United Status Clfice of EPA 230-11-85-017
Environmental Protection Policy Analysis
Agency Washington, DC 20460
W«t»f
EpA A Methodological Approach
to an Economic Analysis of
the Beneficial Outcomes of
Water Quality
Improvements from Sewage
Treatment Plant Upgrading
and Combined Sewer
Overflow Controls
Environmental Benefits
Analysis Series
-------
A Methodological Approach to
an Economic Analysis of the Beneficial Outcomes
of Water Quality Improvements from Sewage
Treatment Plant Upgrading and Combined Sewer
Overflow Controls
Prepared for
Office of Policy Analysis
U.S. Environmental Protection Agency
Washington, D.C.
by
Meta Systems Inc
10 Bolworthy Street
Cambridge, Massachusetts 02138
-------
Preface
This report is submitted by Meta Systems Inc in fulfillment of EPA
contract §68-01-6596 700-E. This report estimates the benefits and costs of
upgrading two sewage treatment plants and of constructing combined sewer
overflow controls in the Boston Harbor area.
We are grateful for the review and comments of dark Binkley, Yale
University; A. My rick Freeman, Bowdoin College; and Leon Abbas, North
Carolina State University. We wish to give special thanks to the following
people who provided technical assistance and/or data for the study:
Peter Harrington—Massachusetts Department of Environmental Quality
Engineering;
Michael Hickey and David Chadwick—Massachusetts Division of Marine
Fisheries;
Jean M. Haggerty, Al Ferullo, and Paul DiPetro—Metropolitan District
Commission;
F. Williams Sieling, Mark M. Bundy, and Christopher Bonzack—Maryland
Department of t&tural Resources;
Dana Wallace—Maine Department of Marine Resources;
While we are indebted to all of the above for their contributions, final
responsibility for the analysis, results and conclusions rests solely with
the authors.
-i-
-------
Table of Contents
Section Page
Number Number
1. Summary and Conclusions
1.1 Pollution Sources 1-3
1.2 Water Quality 1-12
1.3 Benefit Categories and Receptors 1-15
1.4 Summary of Study Findings 1-17
1.5 Specific Benefit Estimates 1-24
1.5.1 Recreation 1-24
1.5.2 Health 1-26
1.5.3 Commercial Fisheries 1-26
1.5.4 Intrinsic Benefits 1-27
1.5.5 Ecological impacts 1-27
1.5.6 Secondary Effects 1-28
1.5.7 Charles River Basin 1-29
1.6 Guide to the Report 1-29
2. Municipal Sewage Treatment Plant Operations, Options
and Water Quality Impacts
2.1 Current STP Performance 2-3
2.2 STP Options and Costs 2-9
2.3 Areas Impacted by STP Discharges 2-11
References 2-19
3. Combined Sewer Overflow Control in Boston Harbor
3.1 Scope of the Combined Sewer Overflow Problem 3-1
3.2 Neponset River Estuary 3-6
3.3 Dorchester Bay 3-10
3.4 Inner Harbor 3-13
3.5 Charles River Basin 3-15
3.6 Quincy Storm Sewers 3-16
3.7 Summary of Options 3-19
References 3-23
4. Water Quality Impacts
4.1 Water Quality Impacts of STP Dischargers 4-1
4.2 Water Quality impacts of Combined Sewer Overflows . . . 4-4
4.3 Estimated Water Quality Impacts of the STP and CS
Treatment Options 4-6
References 4-9
5. Approaches to Measuring Benefits from Water Quality Improvement
5.1 Theoretical Concepts 5-1
5.2 Study Methodology 5-9
References 5-11
-ii-
-------
Table of CDntents
(continued)
Section Page
Number Number
Recreation Benefits
6.1 Data Needs and Data Bases 6-4
6.1.1 Swimming Attendance 6-4
6.1.2 Recreation Studies 6-8
6.1.3 Water Quality Data for log it Model 6-9
6.1.4 User (Unit) Day Value 6-10
6.1.5 Water Quality impact 6-11
6.2 Benefits 6-12
6.2.1 Swimming—increase in Participation 6-13
6.2.1.1 Regional Participation Model 6-14
6.2.1.2 Benefit Estimates 6-17
6.2.1.3 Higher Valued Experience 6-18
6.2.1.4 Limits of Analysis 6-19
6.2.2 Travel Cost Model—Conditional Logit Analysis . . 6-20
6.2.2.1 Methodology 6-21
6.2.2.2 The Conditional Multinomial Log it Model. 6-25
6.2.2.3 Model Results 6-30
6.2.2.4 Benefit Estimates 6-32
6.2.2.5 Limits of Analysis 6-36
6.2.3 Swimming—Beach Closings 6-39
6.2.3.1 Boston Harbor Beaches 6-40
6.2.3.2 Nantasket Beach 6-41
6.2.3.3 Benefit Estimates 6-44
6.2.3.4 Limits of Analysis 6-44
6.3 Recreational Boating 6-45
6.3.1 Increased Participation 6-46
6.3.2 Benefits Estimates 6-48
6.3.3 Limits of Analysis 6-48
6.4 Recreational Fishing 6-50
6.4.1 Components of Recreational Fishing 6-51
6.4.2 Benefits Estimates 6-53
6.4.3 Limits of Analysis 6-55
6.5 Boston Harbor Islands 6-55
6.5.1 Increased Participation 6-56
6.5.2 Limits of Analysis 6-r '
6.6 Summary of Recreation Benefits 6-57
References 6-61
-iii-
-------
Table of Contents
(continued)
Section page
Number Number
7. Health Benefits
