Watershed Analysis of the
Mystic River and Neponset River Watersheds

Task 4C Technical Report
Neponset River Watershed Property Parcel Analyses

November 13,2023

Prepared for:
U.S. EPA Region 1

Blanket Purchase Agreement: BPA-68HE0118A0001-0003
Requisition Number: PR-OA-22-00343
Order: 68HE0123F0002


-------
Watershed Analysis of the Mystic River
and Neponset River Watersheds

Task 4C
November 13, 2023

Table of Contents

1	Introduction	3

2	Methodology	4

2.1	D ata Inventory	4

2.2	P arcel Analysis	4

2.2.1	GIS-based Spatial Analysis

2.2.2	Python-Based Summary Analysis.

2.3	Refinements from CRW Analysis.

2.3.1 Public/Private Classification.

2.4	Quality Assurance / Quality Control (QA/QC)	

2.5	Limitations	

3	Results and Discussion	

3.1	All Parcels	

3.2	Commercial, Industrial, Institutional, and Multi-Fa
3.2.1 Analysis of CUM Parcels by IC Area.

4	Conclusions	29

5	References	30

Appendix A	31

i


-------
Watershed Analysis of the Mystic River
and Neponset River Watersheds

Task 4C
November 13, 2023

Figures

Figure 2-1. Parcel analysis workflow, including relevant requirements, inputs, and outputs	6

Figure 2-2. Distribution of parcel area by Use Group within the Neponset River Watershed	9

Figure 2-3. Map of parcel Use Groups within the Neponset River Watershed	10

Figure 3-1. Private/Public summaries for all parcels within the Neponset River Watershed	18

Figure 3-2. Private CUM parcel count and total TP load by parcel IC area. Note that a threshold of > Oac IC

includes all private CUM parcels	23

Figure 3-3. Private CUM parcel count and total TN load by parcel IC area. Note that a threshold of > Oac

IC includes all private CIIM parcels	24

Figure 3-4. Percentage of watershed TP load that can be captured from IC runoff, assuming a 60% treatment
efficiency, and the corresponding number of private CIIM parcels based on IC threshold. Labels for IC
thresholds correspond to the dark blue dots	28

Tables

Table 2-1. Data used in the parcel analysis	4

Table 2-2. Parcel-level summary attributes calculated	5

Table 2-3. Use Groups assigned in the CRW and Neponset analyses	7

Table 2-4. Reclassification table for Multi-Use parcels	8

Table 2-5. Reclassification table for major Use Code category	8

Table 2-6. Distribution of public and private parcel areas by Use Group	11

Table 2-7. OWNER1 keywords for initial public/private parcel filtering	15

Table 3-1. Stormwater annual average TP and TN loads (1992-2022) for municipalities within the Neponset

River Watershed	17

Table 3-2. Summary of parcel attributes by Use Group and Public/Private designation with TP*	19

Table 3-3. Summary of parcel attributes by Use Group and Public/Private designation with TN*	20

Table 3-4. Summary of private commercial, industrial, institutional, and multifamily parcel attributes.... 22
Table 3-5. Summary of private commercial, industrial, institutional, and multifamily parcels with IC > 0.25

ac	25

Table 3-6. Summary of private commercial, industrial, institutional, and multifamily parcels with IC > 0.5

ac	25

Table 3-7. Summary of private commercial, industrial, institutional, and multifamily parcels with IC > 0.75

ac	26

Table 3-8. Summary of private commercial, industrial, institutional, and multifamily parcels with IC > 1 ac

			26

Table 3-9. Summary of private commercial, industrial, institutional, and multifamily parcels with IC > 2 ac

	27

Table 3-10. Summary of private commercial, industrial, institutional, and multifamily parcels with IC > 5

ac	27

Table 4-1. Summary of private CIIM parcels installing SCMs based on parcel IC area and the reduction
achieved in watershed total TP load	29

11


-------
Watershed Analysis of the Mystic River
and Neponset River Watersheds

Task 4C
November 13, 2023

1 INTRODUCTION

The Neponset River Watershed is highly residential, drains into Boston Harbor, and faces multiple water
quality impairments including pathogens and nutrients (phosphorus and nitrogen) from human activity and
urban development. These impairments are evidenced by high fecal bacteria levels. Twenty-four waterbody
segments have TMDLs for fecal bacteria, with illicit sewer connections, failing septic systems, and
stormwater carrying fecal matter from domestic and wild animals identified as the primary sources
(MassDEP, 2012, 2002) with an additional eighteen segments listed as requiring a TMDL (MassDEP, 2023).
Stormwater runoff also carries excess nutrients into waterbodies which can cause algal blooms and
macrophyte growth that contribute to anoxic bottom waters that do not support aquatic life, reduce water
clarity, degrade waterbodies' aesthetic quality, and impair designated uses such as fishing, swimming, and
boating, as evidenced in the nearby Charles River and Mystic River Watersheds where stormwater is the
primary source of these pollutants (U.S. EPA, 2022a; USEPA, 2020). Within the Neponset River
Watershed, there are twenty-three segments on the 303(d) list with evidence of nutrient pollution and
eutrophication based on impairments for either nutrients, dissolved oxygen, turbidity, aquatic plants
(macrophytes), or algae (MassDEP, 2023). On August 24, 2020, the U.S. Environmental Protection Agency
(EPA) received a residual designation petition from the Conservation Law Foundation for the Neponset
River Watershed. The petition requests that EPA permit stormwater from commercial, industrial,
institutional, and multi-family residential (CUM) properties of one acre or greater under the National
Pollutant Discharge Elimination System (NPDES) program to meet water quality standards (WQS) in the
Neponset River.

This report presents a methodology to develop, and analyses of, parcel-level stormwater TP and total
nitrogen (TN) loading within the Neponset River Watershed that may be used by EPA Region 1 to support
decision making regarding the residual designation petition. Long-term stormwater runoff and pollutant
loading developed for the Neponset River Watershed (Paradigm Environmental, 2023a) forms the basis
against which parcel-level stormwater management strategies can be evaluated. Key information presented
in this report includes an analysis of CUM parcels, their characteristics such as the amount of impervious
cover (IC), and their estimated stormwater nutrient loads. IC is the largest source of stormwater runoff within
the watershed and findings from an analysis of the relationship between the number of CUM parcels, IC
area, and TP load are presented to demonstrate potential strategies designating the fewest number of parcels
while achieving the largest possible pollutant reduction benefits. The results of this report can be further
refined using other considerations, such as where and how communities facing environmental justice (E J)
concerns may be impacted, and apportioned into each municipality within the watershed where, ultimately,
progress can be made towards improving water quality.

3


-------
Watershed Analysis of the Mystic River
and Neponset River Watersheds

Task 4C
November 13, 2023

2 METHODOLOGY

The general methodology presented here follows the process used in the Charles River Watershed (CRW)
Total Phosphorous analysis (U.S. EPA, 2022a Appendix 4, 2022b). Key steps, refinements, and quality
assurance checks are detailed in the following subsections.

2.1 Data Inventory

Readily available data necessary for parcel analysis were collected, reviewed, and assessed. Data were
obtained from online repositories as well as from EPA staff. Table 2-1 provides an inventory of the GIS data
collected and indicates the use of that dataset.

Table 2-1. Data used in the parcel analysis

Name

Use

Source

Source Link

Source
Date

2016 Land use and
land cover

IC calculation

MassGIS

https://www.mass.gov/info-
details/massgis-data-2016-
land-coverland-use

f

May 2019

Hydrologic
Response Units
(HRUs)

Loading rate and
load calculations

Task 4A and 4B (Paradigm Environmental, 2023a)

Parcel boundaries

For summary results

MassGIS

https://www.mass.gov/info-
details/massgis-data-
property-tax-parcels

Feb 2023

L3 Tax Assessor
Table

Parcel details

MassGIS

Feb 2023

Municipal
boundaries

For summary results

MassGIS

https://www.mass.gov/info-
details/massgis-data-
municipalities

April 2022

Neponset River
Watershed

For summary results

w

MassGIS

https://www.mass.gov/info-
details/massgis-data-major-
watersheds

June 2000

Subwatershed
boundaries

For summary results

MassGIS

https://www.mass.gov/info-
details/massgis-data-
drainage-sub-basins

December
2007

Municipal separate
storm sewer system
(MS4) boundaries

For summary results

EPA

Admin Record



2.2 Parcel Analysis

Parcel analysis includes two main components: 1) GIS-based spatial analysis and 2) summary analysis using
a python-based tool. These steps are described below and were formulated to be as accurate, transparent,
and reproducible as feasible. The parcel analysis workflow, required inputs, and outputs is shown in Figure
2-1. Results of the parcel analysis are summary attributes for each parcel, as shown in Table 2-2, and
additional summaries aggregating the parcel data with other conditions and spatial scales as detailed in
Section 3.

