Characterizing Groundwater - Surface Water Interactions

Characterizing Groundwater –

Surface Water Interactions

Robert Ford

Office of Research and Development

National Risk Management Research Laboratory, Cincinnati, OH and Ada, OK

USEPA Ground Water Forum Meeting, August 8, 2019

Disclaimer

1

The findings and conclusions in this

presentation have not been formally

disseminated by the U.S. EPA and

should not be construed to represent

any agency determination or policy.

SHC 3.61.1 Contaminated Sites - Technical Support

Plan for Presentation

2

• Context for evaluating water and

contaminant flux from upland

groundwater to downgradient surface

water bodies

• Tools for assessing hydraulic pathway

from groundwater to surface water

• Tools Implementation – Site Case

Study (Arsenic)


Conceptual Site Model

3 SHC 3.61.1 Contaminated Sites - Technical Support

Common Scenarios at Contaminated Sites

• There is a GW plume at a site that is near a

surface water body.

‒ Is the GW plume impacting the SW

body or does the potential exist?

• There is an observed impact within a

surface water body adjacent to a

contaminated site.

‒ Is the impact related to GW plume

discharge?





• CSM needs to be informed by knowledge of

several components

‒ Site hydrology

‒ Contaminant transport characteristics

‒ Ecological/human exposure endpoints

• Interaction of these factors dictates location

and magnitude of exposure



Effective CSMs - Site Hydrology Issues

• Hydraulic connection between GW plume and

surface water body

‒ Does it exist?

‒ If so, is it continuous or episodic?

‒ When connected, does the direction of water

exchange vary?

• Questions need to be addressed to understand

timing and location of contaminant discharge



Connected

Gaining

Connected

Losing

Disconnected



• Not uncommon to have deep

unsaturated zone

• May be an episodic situation for

semi-arid climates with extended

dry-wet periods

• Need to develop good

understanding of local GW table

elevation and seasonal variation

‒ Episodic (e.g., quarterly) manual

measurements of GW table

insufficient to assess situation

Disconnected



• GW contribution to SW flow may

vary seasonally or due to

external forces

‒ Flow management in SW body

‒ GW extraction system

operations

• Need to define gaining period &

location in relation to GW plume

• May also vary along reach of SW

body

Losing

Gaining



• Site topography and

stream morphology

influence GW flow

direction and magnitude

• May need to

characterize this spatial

variability relative to GW

plume dimension

• GW is not a static

system, but may

respond more slowly to

changes in water budget

(continuous logging)

Latitu

de

Longitude

Ele

vation Stream

Aquifer

GW Potentiometric Surface



• An effective CSM depends on understanding

contaminant transport

• Typically attempt to combine some level of

knowledge of GW flow with measurements of

contaminant concentrations in GW and SW

• Contaminant non-detects that occur along some

assumed flow path could mean two things:

‒ Plume edge does not reach SW

‒ Monitoring location is not in the flow path

• Hydrologic measurements within the GW/SW

transition zone bridge upland GW-to-SW pathway

Characterization Tools

12

Upland Groundwater


Characterization Tools – Upland GW

13

• Install monitor wells or piezometers

‒ Determine groundwater elevation

‒ Determine aquifer properties

‒ Measure groundwater chemistry

• Determine flow direction and magnitude

‒ Calculate groundwater potentiometric surface

from a network of wells/piezometers (sitewide)

‒ Calculate flow gradient and direction for a

subset of wells/piezometers (targeted)

❖ 3PE: A Tool for Estimating Groundwater

Flow VectorsSHC 3.61.1 Contaminated Sites - Technical Support


14

• EPA 600/R-14/273

September 2014

• Provides background and

technical guidance on

appropriate application of

evaluation technology

• Provides spreadsheet-

based analysis tool for

calculating flow gradient,

velocity, and direction

from measured

groundwater elevations



15

3PE – Three Point Estimator

• Implementation of a three-point mathematical solution

to calculate horizontal direction and magnitude of

groundwater flow

• Applicable within portions of the groundwater flow

field with a planar groundwater potentiometric surface

• Groundwater seepage velocity estimated using

Darcy’s Law

‒ hydraulic gradient from 3PE calculation

‒ estimates of hydraulic conductivity and effective

porosity




“3 Points”

monitor wells/piezometer

locations

“3 Points” – measured groundwater elevations

Estimated/measured

aquifer properties


17

• Example of several

triangles adjacent to SW

• Evaluate seasonal

influences based on

synoptic manual WL

measurements

• Develop a sense of

seasonal and spatial

variability of GW flow



18

• Track time-dependent

changes through use

of logging pressure

transducers

• Modification of 3PE to

view time-dependent

changes in gradient

• 3PE Workbook demo


Characterization Tools

19

GW/SW Transition Zone

(Surface Water Body)