7.1 Swimming-related Health Benefits 7-2
7.1.1 Benefit Measurement Approach 7-2
7.1.2 Benefit Estimates 7-6
7.1.3 Limits of Analysis 7-6
7.2 Shellfish Consumption 7-8
References 7-11
8. Commercial Fisheries Benefits
8.1 Lobstering and Finfishing 8-2
8.2 Commercial Shellfishing industry 8-7
8.2.1 Pollution Abatement Impacts 8-10
8.2.2 Benefit Assessment Methodology 8-14
8.2.3 Benefit Estimates 8-17
8.2.4 Limits of Analysis 8-25
References 8-27
9. Intrinsic Benefits
9.1 Methodology 9-3
9.2 Benefits Estimates 9-4
9.3 Limits of Analysis 9-5
References 9-6
10. Ecological Effects
10.1 CSO and Secondary Treatment Options 10-1
10.2 Ocean Outfall Option 10-5
10.2.1 Plankton 10-6
10.2.2 Benthos 10-7
10.2.3 Finfish/Lobsters 10-7
10.2.4 Endangered or Threatened Species 10-10
References 10-12
11. Secondary Effects
11.1 Methodology 11-2
11.2 Benefit Estimates 11-4
11.3 Limits of Analysis 11-12
References 11-13
-iv-
-------
Table of Contents
(continued)
Section
Number
12. Charles River Basin Benefits
12.1 The Charles River 12-1
12.2 Boating 12-4
12.2,1 Methodology 12-5
12.2.2 Benefit Estimates 12-7
12.2.3 Limits of Analysis 12-8
12.3 Intrinsic (Non-User) and User Benefits 12-10
12.3.1 Benefit Methodology and Estimates 12-11
12.3.2 Limits of Analysis 12-12
12.4 Summary 12-13
References 12-14
Appendices:
A. Correlating STP Performance and Operation
to Boston Pfcrbor Water Quality
A.I Influent, Effluent, and Sludge Characteristics A-2
A.2. Pollutant Transport from STP Outfalls A-4
References A-11
B. Recreation Benefit Computations
B.I Seasonal Swimming—Increased Participation B-l
B.2 Seasonal Beach Capacity and Current Attendance B-3
B.3 Lower Bound Estimate for Increased Participation .... B-5
B.4 The Conditional Multinomial log it Model, in Brief . . . B-7
B.5 Beach Closings B-12
B.6 User Day Values B-13
B.7 Sources of Recreation Data B-17
References B-27
C. Swimming Health Benefit Calculations
C.I Number of Cases of Gastrointestinal Illness C-l
C.2 Reduced Cases of Gastrointestinal Illness C-2
C.3 Population at Risk C-4
References C-8
D. Commercial Fisheries Benefit Computations
D.I Demand Function Estimation D-l
D.2 Demand Function Computations D-2
D.3 Supply Cost Data and Computations
and Producer Surplus Computation Example D-5
References D-20
E. Charles River Boating Benefits E-l
-v-
-------
List of Tables and Figures
Table
Number
1-1 Oasts and Potential Reduction
in STP Effluent Pollutants for the STP Options 1-10
1-2 Incremental Oasts and Ibtential Reductions in Pollutant
Loadings for the CSO Options 1-13
1-3 Estimated Water Quality Impacts of the STP and CSO
Treatment Options 1-16
1-4 Pollution Control Program and Receptors 1-18
1-5 Annual Benefits and Costs of Combined Sewer Overflow
Controls 1-19
1-6 Annual Benefits and Costs of Combined Sewer Overflow
Controls and Ocean Outfall Control Option 1-20
1-7 Annual Benefits and Costs of Combined Sewer Overflow
Controls and Secondary Treatment Option 1-21
2-1 Comparison of STP Loading for Deer and
Nut Islands Combined 2-6
2-2 Costs of the Two STP Options 2-12
2-3 Pollutant concentrations in Effluent for STP Options .... 2-13
3-1 CSO Planning Area Characteristics 3-7
3-2 Combined Sewer Overflow Project Costs:
Neponset River Estuary 3-9
3-3 Combined Sewer O/erflow Project Costs:
Dorchester Bay 3-12
3-4 Combined Sewer O/erflow Project Costs:
Inner Harbor Planning Area 3-14
3-5 Combined Sewer O/erflow Project Costs:
Charles River Basin 3-17
3-6 Potential Storm Sewer and Infiltration/Inflow Project
Costs for City of Quincy 3-20
3-7 Incremental Costs and Potential Reductions in Pollutant
Loadings for the CSO Options 3-22
4-1 Effluent Concentrations and Dilution Ratios Used in the
Water Quality Impact Analysis 4-5
4-2 Estimated Water Quality impacts of the CSO and STP
Treatment Options 4-7
4-3 Estimates of Pollution Reduction at Receptor Sites in
Study Area 4-8
-vi-
-------
List of Tables and Figures
(continued)
Table Page
Number Number
5-1 A Spectrum of Water Quality Benefits 5-2
5-2 Economic Benefit Categories 5-4
5-3 Benefit Categories and Methodologies
for Boston Harbor Study Area 5-10
6-1 Seasonal Swimming Supply 6-5
6-2 Increased Swimming Participation—Regional Participation
Model 6-16
6-3 Annual Benefit of Increased Swimming Participation for all
Boston Harbor Beaches 6-17
6-4 Conditional logit Model Estimates 6-31
6-5 Per Capita Annual Benefit Estimates from Conditional
log it Model 6-34
6-6 Increased Participation Estimates from Conditional Logit
Model 6-35
6-7 Annual Benefit Estimates from Conditional Logit Model .... 6-37
