Welcome to the CLU-IN Internet Seminar
“Opportunities for Bringing Rapidly Emerging Technologies to
Revolutionize Modeling of Chemical Contaminants in Coastal Waters”
Presenter:
Dr. Joel Baker ([email protected])
Moderator:
Kira Lynch, US EPA Region 10 ([email protected])
Agency Seminar Series at US EPA Region 10
Sponsored by
University of Washington Superfund Research Program
Delivered: October 4, 2012, 11:00AM-12:30PM, PDT
1
Visit the Clean Up Information Network online at www.cluin.org
Housekeeping
•
•
•
•
Please mute your phone lines, Do NOT put this call on hold
Q&A
Turn off any pop-up blockers
Move through slides using # links on left or buttons
Download slides as
PPT or PDF
Go to slide 1
Move back 1 slide
Move forward 1 slide
Go to
last
slide
Go to
seminar
homepage
Submit comment or
question
Report technical
problems
• This event is being recorded
• Archives accessed for free http://cluin.org/live/archive/
2
Opportunities for Bringing Rapidly Emerging
Technologies to Revolutionize Modeling of
Chemical Contaminants of Coastal Waters
Dr. Joel Baker
Director, UW Puget Sound Institute
University of Washington Tacoma
3
Introduction and Perspective
May, 1982: Duluth, Minnesota
4
Introduction and Perspective
October, 2012: Tacoma, WA
5
The Information Technology
Revolution
NJTechReviews
6
The Information Technology
Revolution
2010 Map of the Global Internet by Cisco Systems
7
The Information Technology
Revolution
seattle traffic report - Google Maps
http://maps.google.com/maps?oe=utf-8&rls=org.mozilla:en-US:.
To see all the details that are visible on the
screen, use the "Print" link next to the map.
Map data © 2012 Googl e -
8
Modeling Chemical Contaminants in
Aquatic Ecosystems:
Seminal Papers in PCB Modeling
9
Modeling Chemical Contaminants in
Aquatic Ecosystems: Karickhoff et al. 1979
10
Modeling Chemical Contaminants in
Aquatic Ecosystems: Karickhoff et al. 1979
11
Modeling Chemical Contaminants in Aquatic
Ecosystems: Thomann and DiToro, 1983
12
Modeling Chemical Contaminants in Aquatic
Ecosystems: Mackay, 1989
13
Modeling Chemical Contaminants in Aquatic
Ecosystems: Mackay, 1989
14
Modeling Chemical Contaminants in Aquatic
Ecosystems: Gobas and Mackay, 1988
15
Modeling Chemical Contaminants in Aquatic
Ecosystems: Gobas and Mackay, 1988
16
Modeling Chemical Contaminants in Aquatic
Ecosystems: Current Models
R.A. Park et al., 2010
17
Modeling Chemical Contaminants in Aquatic
Ecosystems: NY/NJ Harbor CARP Model
Management Question
è Which sources of contaminants need to be reduced or
eliminated to render future dredged material clean?
18
Modeling Chemical Contaminants in Aquatic
Ecosystems: NY/NJ Harbor CARP Model
19
20
21
Summary of 2,3,7,8-TCDD
interim “clean bed” analysis
22
The Information Technology
Revolution
?
CARP
NJTechReviews
23
Premise of Today’s Talk
Tools to model contaminant behavior and
effects in aquatic ecosystems have not kept up
with the information technology revolution
24
Corollaries
1. We assume that technology is frozen in time
to what tools we had available in grad school
(computers, IT, and analytical chemistry)
2. Innovation and experimentation may be seen
at odds with stability and confidence
25
Wait!
Is this really a problem?
What are we missing with current
models?
26
27
Non-Spherical Cows
1. Phase partitioning in the water column
28
• Use PCB and PAH distribution coefficients
measured in the Chesapeake Bay to explore the
mechanism driving observed variability
– three-phase partitioning?
– slow sorption kinetics?
– highly sorbent particles?
