development and application of a multi-media screening ......2010/11/30 · limnotech, ann arbor,...

Atmospheric Toxics Webinar SeriesGreat Lakes Air Deposition ProgramNovember 30, 2010

J.V. DePinto, Todd M. RedderH. Tao, C.L. Turner

LimnoTech, Ann Arbor, MI

D.L. Swackhamer, D.L. CarlsonEnvironmental Health Sciences, University of Minnesota

Development and Application of a Multi-media Screening Model for

Chemicals of Emerging Concern in the Great Lakes Basin (GLMOD)

1

Presentation Outline

Project Objectives and OverviewGLMOD development

Conceptual modelConfiguration to Great LakesObject-oriented approach

Historical PCB field testing (6 congeners)Loads & boundary conditionsModel-data comparisons

Diagnostic simulations to demonstrate value of GLMOD

PCB congenersPBDE congeners

Summary

2

GLMOD Project Objectives

Develop a physically-based, multi-media, time-dependent model to support assessment and management of chemicals of emerging concern in Great Lakes BasinTier 2 screening model

Establish quantitative source-receptor relationshipsIdentify critical source areasCompare potential exposure pathwaysIdentify locations of highest exposure and riskEvaluate temporal trends

Assist in prioritizing research and monitoring programs

3

GLMOD Compartments and Inter-Media Pathways

•Wet •Deposition

•p•-•Dry •Deposition

•p•-•Resuspension

•Convection

•Convection

•Convection

•p,d•-•Decay

•p,d•-•Decay

•Ground •Water

•d•-•Interflow

•d•-•Baseflow

Air

Vegetation SoilLake Segment

Sediment

Surface Vegetation

VadoseZone

Surface Layer

Active Layers (4-cm x 3 layers)

Boundary Layer

Troposphere

Deep Sediment (not modeled)

Ground Water

Emissions toatmosphere

Discharges towatershed

Direct discharges to lakes

Fish Sub-Model •Bioaccumulation factors (BAFs)

Effects Sub-Model•Ecological•Human health

Conc. in Water

4

Schematic diagram of GLMOD intra- and inter-compartmental processes

Air

Vegetation Soil Surface water

Sediment

Surf

ace

Vege

tatio

n

Vado

seZo

neSu

rfac

e La

yer

Upp

er

Mix

ed

Laye

r

Activ

e La

yer

Boun

dary

La

yer

Trop

osph

ere

Dee

p Se

dim

ent

(not

m

odel

ed)

Stra

tosp

here

(not

m

odel

ed)

p-Settling

p-Resuspension

t-Net Runoff(d-Runoff, p-Erosion)

Notes:d- = dissolved/vapor phase processp- = particulate/solid phase processt- = total (dissolved + particulate)

d-Pore Water Diffusion

p-Burial

d-Diffusion

d-Diffusion

d-Absorption(Volatilization)

p,d-Wet Deposition

p-Dry Deposition


p,d-Decay

d-Sorptionp-Desorption



d-Percolation

t-Net Washoff(d-Washoff, p-Litterfall)


d-Diffusion

d-Absorptionp,d-Wet

Depositionp-Dry

Deposition

d-Diffusion

d-Absorptionp,d-Wet

Depositionp-Dry

Deposition

p-Resuspension

Convection

Convection

Convection

p,d-Wind-Driven

Advection

p,d-Point Source

Emissions

p,d-Point Source Discharges

p,d-Flow-Driven Advection

p,d-Flow-Driven

Advection

p,d-Wind-Driven

Advection

p,d-Decay

p,d-Decay

d-Dispersion& Diffusion

Dispersion& Diffusion


Ground Water

d-Interflow

d-Baseflow

Air

Vegetation Soil Surface water

Sediment

Surf

ace

Vege

tatio

n

Vado

seZo

neSu

rfac

e La

yer

Upp

er

Mix

ed

Laye

r

Activ

e La

yer

Boun

dary

La

yer

Trop

osph

ere

Dee

p Se

dim

ent

(not

m

odel

ed)