6-8 Annual Benefit of Averted Beach Closings at 200 MPN/100 ml. . 6-42
6-9 Annual Value of Averted Beach Closings at 500 MPN/100 ml . . 6-43
6-10 Annual Saltwater Boating Benefits 6-49
6-11 Annual Recreational Fishing Benefits 6-54
6-12 Annual Benefits for Recreation on Boston Harbor Islands . . . 6-58
6-13 Annual Recreation Benefits 6-60
7-1 Annual Reduction in Cases of Gastrointestinal Illnesses . . . 7-5
7-2 Swimming Health Benefits 7-7
8-1 Characteristics of Boston Harbor Shellfish Areas 8-9
8-2 Estimated Potential Impacts of Pollution Abatement Options
on Boston Harbor Shellfish Areas 8-11
8-3 Estimated Changes in Price of Soft Shelled Clams
Associated with Alternative Abatement Options and
with Assumed Price Elasticities of Demand 8-21
8-4 Estimated Ibtal Benefits Associated with Alternative
Abatement Options and with Assumed Price
Elasticities of Demand 8-22
9-1 Annual Intrinsic Benefits '. 9-5
11-1 Multipliers Showing Direct, indirect, and
Induced Effects Per $1 Change in Output 11-5
11-2 Secondary Effects Estimates 11-7
11-3 Comparison of Multipliers with and without
Direct Effects per $1 Change in Output 11-10
12-1 Annual Recreation Boating Benefits 12-8
12-2 Annual Estimated Willingness to Pay
for Fishable Charles River 12-12
-vii-
-------
List of I&bles and figures
(continued)
Table
Number
A-l MDC Treatment Facilities Current Pollutant Removals
for Wastewater Effluents A-3
B-l Current Seasonal Attendance Figures B-4
B-2 Sites included in the log it Model B-ll
B-3 User Day Values B-14
C-l Water Quality Ptecal Coliform levels C-3
C-2 Calculation of Number of Highly Credible Gastrointestinal
Cases for Ttenean Beach C-5
D-l Cost Data for a Typical Maine Clam Digging Firm D-9
D-2 Costs for a T/pical Massachusetts Shellfishing Firm
Operating in Unrestricted Areas D-10
D-3 Per Bushel tonlabor f&rvest Costs
for Boston Harbor Restricted Areas D-ll
D-4 Per Bushel Costs for Rmspecialized Items D-12
D-5 Per Bushel Specialized Costs for Subordinate Diggers .... D-13
D-6 Per Bushel Specialized Costs for Master Diggers D-14
D-7 Changes in Per Bushel Nonlabor Costs for Boston Harbor
Restricted Areas Due to Pollution toatement D-16
D-8 Comparison of Nonlabor Costs and Prices D-17
-viii-
-------
List of Tables and Figures
(continued)
Figure Page
Number Number
1-1 Boston Harbor Study Area 1-4
1-2 Schematic of Sources of Pollutant
Loadings to Boston Harbor 1-6
1-3 Area Served by the MDC Sewerage System 1-8
1-4 Location of Combined Sewers Overflow and Storm Sewer
Outlets within the Study Area 1-11
1-5 Current STP Dispersion Patterns and CSO Outlets 1-14
2-1 Area Served by the MDC Sewerage System 2-2
2-2 Schematic of Sources of Pollutant Loadings
to Boston Harbor 2-4
2-3 Location of Sewage Treatment Plants in Boston Harbor
Study Area 2-7
2-4 Dispersion of Current STP Discharges 2-14
2-5 Dispersion of Proposed Ocean Outfall Discharges 2-16
2-6 Dispersion of Proposed Secondary Treatment Discharges .... 2-18
3-1 Combined Sewer O/erflow and Storm Sewer Project Planning
Areas 3-2
3-2 Water Quality Standard Classifications in Boston Harbor . . . 3-4
4-1 Receptor Areas for the Boston Harbor Study 4-2
5-1 The Demand function and the Consumer Surplus
Welfare Measure 5-5
6-1 Receptor Areas for Boston Harbor Study Area 6-2
6-2 Effects and Responses to STP, CSO and Sewer Controls .... 6-52
8-1 Commercial Finfishing and Shellfishing Resources in
Boston Harbor 8-8
8-2 Typical Demand and Supply Curves for the Shellfish
Industry 8-15
12-1 Map of Charles River Basin 12-2
A-l Example of DISPER Output A-8
D-l Assumed Shape of Supply Curve for Boston Area
Soft Shelled Clam Market D-7
-ix-
-------
Section 1
Summary and Conclusions
The purpose of this project is to determine the feasibility and
usefulness of an economic analysis of the beneficial outcomes of water
quality improvements that should result from upgrading-sewage treatments
plants (STPs) and from combined sewer overflow (CSO) controls. This
report uses Boston Harbor, Boston, Massachusetts, to serve as a case study
which demonstrates the application of a variety of benefit estimation
techniques in order to develop a range of benefit values associated with \^~
the uses of the Harbor which would be affected by the various pollution
control treatment alternatives. It contains pertinent data and
computations to demonstrate the application of the techniques. This
report may also serve as an Appendix to the EPA's Marine CSO Handbook,
which states can use as an example of how to perform benefit analysis.