4


-------
Watershed Analysis of the Mystic River
and Neponset River Watersheds

Task 4C
November 13, 2023

Table 2-2. Parcel-level summary attributes calculated

Attribute

Description

LocJD

Unique parcel ID

Parcel Type

Tax classification (e.g., TAX, FEE, ROW, WATER)

Municipality Name

City or Town name

Owner

Property owner's name

Owner Address

Property owner's address

Owner City

Property owner's city

Owner Country

Property owner's country

Owner State

Property owner's state

Owner Zip Code

Property owner's zip code

Lot Size (ac)

Deed area (converted to acres in this analysis)

Site Address

Site address

Site Zip Code

Site address zip code

Units

Number of units on the property

Year Built

Year building was built

Public/Private

Owner type (public or private) based on filtering described in
Section 2.3.1

FY

Year of data

MS4 (boolean, 1 = within MS4)

Inside or outside of the MS4 area

CSA (boolean, 1 = within CSA)

Inside or outside of the Combined Sewer Area (CSA)

Subbasin

Subbasin name(s) and percentages if multiple

Area in Major Basin (%)

Percentage of parcel area with the watershed

Use Code

Use code from Tax Assessor

L3 Use Description

Use description from Tax Assessor

Dept. Revenue Description

Use description from MA Dept. of Revenue (from Use Code)

Parcel Use Group

Land use classification in the current analysis (from Use Code)

MassGIS Land Use

Land use from MassGIS 2016 LULC

Total Area (ac)

Parcel area calculated in the current analysis

IC Area (ac)

The impervious cover area from the 2016 data

IC Percent

Percent impervious cover calculated from MassGIS 2016 LULC

Wetland Area (ac)

Wetland area on the parcel

Wetland Percent

Wetland percent on the parcel

Water Area (ac)

Water area on the parcel

Water Percent

Water percent on the parcel

Forest Area (ac)

Forest area on the parcel

Forest Percent

Forest percentage on the parcel

Pervious Area (ac)

Pervious area

Pervious Percent

Percent pervious

5


-------
Watershed Analysis of the Mystic River
and Neponset River Watersheds

Task 4C
November 13, 2023

Attribute

Description

Total Pervious Load (Ib/yr)

TP arid TN load from the parcel's pervious cover

IC Load (Ib/yr)

TP and TN load from the parcel's impervious cover

Total parcel load (Ib/yr)

Total TP and TN load from parcel

Parcel Owner
Names

Public/Private
Keywords

Raster Datasets

HRUs ^ Impervious ^ Land Use

LOCJD y/' MS4 ^Municipality^ Subbasins

Watershed /WaterWetland/ Zip Code

Initial
Public/Private
Designation

Parcel GIS
Attributes

L3 Assessor Table

Combined Raster

HRU Loading
Rates

Use Group &
Public/Private
Designation

Legend

Python/Excel
Tool

Input/Output

Parcel Summaries

Python (£ v3.6)

Dependencies

pandas
numpy
matplotlib

User Specified IC
Threshold

Aggregated Parcel
Summaries &
Figures

Figure 2-1. Parcel analysis workflow, including relevant requirements, inputs, and outputs.

6


-------
Watershed Analysis of the Mystic River
and Neponset River Watersheds

Task 4C
November 13, 2023

2.2.1 GIS-based Spatial Analysis

The GIS-based processing is raster-based and assembles the required data for parcel-level analysis and
summaries at other spatial scales. The output of this step is a combined raster layer and attribute table listing
the unique combinations of all input rasters. The major GIS processing steps are:

1.	Ensure all layers use the same projection (EPSG: 26986, Massachusetts State Plane Coordinate
System, Mainland Zone).

2.	Clip all layers to the area of interest (Neponset River Watershed boundary).

3.	Convert all polygon layers to rasters with 1-square meter cells (e.g., parcel layer). The raster cell
values will be a unique identifier (e.g., LOC_ID for parcels) to allow the joining of attributes in later
steps.

4.	Overlay all rasters using the ESRI Combine tool. The output is a combined raster and an attribute
table with a unique identifier for each unique combination of input raster values.

The combined raster attribute table is converted into an Excel spreadsheet and the pollutant load calculated
based on the HRU by multiplying the area of each unique combination by the appropriate loading rate. The
development of loading rates is described in the Task 4A-B memo (Paradigm Environmental, 2023a).

Parcel Preprocessing

One additional processing step was performed on the tax parcel polygon layer to create a second input to
the python-based processing described below. Namely, once all the parcels were clipped to the watershed,
all the unique "Use_Codes" were mapped to a single use group similar to those used in the CRW TP analysis
(Table 2-3). This allows users to easily update parcel classifications as additional parcel details or corrections
become available, without having to modify the python code. The Use Codes are generally, but not always,
standardized codes set by the MA Department of Revenue and provide a greater number of categories than
the 2016 LULC dataset (MA Dept. of Revenue Division of Local Services, 2016). For example, the CRW
TP analysis grouped several Code 9 Use Code values into institutional groups; these categories do not exist
in the MassGIS 2016 LULC dataset but are important for evaluating loading from CUM parcels. As well as
using the CRW Use Code groups, a right-of-way group and water group were added as the current analysis
considers all parcels within the Neponset River Watershed. A public or private designation was assigned to
each parcel during this preprocessing step (detailed in Section 2.3.1).

Table 2-3. Use Groups assigned in the CRW and Neponset analyses

CRW Use Group

Neponset Use Groups

Public/Private

Agriculture

Agriculture

Private

Commercial

Commercial

Public or Private

Industrial

Industrial

Public or Private

Open Land

Open Land

Public or Private

Local Institutional

Local Institutional

Public

Private Institutional

Private Institutional

Private

State Institutional

State Institutional

Public

MultiFamily Residential

MultiFamily Residential

Public or Private

Single Family Residential

Single Family Residential

Public or Private

Federal Institutional

Federal Institutional

Public

Two Family Residential

MultiFamily Residential

Public or Private

Three Family Residential

MultiFamily Residential

Public or Private

MultiFamily Residential (4-8)

MultiFamily Residential

Public or Private

7


-------
Watershed Analysis of the Mystic River
and Neponset River Watersheds

Task 4C
November 13, 2023

CRW Use Group

Neponset Use Groups

Public/Private

MultiFamily Residential (>8)

MultiFamily Residential

Public or Private



Right-of-Way

Public or Private



Water

Public or Private

The main steps in parcel Use Group classification are listed below. In general, each step reclassifies
unknowns from the preceding steps.

