Modelling the response of fish to major infrastructure upgrades in wastewater treatment plants

Mark ServosCanada Research Chair in Water Quality Protection

University of Waterloo

Rainbow darter (Etheostoma caeruleum)

Arlos, M.J., W.J. Parker, J. Bicudo, P. Law; K.A. Hicks, M. Fuzzen, S. Andrews, M.R. Servos. 2018a. Modeling the exposure of wild fish to endocrine active chemicals: potential linkages of total estrogenicity to field-observed intersex. Water Research 139:187-197.

Arlos, M.J., W.J. Parker, P. Law, J. Bicudo, P. Marjan, S.A. Andrews, M.R. Servos. 2018b. Multi-year prediction of estrogenicity in municipal wastewater effluents. Science of the Total Environment 610-611C:1103-1112.

Patricija Marjan, Meghan Fuzzen, Keegan Hicks,Maricor Arlos, Katie McCann, Alex Crichton, Samantha Deeming, Paulina Bahamonde, Brendan Smith, Chris Robinson, Carolyn Brown, Jennifer

Ings, Gerald Tetreault, Heather Loomer, Jenn Kormos, Ken Oakes, Xu Zhang, Chloe Wang, Paul Togunde, Hadi Dhiyebi, Leslie Bragg, and many others

Also the generous support of our partners: GRCA, Region of Waterloo, City of Guelph, OMECC, ECCC, NSERC, CWN, CRC,CREATE WATER, GWF, etc.

GuelphWaterloo

Lake Erie

Grand River Watershed in Southern Ontario

McMaster

The Grand River Watershed, southern Ontario, Canada Drains into Lake Erie – 6,985 km2

Population ~one million 70% agricultural Inputs from 30 municipal WWTPs

50,000 to 100,000 people served

Drinking Water Treatment Plant

Sewage Treatment Plants

>100,000 people served

5000 to 50,000 people served0 to 5000 people served

Wastewater Treatment Plants in the central Grand Riverare a major concern for water quality?

Waterloo (2o partial-nitrifying)(major upgrades delayed until 2017)

Kitchener (2o non/partial-nitrifying)(major upgrades in 2012 and continuing)

Rapidly growing population >1 M people

Galt Cambridge

Paris

Brantford

Grand Valley

Fergus

Elmira

Elora

Caledonia

Guelph (3o nitrifying)

Toxicant Cellular Organism Population Physiological

vtg expression (gene expression studies)

Clear biological signals across many levels of biological organization

7

Community Ecosystem

Steroid hormone production, delayed sperm development

Intersex conditions

Changes in community assemblages

Processing of nutrients in the food web

Bahamonde et al. 2014Fuzzen et al. 2016Marjan et al. 2017

Tetreault et al 2011Fuzzen et al. 2016Marjan et al. 2018

Tetreault et al 2011Tanna et al. 2011Fuzzen et al. 2015Hicks et al. 2017

Tetreault et al 2011Hicks et al. 2017b

Loomer et al.2016Hicks et al. 2017c

Reduced reproductive

success (lab study)Fuzzen et al. 2015

Male rainbow darter

Female rainbow darter

Intersex condition is the most consistent

Toxicant Cellular Organism Population Physiological Community Ecosystem

Healthy Male Exposed Male

Intersex

Fuzzen et al. 2016, PLOS One

Continuing to following responses to upgrades, 2007-2019

2018

Waterloo (delayed to 2017)

Kitchener (major upgrade in 2012 and continuing

Date

July 2009 Fall 2010 July 2011 Fall 2013 Fall 2014 Fall 2015

Coce

ntrat

ion (

ng/L

E2e

q)

0

5

10

15

20

25A

B

C

AB

D

BC

Date

Fall 2010 Fall 2011 Fall 2012 Fall 2013 Fall 2014 Fall 2015

Conc

entra

tion (

ng/L

E2e

q)