Where feasible, the study provides dollar estimates of the economic
benefits of the treatment alternatives for the two primary benefit <—~~
categories (recreation and commercial fishing) as well as for other ~^
relevant benefits.
The STP treatment options considered here include upgrading from primary
to secondary treatment and upgrading the existing primary treatment with an
ocean outfall. One of the STP options considered follows from the legal
mandate of the 1972 and 1977 Clean Water Act and Amendments, the Environmental
-------
1-2
Protection Agency (EPA) standards and procedures for the treatment and
disposal of municipal wastes. These regulations call for treatment at the
secondary level (which includes more BOD and SS removal in addition to basic
primary treatment) and a cessation of sludge disposal in the ocean.
The second STP option is an ocean outfall in conjunction with upgrading
existing primary facilities. Plans have been made by the Metropitan District
Ctommission (MDC) to repair and rehabilitate the STPs so that they will
function properly at an upgraded primary treatment level. In addition, the
MDC has applied for a variance under section 301 (h) of the Clean Water Pet
from secondary treatment requirements. The application is based on an
improved discharge whereby the two existing plants will improve their
operation of primary treatment, and effluent will be discharged at an ocean
outfall in Massachusetts Bay via a tunnel 12.1 km (7.5 miles) from Boston
Harbor. Since the initiation of this study, the proposed ocean outfall has
been tentatively denied by the EPA Administrator (in June, 1983) .
The selection of these options does not constitute endorsement of these
proposals over other STP options, nor is this study a part of the formal
301{h) evaluation efforts. Rather, since the purpose of this study is to
determine the feasibility and usefulness of an economic analysis of the
beneficial outcomes of improved water quality, the two STP options are
analyzed here as representative of the options under cons ieration at the time
the study was initiated.
The CSO control options are derived from studies done for the
Massachusetts District Commission as well as studies done for the town of
-------
1-3
Quincy. They include control of pollution due to combined sewer overflows,
stormwater discharges and dry weather overflows all of which contribute
significantly to the CSO problems in the Boston Harbor area.
Boston Harbor is surrounded by a major urban center and, despite its
serious water quality problems, provides the setting for many and diverse
water Uses including a fishing and shipping port, recreational boating,
swimming and beach activities, shellfishing, finfishing, and, especially
recognized in recent years, an aesthetic focal point for commercial,
residential and recreational activities. Figure 1-1 shows the geographic
features of the study area.
Due to the complexity of the situation, the constraints of the data, and >
i
the evolving nature of benefits analysis the results of this study should be I
viewed with caution. Every effort is made to assess the reliability of both
the data and methods used. In the individual chapters of the report specific
sections on the limitations of the analysis are provided.
This chapter provides a brief overview of the treatment alternatives,
receptors, benefit categories, and benefit methodologies. A comparison of the
benefits and costs of the alternatives is presented and the results of the
study summarized.
1.1 PDllution Sources
Two major sources of pollutant loadings to Boston Harbor are 1) the Nut
Island and Deer Island Sewage Treatment Plants (STPs) , owned and operated by
-------
Charles River .-:/•
Figure 1-1.. Boston Harbor Study Area
\
/* Presidents Roads'^
/ «
/ f
I'f'.nd ,
\
Tlic
Citlepf It- IX
tiCtor?!. |, ^ -
- - - " Naj\tasket Roads
HULL
• Wfrmogih ' ^*y>
r»f. •!»•> f}^*J
HINGHAM
-------
1-5
the Massachusetts Metropolitan District Commission (MDC) , and 2) the combined
sewer overflows (CSOs) located along the Harbor shoreline. The pollutants
which are released from these sources serve as parameters for describing the
environmental condition of the waters of Boston Harbor. Figure 1-2 is a
schematic presentation of how the pollutant loadings enter the harbor from
these sources. The following water quality parameters are considered in this
report:
Parameter Reason for Consideration
Coliform (fecal and/or total) important criteria for
swimming and shellfishing
needs; indicator of domestic
sewage pollution
BOD (Biochemical oxygen demand); conventional pollutants;
SS (suspended solids); standard wastewater
oil and grease characteristics
Heavy metals and toxics potentially dangerous to
(copper, mercury, nickel, etc.) aquatic life
Once these pollutants are released into the Harbor, they mix with ambient
waters, and can seriously compromise water quality and, consequently,
adversely affect the ecological habitat, recreation, aesthetic, and commercial
fishing activities, and personal health. The heavy metals and other toxic
pollutants affect the functioning of Harbor marshlands and influence the
abundance and diversity of shellfish and finfish in the waters. The
mechanisms and effects as related to levels of pollutant control are not
known, however. Thus, this report presents information on current loadings of
toxic pollutants from the STPs and qualitatively describes the ecological
habitat and potential effects for these pollutants.
-------
Figure 1-2. Schematic of Sources of Pollutant Loadings to Boston Harbor
stormwatersi
domestic,
commerciali S
industrial
waatewaters
combined
sewers
to ^
STPs ""
• (influent)— *-
sewage
treatment
process
J_£C1 A- \ ^^
BOSTON
HARBOR
WATERS
V
disinfection
(largely of bacteria .
and other pathogens)
-------
1-7
Rsrty-three towns and cities in the Boston Metropolitan area belong to the
Metropolitan Sewage System and send their domestic, commercial and industrial
wastewater to the two sewage treatment plants for treatment and disposal (see
Figure 1-3). At present, both plants are designed to carry out primary
treatment which is essentially a screening, sedimentation and chlorination
procedure. The treated effluent and concentrated, digested sludges are then
discharged into the ffirbor. System malfunctions are common, however,
resulting from such factors as outfall pipe deterioration, inadequate holding
capacity and lack of normal required maintenance due to, among other things,
difficulties in obtaining funds for repairs and suitable replacements for
malfunctioning components. As a result, the two STPs have not been
functioning properly in accordance with their designs, leading to raw sewage
bypassess directly into the ffirbor, improperly timed sludge releases, sewer
backups from the STPs, and less than design-level treatment performance, all
of which adversely affect water quality.