Stra

tosp

here

(not

m

odel

ed)

p-Settling

p-Resuspension

t-Net Runoff(d-Runoff, p-Erosion)

Notes:d- = dissolved/vapor phase processp- = particulate/solid phase processt- = total (dissolved + particulate)


p-Burial

d-Diffusion

d-Diffusion

d-Absorption(Volatilization)

p,d-Wet Deposition

p-Dry Deposition


p,d-Decay




d-Percolation

t-Net Washoff(d-Washoff, p-Litterfall)


d-Diffusion

d-Absorptionp,d-Wet

Depositionp-Dry

Deposition

d-Diffusion

d-Absorptionp,d-Wet

Depositionp-Dry

Deposition

p-Resuspension

Convection

Convection

Convection

p,d-Wind-Driven

Advection

p,d-Point Source

Emissions

p,d-Point Source Discharges

p,d-Flow-Driven Advection

p,d-Flow-Driven

Advection

p,d-Wind-Driven

Advection

p,d-Decay

p,d-Decay


Dispersion& Diffusion


Ground Water

d-Interflow

d-Baseflow

5

Conceptual mass balance diagram for surface water/sediment compartments

Surface water

Sediment

Upp

er

Mix

ed

Laye

r

Activ

e La

yer

Dee

p Se

dim

ent

(not

m

odel

ed)

p-Settlingp-Resuspension

p,d-Runoff from

watershed

Burial

d-Diffusion

d-Volatilization d-Absorptionp,d-Wet Deposition

p-Dry Deposition


p,d-Decay

d-Sorption

p-Desorption

p,d Advection-dispersion to

downstream WC segmentp,d Advection-

dispersion from upstream WC

segment

Total Chemical

ParticulatePhase

Chemical

DissolvedPhase

Chemical

p,d Direct Point Source

p,d-Decay

d-Sorption

p-Desorption

Total Chemical

ParticulatePhase

Chemical

DissolvedPhase

Chemical

6

GLMOD Segmentation: Air, Water, & Land

7

GLMOD Implementation - Object Model

“Watersheds” Class

•geometry & properties•forcing functions

“WShed Chemicals” Class

•chemical mass/conc.•media-specific props.

“Chemicals” Classglobal chemical props.

“Lake Segments” Class


“Lake Chemicals” Class


“Air Interfaces” Classlake/wshed interfacial relationships

“Air Segments” Class


“Air Chemicals” Class


8

GLMOD Development and Field Testing

Field test model using six PCB congeners:Wealth of historical PCB data (e.g., Hornbuckle, et al. 2005)Use PCB congeners as “tracers” for constraining inter- and intra-media transfer/transport ratesDiagnostic analysis of transport, fate, and bioaccumulation as function of chemical properties and basin geography

Congener

Substitution Pattern

# Chlorines

# ortho Chlorines

Molecular Mass

(g/mol)

Molar Volume

(cm3/mol)

Henry's Law Constant KH

(Pa m3/ mol)

Enthalpy of KH

(kJ/mol)

Octanol-Water Partitioning

(log KO W)

Enthalpy of KO W

(kJ/mol)

Organic Carbon

Partitioning (log KO C)

PCB 18 2,2',5 3 2 257.55 230.1 38 35 5.31 25 4.86

PCB 52 2,2',5,5' 4 2 291.99 247.6 35 31 5.79 20 5.35

PCB 118 2,3',4,4',5 5 1 326.44 265.1 10 50 6.64 20 6.23

PCB 180 2,2',3,4,4',5,5' 7 2 395.33 300.1 12 144 7.28 10 6.89

PCB 201 2,2',3,3',4,5',6,6' 8 4 429.77 317.6 16 145 7.34 5 6.95

PCB 206 2,2',3,3',4,4',5,5',6 9 3 464.22 335.1 10 191 7.97 0 7.60

9

Historical PCB Field Test Simulation

Air emission loading:Population, yield-based loadSeasonal variations, 13% / year decline