Characterization Tools –

Transition Zone

20

• Qualitative Tools or Approaches (“Where”)

‒ Visual observations in surface water body

(discolorations, sheens)

‒ Spatial chemistry sampling for contaminants or

plume indicators

‒ Spatial geophysical measurements (resistivity,

electromagnetic surveys)

‒ Spatial temperature contrast measurements

Critical first step to defining CSM and devising a

site characterization network



Transition Zone

21

• Quantitative Tools (“How Much & Direction”)

‒ Flow balance calculations to estimate GW

contribution to baseflow (quantity)

‒ Piezometer-Stilling Well installations in surface

water body (direction, quantity estimate)

‒ Seepage meter measurements: snap-shots or

continuous (quantity and direction)

‒ 2D-3D Groundwater-Surface Water flow models

(major undertaking; data intensive)

‒ Quantify Seepage Flux using Sediment

Temperatures



Transition Zone

22

• MWs (6,7,9) and SW

piezometer (8) can be

combined to track flow

into the SW body

• PZ elevations can be

surveyed into the

existing monitoring

network to examine

horizontal gradients



Transition Zone

23

• Nested piezometer-

stilling well

installations also

allow simultaneous

measurement of

vertical gradients

• Helps in defining

potential flow paths

and discharge into

SW body



Transition Zone

24

• Quantitative Tools (“How Much & Direction”)

‒ Flow balance calculations to estimate GW

contribution to baseflow (quantity)

‒ Piezometer-Stilling Well installations in surface

water body (direction, quantity estimate)

‒ Seepage meter measurements: snap-shots or

continuous (quantity and direction)

‒ 2D-3D Groundwater-Surface Water flow models

(major undertaking; data intensive)

‒ Quantify Seepage Flux using Sediment

Temperatures



Transition Zone

25

• EPA 600/R-15/454

December 2014

• Provides background and

technical guidance on

appropriate application of

technology

• Illustrates use of

spreadsheet-based

analysis tools for

calculating seepage flux

magnitude and direction

from sediment

temperature profile dataSHC 3.61.1 Contaminated Sites - Technical Support

Modeling Seepage Flux

26

• Heat conduction

influenced by GW-SW

temperature gradient

• Heat convection

influenced by flow up

(discharge) or flow down

(recharge)

• Shape of temperature

profile influenced by

magnitude and direction

of GW flow

Adapted from: Conant (2004) Ground Water,

42:243-257SHC 3.61.1 Contaminated Sites - Technical Support



10 15 20Temperature

Sediment

Surface

Water

Heat

Conduction

0

De

pth

High

qz

Water

Advection(Heat Convection)

Low

qz

Ave

rage

Gro

un

dw

ate

r Te

mp

era

ture

Ave

rage

Su

rfa

ce

Wate

r Te

mp

era

ture

10 15 20Temperature

0

Dep

th

Ave

rage

Gro

un

dw

ate

r Te

mp

era

ture

Ave

rage

Su

rfa

ce

Wate

r Te

mp

era

ture

Water

Advection(Heat Convection)

Heat

Conduction

High

qz

Low

qz

Adapted from: Conant (2004) Ground Water, 42:243-257

Discharge (flow up) Recharge (flow down)

Temperature Profiles (Summer)


28

• Steady-State and Transient Model Systems‒ temperature contrast across vertical boundaries

‒ sediment properties (heat transport, transmissivity)

‒ direction and magnitude of seepage flow


T0

T1

T2

T3

Dep

th

Temperature Time

Temperatu

re

Discharge

Recharge

Steady-State Transient


29

• Spreadsheet-based models that implement calculations

using several derived analytical solutions

• “Steady-State” Models

‒ Schmidt et al (2007) 2 sediment depths + regional GW

temperature

‒ Bredehoeft and Papadopulos (1965) 3 sediment depths

• “Transient” Models

‒ McCallum et al (2012) 2 sediment depths, diurnal

amplitude ratio and phase shift

‒ Hatch et al (2006) 2 sediment depths, only diurnal

amplitude ratio

• Output from models is equivalent to Darcy Flux (specific

discharge)SHC 3.61.1 Contaminated Sites - Technical Support


30

• Steady-State Workbook - Spreadsheet-based calculation tool


Water & Sediment

Properties

Measured

TemperaturesSensor

Spacing

Calculated

Flux!