0

2

4

6

8

10

12

14

16

A

A

BB B

C

Kitchener

Waterloo

2006

2007

2008

2009

2010

2011

2012

2013

2014

Ann

ual T

onn

age

0

100

200

300

400

500

600

700

Total ammonia Total nitrate

2006

2007

2008

2009

2010

2011

2012

2013

2014

Ann

ual T

onna

ge

0

100

200

300

400

500

600

700

Total ammoniaTotal nitrate

Estr

ogen

icity

-YE

S Upgrades

Intersex incidence in rainbow darter in the Grand River

Hicks et al., 2017ES&T 51(3): 1811-1819

Ethinylestradiol(Synthetic estrogen)

Natural and synthetic estrogens

Estradiol Estrone Estriol

Lab

Field?Effects

Directed Assessment

Toxicant Cellular Organism Population Physiological Community Ecosystem

Highly associated with the presence of estrogens

Very difficult analysis

SOURCE TRANSPORT & FATE EFFECTS

Modeling the exposure of rainbow darters)to endocrine disrupting chemicals: linkages of stressor concentrations (estrogens) to physiological consequences?

1 2 3

Mode

lling

Part

1: So

urce

Multi-year prediction of estrogenicity in municipal wastewater effluents

Population Demographic Profile

Usage Rates Excretion Rates Sewer Conversionof E2 to E1

Influent Estimates Effluent EstimatesWWTP Removal Rate

𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸 𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸 (𝐸𝐸𝐸𝐸𝐸𝐸)𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚 = �𝐶𝐶𝑚𝑚 × 𝐸𝐸𝐸𝐸𝐸𝐸

Jan-07 Jan-08 Jan-09 Jan-10 Jan-11 Jan-12 Jan-13 Jan-14 Jan-15 Jan-16

Am

mon

ia (m

g/L)

0

10

20

30

40b)

Jan-07 Jan-08 Jan-09 Jan-10 Jan-11 Jan-12 Jan-13 Jan-14 Jan-15 Jan-16

Am

mon

ia (m

g/L)

0

10

20

30

40b)

Kitchener

Waterloo

Ammonia (indication of treatment)

Post-upgrade

Mode

lling

Part

1: So

urce

Jan-07 Jan-08 Jan-09 Jan-10 Jan-11 Jan-12 Jan-13 Jan-14 Jan-15 Ja

E2

Equ

ival

ence

(ng/

L)

0

5

10

15

20

25

(i)

(ii)

a)

Jan-07 Jan-08 Jan-09 Jan-10 Jan-11 Jan-12 Jan-13 Jan-14 Jan-15 Jan-16

E2

Equ

ival

ence

(ng/

L)

0

2

4

6

8

10

12

14

16

(i)

(ii)

a)

Kitchener

Waterloo

Estimated Total Estrogenicity (E2 Equivalency)

Post-upgrade

Mode

lling

Part

1: So

urce

𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸 𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸 (𝐸𝐸𝐸𝐸𝐸𝐸)𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚 = �𝐶𝐶𝑚𝑚 × 𝐸𝐸𝐸𝐸𝐸𝐸

26-Feb-10

14-Sep-10

14-Aug-10

12-Sep-11

18-Oct-12

17-Jun-132-Jul-13

20-Aug-13

6-Nov-13

1-Dec-13

21-Jan-15

18-Mar-15

16-Sep-15

19-Nov-15

E2

Equ

ival

ence

(ng/

L)

0

5

10

15

20

25

30 Measured - YES assayPredicted

7-Jul-09

14-Sep-10

13-Jun-113-Jul-11

11-Nov-13

18-Mar-14

9-Dec-14

23-Jan-15

18-Mar-15

18-Sep-15

19-Nov-15

E2

Equ

ival

ence

(ng/

L)

0

5

10

15

20

25

30Measured - YES Assay Predicted

Kitchener

Waterloo

Mode

lling

Part

1: So

urce

Post-upgrade

Mode

lling

Part

2: Ri

ver F

ate

1 2 3

Model Segmentation

• 50 segments• 80 km• 4 MWWTPs• 1 water withdrawal• 5 creeks and tributaries• 8 years • 9 sites with biological

data

Hydraulics Validation

• Water levels• Chloride concentration

Model Platform

• WASP 7.5 by the US EPA

Mode

lling

Part

2: Ri

ver F

ate

c) Chloride at Segment 50

Jan 07 Jan 08 Jan 09 Jan 10 Jan 11 Jan 12 Jan 13 Jan 14

Con

cent

ratio

n (m

g/L)