The two STP options consist of secondary treatment and upgraded primary
treatment with an ocean outfall. The secondary treatment option includes more
BOD and SS removal than the current primary treatment facilities and a
cessation of sludge disposal in the ocean. The ocean outfall option includes
repair and rehabilitation of the existing primary treatment facilities and
discharge of the treated effluent into Massachusetts Bay by way of a tunnel
from Deer Island. These two options were picked from the many proposals being
studied at the time of this report as representative of the proposals and not
as an endorsement of one proposal over another.
-------
1-8
Figure 1-3. Area Served by the MDC Sewerage System
MASSACHUSETTS
communities which will
possibly be added to
the system
jt a portion r : Hingham
is presently part of
the system
-------
1-9
l&ble 1-1 compares the annual costs of the STP options and shows very
approximate percentages for reductions in effluent pollutants, including
BOD5» suspended solids (SS), and metals, over existing concentrations.
In its effort to develop a comprehensive plan for CSO control in Boston
Harbor, the MDC has designated four CSO planning areas: 1) Dorchester Bay,
2) Neponset River, 3) Inner Barbor (including Constitution Beach) and
4) Charles River Basin. The four areas are defined on the basis of existing
water use and coastal use patterns. The water quality of all four planning
areas is compromised by pollution from combined sewer overflows (CSOs),
stormwater discharges, and dry weather overflows (DWOs). Storm-related
combined sewer overflows vary in duration and frequency. DWOs, caused by
sewer blockages and other malfunctions, are continual discharges of sanitary
wastewater and are considered by the MDC to be the single most important
source of pollution in Boston arbor. They have thus been included in all the
CSO plans even though they are not officially classified as CSOs under federal
regulations. Combined sewer overflow outlet locations are shown in
Figure 1-4.
Another source of pollutant loadings to Boston Harbor is the Quincy
storm sewers. The Quincy storm sewers discharge waters with fecal coliform,
BOD and SS concentrations that are higher than levels expected from storm
water runoff. Storm water contamination can result from cross-connections
between sanitary and storm drains, due to broken pipes and exfiltration from
sanitary sewers in disrepair, and, possibly, illegal "tie-ins" to the storm
sewer system although the latter has not been documented in Quincy. These
present problems similar to the DWOs in Boston which have been included in the
CSO plans. The Quincy storm sewers lie outside the MDC study area of
-------
1-10
Table 1-1. Costs and Potential Reductions in STP
Effluent Pollutants for the STP Options
(Millions 19823)
Wastewater Costs
Treatment 1 Annualized ,| Annual I Total
STP Options 1 Capital Cost- I O&M Cost 1 Annual Cost
Upgraded Pr imary
With Ocean Outfall 74.9 22.0 96.9
Secondary 85.8 45.2 131.0
1 Approximate Percentage
1 Reduction in t,/"
1 Effluent Pollutants -
£/
60 - 80
5/ Based on 8 1/8 percent interest; 20 year period.
b/ Average potential reductions in effluent pollutants (8005, SS and
metals) over existing concentrations. Range is a very approximate
estimate. For four heavy metals (cadmium, chromium, lead, mercury) the
reduction would be about 30%.
£/ No effluent will be discharged in Boston Harbor. There will be
increases of pollutants in Massachusetts Bay, however. See Section 4
for details.
Source: See Tables 2-2 and 2-3, Section 2.
-------
Figure 1-4
Location of Combined Sewer Overflow
and Storm Sewer Outlets within the
Study Area
-------
1-12
concentrated CSOs. However, they have been included as an option for this
benefit-cost study because they have a significant adverse impact on the water
quality of Quincy's town beaches and Wollaston Beach, a large MDC operated beach
attracting many visitors, located in Quincy.
T&ble 1-2 shows the annual costs of the CSO options along with the
approximate percentage reduction in pollutant loadings, including fecal coliform,
floatable and suspended solids and oil and grease. The top part of the table
presents the four CSO plans as designated by the MOC. The bottom part shows the
options used in the benefit-cost analyses in this study (for a detailed
discussion of the CSO options see Section 3) . The options as defined in the
lower half of the table correspond more appropriately with the benefit estimates
associated with the uses of the Harbor. For example, all the swimming and
shellfishing uses affected by the GSOs (and therefore the corresponding benefits
estimates) can be captured by including only the Constitution Beach portion of
the Ihner Harbor Plan plus the Darchester Bay, NSponset River, and Quincy Bay
Plans. The CSO options in the table reflect incremental increases in annual
costs.
1.2 Water Quality
Currently, the CSOs and STPs jointly affect some of the same harbor areas
(see Figure 1-5). However, the CSOs generally affect the areas closest to the
shore including the shoreline swimming beaches and fishing and boating areas near
the shore. In comparison, the STPs have the greatest impact on water surrounding
the STP outfalls and thus mostly influence the central parts of the harbor,
particularly the Boston Harbor Islands. Beaches in the towns of Quincy,
Weymouth, Hing ham and Hull are also affected.