Tributary loading:Population, yield-based load

based primarily on LMMBS data

Exponential decline of 15% / year“Hot spots” represented with higher yield coefficients (e.g., Fox River)

10

100

1000

100 1000 10000 100000 1000000 10000000

Local Population

Ann

aul A

vera

ge P

CB

Con

cent

ratio

n, p

g/m

3

0.000

0.002

0.004

0.006

0.008

0.010

0 100 200 300 400

Population Density (#/km2)To

tal P

CB

Loa

d Yi

eld

(kg/

km2/

yr)

6 congeners (18, 52, 118, 180, 201, 206)Historical simulation of PCB washout(1980–1999)

10

0

1,000

2,000

3,000

4,000

5,000

6,000

7,000

8,000

9,000

10,000

1948 1953 1958 1963 1968 1973 1978 1983 1988 1993 1998

Year

Lake

Ont

ario

Out

flow

(m3 /s

)

Model Data

Absolute relative error = +3.9%for 1948-1998 period

Results of Basin Hydrology Sub-model

11

PCB Hindcast Field Test Results: Lake Michigan (south): Congener 52

0.0

0.5

1.0

1.5

2.0

2.5

3.0

1980 1985 1990 1995 2000 2005Year

Log[

Con

cent

ratio

n (n

g/g)

]Fish dataGLMOD simulation

-1.0

-0.5

0.0

0.5

1.0

1980 1985 1990 1995 2000 2005Year

Log[

Con

cent

ratio

n (p

pb)]

Sediment dataGLMOD Simulation

Fish

Res

ults

Sedi

men

t Re

sult

s

12

PCB Hindcast Field Test Results: Lake Ontario: Congener 118

Fish

Res

ults

Sedi

men

t Re

sult

s

0.00.51.01.52.02.53.03.5

1980 1985 1990 1995 2000 2005Year

Log[

Con

cent

ratio

n (n

g/g)

]Fish dataGLMOD simulation

-1.0-0.50.00.51.01.52.02.5

1980 1985 1990 1995 2000 2005Year

Log[

Con

cent

ratio

n (p

pb)]

Sediment dataGLMOD Simulation

13

Diagnostic Model Analysis Demonstrates:

Quantifying the relative distribution of chemicals among media (air, water, sediment, fish, etc.) and specific geographic segments (e.g., Lake Michigan northern basin);Quantifying the residence time of chemicals in various Great Lakes media and segments;Quantifying the potential for long-range transport of chemicals from their release point;Identifying the geographic location of exposure “hotspots”; andQuantifying the rate of change of chemical concentrations in fish, water, and air once air emissions or land-based loadings have changed.

14

PCB Diagnostic Simulations

Diagnostic SimulationsDiagnostic #1: Constant emission loading of 1,000 kg/yr (for each PCB congener) to the air segment above the Chicago metropolitan area for a 20-year period;Diagnostic #2: Constant emission loading of 1,000 kg/yr (for each PCB congener) to the air segment above the Detroit metropolitan area for a 20-year period; andDiagnostic #3: Constant loading of 1,000 kg/yr (for each PCB congener) from the Fox River basin to Green Bay.

Each diagnostic simulates all six PCB congeners in all media for a 20 year period with zero initial conditions

15

Diagnostic #1: Emission to air over Chicago Air Simulation Results (July - year 20)

Constant load of 1,000 kg/yr for each congener

16

Diagnostic #1: Emission to air over Chicago Lake Michigan (south) results

Wat

er R

esul

tsSe

dim

ent

Resu

lts

0.0

1.0

2.0

3.0

4.0

5.0

0 2 4 6 8 10 12 14 16 18 20

Simulation Year

Wat

er C

once

ntra

tion

(pg/

l)PCB #18 PCB #180 PCB #206

0.0

0.5

1.0

1.5

0 2 4 6 8 10 12 14 16 18 20

Simulation Year

Sedi

men

t Con

c. (p

pb)

17

0.10

1.00

10.00

100.00

Superi

or

Michiga

n

Green B

ay

Huron

Sagina

w Bay

St. Clai

r

Erie

Ontario

Fish

Con

cent

ratio

n (n

g/g)

