Cleaning Up a Salt Spill:

Predictive Modelling and Monitoring Natural Attenuation to Save Remedial Costs

Outline

• Introduction• Remedial Alternatives• Alternate Approach: Site Assessment and Predictive

Modelling• Importance of Predictive Modelling for Remediation by

Natural Attenuation

Introduction

• Site located in central Alberta.• Pipeline break on September 6, 2002 caused by corroded

pipe.• Approximately 5,250 m2 affected area.• Initial spill response (by others):

– standing fluid (produced water and oil) removed by vacuum truck;

– trenches excavated along and downslope of spill area;– limited soil and trench water quality data collected; and– geophysics survey (EM 38).

• Shallow groundwater table present.

Site Plan

Remedial Alternatives (16,000 m3 in situ)

• Traditional dig and dump ($2.1M to $2.5M).• Pump and deep well disposal ($1.0M to $1.3M).

EM38 Survey (September 2002)

Base: ESSIS 2002 EM38

EM38 Survey (September 2002, 2003)

Site Assessment

– Electromagnetic (EM) survey 2004.– Groundwater flow regime.– Soil and groundwater quality data.– Vegetation survey.– Predictive modelling vs. observed water quality.

Site Assessment: EM Survey (2003, 2004)

Site Assessment: Groundwater Flow Regime

K=1 m/dayI = 5%

Site Assessment: Groundwater Flow Net

Site Assessment: Soil Quality Data

• Fall 2002 (spill response info in source area):– maximum chloride (Cl) concentration of

39,100 mg/Kg; and– maximum sodium adsorption ratio (SAR) of

49.7 dS/m and electrical conductivity (EC) of 29 dS/m.

Site Assessment: Soil Chloride Concentration (Spring 2005)

(source area) (high EM38 area)

Site Assessment: Soil SAR Concentration (Spring 2005)

(source area) (high EM area)

Site Assessment: Groundwater Quality Data

Spring 2005• Spill Area:

– chloride in shallow well (523 mg/L); and– chloride in deep (bedrock) well (27 mg/L).

• High EM38 Area:– chloride in shallow well (747 mg/L); and– chloride in deep (bedrock) well (6 mg/L).

Site Assessment: Dissolved Chloride Concentration Profile Along Plume

C o ncentrat io n vs D istance

05MW10

05MW0905MW06

05MW04

05MW11

0

100

200

300

400

500

600

700

800

900

0 20 40 60 80 100 120 140 160 180 200D istance fro m pipeline break area

2005

Site Assessment: Vegetation Assessment

• White spruce trees dying because of waterlogged conditions.

• Waterlogged conditions predate 2002 pipeline spill.• Some regeneration occurring, but growth is inhibited

because of salt water spill.• Restoration plan for site reclamation provided to

client.

Site Assessment: Air Photo Review

Drainage direction

Site Assessment: Air Photo Review

Site Assessment: Air Photo Review

Site Assessment: Air Photo Review

So How Will We Clean Up This Site?

• Trees not killed by salt spill.• Chloride concentrations not affecting bedrock aquifer.• Chloride plume moving, but decreasing over time.• What’s the process? Can this be sustained and at

what rate?

Conceptual Salt Leaching Model

Chloride in soil and

groundwater

Chloride in soil and

groundwaterMixing Zone

Chloride Transport rate

Q(infiltration rate)

Conceptual Transport Model

Groundwater Flow into Centroid = Vi Final

Volume = Vi + Vr +Vu

Infiltration Through Recharge = Vr

Upward Groundwater Flow = Vu

Mi/Vi C1

Vr,1

Vu,1

C3

Vu,2 Vu,3

C2

Vr,2 Vr,3

Conceptual Transport Model

Time Step 1 Time Step 2 Time Step 3

Model Calculation Results

Dissolved Chloride Concentration Profile Along Plume

39100

5661

706

61

20

1

10

100

1000

10000

100000

0 50 100 150 200Distance from Pipeline Break Area

Chl

orid

e C

once

ntra

tion

(mg/

L)

Drinking Water Guidelines

Fall 2002 (Estimated)

Fall 2003 (Predicted)

Fall 2004 (Predicted)

Fall 2005 (Predicted)

Fall 2006 (Predicted)

Groundwater Quality

Dissolved Chloride Concentration Profile Along Plume

05MW11

05MW04

05MW06 05MW09

05MW10

0100200300400500600700800900

0 50 100 150 200Distance from Pipeline Break Area

Chl

orid

e C

once

ntra

tion

(mg/

L)

Spring 2005

Spring 2006

Fall 2006

Drinkingw ater Guidelines

Model Calculation Results (Comparison to Analytical Data)

Dissolved Chloride Concentration Profile Along Plume

05MW11

05MW04

05MW06 05MW09

05MW10

0100200300400500600700800900

0 50 100 150 200Distance from Pipeline Break Area

Chl

orid

e C

once

ntra

tion

(mg/

L)

Spring 2005

Spring 2006

Fall 2006

Drinkingw ater Guidelines

Fall 2004 (Predicted)

Fall 2005 (Predicted)

Fall 2006 (Predicted)

Fate and Transport: What processes are occurring?

• Upward vertical groundwater flow direction (Vu):– salts not being transported deeper; and– enhances soil flushing.

• Horizontal flow 45 m/year (Vi):– provides natural flushing capacity.

• Infiltration rate (Vr):– provides groundwater recharge; and– enhanced by ponding (drainage course

disrupted).

Fate and Transport: So what does this tell us?

• Met remedial objectives within four years of the spill.• Natural attenuation effective for remediating this salt

spill.

Revisit to Remediation Options

• Traditional Dig and Dump ($2.1M to $2.5M).• Pump and Treat $1.0M to $1.3M.• Natural Attenuation:

– $50,000 for site assessment;– $10,000 for predictive modelling; and– $20,000 for ongoing EM survey and groundwater quality

monitoring (3 years).• Preferred Remediation Option?

– cost Saving of $900,000 to $2.3M;– remediation complete within 4 to 6 year timeframe; and– minimal surface disturbance.

Important Considerations for Predictive Modeling

• Site Assessment: EM survey and attenuation with time (mass spreading with time).

• What’s the process? Can this be sustained and at what rate?

• Impact of Cl off site?

Next Steps for the Site

• Continue monitoring (EM survey, groundwater quality).• Evaluate EC/SAR in soils and need for amendments.• Engage stakeholders [Alberta Environment (AENV)/landowner].• Implement restoration plan:

– improve surface drainage after remediation goal achieved; and

– revegetate area and weed control .

Questions??