-------
1-13
Table 1-2. incremental Costs and Potential Reductions
in Pollutant Loadings for the CSO Options
(Millions 19829)
I MDC PLANNING AREA DESIGNATION 1
Treatment
Alternative/ 1 Annualized I Annual
Receptor I Capital Cost £/ |O&MCost
Inner Harbor
a) including
Constitution
b) Constitution
only
Dorchester Bay
Neponset River
Charles River Basin
Implementation of
all MDC design-
ated CSO plans
14.63 1.97
0.04 0.01
4.97 0.37
0.61 0.10
8.87 1.56
35.44 4.00
(Percentage
(Reduction in
1 Total (Pollutant
1 Annual Cost I Loadings V
16.61
0.05
5.34
0.71
10.43
33.39
50 - 99
70 - 99
60 - 98
65 - 100
50 - 100
1 STUDY AREA DESIGNATION I
Inner Harbor
Constitution
Beach only
Dorchester Bay/
Neponset River
Quincy Storm
Sewers £/
Pbove three plans
combined
Charles River
0.04 0.01
5.59 0.47
0.27 -.02
5.90 0.46
8.87 1.56
0.05
6.06
0.25
6.36
10.43
50 - 99
60 - 99
60 - 99 I/
50 - 99
65 - 100
3/Based on 8 1/8 percent interest; 20 year period.
—/From Contractor reports.
£/Quincy plan is currently undergoing extensive revision,
I/Assumed to be the same as Dorchester Bay Area.
-------
Charles
I • —
• •/*'Presidents Roads
•/v&tf -•
"'••'••• y-\.«
.". Najntasket Roads
Figure 1-5. Current STP Dispersion
Patterns and CSO Outlets
DORCHESTER
•l«ir .
^^
QUINCY
HULL
HINGHAM
Areas of heaviest
STP loadings
Areas of moderate
STP loadings
Areas of slight
STP loadings
CSO/Storm sewer
outlets
-------
1-15
The various STP and CSO treatment options will reduce pollutant loadings
to the Harbor waters. The change in ambient water quality at various
locations throughout the Harbor will depend on the change in reduced loadings
but also on the dispersion pattern in the Harbor from the point of discharge
to the receptor areas where recreation, boating and fishing take place.
Several water quality models were used in the various contractor reports
delineating the STP and CSO options. We use the results of these models to
predict improvement in water quality related to percent reduction in pollutant
loadings for the different treatment options at each receptor point in the
study area. (See Section 4.) These estimates are presented in Table 1-3.
The accuracy of the water quality models depends on both the data and
methodologies available. Complexities due to currents, tides and weather make
the transport and fate of pollutant discharges difficult to model. The
results currently available preclude estimation of absolute changes in water
quality but the relative percentage changes, as shown in Table 1-3, are
adequate for the benefit estimation procedures used in this study.
1.3 Benefit Categories and Receptors
t
The benefit categories for which benefit estimates have been computed in
this study have been determined by those uses of Boston Harbor that are
affected by the pollution sources discussed above (STPs and CSOs). A term
often used to describe areas or uses which are adversely affected by pollution
sources and which would benefit from pollution abatement options is
"receptors." The receptors or benefit categories in this study include
recreation activities such as swimming, boating and fishing, commercial
finfishing and shellfishing, the ecological habitat of the harbor and
non-users who would be willing to pay, nonetheless, for pollution control
-------
1-16
Table 1-3. Estimated Water Quality impacts of the STP and CSO
Treatment Options
1 Percent Pollution Reduction by
I Treatment Option
Receptor Area
Constitution Beach
Dorchester Bay
Cuincy Bay
Hingham Bay
Outer Harbor Islands
Brewsters Islands
ten task et Beach
Massachusetts Bay
Charles River
1 Combined
1 Sewer 0/erflow/
1 Storm Sewer
50 to SO
60 to 90
60 to 90
—
—
—
—
—
50 to 80
1 1
1 Deep Ocean
1 Outfall
1
5 to 10
10 to 25
10 to 20
15 to 40
60 to 90
-10 to -15
-5 to -10
-35 to -45
1
1 Secondary
1 Tteatmen t
1
0 to 5
5 to 15
10 to 20
15 to 40
30 to 80
30 to 40
0 to 5
15 to 20
1
Nate: Positive figures denote improved water quality. Negative figures
denote degradation in water quality.
-------
1-17
(intrinsic benefits). Alternative pollution control programs and the
affected receptors are shown in Table 1-4.
The benefits of improved water quality resulting from implementation of
pollution control options in Boston Harbor accrue to users and non-users
alike, and are presented below with a summary discussion of specific benefit
estimates. The techniques used in this report to measure benefits to society
from implementation of pollution control plans are based on the theory of
welfare economics and the concept of willingness to pay. This economic theory
is founded on the principle that the demand" for water quality is the sum or
aggregate of how much individuals would be willing to pay to receive
additional increments of improved water quality. Section 5 discusses the
theoretical concepts, benefits categories and the various methodologies used
to estimate benefit values for the different treatment alternatives.