0

50

100

150

200

250

300

ObservedPredicted

b) Chloride at Segment 21

Jan 07 Jan 08 Jan 09 Jan 10 Jan 11 Jan 12 Jan 13 Jan 14

Con

cent

ratio

n (m

g/L)

020406080

100120140160180

Observed Predicted

a) WL at Segment 37

Jan 07 Jan 08 Jan 09 Jan 10 Jan 11 Jan 12

Wat

er L

evel

(m)

0.0

1.0

2.0

3.0

4.0

5.0PredictedObserved

Hydraulics Validation

• Water levels• Chloride concentration

Mode

lling

Part

2: Ri

ver F

ate

Waterloo

Kitchener

2007 2008 2009 2010 2011 2012 2013 2014 2015 2016

Estr

ogen

Con

cent

ratio

n, n

g/L

0

2

4

6

8

10

12

14

Flow

(m3 /

s)

0

40

80

120

160

200

E1 E2 EE2 River Flow

2007 2008 2009 2010 2011 2012 2013 2014 2015 2016

Estr

ogen

Con

cent

ratio

n, n

g/L

0.0

0.5

1.0

1.5

2.0

2.5

Flow

(m3 /

s)

0

40

80

120

160

200

E1 E2 EE2 River Flow

a) Segment 23

b) Segment 42

• Very dependent on flow and treatment change• Transformation important for some chemicals

Mode

lling

Part

2: Ri

ver F

ate Temporal Patterns

Waterloo

Kitchener

June 1, 2012Pre-upgrade

Mode

lling

Part

2: Ri

ver F

ate Spatial Patterns – Summer Low Flow

Waterloo

Kitchener Kitchener

Waterloo

June 1, 2012Pre-upgrade

June 1, 2014Post-upgrade

Mode

lling

Part

2: Ri

ver F

ate Spatial Patterns – Summer Low Flow

Waterloo

Kitchener Kitchener

Waterloo

Mode

lling

Part

3: Lin

kage

s to E

ffect

sModelling the exposure of rainbow darters to estrogens: comparison to field data?

1 2 3

0 .0 0 1 0 .0 1 0 .1 1 1 0 1 0 00

2 0

4 0

6 0

8 0

1 0 0

T o ta l E s tro g e n ic ity , n g /L E 2 E q u iva le n ce

Inte

rse

x I

nci

de

nce

, % 2 3

1 2

4 4

3 54 2

5 0

3 1

2 0

0 8

0 .0 0 1 0 .0 1 0 .1 1 1 0 1 0 00

2

4

6

T o ta l E s tro g e n ic ity , n g /L E 2 E q u iva le n ce

Inte

rse

x S

eve

rity

2 3

1 2

4 4

3 54 2

5 0

3 1

2 0

0 8

a )

b )

Mode

lling

Part

3: Lin

kage

s to E

ffect

s

IntersexWaterloo 2016

Segment

Moving Forward• Studies that follow changes in the Grand River

in response to the on-going major MWWTP upgrades (NSERC CRD).

• Refine (processes) and expand (spatially) the models

• Models can now be used to test scenarios:• additional emerging contaminants?• effectiveness of treatment,• impacts of population growth, • changes in hydrology (e.g., related to

climate change). • Extend predictions to other watersheds,

cumulative effects.

Major Assumption

28

The estrogens estrone (E1), estradiol (E2), and ethinylestradiol (EE2) are the major contributors to the total estrogenicity

0.01

0.10

1.00

10.00

E2

Equ

ival

ence

(ng/

L)

E2 EE2E1

Effec

ts di

rect

ed an

alys

is

1E-10

1E-09

1E-08

E2 E

quiv

alen

ts (n

g/L)

Retention Time (min)

E2EE2E1 DES

OP NPE3

BPA

Kitchener WWTP effluent in 2012

Standard Mixture