31

• Transient Workbook - Spreadsheet-based calculation tool


Water & Sediment Properties

Sensor Spacing

Measured Temperatures (24-hour period)

Calculated

Flux!


32

Temperature Profile Data

• Sensors have non-volatile

memory & programmed

for unattended data

acquisition

• Temperature monitoring

network installed in 1-2

days

• Deployed for 2-3 months

& retrieved in 1 day – data

downloaded and analyzed

using Workbook ToolSHC 3.61.1 Contaminated Sites - Technical Support


33

• Data collection can be configured to allow potential use of

both model types


+60cm

+0cm

-30cm

-60cm

-120cm

-150cm

12 14 16 18 20 22 24

-150

-100

-50

0

50

Temperature (°C)

De

pth

Be

low

Su

rfa

ce

(cm

)

Continuous temperature logs…

Give daily

temperature profiles

Tools Development &

Implementation

34

Standard Operating Procedures (Internal EPA/ORD)

• Upland Groundwater‒ Elevation Surveys (very critical in low gradient areas)

‒ Slug Tests (manual, pneumatic) to assess hydraulic

conductivity of screened aquifer interval

‒ Manual Water Level measurements

‒ Use of Automated Pressure Transducers/Data Loggers for

continuous records of water level measurements

• These measurements all present potential sources of

error that need to be controlled as much as possible

• Presumes that the well/piezometer was properly

constructed and developed to insure representative of

aquifer condition


Tools Development &

Implementation

35

Standard Operating Procedures (Internal EPA/ORD)

• Seepage Flux (Surface Water Body)

‒ Installation of Temporary Piezometers with Stilling Wells to assess

vertical gradient

‒ Thermal Conductivity measurement for saturated sediments

(important model input parameter)

‒ Snap-Shot Temperature Profile measurement for submerged

sediments (still a work in progress; issues with thermal

conduction)

‒ Sediment Temperature Profile Logging using commercial

temperature logging devices (range of options; deployment

configuration is important to insure usable data)

• Current EPA/ORD recommendation is to always try to

collect an independent measure of vertical gradient


Tools Development &

Implementation

36

Steven Acree

Methods and best practices for measuring groundwater

hydraulics; 3PE Workbook (with Milovan Beljin)

Robert Ford

Methods and best practices for measuring seepage flux

in surface water bodies

Bob Lien

Seepage Flux Workbooks

Randall Ross

Equipment development for sediment temperature profile

data acquisition; 3PE Workbook (with Milovan Beljin)SHC 3.61.1 Contaminated Sites - Technical Support

Factors Affecting Contaminant

Transport and Exposure Route


Contaminant Transport Issues

• Contaminant properties dictate whether it will remain

mobile in water, attached to sediment, and/or change

chemical form

‒ Does contaminant partition to aquifer/sediment solids?

‒ Does it biodegrade? Product non-toxic and/or

immobile?

‒ Does chemical form change due to shifts in water

chemistry?

• These factors interact with flow dynamics and may

govern best locations and timing of sampling efforts



38 SHC 3.61.1 Contaminated Sites - Technical Support Chris Eckley (Region 10), Todd Luxton (ORD)

Hydrologic Fluctuations

• Contaminated sediment

exposure to air during

baseflow can affect Hg

chemistry

• Hg-methylation linked to

microbial conversion of

sulfur and organic carbon

• Patterns in Methyl-Hg

production during gaining

periods may be

misinterpreted as GW flux Gaining

Re-submerged

Sulfate ReductionHg Methylation

SW

GW

Losing

Exposure to Air

Sulfide OxidationInorganic Hg

SW

GW

High Flow

Low Flow




Reduced GW Plume

• SW body with varying

water depth in which

oxygen reaches

sediments in shallow

locations but not deep

• Oxidation & attenuation of

Fe and As in sediments

for shallow depths

• Unhindered transport of

As into SW for deeper

depths

Break Time

40

Questions or Discussion?