1.4 Summary of Study Findings
A summary of annual benefits and costs for the different control scenarios
is presented in Tables 1-5 through 1-7. The control scenarios include the
MDC's recommended plans for CSO control and also the benefits of implementing
CSO controls along with the STP options. The tables report the dollar
estimates for the benefit categories and receptor areas for Boston Harbor. An
indication of those benefits which were not monetizable in this economic
analysis is also included to emphasize the full range of impact of these
pollution sources and their consequent clean-up, (he way to consider this
potentially large non-monetizable portion from the point of view of the
decision maker is an implicit evaluation of what they must be worth if it is
decided to implement the controls by considering the difference between the
-------
1-18
Table 1-4. Pollution Control Program and Receptors
Pollution Control
Option
I Predicted a/
I Par cent Cleanup"
I Receptor s/Benef i t Categories
STP
Ocean Outfall
Secondary
CSO
Inner Harbor
(includes Constitution)
Dorchester Bay and
Neponset River
Quincy Storm Sewers
Charles River
10 to 30
-40 to -10
5 to 30
20 to 70
70
80
80
70
Beaches: Weymouth, Hingham, Hull
Boating and fishing
Siellfishing
Intrinsic and Biological
Beaches: tentasket,
Brewsters Islands
Boating and Fishing
Intrinsic and Biological
Beaches: Constitution,
Dorchester Bay,
Cuiricy Bay,
Hingham Bay
aiellfishing
Intrinsic and Ecological
Recreation: Cuter Harbor
Islands
Boating and Fishing
Intrinsic and Biological
Beach: Constitution
Boating and Fishing
Shellfishing
Intrinsic and Biological
Beaches: Castle Island,
Pleasure Bay, Carson,
Malibu, Tfenean
Boating and Fishing
aiellfishing
Intrinsic and Ecological
Beaches: Wollas to n , QJ i nc y
Boating and Fishing
Siellfishing
Intrinsic and Ecological
Boating
Intrinsic
a/ see Tables 4-2 and 4-3.
-------
Table 1-5. Annual Benefits and Costs of Combined Sewer Overflow Controls
(Millions 19829)
Benefit Estimates by Category
tellutlon
Control
Option
Recreational Recreational
Swimming!/ Boating Pishing
Health!*/
Oammerclal
She 11-
PlahlngS/
Intrinsic!/
Ecological
ITbtal
1 Annual
ICostsV
TOTAL I
Oomhlncd Sewer Overflows
Conntltutlon
Beach
Dorchester Bay/
Neponset River
Qulncy Bay
Hlngham Bay
Massachusetts Bay/
Nan task et
Entire Harbor
(not including
Charles River)
Charles River
Pour MDC CSO Plans
(Constitution,
Dorchester, Neponset,
Charles River)
Range:
Moderate:
Range:
Moderate:
Range:
Moderate:
Range:
Moderatei
Range:
Moderate:
Range:
Moderate:
Range:
Moderate:
Range:
Moderate:
0.91-1.36 Not available for this
1.14 option since boating
and fishing are only
6.21-9.29 calculated harbor-wide
7.75 for combined STP and
CSO options.
S. 29-7.91
6.60
-0-
-0-
12.OS-18.OS/
15.02
-0- .05-. 96 -0-
.51
7.12-10.65 .05-.96
8.89 .51
.005-. 077
.041
.021-. 117
. 169
.086-1.275
.681
-0-
-0-
.124-1.716
.92
-0-
.027-. 394
.21
O-.OOS
.001
.001-.009
.DOS
0-.004
.002
-0-
-0-
.001-. 018
.010
-0-
.001-. 014
.008
Based on total
recreational
benefits. Not
available for
this option
since boating
and fishing
benefits are
only calculated
harbor-wide
for combined
STP and CSO
options.
3.14-6.28*
4.71
3.14-6.28*
4.71
Cannot be
quantified
but Includes
value of
highly
productive
saltmarshes
In Boston
Harbor.
Inese marshes
In turn
support many
species of
fish and
invertebrates
as well as
animals.
shoreblrds
and
waterfowl.
0.92-1.44
1.18
6.23-9.62
7.92
5.38-9.19
7.28
-0-
-0-
12.18-19.71*
15.95 1
3.19-7.24
5.22
10.34-18.3
14.3
o.o si!/
6.06
0.25-
6.06V
6.16-
2. 17i/i/
10.43
16.54i/
VO
S/ Moderate benefits represent best estimates except for those categories where best estimate is marked by •. Range Includes high and low
estimate.
!>/ Swimming benefits based on conditional log It model. For Qulncy, Hlngham and Nantasket beaches, benefits from Increased participation are
added since loglt model did not Include these beaches. All benefits are derived using user day value from logit model.
£/ Includes general recreation benefits at Boston Harbor Islands.
2/ Health benefits for individual areas based on swimming; for entire harbor benefits based on shellfish consumption are also included.
S/ Commercial fishing benefits based on shellfIshlng; estimates for flnflshing and lobstering not available.
y Intrinsic benefits based on SO percent of all recreational benefits; except for Charles River, which includes willingness to pay for user
and non-user values.
9/ Annuallzed capital costs (assuming 8 1/8 percent interest, 20-year period) plus annual operation and maintenance costs.
"/ Excludes cost of Inner Harbor CSO plan except for Constitution Beach portion; total annual cost of Inner Harbor CSO plan Is SIS.61 million.
y Cost estimates for Qulncy storm sewers are still preliminary. High estimate Is equivalent to costs for CSO control In Dorchester Bay.