1.	Assign from Use Code based on existing CRW classification

2.	Assign institutional categories from Use Code 9 parcels unique to the Neponset River Watershed

3.	Assign "Unknown" if no Use Code and no Owner

4.	Assign Multi-Use (Table 2-4)

5.	Assign from major Use Code category (first digit, Table 2-5)

6.	Assign from PolyType (only Right-of-Way and Water)

7.	Remaining Unknowns classified from the dominant MassGIS 2016 LULC category.

8.	Manually reclassify select parcels.

Table 2-4. Reclassification table for Multi-Use parcels

2 Digit Use Code

Use Group

01

MultiFamily Residential

03

Commercial

04

Industrial

06

Open Land

07

Open Land

08

Open Land

09

Private Institutional

Table 2-5. Reclassification table for major Use Code category

1 Digit Use Code

Use Group

1

MultiFamily Residential

2

Open Land

^ 3 l

Commercial

4

Industrial

5

Unknown

6

Open Land

7

Open Land

8

Open Land

9

Private Institutional

8


-------
Watershed Analysis of the Mystic River
and Neponset River Watersheds

Task 4C
November 13, 2023

Results of the parcel preprocessing, in terms of total distribution within the Neponset River Watershed, are
shown in Figure 2-2 and mapped in Figure 2-3. This analysis shows that a third (33%) of the watershed area
is made up of single-family residences; combined multi-family residences make up 6%. Public institutional
groups (local, state, and federal) make up 6.4% of the watershed area with private institutional totaling 3.6%.
In total, commercial, industrial, and private institutional constitute 13.5% of the watershed area. The
distribution of public/private parcels, grouped by Use Code group, is shown in Table 2-6. In total, the current
classification has 33.9% public parcels (largely open land) and 66.1% private by area. By parcel count,
however, the watershed is dominated by private parcels (95.8%).

MultiFamily Residential

Public Institutional

Figure 2-2. Distribution of parcel area by Use Group within the Neponset River Watershed.

9


-------
Watershed Analysis of the Mystic River
and Neponset River Watersheds

Task 4C
November 13, 2023

~ Neponset River
Watershed
Waterbody

Parcel Use Group
¦I Agriculture
Commercial
Industrial
MultiFamily
Residential
Open Land
Private Institutional
Public Institutional
Right-of-Way
Single Family
Residential
Water

Figure 2-3. Map of parcel Use Groups within the Neponset River Watershed. Note that the resolution of this map
parcel-scale; waterbodies exist within non-Water class parcels and have been added to aid
visualization.

is

10


-------
Task 4C
November 13, 2023



Private

Agriculture

66

Q QO

126.77

Ot4Z

Commercial





5,189.70

&S2

Industrial

507

0t64

2,211.27

2m

MultiFamily Residential

13,315

16t74

6,211.75

8r39

Open Land



6t44

9,S11.63

4^t44

Private Institutional

o^3

4t©3

2,667.03



Public Institutional

0

OtOO

OtOO

OtOO

Right of Way

434

0t4&

54S2



Single Family Residential

53,929 67tS1 21,987.16 |

Water

0

OtOO

OtOO

OtOO

Subtotal

76,219

95 g

51,326.21

€8t5

Public

Agriculture

0

OtOO

OtOO

OtOO

Commercial

4

0t04

40t46

0t04

Industrial

0

OtOO

OtOO

OtOO

MultiFamily Residential

0

Q QQ

OtOO

OtOO

Open Land

O^Q

4r©3

12,303.27

46t43

Private Institutional

0

OtOO

OtOO

OtOO

Public Institutional



2 go

1,820.08

&A2

Right of Way

164

©t34

5,162.53

7t39

Single Family Residential

0

OtOO

OtOO

OtOO

Water

43

OtO&

1,021.75

2 35

Subtotal

3,308

4,2

23,618.10

34r5

T otal

79,527

400

71,911.31

400

Public/
Private

Use Group

Count

Count(%)

Total Area (ac)

Total Area (%)



Agriculture

65

0.08

126.74

0.17



Commercial

2,564

3.22

5,186.25

6.92



Industrial

507

0.64

2,241.27

2.99



MultiFamily Residential

13,233

16.64

4,425.76

5.91

Private

Open Land

4,967

6.25

9,841.80

13.13

Private Institutional

813

1.02

2,667.03

3.56



Right-of-Way

113

0.14

54.43

0.07



Single Family Residential

53,929

67.81

24,987.16

33.34



Water

0

0.00

0.00

0.00



Subtotal

76,191

95.8

49,530.46

66.1

Public

Agriculture

0

0.00

0.00

0.00

11

Watershed Analysis of the Mystic River
and Neponset River Watersheds

Table 2-6. Distribution of public and private parcel areas by Use Group


-------
Watershed Analysis of the Mystic River
and Neponset River Watersheds

Task 4C
November 13, 2023



Commercial

4

0.01

10.46

0.01



Industrial

0

0.00

0.00

0.00



MultiFamily Residential

0

0.00

0.00

0.00



Open Land

810

1.02

12,303.27

16.42



Public Institutional

2,288

2.88

4,820.08

6.43



Right-of-Way

191

0.24

7,258.29

9.68



Single Family Residential

0

0.00

0.00

0.00



Water

43

0.05

1,021.78

1.36



Subtotal

3,336

4.2

25,413.88

33.9

Total

79,527

100

74,944.34

100

12


-------
Watershed Analysis of the Mystic River
and Neponset River Watersheds

Task 4C
November 13, 2023

2.2.2 Python-Based Summary Analysis

The raster attribute table output from GIS processing was further processed using a tool developed in python
(Figure 2-1). Python is a commonly used programming language for data management and analysis due to
its ease of use and readability; all python software and packages used in this analysis are freely available and
open source. The output of this processing is an Excel compatible spreadsheet file including every parcel and
additional summaries based on aggregating the data with different conditions and spatial scales. The major
python processing steps include:

1.	Read the combined raster attribute table into a data frame using the pandas python package.

o This is a powerful data structure that allows for a wide variety of data manipulation and

evaluation,
o Efficient for large-size datasets.

2.	Join relevant attribute tables.

o The parcel LOC_ID is used to join parcels with the preprocessed parcel data and the L3
Assessors Table, which provides parcel details such as site address, owner information, and
year built.

o Performing this step outside of a GIS environment should reduce processing time and
potential data overlap errors.

3.	Calculate parcel-level information.

o The information shown in Table 2-2 is calculated for each parcel.

4.	Generate parcel-level summary tables.

o A summary spreadsheet file is generated and saved. The file includes all necessary

information as shown in Table 2-2 for each property parcel,
o The code can also optionally generate a summary table for multiple attributes (e.g., the
unique intersection of parcel, municipality, and subbasin). This is useful for additional QA
of area and loads.

5.	Generate additional summaries.

o Additional summaries of the calculated parcel data are created by aggregating with other

conditions and spatial scales,
o For example, TP loading from IC areas can be summarized for parcels of varying IC areas
and Use Groups.

Outputs of the python-based processing are presented and evaluated in the Results and Discussion section.
All input data, processing codes, and outputs will be made available to EPA; outputs can be visualized by
joining with the LOC_ID in the parcel polygon layer.

2.3 Refinements from CRW Analysis

Parcel analysis for the Neponset River Watershed is based on the general methodology used in the Charles
River Watershed. Several refinements were made to improve transparency and consistency including:

1.	High-resolution (i.e., 1 squared meter cell size) raster-based GIS data processing

2.	A consistent source of land use classification (MassGIS 2016 LULC)

3.	Binary public or private designation at parcel-scale

4.	The long-term annual average loading rates from the Opti-Tool HRU time series (for a period of
2007-2016)

5.	The pollutant loading from the pervious areas is based on the HSG classification, providing
additional resolution for load calculation.

13


-------
Watershed Analysis of the Mystic River
and Neponset River Watersheds

Task 4C
November 13, 2023

The GIS processing performed in the CRW TP analysis was based primarily on polygon datasets, which can
lead to errors or require additional processing with polygons being duplicated and/or overlapped. For the
Neponset River analysis, the raster-based approach (Section 2.2.1) spatially aligns all datasets to eliminate
errors such as polygon slivers or overlap of multi-part polygons.

The impervious cover is based on a rasterized version of the MassGIS 2016 LULC dataset, which contains
the most recent and detailed IC data currently available. The land cover information in this layer is consistent
with the Coastal Change Analysis Program (C-CAP)'s high-resolution land cover classification scheme and
the entire LULC layer has been thoroughly vetted and documented by MassGIS (MassGIS, 2016). The
CRW analysis used both a MassGIS 2005 IC raster and the 2016 LULC polygon dataset and stated that the
2005 IC raster was overly simplified. However, the 2016 land cover classifications are based on 1-meter
aerial imagery and can be converted back to a 1-meter raster with essentially no loss in accuracy. This was
evaluated for the Mystic River Watershed and shown in the Task 3C memo (Paradigm Environmental,
2023b). Additionally, loading rates were calculated from the Opti-Tool Hydrologic Response Units (HRUs)
that are primarily based on the MassGIS 2016 LULC dataset and average annual SW HRU pollutant load
export rates that were developed in subtasks 4A and 4B (Paradigm Environmental, 2023a).