Walleye

Lake Trout

Diagnostic #1: Emission to air over ChicagoFish Simulation Results (after 20 years)

Congener #180

Constant load of 1,000 kg/y for each congener

18

0.01

0.10

1.00

10.00

100.00

Superi

or

Michiga

n

Green B

ay

Huron

Sagina

w Bay

St. Clai

r

Erie

Ontario

Fish

Con

cent

ratio

n (n

g/g)

Walleye

Lake Trout

Diagnostic #2: Emission to air over DetroitFish Simulation Results (after 20 years)

Congener #180


19

0.01

0.10

1.00

10.00

100.00

1,000.00

Superi

or

Michiga

n

Green B

ay

Huron

Sagina

w Bay

St. Clai

r

Erie

Ontario

Fish

Con

cent

ratio

n (n

g/g)

Walleye

Lake Trout

Diagnostic #3: Fox River Load to Green BayFish Simulation Results (after 20 years)


Congener #180

20

Example Emerging Chemical (PBDE) Simulation

Polybrominated diphenyl ethers (PBDEs)class of chemicals used as flame retardants, plastics in consumer electronics, etc.

Seven PBDE congeners simulated#28, #47, #99, #100, #153, #154, #209Used range of measured chemical properties

Synthetic air emission loading scenario:Load proportional to local population densityTotal annual emission load: ~47,200 kg/yrRun to steady-state (40-year simulation)

Predict relative distribution of PBDEs in:Air boundary layerWatershed (soil, vegetation, etc.)Lake surface water, bottom sediments

21

PBDE Diagnostic: Air Concentrations (Congener #100, Year 20 - July)

Emission loading: ~47,200 kg/yr

22

PBDE Diagnostic: Lake Water Results

Water Column Concentration (year 40)

0.00

0.10

0.20

0.30

0.40

Superior Michigan Huron Erie Ontario

Bulk Con

centration

(ng/l)

PBDE #28 PBDE #100 PBDE #209

23

PBDE Diagnostic: Lake Sediment Results

Sediment Mass Inventory (year 40)

0.00

0.50

1.00

1.50

2.00

2.50

Superior Michigan Huron Erie Ontario

Mass in Sed

imen

t (kg/km

2 )PBDE #28 PBDE #100 PBDE #209

24

Value of a Great Lakes Multi-Media Screening Level Model (GLMOD)

Assess the potential basin-wide and location-specific impact of emerging chemicals

Framework within which to synthesize all research and monitoring data

Assess relative importance of various exposure pathways

Based on chemical properties, receptor locationRelative contributions of sources inside and outside the basin

Assess progress toward achievement of specific risk reduction targets of emerging chemicals in the Great Lakes

Estimate concentrations of bioaccumulative chemicals in sport fish and compare them to toxicity benchmarks

Assist in prioritizing research and monitoring programs

determining chemical properties, measuring sources, evaluating exposure pathways, evaluating trends

25

Summary

Developed a physically-based, process oriented multi-media model for transport, exposure, and effects of chemicals in the Great Lakes basin

Prototype developed using object-oriented programming approach

Field testing for six PCB congeners spanning range of Kow

and Henry’s constants

Model can serve as a “tier 2” screening model for chemicals of emerging concern in the Great Lakes Basin

Next StepsUse to synthesize and interpret developing CEC database and prioritize further research and monitoring

Link to CHARM model (NOAA-GLERL) and evaluate climate change scenarios

26

Questions?

Acknowledgements:Funding: EPA-Great Lakes Air Deposition (GLAD) Program

http://www.glc.org/glad/

Project Officer: Jon Dettling, Great Lakes Commission

Sharing Large Basin Runoff Model (LBRM) for hydrology of basin: Tom Croley (NOAA-GLERL)

Contact Information: Joe DePintoLimnoTechAnn Arbor, MI [email protected]

27

development and application of a multi-media screening ......2010/11/30 · limnotech, ann arbor,...

Documents