Case Study

41

Monitoring Contaminant Flux at

GW-SW Transition Zone

• Initial Site Characterization to

Inform Remediation Design

•Monitoring Remedy Performance


Case Study

42

• Historical, un-

lined landfill

• Arsenic

contamination in

GW derived from

waste and natural

sources

• Contaminated

groundwater

discharging to

part of adjacent

recreational lake

SHC 3.61.1 Contaminated Sites - Technical Support Ford et al (2011) Chemosphere, 85: 1525-1537

MoundedMaterial

SanitaryLandfill

Incinerator

Plow Shop Pond

Red Cove

N2

N3RSK8-12N5

N7

N6

SHL-1

SHL-24

SHP-99-35X

SHL-12SHL-17

SHL-15

SHP-95-27XSHL-7

N4

Shepley’sHill

Location of Commercial Development

RailroadYards

N

(Former) Fort Devens Superfund Site

Case Study

43

Monitoring

Approach

• GW hydrology

and chemistry

• Flow gradient and

seepage flux in

cove

• SW chemistry

• Sediment

chemistry


Nested Piezometers, Cove Piezometers

Seepage Flux, Chemistry (Water & Sediment)

N

Case Study

44 SHC 3.61.1 Contaminated Sites - Technical Support EPA-600-R-09-063

Flow Net Analysis – GW Table from Site Wells

N

Case Study

45

• Arsenic plume

flowing from

landfill toward

cove

• Nested

piezometers

used to evaluate

magnitude &

distribution of

arsenic flux

SHC 3.61.1 Contaminated Sites - Technical Support EPA-600-R-09-063

N

Case Study


Picture of cove from north shore Picture at central

cove from boat next

to contaminated

seepage area

April 2007

April 2007

Case Study


Seepage Flux

Aug-Sep 2011

9TB (PZ13)

2.0 ± 0.9 cm/d

5TB (PZ5)

14.3 ± 1.0 cm/d

N

Case Study

48

• Sediment arsenic

concentrations variable

within cove – correlate

with iron

• PZ5 location shows

sustained discharge

with plume chemistry

signature in deep SW

• PZ13 location shows

variable discharge-

recharge & no plume

chemistry signature in

deep SWSHC 3.61.1 Contaminated Sites - Technical Support EPA-600-R-09-063

192215 192220 192225 192230 192235

208

210

212

214

216

218IC SW01 MC SW02B SW04

RCTW 8 RCTW 4RCTW 9

RCTW 10

Easting (meters)

Ele

va

tio

n (

ft A

MS

L)

Contaminated

Sediment

GW DischargeHigh As, Fe, K

Low DO

Sediment RecyclingHigh As, Fe – Low K

Variable DO

192215 192220 192225 192230 192235

208

210

212

214

216


RCTW 8 RCTW 4RCTW 9

RCTW 10

Easting (meters)

Ele

va

tio

n (

ft A

MS

L)

Contaminated

Sediment

192215 192220 192225 192230 192235

208

210

212

214

216


RCTW 8 RCTW 4RCTW 9

RCTW 10

Easting (meters)

Ele

va

tio

n (

ft A

MS

L)

Contaminated

SedimentPZ5PZ13

What influences SW concentrations?

Case Study

49

Non-Time Critical Removal Action


Hydraulic Barrier Wall (2012) Sediment Removal in Cove (2013)

N

Case Study


GW Potentiometric Surface9-10 July 2013 (0.2-ft contour)

N • Monitoring Remedy

Performance

‒ Does remedy influence

GW-SW hydraulics?

‒ Does groundwater

show recovery trend?

‒ Is there a decrease in

groundwater and

contaminant flux?

Case Study


Case Study

52

• GW arsenic concentrations decreasing in aquifer at

primary area of contaminant flux (RSK12)

• Arsenic concentrations less changed southwest of cove

(RSK15)


West of Cove (RSK12) Southwest of Cove (RSK15)

2005 2008 2011 2014 2017

0

200

400

600

800

1000

Upland (water table)

Upland (mid-depth)

Upland (above bedrock)

Ars

enic

(

g/L

) filtere

d

Calendar Year

Barrier Wall

Installed

2005 2008 2011 2014 2017

0

200

400

600

800

1000

Calendar Year

Case Study

53

• GW Flux = (3PE Seepage Velocity) x (Porosity)

• Arsenic Flux = (GW Flux) x (GW Concentration)


0 4 8 12 16 20 24

56

58

60

62

64

66

68

RSK12

RSK11

RSK10

RSK9

RSK8

RSK15

RSK14

RSK13

GW Flux 0.15 m/d (15.2 cm/d)

Kx-y(avg) 19.8 m/d

GW Arsenic 710 g/L (Median)