2.3.1 Public/Private Classification

For the Neponset River Watershed, a binary public/private classification based on similar Use Codes as
those used in the CRW analysis was developed. The public/private designation uses keyword filters on the
"OWNER1" attribute as an initial designation which is then further refined by looking at the assigned Use
Group (Table 2-3). For example, any local, state, or federal institutional groups that were not classified as
public in the keyword filtering were changed to public. These keywords are listed in Table 2-7 and were
selected by visual inspection of unique code 9 owner names remaining after filtering out CRW use codes for
local, state, and federal institutional groups. Binary classification is important to help distinguish between
parcels already subject to regulation.

The main steps in creating the public/private designation are:

1.	Owner keyword filtering (performed in python script)

2.	Update RoW based on parcel PolyType

3.	Classify parcels with a "Water" Use Group as public

a. These parcels are predominately water and account for only 0.1% of the total TP and TN
loads (Table 3-2)

4.	Classify unknowns based on Use Group

5.	Update Residential

a. If residential and not Use code 9XX: private

6.	Manual reclassification of selected parcels

14


-------
Watershed Analysis of the Mystic River
and Neponset River Watersheds

Task 4C
November 13, 2023

Table 2-7. OWNER1 keywords for initial public/private parcel filtering

Public Keywords

CITY OF

MASS BAY TRANS AUTHORITY

MASSACHUSETTS COMM OF
METRO DIST COMM

COMM OF MASS

MASS BAY TRANS, AUTHORITY

MASSACHUSETTS COMMON OF

COMMNWLTH OF
MASS

MASS BAY TRANS. AUTH

MASSACHUSETTS
COMMONWEALTH

COMMONWEALTH
OF

MASS BAY TRANS. AUTHORITY

MASSACHUSETTS
COMMONWEALTH OF

COMMWLTH OF
MASS

MASS BAY TRANSP AUTH

MASSACHUSETTS DEPARTMENT

COMWLTH OF
MASS

MASS BAY TRANSPORT AUTH

MASSACHUSETTS GOVT LAND
BANK

COUNTY OF

MASS BAY TRANSPORTATION

MASSACHUSETTS PORT AUTH

DCR

MASS BAY TRANSPORTATION
AUTH

MASSACHUSETTS PORT
AUTHORITY

DEPARTMENT OF

MASS BAY TRANSPTN AUTHOR

MASSACHUSETTS TURNPIKE
AUTHORITY

DEPT OF

MASS DOT

MBTA

DEPT. OF

MASS ELECTRIC CO

POST OFFICE

FIRE + POLICE
BUILDING

MASS PORT AUTHORITY

REDEVELOPMENT ASSOC

FIRE DEPARTMENT

MASS TURNPIKE AUTHORITY

REDEVELOPMENT AUTH

HOUSING AUTH

MASS WATER RESOURCE
AUTH

TOWN OF

LIBRARY

MASSACHUSETT PORT
AUTHORITY

TOWN OFFICE BUILDING

M BT A

MASSACHUSETTS BAY TRANS
AUTH

U S POST OFFICE

MASS BAY
AUTHORITY

MASSACHUSETTS BAY
TRANSIT AUTHORITY

UNITED STATES OF AMERICA

MASS BAY TRAN
AUTHORITY

MASSACHUSETTS BAY
TRANSIT AUTHORITY

UNITED STATES POSTAL SERVICE

MASS BAY TRANS
AUTH

MASSACHUSETTS COMM OF

UNITED STATES PROPERTY

Private Keywords

COMPANY

CONDO

LLC

TRUST

TRS

CREDIT UNION

REALTY

RLTY

INC

INCORPORATED





15


-------
Watershed Analysis of the Mystic River
and Neponset River Watersheds

Task 4C
November 13, 2023

2.4	Quality Assurance / Quality Control (QA/QC)

Several QA/QC steps have been performed to ensure the highest level of accuracy feasible. The greatest
source of uncertainty in this analysis is the parcel attributes from the T3 Tax Assessor table. As seen in Table
2-7, owner names and other details are not standardized and may have typos or other inconsistencies that
make automated processing difficult. To address this, parcels that remain unclassified as public/private after
the processing described in Section 2.3.1 were manually examined and updated.

Additional checks include:

•	Ensuring parcels have a single MassGIS 2016 land use category by intersecting parcel boundaries
and reclassified land use categories.

•	Evaluating the fraction of public and private ownership for feasibility

•	Checking the sum of all parcel areas within a municipality equals the sum of municipality area and
similar checks for other boundaries such as sub-watersheds.

2.5	Limitations

One limitation of this analysis is that parcels crossing the boundary of the Neponset River Watershed will
only be evaluated for the portion of their area within the watershed. This is not expected to impact the
calculation of load within the watershed but will impact the calculation of area for different land uses at the
parcel scale. For example, a parcel may have IC outside of the watershed boundary, but only the IC area
and load within the watershed will be accounted for. The percentage of IC area will be calculated as the IC
area within the watershed divided by the total parcel area within the watershed. The impact of splitting
parcels on the watershed boundary should be negligible given that this is a small portion of the total number
of parcels.

16


-------
Watershed Analysis of the Mystic River
and Neponset River Watersheds

Task 4C
November 13, 2023

3 RESULTS AND DISCUSSION

This section describes the calculated TP and TN loads from private and public properties based on the parcel
Use Group. Analyses include the proportional impact of different private property classes, the proportional
impact of different property sizes based on the amount of impervious cover on each property, and a range
of the optimal IC size thresholds to reduce the greatest amount of TN and TP while potentially designating
the fewest number of properties. The analyses presented here do not distinguish between parcels inside or
outside of MS4 areas. Only 5% of the area within the Neponset River Watershed is outside MS4 boundaries
and this area is predominately undeveloped and pervious (see Figure 2-9 and Table 2-9 in Paradigm
Environmental [2023a]).

Total baseline nutrient loads used in this parcel analysis are from the 1992-2022 time period and represent
unattenuated stormwater TP and TN loads of 34,367 lb/yr and 253,686 lb/yr, respectively (Table 3-1). Note
that there is 1,254 ac of parcel area outside of any municipality (denoted as "No Data" in Table 3-1). TP
load from these areas is 0.05% of the watershed total but was not assigned to any municipality; this
corresponds to the approach for calculating municipality loading used in the loading analysis (Paradigm
Environmental, 2023a).

Table 3-1. Stormwater annual average TP and TN loads (1992-2022) for municipalities within the Neponset River
Watershed

Municipality

Unattenuated Annual Average (1992 2022) Load (lb/yr)

Public (%)

Private (%)

TP

TN

BOSTON

30%

|70%

5,190.56

37,058.56

CANTON

29%

71%

4,660.38

35,662.07

DEDHAM

34%

\ 66%

1,240.24

9,340.38

DOVER

18%

82%

415.84

2,815.36

FOXBOROUGH

21%

79% ^

1,112.64

8,562.31

MEDFIELD

27%

73%

447.38

3,088.03

MILTON

34%

66%

3,289.64

23,747.69

NORWOOD

26%

74%

4,439.66

34,070.60

QUINCY

31%

69%

1,813.70

13,204.50

RANDOLPH

49%

51%

320.40

2,315.00

SHARON

36%

64%

2,649.47

18,981.10

STOUGHTON

30%

70%

2,253.34

16,534.24

WALPOLE

31%

69%

4,350.17

31,983.97

WESTWOOD

29%

71%

2,165.69

16,148.12

No Data1

61%

39%

18.19

173.57

Total

34,367.30

253,685.50

^ote that there are 274 ac of parcel area within the Neponset River Watershed that are not covered by a municipal
boundary. These areas represent 0.05% and 0.07% of the total TP and TN loads, respectively.

