RECEIVING WATER QUALITY MODEL CALIBRATION FOR MWRA CSO PROGRAM

(1)

RECEIVING WATER QUALITY MODEL CALIBRATION FOR MWRA CSO

PROGRAM

Dominique Brocard (AECOM), Rachael Hager (AECOM), Salma Taha (AECOM), Betsy Reilley (MWRA)

(2)

Outline

– Purpose

– Models and Approaches – Charles River Model

– Model Coverage and Discretization – Flow inputs

– Boundary Condition – Model calibration

– Alewife Brook/Upper Mystic River Model – Model Coverage and Discretization – Flow inputs

– Boundary Condition – Model calibration

(3)

Purpose

– To confirm the receiving water quality benefits of the MWRA CSO program predicted by the CSO Long Term Control Plan (LTCP)

– Concentrate on receiving waters with Massachusetts Water Quality Standards Variances

– Charles River

– Alewife Brook / Upper Mystic River – Concentrate on Bacteria

– E. coli

– Enteroccus

– This presentation is limited to the calibration of the water quality models that are being used to assess current conditions and evaluate alternatives

(4)

Models and Approaches

(5)

Models

– Two Models receiving water models:

– Charles River

– Alewife Brook/Mystic River – MWRA CSO model

– Stormwater models

(6)

Calibration Approach

– Calibration Parameters – Die-off rates

– Stormwater counts – Calibration Data

– MWRA stream monitoring – 2017, 2018 and 2019

– 17 station in Charles

– 16 stations in Alewife Brook / Upper Mystic River – Calibration Approach

– Vary calibration parameters within justifiable ranges to achieve the best match with the calibration data

– Document sensitivity

(7)

Calibration Approach

– Weight of Evidence Approach – Peak bacterial counts

– Shape of bacterial count variations with time – Quantitative Assessments

– Average counts

– Wilmot Index of Agreement

• P = predicted

• O = observed

(8)

Water Quality Standards

Parameter

Class B Criteria for Non-Bathing Beach Waters⁽¹⁾ Existing Class B Criteria

6-month Geometric Mean

(colonies/100 mL) Single Sample Maximum (colonies/100 mL)

E. coli 126 235

Enterococcus 33 61

– To put measured/modeled bacterial counts in perspective – Current Standards

– New Standards are forthcoming

(9)

Charles River Model

(10)

Model Coverage and Discretization

– From Watertown Dam to New Charles River Dam – Delft-3D in two-dimensional mode. 4,400 grid cells

(11)

Flow Inputs

– Stormwater from Cambridge, BWSC and USGS models – CSOs from MWRA Collection System Model

(12)

CSO Quality

– Cottage Farm and Prison Point inflow monitoring to characterize untreated CSOs

Very different bacterial counts at Cottage Farm and Prison Point

Cottage Farm⁽¹⁾ Prison Point⁽²⁾

Number of Measurements 31 16

Number of Storms 7 6

E. coli

(#/100 mL) Arithmetic Average 1,306,000 175,000

Enterococcus

(#/100 mL) Arithmetic Average 206,000 52,000

(1) Data collected between October 2017 and August 2019 (2) Data collected between January 2018 and December 2019

(13)

CSO Quality

– Sanitary fraction from Collection System Model

– Bacterial counts in sanitary and stormwater fractions

Sanitary Stormwater Enterococcus 1,000,000 5,600

E.coli 6,300,000 17,000

0 50,000 100,000 150,000 200,000 250,000

11/9/18 12:00 11/10/18 0:00 11/10/18 12:00

Enterococcus Count (#/100 mL)

Prison Point Cottage Farm

Cottage Farm - Measurements Prison Point - Measurements

(14)

Stormwater Quality

Date 10/7/2019 10/27/2019 11/18/2019 11/24/2019 12/13/2019 Averages

Depth (in) ⁽¹⁾ 0.16 1.43 0.24 1.51 1.41

Duration (hr) 2.5 10.5 6 17 17.25

Peak Int. (in/hr)⁽²⁾ 0.16 0.56 0.12 0.6 0.24

Prior Dry Days 2 3 5 1 2.2

E. coli

CAM3 42,000 3,017 4,367 15,650 16,258 16,667

CAM4 542 2,308 11,288 54,167 17,076

Enterococcus

CAM3 6,017 2,465 5,350 9,650 5,870 3,674

CAM4 1,273 1,153 1,603 1,877 1,477

(1) Somerville Marginal Data

(2) 15-min peak intensity

– 2019 Monitoring

(15)

Upstream Boundary Condition

– Flow from upstream brings large quantities of Enterococcus and E. coli

(16)

Upstream Boundary Condition

– Buildup/washoff model based on USGS flows at Waltham Gauge

Build-up Rate Washoff Coeffi-

cient

Washoff Exponent

Die-off Rate

Base Flow Count

Ave Meas.