29
[Mass on Filter/Volume Filtered]
Kd= ----------------------------------------------[Mass on XAD/Volume Processed][TSS]
9
8
6
d
Kd
Log
Log K
7
5
4
3
PAHs
PCBs
2
1
3
4
5
6
7
8
9
Log KOW
30
Pyrene
N = 119
Baltimore Harbor
Surface Waters
20
uency
15
31
16
14
Pyrene Fraction Disso
N = 119
Baltimore Harbor Surface W
ncy
12
10
32
deled
ry)
Residual solid phase concentration (ng/g-dry)
Investigating the sources of variability in partitioning
200
100
0
33
Investigating the sources of variability in partitioning
1. The presence of colloids
KDOC = 0.08Kow
High variation due to:
the nature of DOC
the methods used
Environmental Science and Technology, 2000, 34, 4663-4668
34
Investigating the sources of variability in partitioning
1. The presence of colloids
Baltimore Harbor
Dissolved Organic Carbon
DOC, mg/L
DOC (mg/L)
8
70% between 3.5 and 5.5 mg/L
6
4
20
40
60
80
Total Suspended Solids (mg/L)
35
Investigating the sources of variability in partitioning
2. Kinetics of Partitioning
Laboratory PCB congener sorption experiments
• Gas-phase equilibration maintains constant dissolved
PCB congener concentrations.
• Stationary-phase chrysophyte Isochrysis galbana
• 18 congeners studied over 120 hours
C-18or PUFSorbent
5mImpactor
ActivatedCarbonTrap
Gas FlowM
eters
W
ater Saturator
Reactor Vessel
High-PurityAir
HOCEquilibratedBubbles
SamplingPort
TemperatureRegulated
Generator Columns
HeatedBlock
Air Stone
36
500
400
PCB
PCB
PCB
PCB
PCB
#3
#10
#15
#180
#170
PCB
PCB
PCB
PCB
PCB
PCB
#52
#101
#97
#77
#118
#105
200
PCB in Algae (ug/g)
PCB Concentration in Algae
300
100
0
500
400
300
200
100
0
0
20
40
60
80
100
120
Time (hours)
37
U p ta k e R a te C o n s ta n t
L /k g - h r
105
104
103
D e s o rp tio n R a te C o n s ta n t
1 /h r
0 .0 4
0 .0 3
0 .0 2
0 .0 1
5
6
7
Log Kow
38
15 minutes
72 hours
7.5
Log Koc
Observed Log KOC
7.0
6.5
6.0
5.5
5.0
4.5
4.5
5.0
5.5
6.0
6.5
7.0
7.5
Log Kow
39
Investigating the sources of variability in partitioning
3. Types of aquatic particles
1.0
Fraction Plankton
Fraction dissolved pyrene
CHARM01 (Fall '99)
CHARM02 (Winter '00)
0.8
CHARM03 (Summer '00)
0.6
0.4
0.2
0
20
40
60
TSP
40
41
1. Phase partitioning in the Water
Column
The observed variations in dissolved-particulate distributions of PCBs,
PAHs, etc. are large and real.
Although organic colloids likely moderate dissolved HOC concentrations,
DOC does not vary enough to explain the observed partitioning.
In studies with well-characterized solids, sorption kinetics
are sufficiently fast (at least on a log-log plot).
Remarkably large (i.e., order of magnitude) variations in HOC-solid
interactions among particle types.
42
Non-Spherical Cows
2. Interactions among particles
43
Physical characteristics of flocs
• Lower settling
velocity
• Lower bulk density
• Higher contact
area (porosity)
http://www.water-technology.net/contractor_images/cu_water/flocke.jpg
44
How are flocs formed?
Yao and O’Melia (1971)
45
Flocculation and PCB Models
• The model simulated the floc size among 2 to 1000 µm
• The multi-class flocculation model equations are based on the concept
of O’Melia (1982)
• The floc porosity and settling velocity are based on the concept of
Winterwerp (1998)
• The floc settling velocity, floc density, stickiness coefficient, and fraction
of organic carbon (fOC) are calculated simultaneously and temporally at
each class of flocculation particle
• The PCB mass transfer coefficient is varied with floc properties
46
Total Volume Concentration
Particulate PCB
PC B 52
140
F ig u re 6 c , d
120
3
C p ( u g /m
)
100
5 e -4
80
60
40
M odel Cp
M e a s u re d C p
20
4 e -4
0
0
10
20
30
40
50
14
3
T V C ( m/m )
T im e (h o u r)
3 e -4
3
12
3
C d ( u g /m
)
10
2 e -4
8
6
4
M odel Cd
M e a s u re d C d
2
Total1 Suspended
Solids
e -4
0
0
10
20
30
T im e (h o u r)
Dissolved PCB
0
0
10
20
30
T im e (h o u r)
F ig u re 3 d
40
40
50
P re d ic te d T V
M e a s u re d T V
50
47
Non-Spherical Cows
3. Chemical release during resuspension
48
Desorption Rates
Engineering Performance Standards for Dredging
Volume 2: Technical Basis and Implementation of the
Resuspension Standard
Given the length of time required for PCBs to reach equilibrium for desorption, it is
unlikely that there will be large release of dissolved phase PCBs as a result of dredging
activities.