17


-------
Watershed Analysis of the Mystic River
and Neponset River Watersheds

Task 4C
November 13, 2023

3.1 All Parcels

A total of 79,527 parcels were analyzed within the Neponset River Watershed. More than two thirds of the
parcels are single family residential (68%) and multi-family residential parcels make up another 17%. By
land area these two categories make up 42%. Open land makes up 30% of the parcel area and is
approximately evenly split between public and private.

Figure 3-1 illustrates the distribution of summary attributes by Public/Private designation for all parcels in
the Neponset River Watershed. Private parcels account for 96% of parcels and 66% of total parcel area.
Twenty-one percent of total parcel area is impervious cover, with private parcels having 1.8 times as much
total IC area as public parcels (64% of total IC area). In terms of nutrient loading, private parcels contribute
70% of the total TP and 69% of the total TN. Loading from IC within private parcels amounts to 81% of the
total TP and 82% of the total TN load from private parcels. Private IC load represents 57% of total TP and
total TN from all parcels. These results indicate that private parcels contribute nearly three quarters of the
nutrient load and may require further stormwater controls for the watershed to meet its water quality goals.
Table 3-2 and Table 3-3 provide additional details summarizing the IC area and load for parcels by Use
Group and Private/Public designation for TP and TN, respectively.

¦ Private ¦ Public
Total TP (34367.3 Ib/yr)

Parcel Count (79527)

Parcel Area (74944.3 ac)

¦ Private IC ¦ Private Pervious
¦ Private ¦ Public	¦ Public IC	¦ Public Pervious

Figure 3-1. Private/Public summaries for all parcels within the Neponset River Watershed.

18


-------
Watershed Analysis of the Mystic River	Task 4C

and Neponset River Watersheds	November 13, 2023

Table 3-2. Summary of parcel attributes by Use Group and Public/Private designation with TP*











IC Area



TP Load (Ib/yr)



Public/
Private

Use Group

Count

Count
(%)

Total Area
(ac)

Area
(%)

Acre

% IC of
Total
Area

Parcel
Avg.
(ac)

IC

Pervious

Total

Total

(%)

Parcel
Total
Avg.

Private

Agriculture

65

0.08

126.74

0.17

20.56

16.22

0.32

36.09

23.32

59.41

0.17

0.91

Commercial

2,564

3.22

5,186.25

6.92

2,266.98

43.71

0.88

4,083.84

302.19

4,386.03

12.76

1.71

Industrial

507

0.64

2,241.27

2.99

1,020.67

45.54

2.01

1,835.81

91.33

1,927.14

5.61

3.80

MultiFamily Res.

13,233

16.64

4,425.76

5.91

1,695.04

38.30

0.13

3,977.59

311.37

4,288.96

12.48

0.32

Open Land

4,967

6.25

9,841.80

13.13

413.04

4.20

0.08

750.19

1,086.79

1,836.99

5.35

0.37

Private Inst.

813

1.02

2,667.03

3.56

410.64

15.40

0.51

742.43

296.06

1,038.49

3.02

1.28

Right-of-Way

113

0.14

54.43

0.07

26.13

48.01

0.23

36.64

2.81

39.45

0.11

0.35

Single Family Res.

53,929 67.81 24,987.16 33.34 4,105.84

16.43

0.08

8,082.22 2,345.06 10,427.28 30.34

0.19

Water

0

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

Subtotal

76,191

95.8

49,530.46

66.1

9,958.90

20.1

—

19,544.82

4,458.92

24,003.73

69.8

—

Public

Agriculture

0

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

Commercial

4

0.01

10.46

0.01

7.85

75.02

1.96

14.14

0.36

14.50

0.04

3.62

Industrial

0

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

MultiFamily Res.

0

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

Open Land

810

1.02

12,303.27

16.42

93.06

0.76

0.11

161.73

1,476.44

1,638.17

4.77

2.02

Public Inst.

2,288

2.88

4,820.08

6.43

840.61

17.44

0.37

1,461.28

429.17

1,890.44

5.50

0.83

Right-of-Way

191

0.24

7,258.29

9.68

4,689.99 64.62

24.55

6,557.05

255.43

6,812.48 19.82

35.67

Single Family Res.

0

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

Water

43

0.05

1,021.78

1.36

1.87

0.18

0.04

3.00

4.97

7.97

0.02

0.19

Subtotal

3,336

4.2

25,413.88

33.9

5,633.38

22.2

—

8,197.20

2,166.37

10,363.57

30.2

—

Total

79,527

100

74,944.34

100

15,592.28

20.8

-

27,742.02

6,625.28

34,367.30

100

-

* A darker color gradient represents increasing value within a column.

19


-------
Watershed Analysis of the Mystic River	Task 4C

and Neponset River Watersheds	November 13, 2023

Table 3-3. Summary of parcel attributes by Use Group and Public/Private designation with TN*













IC Area

TN Load (Ib/yr)

Public/
Private

Use Group

Count

Count
(%)

Total
Area (ac)

Total
Area
(%)

Acre

% IC
of
Total
Area

Parcel
Avg.
(ac)

IC

Pervious

Total

Total

(%)

Parcel
Total
Avg.



Agriculture

65

0.08

126.74

0.17

20.56

16.22

0.32

301.04

164.12

465.16

0.18

7.16



Commercial

2,564

3.22

5,186.25

6.92

2,266.98

43.71

0.88

34,061.46

2,468.02

36,529.48

14.40

14.25



Industrial

507

0.64

2,241.27

2.99

1,020.67

45.54

2.01

15,528.90

725.58

16,254.48

6.41

32.06



MultiFamily Res.

13,233

16.64

4,425.76

5.91

1,695.04

38.30

0.13

24,118.65

2,214.67

26,333.32

10.38

1.99

Private

Open Land

4,967

6.25

9,841.80

13.13

413.04

4.20

0.08

5,370.44

7,398.24

12,768.68

5.03

2.57

Private Inst.

813

1.02

2,667.03

3.56

410.64

15.40

0.51

6,214.73

2,024.01

8,238.74

3.25

10.13



Right-of-Way

113

0.14

54.43

0.07

26.13

48.01

0.23

270.13

20.30

290.43

0.11

2.57



Single Family Res.

53,929

67.81

24,987.16

33.34

4,105.84

16.43

0.08

58,446.03

16,377.44

74,823.47

29.49

1.39



Water

0

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00



Subtotal

76,191

95.8

49,530.46

66.1

9,958.90

20.1

—

144,311.37

31,392.38

175,703.75

69.3

—



Agriculture

0

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00



Commercial

4

0.01

10.46

0.01

7.85

75.02

1.96

119.68

3.23

122.91

0.05

30.73



Industrial

0

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00



MultiFamily Res.

0

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

Public

Open Land

810

1.02

12,303.27

16.42

93.06

0.76

0.11

1,343.55

10,888.38

12,231.94

4.82

15.10

Public Inst.

2,288

2.88

4,820.08

6.43

840.61

17.44

0.37

12,138.61

3,221.03

15,359.64

6.05

6.71



Right-of-Way

191

0.24

7,258.29

9.68

4,689.99

64.62

24.55

48,290.44

1,906.40

50,196.84

19.79

262.81



Single Family Res.



0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00



Water

43

0.05

1,021.78

1.36

1.87

0.18

0.04

23.95

46.47

70.42

0.03

1.64



Subtotal

3,336

4.2

25,413.88

33.9

5,633.38

22.2

—

61,916.23

16,065.52

77,981.75

30.7

—

Total

79,527

100

74,944.34

100

15,592.28

20.8

-

206,227.60

47,457.90

253,685.50

100

-

* A darker color gradient represents increasing value within a column.