Ave Model

IA

a (#/mi²/day)

Winter/

Fall Ratio

α β K

(day^-1)

CB

#/100ml

Entero 2017 1.7 x 10¹¹ 0.2 / 0.5 8 x 10^-4 1.4 0.5 45 405 408 0.92 Entero 2018 1.7 x 10¹¹ 0.2 / 0.5 8 x 10^-4 1.4 0.5 45 432 423 0.91 E. coli 2017 3.5 x 10¹¹ 0.2 / 0.5 8 x 10^-4 1.4 0.5 134 997 1,094 0.87 E. coli 2018 3.5 x 10¹¹ 0.2 / 0.5 8 x 10^-4 1.4 0.5 134 975 879 0.93

(17)

Upstream Boundary Condition

0 250 500 750 1,000 1,250 1,500

10 100 1,000 10,000 100,000

04/20/17 04/30/17 05/10/17 05/20/17

Flow (cfs)

Date

Enterococcus Count (#/100 mL)

Enterococcus - Station 012 Enterococcus - Station 001

Enterococcus Model Base Flow

Watertown Dam Flow Rainfall

Rainfall, 1 inch/hr.

(18)

Upstream Boundary Condition

0 250 500 750 1,000 1,250 1,500

10 100 1,000 10,000 100,000

05/20/17 05/30/17 06/09/17 06/19/17

Flow (cfs)

Date

Enterococcus - Station 012 Enterococcus - Station 001 Enterococcus Model Base Flow

Watertown Dam Flow Rainfall Rainfall,

1 inch/hr.

(19)

Upstream Boundary Condition

0 250 500 750 1,000 1,250 1,500

10 100 1,000 10,000 100,000

09/07/17 09/17/17 09/27/17 10/07/17

Flow (cfs)

Date

Enterococcus Station 001 Enterococcus Model Enterococcus - Station 012 Base Flow

Watertown Dam Flow Rainfall

Rainfall, 1 inch/hr.

(20)

Downstream Boundary Condition

– Water level is kept approximately constant

– Small variations due to discharges at low tide and pumping before storms

0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00

5/1/17 5/6/17 5/11/17 5/16/17 5/21/17 5/26/17 5/31/17

Water Level (m)

Water Level at New Charles River Dam

USGS (Meas)

– In model: USGS water

levels specified as boundary condition

(21)

Water Quality Calibration

Upstream CSOs

(22)

Water Quality Calibration

(23)

Water Quality Calibration Water Quality Calibration

Upstream CSO

(24)

Water Quality Calibration

(25)

Alewife Brook / Upper Mystic River Model

(26)

Model Coverage and Discretization

– From Amelia Earhart Dam to Lower Mystic Lake

– InfoWorks ICM

– Based on FEMA Model

– The FEMA model covers the entire watershed.

– 278 cross-section

(27)

CSO Quality

FACILITY LOCATION SAMPLE TIME LOCAL E. COLI

#/100ML ENTEROCOCCUS

#/100ML

CAMB-CSO 401A 8/29/19 0:20 54,800 36,500

CAMB-CSO 401A 8/29/19 0:40 86,600 61,300

CAMB-CSO 401A 8/29/19 1:20 86,600 54,800

CAMB-CSO 401A 10/17/19 0:31 130,000 54,800 CAMB-CSO 401A 10/17/19 0:46 36,500 22,500 CAMB-CSO 401A 10/17/19 1:01 21,900 17,900 CAMB-CSO 401A 10/17/19 1:16 13,100 30,800 CAMB-CSO 401A 10/17/19 1:31 17,200 16,100

SOM-CSO 001A 8/29/19 0:52 72,700 38,700

SOM-CSO 001A 8/29/19 1:09 81,600 22,500

SOM-CSO 001A 10/17/19 2:18 61,300 13,700

SOM-CSO 001A 10/17/19 3:06 43,500 13,300

– Same approach as for the Charles River: CSO counts calculated from based on

sanitary fractions derived by the collection system model.