Arsenic Flux 108 mg / d-m2

Upland Ground Surface

Upland Bedrock Surface

Upland GW Elevation

Cove Piezometer

Cove SW Elevation

Cove Sediment Surface

Ele

vation, m

NA

VD

88

Relative Distance, m

14 September 2011

View from upland out to cove

Case Study


Median Flux Reduction Factors

Flow 2.9 Barium 7.6

Arsenic 4.3 Ammonium 12.8

2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019

0.00

0.04

0.08

0.12

0.16

0.20

Barrier Wall

Calendar Year Groundwater Flux, m / d

0

25

50

75

100

125

150

GW Arsenic Flux, mg / d-m2

Measured

Interpolated

Case Study

55

• Compare upland GW flux to cove seepage flux

‒ Darcy Flux (3PE) = “Effective Porosity” x “GW Velocity”

• Flow conservation indicates independent measures should

be comparable


Shoreline19 August 2008

13.6 cm/d Darcy Flux (3PE)13.8 cm/d Seepage Flux (Temperature)

Case Study


Sediment

Temperature

Profile

Method

Comparison

over entire

monitoring

period…

170 180 190 200 210 220 230 2400

5

10

15

20 Middle of Cove ( June - August )

Pre-Installation ( 2008 )

Post-Installation ( 2014 )

Upland GW Flux

Ca

lcu

late

d S

ee

pa

ge

Flu

x (

cm

/d)

Calendar Day

2008 2009 2010 2011 2012 2013 2014 2015 2016 20170

5

10

15

20

25 GW Flux

Seepage Flux

Wa

ter

Flu

x,

cm

/d

Calendar Year

Case Study


2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019

0

25

50

75

100

125

150

Calendar Year

Interpolated GW Arsenic Flux, mg / d-m2

Measured Cove Arsenic Flux, mg / d-m2

Application Illustration


Outcome

• GW plume diverted by

hydraulic barrier & removal of

existing contaminated

sediment

• Fish nest building observed

immediately after and

continues (2014-2018)

• Performance metric of GW

contaminant flux reduction

was realized and could be

assessed in multiple ways

BEFORE

AFTER

Application Illustration

59

• Methods to assess groundwater flow and

seepage flux are relatively easy to implement

and provide for great flexibility in site monitoring

• There is a range of equipment choices and

mathematical tools that can be matched up with

available resources

• Knowledge gained from determination of water

flux benefits assessments of degradation,

design of reclamation efforts, and monitoring of

restoration success.


Acknowledgements

60

Engineering Technical Support Center

John McKernan, [email protected]

Ed Barth (Acting), [email protected]

Groundwater Technical Support Center

David Burden, [email protected]

EPA Region 1 – Carol Keating, Bill Brandon, Ginny Lombardo, Jerry Keefe, Dan

Boudreau, Tim Bridges, Rick Sugatt, David Chaffin (State of Massachusetts)

Workbook Beta Testing – Region 1 (Bill Brandon, Marcel Belaval, Jan Szaro),

Region 4 (Richard Hall, Becky Allenbach), Region 7 (Kurt Limesand, Robert

Weber), Region 10 (Lee Thomas, Kira Lynch, Bruce Duncan, Piper Peterson,

Ted Repasky), Henning Larsen and Erin McDonnell (State of Oregon)

EPA ORD – Jonathon Ricketts, Patrick Clark (retired!), Kirk Scheckel, Todd Luxton,

Mark White, Lynda Callaway, Cherri Adair, Barbara Butler, Alice Gilliland

US Army – Robert Simeone

Don Rosenberry (USGS – Lakewood, CO) – verification studies at Shingobee

Headwaters Aquatic Ecosystems Project


Supplementary Information

61

Published Seepage Flux Calculation Tools

Software Platform: MATLAB®

Ex-Stream

Swanson and Cardenas (2011) Computers and Geosciences, V37, PP1664-1669.

VFLUX

Gordon et al. (2012) Journal of Hydrology, V420-421, PP142-158.

VFLUX2

Irvine et al. (2015) Journal of Hydrology, V531, PP728-737.

FLUX-BOT

Munz and Schmidt (2017) Hydrological Processes, V31, PP2713-2724.

Software Platform: Excel®

Flux-LM

Kurylyk et al. (2017) Hydrological Processes, V31, PP2648-2661.

Software Platform: FORTRAN with Microsoft Windows™ GUI

1DTempPro

Voytek et al. (2014) Groundwater, V52, PP298-302.

1DTempPro V2

Koch et al. (2016) Groundwater, V54, PP434-439.


Documents

Characterizing Groundwater - Surface Water Interactions