20


-------
Watershed Analysis of the Mystic River
and Neponset River Watersheds

Task 4C
November 13, 2023

3.2 Commercial, Industrial, Institutional, and Multi-Family Parcels

Commercial, Industrial, Institutional, and Multi-Family parcels make up 22% parcels by count and 19% of
area in the Neponset River Watershed (Table 3-4). The greatest number of CUM parcels are Multifamily
Residential. On average, however, these parcels have the lowest IC area and TP and TN loads. Industrial
parcels have the highest average IC and load values, followed by Commercial and Institutional. CUM
parcels make up 35% of total IC area and 38% and 39% of total TP and TN load from all IC, respectively.
These parcels could be candidates for additional stormwater controls while excluding the designation of the
nearly 54,000 private single family residential parcels.

21


-------
Watershed Analysis of the Mystic River	Task 4C

and Neponset River Watersheds	November 13, 2023

Table 3-4. Summary of private commercial, industrial, institutional, and multifamily parcel attributes in the Neponset River Watershed







IC Area

TP Load (Ib/yr)

TN Load (Ib/yr)

Use Group

Count

Total
Area (ac)

Acre

% IC of
Total
Area

Parcel
Avg.
(ac)

IC

Pervious

Parcel
Total
Avg.

IC

Pervious

Parcel
Total
Avg.

Commercial

2,564

5,186.25

2,266.98

43.71

0.88

4,083.84

302.19

1.71

34,061.46

2,468.02

14.25

Industrial

507

2,241.27

1,020.67

45.54

2.01

1,835.81

91.33

3.80

Hl5,528.90

725.58

32.06

MultiFamily Residential

13,233

4,425.76]

1,695.04

38.30

0.13

3,977.59

311.37

0.32

24,118.65

2,214.67

1.99

Private Institutional

813

2,667.03

410.64

15.40

0.51

742.43

296.06

1.28

6,214.73

2,024.01

10.13

Subtotal

17,117

14,520.32

5,393.33

37.14

—

10,639.67

1,000.94

—

79,923.73

7,432.29

—

Watershed Total (%)

21.5

19.4

34.6



-

38.4

15.1

-

38.8

15.7

-



22


-------
Watershed Analysis of the Mystic River
and Neponset River Watersheds

Task 4C
November 13, 2023

3.2.1 Analysis of CUM Parcels by IC Area

Because designating all CUM parcels may be impractical due to the high number of multifamily residential
parcels, it may be possible to designate fewer CUM properties based on the amount of IC area while still
targeting the majority of the nutrient load. The relationship between the number of parcels, the amount of
IC area within a parcel, and the total load was evaluated for private CUM parcels by varying thresholds of
IC area as shown Figure 3-2 and Figure 3-3 (A presents similar plots by individual parcel use group). These
plots show that while the IC threshold is relatively large (e.g., > 2 ac), the number of parcels potentially
designated is relatively small, but accounts for approximately half of the private CUM total load. As the IC
threshold decreases below 1 ac, the number of parcels potentially designated sharply increases, but with
lower increases in the total load. IC thresholds below > 0.25 ac exhibit a large increase in the number of
parcels potentially designated because more multifamily residential parcels are included (these parcels have
an average IC area of 0.1 ac, as shown in Table 3-4).

17500

15000

12500

10000

o
O

a>
o

£ 7500

5000 -

2500

\

n
ii

*
i#
®

\

-• - Count
TP

- 12000

¥

V

—I	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1—

-7? -Vs -ft -Vs iY -T? -T? -T/ -Tp -7? ^ ^ ^^	^

Private CUM IC (ac)

10000

8000 -c
>

~o
CD
O
_l

~5

6000 £

O

0

-t—>

CO
>
' 1	

4000 °"

-2000

Figure 3-2. Private CUM parcel count and total TP load by parcel IC area. Note that a threshold of > Oac IC includes
all private CIIM parcels.

23


-------
Watershed Analysis of the Mystic River
and Neponset River Watersheds

Task 4C
November 13, 2023

17500

15000 -

12500

_ 10000

o
O

"33
o

03
CL

7500

5000 -

2500 -

0-

I

I®

16000 -
14000 -

I \
l ®

\ V

i

12000 -

i

i
\

I*
1 *

1 ®

10000 -

\
i
i

:\

8000 -

i
i
i

*

\

\

k

i ®

\

\

\

\

\

*

6000 -
4000 -
2000 -

r

- •

o.o

T"

0.5

-• - Count
TN

-	80000

-	70000

-	60000

-	50000

-	40000
30000
20000

h10000
0

1.0

¥

V





- 80000

- 60000

£

¦o

co
o

_i

iS
,o

- 40000 o
a)

>

- 20000

S*	~v> "Tp	Q n'' v3 'v5 ^ s85N8>ri?)

' " " " " " " " " " 1, 1/ 1/ 1, 1/ 1, 1/ 1/ 1, 1/ 1/

Private CIIM IC (ac)

Figure 3-3. Private CIIM parcel count arid total TN load by parcel IC area. Note that a threshold of > Oac IC includes
all private CIIM parcels.

The parcel count, load, IC relationship was further analyzed for IC thresholds of 0.25 ac, 0.5 ac, 0.75 ac, 1
ac, 2 ac, and 5 ac as shown in Table 3-5 to Table 3-10. These tables show that, for a small percentage of the
total number of parcels, a larger proportion of the IC nutrient load can be controlled. For example, private
CIIM parcels with > 0.25 ac of IC represent 3% (2,430) of the total number of parcels but account for 29%
of the IC TP load and 30% of the IC TN load from all parcels. Larger IC thresholds require designating
fewer parcels, but with the potential to treat less of the nutrient load.

24


-------
Watershed Analysis of the Mystic River
and Neponset River Watersheds

Task4C
November 13, 2023

Table 3-5. Summary of private commercial, industrial, institutional, and multifamily parcels with IC > 0.25 ac in the Neponset River Watershed







IC Area

TP Load (Ib/yr)

TN Load (Ib/yr)

Use Group

Count

Total Area
(ac)

Acre

% IC of
Total
Area

IC

Pervious

Total

IC

Pervious

Total

Commercial

1,222 4,782.05 2,107.26

44.07

3,786.92 270.06 4,056.99 31,680.82 2,231.12 33,911.94

Industrial

410

2,124.38

1,011.09

47.59

1,818.89

83.05

1,901.94

^ 15,387.13

659.17

16,046.30

MultiFamily Residential

536

2,100.06

699.39

33.30

1,625.85

179.31

1,805.16

9,964.84

1,279.05

11,243.89

Private Institutional

262

1,905.30

369.82

19.41

668.43 .

216.99

885.42

5,599.86

1,478.61

7,078.47

Subtotal

2,430

10,911.79

4,187.56

38.38

7,900.09

749.41

8,649.50

62,632.65

5,647.96

68,280.61

Watershed Total (%)

3.1

14.6

26.9

-

28.5

11.3

25.2

30.4

11.9

26.9

Table 3-6. Summary of private commercial, industrial, institutional, and multifamily parcels with IC > 0.5 ac in the Neponset River Watershed







IC Area

TP Load (Ib/yr)

TN Load (Ib/yr)

Use Group

Count

Total Area
(ac)

Acre

% IC of
Total
Area

IC

Pervious

Total

IC

Pervious

Total

Commercial

780 4,525.19 1,949.60

43.08
47.49

3,499.58 258.49 3,758.06 29,304.48 2,147.02 31,451.50

Industrial

344,

2,076.84

986.25

1,774.50

81.22

1,855.72

15,012.95

644.72

15,657.67

MultiFamily Residential

303

1,777.46

617.43

34.74

1,432.75

143.86

1,576.61

8,795.68

1,053.26

9,848.94

Private Institutional

170

1,741.69

336.81

19.34

608.84

206.18

815.02

5,102.45

1,398.00

6,500.45

Subtotal

1,597

10,121.18

3,890.09

38.44

7,315.68

689.74

8,005.42

58,215.56

5,243.01

63,458.57

Watershed Total (%)

2.0

13.5

24.9

-

26.4

10.4

23.3

28.2

11.0

25.0

25


-------
Watershed Analysis of the Mystic River
and Neponset River Watersheds

Task4C
November 13, 2023

Table 3-7. Summary of private commercial, industrial, institutional, and multifamily parcels with IC > 0.75 ac in the Neponset River Watershed







IC Area

TP Load (Ib/yr)