(28)

Stormwater Quality: 2019 Monitoring

Monitoring Stations Arlington: 2

Cambridge: 4/2 Medford: 3

Somerville: 5

(29)

Stormwater Quality

– No correlation found with sub-catchment parameters – Sub-catchment area

– Percent undeveloped – Undeveloped area – Percent residential – Residential area – Storm depth

– Prior dry days

– Average counts selected

– E. coli: 12,800 / 100 mL

– Enterococcus: 5,600 / 100 mL ₀

5,000 10,000 15,000 20,000 25,000 30,000

0 20 40 60 80 100 120 140 160 180

E. coli and Enterococcus Counts (#/100 mL)

Undeveloped Area (Acres) E. coli Enterococcus

Example Correlation – with Undeveloped Area

(30)

Hydrology Calibration

– Original FEMA model geared towards extreme events

– For continuous simulations, the

hydrology was changed to the SWMM formulation with groundwater routines – Parameters to be specified

– Percent impervious – Catchment width

– Percent routed from impervious to pervious – Evaporation (monthly)

– Evaporation depth

– Percolation coefficient – Percolation threshold

Soil Store Inflow

(31)

Hydrology Calibration – USGS Flow Gauges

USGS 01103010 MYSTIC RIVER AT ARLINGTON, MA August 2016 onward

USGS 01103040 MYSTIC RIVER RT 16 AT MEDFORD, MA November 2015 to November 2017

USGS 01103025 ALEWIFE BROOK NEAR ARLINGTON, MA October 2005 onward

USGS 01102500 ABERJONA RIVER AT WINCHESTER, MA April 1990 onward

(32)

Hydrology Calibration – Alewife Brook Gauge

0.00 10.00 20.00 30.00 40.00 50.00 60.00 70.00

1/2/2018 1/12/2018 1/22/2018 2/1/2018 2/11/2018 2/21/2018

Flow (MGD)

Date

Meter Flow (MGD)

Model Flow-run226 (MGD)

January-February

(33)

Hydrology Calibration – Alewife Brook Gauge

0.00 5.00 10.00 15.00 20.00 25.00 30.00 35.00 40.00 45.00 50.00

5/1/2018 5/11/2018 5/21/2018 5/31/2018 6/10/2018 6/20/2018 6/30/2018

Flow (MGD)

Date Meter Flow (MGD)

Model Flow-run226 (MGD)

May-June

(34)

Water Quality Calibration

1 10 100 1,000 10,000

4/1/2018 4/11/2018 4/21/2018 5/1/2018 5/11/2018 5/21/2018 5/31/2018

Enterococcus Count (#/100 mL)

Date

STA 174 (Meas) STA 074 (Meas)

STA 174 (Model-run019) STA 074 (Model-run019)

(35)

Water Quality Calibration

1 10 100 1,000 10,000

4/1/2018 4/11/2018 4/21/2018 5/1/2018 5/11/2018 5/21/2018 5/31/2018

Date

STA 277 (Meas) STA 277 (Model-run019) STA 276 (Meas) STA 276 (Model-run019)

(36)

Water Quality Calibration

1 10 100 1,000 10,000

6/1/2018 6/11/2018 6/21/2018 7/1/2018 7/11/2018 7/21/2018 7/31/2018

Date

STA 174 (Meas) STA 074 (Meas)

STA 174 (Model-run021) STA 074 (Model-run021)

(37)

Water Quality Calibration

1 10 100 1,000 10,000

6/1/2018 6/11/2018 6/21/2018 7/1/2018 7/11/2018 7/21/2018 7/31/2018

Date

STA 277 (Meas) STA 277 (Model-run021) STA 276 (Meas) STA 276 (Model-run021)

(38)

Summary

– CSO and stormwater quality measurements have been reviewed and analyzed

– Satisfactory calibration for both models

– The models are being used to:

– Assess current conditions – Assess alternatives

• Further CSO reductions

• Stormwater BMPs

(39)

RECEIVING WATER QUALITY MODEL CALIBRATION FOR MWRA CSO PROGRAM