• Analysis assumes first order desorption kinetics during the
first day of resuspension
• Experiments show rapid (nearly instantaneous) release at
onset of resuspension
49
Objectives
• What is the initial release of PCBs from
quiescent river sediment when it is
resuspended (i.e. during high flow or
dredging)?
• How does the frequency and duration of
resuspension events affect PCB desorption?
50
PCB Release from Sediment
• Particulate-bound
• Tracks sediment movement
• Reduced bioavailability(?)
• Engineering controls: solids management
• Dissolved
• Tracks water movement
• Directly bioavailable
• Engineering controls: readsorption (?)
51
Release of Dissolved PCBs from Sediment
• Diffusion
• Bioturbation
• Resuspension
• Amount of sediment resuspended
• Residence time of the particles in the water
column
• Desorption rate
52
Methods: STORM Tanks
• The 1000L tanks produce high levels of bottom
shear stress without generating excessive water
column turbulence
53
Dissolved PCB 49
12
PCB 49 (ng/L)
10
8
6
4
2
0
1
2
3
4
5
6
resuspension time [hours]
54
Release of Resuspended PCBs into the
Dissolved Phase
• After 1 hour of resuspension
– First Resuspension: 20%
– Second and Third Resuspensions: 15%
• After 6 hours of resuspension
– First Resuspension: 40%
– Second and Third Resuspensions: 25%
55
Observations
•
After only one hour, resuspension of 7.4 mg/kg t-PCB
Hudson River sediment under gentle conditions yields:
–
–
–
34 mg/L suspended solids
75 ng/L dissolved t-PCB
300 ng/L particulate t-PCB
•
20% of the PCB mass resuspended is desorbed into the
truly dissolved phase in one hour
•
Higher levels of suspended solids and higher t-PCB
levels in sediments will result in larger dissolved
concentrations
56
Observations
•
A fine fraction of the sediment enriched in t-PCBs is
readily resuspended and does not resettle over 12
hours. This material will likely be transported
downstream.
•
Both desorption kinetics and observed PCB behavior
during resettling are consistent with PCB release being
dominated by fine-grain particles.
57
Lessons Learned (so far…)
1.“Don’t make me come out of retirement to come back here to fix
the loadings estimates” – R. Thomann
2.“Sediment transport is a side show” – D. DiToro
Keep your eye on the ball
3.“If a simulation won’t finish overnight the model is too complex”
The modeling effort must generate something that fits on a
manager’s laptop
4.Complex systems require continual review during development
Building inspectors
58
Final Thoughts
Complex models are too expensive to develop and run too slowly to be useful
Moore’s Law and Silicon Qubits
You can’t calibrate a highly resolved model
Self-learning using real-time observations?
Sediment transport is too hard to model
In situ PSD measurements and highly resolved hydrodynamics
Nobody understand complex models
Pixar studios
59
Dr. Joel Baker
Director, UW Puget Sound Institute
University of Washington Tacoma
[email protected]
60
Links page
• Dr. Joel Baker ([email protected])
• Center for Urban Water at University of Washington Tacoma:
http://www.tacoma.uw.edu/center-urban-waters
• University of Washington Superfund Research Program:
http://depts.washington.edu/sfund/
• US EPA Region 10:
http://www.epa.gov/aboutepa/region10.html
• National Institute of Environmental Health Institute (NIEHS)Superfund Research Program
http://www.niehs.nih.gov/research/supported/srp/
61
Thank you for your time!
Please click here to give the UW-SRP your feedback!
Need confirmation of your participation
today?
Fill out the CLU-IN feedback form and
check box for confirmation email.
If you have additional questions or comments, please contact:
Katie Frevert, University of Washington Superfund Research Program (UW-SRP)
[email protected]
Tel (206)685-5379
62
New Ways to stay connected!
• Follow CLU-IN on Facebook,
LinkedIn, or Twitter
https://www.facebook.com/EPACleanUpTech
https://twitter.com/#!/EPACleanUpTech
http://www.linkedin.com/groups/Clean-UpInformation-Network-CLUIN-4405740
63