TN Load (Ib/yr)

Use Group

Count

Total
Area (ac)

Acre

% IC of
Total
Area

IC

Pervious

Total

IC

Pervious

Total

Commercial

593 4,335.04 1,835.92

42.35
47.44

3,293.69 250.42 3,544.11 27,585.64 2,089.35 29,674.99

Industrial

297

2,017.31

957.00

1,721.85

79.18

1,801.03

14,567.31

630.08

15,197.39

MultiFamily Residential

219

1,561.76

565.98

36.24

1,314.16

119.45

1,433.61

8,062.45

889.96

8,952.41

Private Institutional

126

1,656.41

309.78

18.70

560.20

199.73

759.93

4,694.06

1,352.78

6,046.84

Subtotal

1,235

9,570.53

3,668.68

38.33

6,889.90

648.78

7,538.68

54,909.46

4,962.17

59,871.63

Watershed Total (%)

1.6

12.8

23.5



24.8

9.8

21.9

26.6

10.5

23.6

Table 3-8. Summary of private commercial, industrial, institutional, and multifamily parcels with IC > 1 ac in the Neponset River Watershed







IC Area

TP Load (Ib/yr)

TN Load (Ib/yr)

Use Group

Count

Total
Area (ac)

Acre

% IC of
Total
Area

IC

Pervious

Total

IC

Pervious

Total

Commercial

470 4,084.48 1,730.21

42.36
47.55

3,103.32 240.13 3,343.45 26,005.81 2,014.43 28,020.24

Industrial

253

1,934.04

919.61

1,654.78

76.02

1,730.80

14,000.96

606.60

14,607.56

MultiFamily Residential

1681

1,439.22

521.76

36.25

1,211.01

109.78

1,320.79

7,433.62

812.20

8,245.81

Private Institutional

104

1,579.26

290.18

18.37

525.18

193.24

718.43

4,399.13

1,302.36

5,701.49

Subtotal

k 995

9,036.98

3,461.76

38.31

6,494.29

619.17

7,113.46

51,839.52

4,735.59

56,575.11

Watershed Total (%)

1.3

12.1

22.2

-

23.4

9.3

20.7

25.1

10.0

22.3

26


-------
Watershed Analysis of the Mystic River
and Neponset River Watersheds

Task4C
November 13, 2023

Table 3-9. Summary of private commercial, industrial, institutional, and multifamily parcels with IC > 2 ac in the Neponset River Watershed







IC Area

TP Load (Ib/yr)

TN Load (Ib/yr)

Use Group

Count

Total
Area (ac)

Acre

% IC of
Total
Area

IC

Pervious

Total

IC

Pervious

Total

Commercial

240

3,367.79

1,400.94

41.60

2,508.59

203.15

2,711.74

21,033.47

1,702.59

22,736.06

Industrial

161

1,630.82

782.38

47.97

1,408.65

68.11

1,476.76

11,921.90

546.33

12,468.23

MultiFamily Residential

98

1,096.16

423.35

38.62

981.10

81.18

1,062.28

6,032.13

610.73

6,642.86

Private Institutional

45

1,143.93

210.54

18.40

381.14

117.52

498.66

3,201.30

794.80

3,996.10

Subtotal

544

7,238.71

2,817.21

38.92

5,279.48

469.97

5,749.45

42,188.81

3,654.45

45,843.25

Watershed Total (%)

0.7

9.7

18.1



19.0

7.1

16.7

20.5

7.7

18.1

Table 3-10. Summary of private commercial, industrial, institutional, and multifamily parcels with IC > 5 ac in the Neponset River Watershed







IC Area

TP Load (Ib/yr)

TN Load (Ib/yr)

Use Group

Count

Total
Area (ac)

Acre

% IC of
Total
Area

IC

Pervious

Total

IC

Pervious

Total

Commercial

77 1,977.31 896.16

45.32
49.55

1,597.94 118.95 1,716.90 13,364.99 1,030.01 14,395.01

Industrial

47

855.94

424.09

764.04

38.31

802.36

6,468.40

298.91

6,767.32

MultiFamily Residential

2^1

586.64

195.99

33.41

455.01

50.99

506.00

2,777.10

394.07

3,171.17

Private Institutional

12

472.02

115.67

24.51

207.93

28.73

236.66

1,758.59

248.51

2,007.10

Subtotal

k 161

3,891.91

1,631.91

41.93

3,024.93

236.98

3,261.91

24,369.09

1,971.51

26,340.60

Watershed Total (%)

0.2

5.2

10.5

-

10.9

3.6

9.5

11.8

4.2

10.4

27


-------
Watershed Analysis of the Mystic River
and Neponset River Watersheds

Task 4C
November 13, 2023

Figure 3-4 further illustrates the tradeoff between pollutant reduction and the number of private CUM
parcels with IC area ranging from >20 ac to >0 ac (i.e., all private CUM parcels) that would have to install
SCMs. This figure assumes that runoff from IC within a parcel would be treated by SCMs sized to achieve
a 60% reduction1. The "knee" of the curve, where the slope begins to flatten, indicates the IC threshold
where the fewest number of parcels can provide the greatest benefit in terms of TP reduction. For the
Neponset River Watershed, this appears to lie between parcels with >0.25 ac and >0.75 ac of IC.

Number of Private CIIM Parcels

Figure 3-4. Percentage of watershed TP load that can be captured from IC runoff, assuming a 60% treatment

efficiency, and the corresponding number of private CIIM parcels based on IC threshold. Labels for IC
thresholds correspond to the dark blue dots.

1 In the absence of a specific TP reduction requirement, 60% was used as a starting point in this analysis to align with
MA MS4 permitting.

28


-------
Watershed Analysis of the Mystic River
and Neponset River Watersheds

Task 4C
November 13, 2023

4 CONCLUSIONS

This report presented a methodology for summarizing and evaluating stormwater pollutant load from
parcels within the Neponset River Watershed that may be used by EPA Region 1 to support decisions
regarding the control of stormwater runoff from certain private properties to improve water quality. Findings
from this analysis include:

1.	Private properties contribute nearly three quarters (70%) of the watershed's total TP load.

2.	The majority of TP from private properties is generated from impervious cover (81% of load from
private properties and 57% of the watershed total load).

3.	Private commercial, industrial, institutional, and multi-family residential (CIIM) properties make up
22% of all parcels, but have relatively high percentages of IC and therefore contribute a large
proportion of the watershed TP load (34%)

4.	Selecting private CIIM parcels based on their IC area (which is a dominate factor in the nutrient load
generated) can minimize the number of parcels installing stormwater controls, while providing the
greatest nutrient reduction benefit (Table 4-1).

The findings in this report indicate that unattenuated stormwater runoff from private parcels within the
Neponset River Watershed contributes the majority of the nutrient load. Without additional controls on
stormwater runoff and pollutant loads from these parcels, making meaningful improvements to water quality
could be difficult. Designating stormwater discharges from certain classes of private properties for NPDES
permits based on the amount of IC area will help reduce the burden on MS4s by targeting properties
generating the largest amount of nutrients in stormwater on a per-property scale. In any scenario,
municipalities will still need to engage the private property owners with smaller property size or IC size to
eventually meet water quality goals. However, requiring action on private properties with larger amounts of
IC now through NPDES permitting provides greater flexibility to the communities in deciding which private
properties to target to meet their own MS4 permit obligations.

Table 4-1. Summary of private CIIM parcels installing SCMs based on parcel IC area and the reduction achieved in
watershed total TP load

IC Threshold (ac)

Parcel Count

Total TP Load
(lb/yr)

ICTP Load (lb/yr)

Total TP Treated

(%)*

>0 (All)

17,117

11,640.62

10,639.67

19%

>0.25

2,430

8,649.50

7,900.09

14%

>0.5

11,597

8,005.42

7,315.68

13%

>0.75

1,235

7,538.68

6,889.90

12%

>1

995

7,113.46

6,494.29

11%

>2

544

5,749.45

5,279.48

9%

>5

161

3,261.91

3,024.93

5%

* Percentage calculated as IC load times a 60% treatment efficiency divided by the watershed total TP load of 34,367
lb/yr.

