BLR-SP2, 167-ft

BLR-DP1, 352-ft, broken cup


Big Lost R.

Well Hydrograph MW-6

1460

1462

1464

03/09/07 07/07/07 11/04/07 03/03/08

Wat

er le

vel (

m)

Repaired 2/6/08

Leakyhydrant


Identifiable Recharge SourceBig Lost River Hydrograph

0

50

100

150

200

250

300

350

400

450

500

Jun-94 Jun-95 Jun-96 Jun-97 Jun-98 Jun-99 Jun-00 Jun-01 Jun-02 Jun-03 Jun-04 Jun-05 Jun-06

Dis

char

ge (c

fs)

Subsurface monitoring at BLR-CH


1450

1455

1460

1465

4/8/2005

7/17/2005

10/25/20052/2/2006

5/13/2006

8/21/2006

11/29/20063/9/2007

Hyd

raul

ic h

ead

(m)

0

2

4

6

8

10

12

14

Dis

char

ge (c

ms)

Piezometer, 37-40 m blsTensiometer, 40 m blsBLR discharge

Source and Response


Unsaturated Zone Preferential Flow Unsaturated Zone Preferential Flow


When preferential flow is an important process, traditional approaches to modeling unsaturated flow based on diffuse-flow theory are flawed.

Film-flow theory has provided the foundation for more realistic conceptual models of unsaturated zone preferential flow.

Preferential flow response appears to be most sensitive to availability of water at the source

SourceSource--Responsive Flow (SR)Responsive Flow (SR)

Empirical models were designed to provide simple estimates of preferential flow processes in situations with limited data availability (Nimmo, 2007, 2010a,b).

Free-surface films on walls of macropore, activated in responseto water availability at the source of input

Water supplied may be continuous or intermittent

Within matrix, capillary forces drive diffuse flow


`

(diagram not to scale)

uavg VgV == 2

31 δν

2max 2

1 δνgV =

uusr Vq δ∝


SR Theory SR Theory

SR Modeling approaches SR Modeling approaches


Minimum Travel-Time Model (Nimmo, WRR, 2007)– Maximum velocity Vmax gleaned from first arrival times in

64 tracer tests from independent published studies – Predicts minimum travel time for preferential flow based

on continuous vs. intermittent input at source– Recently verified using 48 additional tracer tests,

predictions to within an order of magnitude accuracy (Ebel and Nimmo, HESS-D, 2010)

Preferential Fluxes Model (Nimmo, VZJ, 2010)– Extends theory to incorporate SR fluxes, coupled to

traditional diffuse flow equations– Simulates fluxes and dynamic water contents– Can be calibrated/evaluated with direct observations

SR Fluxes Model– Couples equations for diffuse and preferential flow

to quantify fluxes (1D vertical)

– Primary degrees of freedom represented by two parameters M(z) [L-1] and f(z,t) [-]

– M – macropore facial area density, propensity for development of preferential flow paths

– f – fractional area of active macropores, controlled by availability of water at the source

),()()(),( tzfzMVz

Ktzq uuD δθ +∂Φ∂

−=


SR Application for INTEC– Uniform basalt unit above A-B interbed, assume

no variations in M and f values with depth

– Source-responsive preferential flow dominates perched zone accretion, assume no diffuse flow

– Decline of perched zone is by saturated diffuse flow, using simple linear recession constant

τδ H

SfMV

dtdH

y

uu −=


f(t) varies based on transmission lossesSy = 0.01 (Johnson et al., Hydrogeol. J., 2002)

= 550 h (regression analysis)

M calibrated to fit observations, evaluated using split sample

Parameterization

τ


0

0.5

1

1.5

2

2.5

3

3.5

4

5/1/20058/1/2005

11/1/2005

2/1/20065/4/2006

Tran

smis

ion

loss

es (C

MS)

0

0.2

0.4

0.6

0.8

1

Act

ive

area

frac

tion

(-)

Transmission lossesActive area fraction


Active Area Fraction f(t)

Results

1457

1459

1461

1463

1465

1467

4/8/2006

5/28/2006

7/17/2006

9/5/2006

10/25/2006

12/14/2006 H

ydra

ulic

hea

d (m

)

Piezo. observation, 37-40 m blsTensio. observation, 40 m blsSimulated

1457

1459

1461

1463

1465

1467

4/8/2005

5/28/2005

7/17/20059/5/2005

10/25/2005

12/14/2005

Hyd

raul

ic h

ead

(m)

CalibrationM = 60 m-1

Evaluation


Sources of Uncertainty – One dimensional model, 3D problem

– Basalt overlain by surficial deposits

– Heterogeneous properties, universal M value may not apply throughout INTEC or INL

– Ignores changes in preferential flow with depth and thickness of hydrogeologic units vary

Regardless, estimates from the simplified SR approach provide useful information


Conclusions and Implications– SR model approximates vertical unsaturated

fluxes through fracture networks

– Parameterization suggests reasonable number of actively connected fractures

– Hypotheses related to eliminating specific recharge sources could be further tested

– Could be combined with Darcian approach for comprehensive model of complex flow system



Ebel, BA, JR Nimmo. 2010. Simple estimation of fastest Ebel, BA, JR Nimmo. 2010. Simple estimation of fastest preferential contaminant travel times in the unsaturated zone: preferential contaminant travel times in the unsaturated zone: Application to Rainier Mesa and Shoshone Mountain, Nevada,Application to Rainier Mesa and Shoshone Mountain, Nevada,HydrolHydrol. & Earth System Sci. Discussions.. & Earth System Sci. Discussions.

Nimmo, JR. 2007. Simple predictions of maximum transport rate Nimmo, JR. 2007. Simple predictions of maximum transport rate in unsaturated soil and rock. in unsaturated soil and rock. Water Water ResourResour. Res. Res..

Nimmo, JR. 2010a. Theory for SourceNimmo, JR. 2010a. Theory for Source--Responsive and FreeResponsive and Free--Surface Film Modeling of Unsaturated Flow. Surface Film Modeling of Unsaturated Flow. Vadose Zone JVadose Zone J..

Nimmo, JR. 2010b. Response to Nimmo, JR. 2010b. Response to GermannGermann’’ss comment on comment on ““Theory Theory for Sourcefor Source--Responsive Responsive …… Unsaturated FlowUnsaturated Flow””. . Vadose Zone J.Vadose Zone J.

Thank you! Thank you!

References Cited:References Cited:

a source-responsive approach for simplified estimates …...

Documents