29


-------
Watershed Analysis of the Mystic River
and Neponset River Watersheds

5 REFERENCES

MA Dept. of Revenue Division of Local Services, 2016. PROPERTY TYPE CLASSIFICATION CODES,
NON-ARM'S LENGTH CODES, AND SALES REPORT SPREADSHEET SPECIFICATIONS.

MassDEP, 2023. Final Massachusetts Integrated List of Waters for the Clean Water Act 2022 Reporting
Cycle. CN 568.1, Massachusetts Department of Environmental Protection, Bureau of Water Resources,
Division of Watershed Management, Watershed Planning Program. Worcester.

MassDEP, 2012. Addendum: Final Total Maximum Daily Loads of Bacteria for Neponset River Basin (CN
121.5).

MassDEP, 2002. Total Maximum Daily Loads of Bacteria for Neponset River Basin.

MassGIS (Massachusetts Bureau of Geographic Information Systems), 2016. Full documentation for
MassGIS ' 2016 Land Use / Land Cover Data.

Paradigm Environmental, 2023a. Task 4A-B Technical Memo Neponset River Watershed Spatial Data and
Mapping Analyses. Prepared for the U.S. EPA.

Paradigm Environmental, 2023b. Task 3C Technical Report Mystic River Watershed Property Parcel
Analyses. Prepared for the U.S. EPA.

U.S. EPA, 2022a. ATTACHMENT 6 Clean Water Act Residual Designation Determination for Certain
Stormwater Discharges in the Charles, Mystic, and Neponset River Watersheds, in Massachusetts:
Charles River Watershed Stormwater Total Phosphorus Analysis.

U.S. EPA, 2022b. Mystic River and Neponset River Watershed Analysis RFQ.

U.S. EPA, 2022c. ATTACHMENT 6 Clean Water Act Residual Designation Determination for Certain
Stormwater Discharges in the Charles, Mystic, and Neponset River Watersheds, in Massachusetts:
Charles River Watershed Stormwater Total Phosphorus Analysis.

USEPA, 2020. Mystic River Watershed Alternative TMDL Development for Phosphorus Management-
Final Report. Boston, MA.

Task 4C
November 13, 2023

30


-------
Watershed Analysis of the Mystic River
and Neponset River Watersheds

Task 4C
November 13, 2023

APPENDIX A

2500 -

2000 -

1500 -

1000 -

500 -

0-

4000

- 3000 -c-
S

X3
03

0
_l

£

h 2000 I

01

E
E
o
O

- 1000

•£> -7> -P lY -T? -T? -!> -Tp -£> Jj§>	^

Commercial IC (ac)

Figure C-1. Private commercial parcel count arid total TP load by parcel IC area. Note that a threshold of > Oac IC
includes all private commercial parcels.

31


-------
Watershed Analysis of the Mystic River
and Neponset River Watersheds

Task 4C
November 13, 2023

500 -

400 -

300 -

o
O

a>
o

CO
CL

200

100 -

0 -



2000

- 1750

1500

h 1250 >-

T3

ra
o

1000 -2
,o

750

- 500

- 250

CO
ZS
T3

—I	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1—

"7? "?> -7? -fi -P -T>	-7>N ft ft ft ft ft ft ft ft ft

Industrial IC (ac)

Figure C-2. Private industrial parcel count and total TP load by parcel IC area. Note that a threshold of > Oac IC
includes all private industrial parcels.

32


-------
Watershed Analysis of the Mystic River
and Neponset River Watersheds

Task 4C
November 13, 2023

800 -

700 -

600 -

500 -

o
O

© 400

o

CO
CL

300 -

200 -

100 -

0 -

t

I
¥

k

- Count
TP



800 -



700 -

600 -

t
•

9
•

t
1
I
I
\

500 -
400 -

300 -

\

\
\

200 -

I

k

100 -

x

V

- 1000

- 800

-600

-400

-200

- 1000

- 800

- 600

—i	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1——f—

-7? -7> & -V> lY -T? J? -T> -ft ^ ^ ^	jP

Private Institutional IC (ac)

400

>

"O
03
O
_l

£
03

c
o

"3

A—'

Ct)

03
>

- 200

Figure C-3, Private institutional parcel count and total TP load by parcel IC area. Note that a threshold of > Oac IC
'dudes all private institutional parcels.

33


-------
Watershed Analysis of the Mystic River
and Neponset River Watersheds

Task 4C
November 13, 2023

12000 -

10000 -

8000 -

o
O

a>

a 6000 H

CL

4000 -

2000 -

0 -

- 4000









"i	1	1	r

3000 §-

"O

cu
o

03

¦4—I'

,o

c
CD
"O

2000 "jg
CH
>

I

03

- 1000

>> & -7? -P $ ^ ^ifi^ififififififififif

MultiFamily Residential IC (ac)

Figure C-4. Private multifamily residential parcel count and total TP load by parcel IC area. Note that a threshold of >
Oac IC includes all private multifamily residential parcels.

34


-------
Watershed Analysis of the Mystic River
and Neponset River Watersheds

Task 4C
November 13, 2023

2500 -

2000 -

1500

o
O

a>
o

CO
CL

1000 -

500 -

0 -

&

*

2500 -

2000 -

1500 -

1000 -

500

k



\

Count
TN

b 35000
30000

-	25000

-	20000

-	15000

-	10000
-5000

0

0.0

0.5

1.0

l

t

\

1 \



V

- 35000

- 30000



~i	1	1	1	1	1	1	1	1	1	1	1	1	r

25000

5

T3

ra
o

b 20000 ^

£
03

'o
CD

b15000 |

o
O

- 10000

- 5000

-7? -Vs -ft -P -fr -P J? ¦%> -ft -7? ^ ^	$

Commercial IC (ac)

Figure C-5. Private commercial parcel count and total TN load by parcel IC area. Note that a threshold of > Oac IC
includes all private commercial parcels.

35


-------
Watershed Analysis of the Mystic River
and Neponset River Watersheds

Task 4C
November 13, 2023

500 -

400 -

300 -

o
O

a>
o

CO
CL

200

100 -

0 -



- 16000

- 14000

- 12000

10000

"O
03
O



—I	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1—

-7> "7> -P if I? -TP -T> "7?	^ ^

Industrial IC (ac)

- 8000

- 6000

-4000

-2000

ra
,o

CO
ZS
T3

Figure C-6. Private industrial parcel count and total TN load by parcel IC area. Note that a threshold of > Oac IC
deludes all private industrial parcels.

36


-------
Watershed Analysis of the Mystic River
and Neponset River Watersheds

Task 4C
November 13, 2023

800 -

700 -

600 -

500 -

o
O

© 400

o

CO
CL

300 -

200 -

100 -

0 -

t

I
*

- Count
TN



•v



k

\



rtttt"-*

8000

- 7000

- 6000

5000

4000 £

>,
x>

TO
O
_J

3

,o

3000

ZJ

-t—'

cn

-2000

- 1000

~i	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1—T—

"7? "?> -7? -fi -P -T>	-7>N ft ft ft ft ft ft ft ft ft

Private Institutional IC (ac)

Figure C-7 Private institutional parcel count and total TN load by parcel IC area. Note that a threshold of > Oac IC
"•eludes all private institutional parcels.

37


-------
Watershed Analysis of the Mystic River
and Neponset River Watersheds

Task 4C
November 13, 2023

12000 -

10000 -

8000 -

o
O

a>

£ 6000

CL

4000 -

2000 -

0 -



Count
TN

- 25000

- 20000

n	1	1	1	1	1	1	1	1	1	1	1	1	r

- 25000

- 15000

20000

"O

cu
o

~aj
,o

c

0
"O

"(/)
CD

10000

- 5000

-T> & -P lY -P -P -T> -Tp -7? jj§ ^

MultiFamily Residential IC (ac)

Figure C-8. Private muitifamily residential parcel count and total TN load by parcel IC area. Note that a threshold of >
Oac IC includes all private muitifamily residential parcels.

38


-------