-------
Table 1-6. Annual Benefits and Coats of Combined Sewer Overflow Controls and
Ocean Outfall Cbntrol Option (Millions 198231
Benefit Estimates by Category
Pollution
Control
Option
Recreational
Swimming]/ Boatlnq
Recreational
Fishing Health!!/
Commercial
Shell-
Fishing*/
IIDtal
1 Annual
IdosttV
Intrinsic!/ Ecological TOTAL 1
(tomb ln<>
-------
Table 1-7. Annual Benefits and Costa of Combined Sewer Overflow Controls and
Secondary Treatment Control Option (Millions 19823)
"Dilution
Oontrol
Option
Benefit Estimates by Category
Recreational
Boating
Recreational
Pishing
Health!/
Commercial
She 11-
Plahlnge/
Intrinsic!/ Ecological TOTAL
llotal
I Annual
ICustsV
Oamhlned Sewer O/er flows
and Secondary Treatment
Constitution
Beach
Dorchester Bay/
Noponset River
Qilncy Bay
Hingham Bay
Massachusetts Bay/
Nantasket
Range i
Moderate:
Range:
Moderate:
Range:
Moderate :
Range:
Moderate:
Range:
Moderate:
.98-1.46
1.22
7.41-11.08
9.25
6.24-9.33
7.78
.215-. 322
.269
-0-
Entire Harbor
(not including
Charles River)
Charles River
Four HOC CSO Plans
(Constitution,
torchester, teponset,
Charles River)
Range: 14.22-22.42E/ 6.46-14.57* .75-9.49
Moderate:
Range:
Moderate:
Range:
Moderate:
18.32
-0-
10.52
.05-.96
.51
5.12
-0-
.007-.096
.051
.032-.477
.255
.146-2.15
1.15
.003-.039
.021
-0-
.198-2.81
1.51
-0-
.022-. 124
.064
-0-
10.7-23.2
17.0
3.14-6.28*
4.71
Potentially
large
beneficial
impact on
shoreline
saltmarshes
supporting
fish and
invertebrates
as well as
animals,
shoreblrds,
and waterfowl.
0.99-1.56
1.27
7.44-11.56
9.51
6.39-11.48
8.93
.22-.36
.29
-0-
32.35-72.61* 137.4-
52.53 143.2W
3.19-7.24
5.22
10.43
*/ Moderate benefits represent best estimates except for those categories where best estimate Is marked by *. Range includes high and low
estimate.
£/ Swimming benefits based on conditional log It model. For Quincy town beaches, benefits from Increased participation are added since logit
model did not Include these beaches. All benefits are derived using user day values from logit model.
£/ Includes general recreation .benefits at Boston Harbor Islands.
i/ Health benefits for individual areas based on swlmmlngi for entire harbor benefits based on shellfish consumption are also Included.
£/ Commercial fishing benefits based on shellfIshing; estimates for flnfishlng and lohsterlng not available.
I/ Intrinsic benefits based on SO percent of all recreational benefits) except for Charles River, which Includes willingness to pay for user
and non-user values.
I/ Annualized capital costs (asiumlng 8 1/8 percent interest, 20-year period) plus annual operation and maintenance costs.
^/ Excludes cost of Inner Harbor CSO plan except for Constitution Rr.ich portion; total annual cost of Inner Harbor CSO pi in Is $16.61 million.
I/ Cost estimates for Quincy storm sewrs are still preliminary. Iliih estimate is equlvjl»nt to costs for CSO control in Dorchester D.iy.
-------
1-22
annual benefits as estimated and the predicted annual costs. Che
result that does stand out is that in addition to either secondary
treatment or an ocean outfall the GSO problem needs to be addressed
if full use restoration and health benefits are to be realized.
Some specific conclusions of this study include:
o Monetizable benefits
-- Swimming benefits and all kinds of recreational benefits
are the largest source of the monetizable benefits. In the
commercial fishing category, we could only estimate
shellfishing benefits. Nonetheless the recreational
categories appear to be especially important for urbanized
areas such as Boston arbor where local population density
and demand for nearby recreational opportunities are high.
— The geographic location of the pollution sources in
relation to the receptor or benefit categories is an
important factor in determining the type and level of
benefits that will be generated by the different treatment
options. In the case of Boston Iferbor most of the recreation
beaches are significantly affected by the CSO discharges and
only moderately affected by the STPs. Ch the other hand,
fishing and boating in Harbor waters are more affected by the
STP discharges. In the case of fishing and boating, however,
a further constraint is marinas and facilities—a constraint
on increased participation in these activities not related to
pollution control.
--In our calculations the CSO options can be broken down by
MDC Planning Area. lor example, benefits related to the
Dorchester Bay and Neponset River Plans and the Constitution
Beach portion of the Inner Harbor Plan are summarized in
Table 1-5. Also, Charles River and Quincy Bay can be
isolated. This separation of plans is possible because of
the geography of Boston Harbor and it would not be possible,
necessarily, for all areas of the country. However, in our
case the separation of plans can assist in the determination
of the most effective way to allocate CSO control funds.
o Non-monetizable benefits
-- Several categories include only a partial estimation of
benefits. The commercial fishing category includes
shellfishing only. Although up to 2.6 million pounds of
lobster and 28.4 million pounds of fish are landed annually
in the port of Boston, benefits related to this activity were
not calculated because of the difficulty of knowing where the
fish were caught and how they might be affected by the
improved water quality.
-------
1-23
— Intrinsic benefits include aesthetic benefits and benefits
such as existence and option value not directly related to
use of the water resource. These are best evaluated by
willingness-to-pay measures. As can be seen in the case of
the Diaries River (Table 1-5) , they can be quite
substantial. For the other areas in this study
willingness-to-pay measures were not available, and the
intrinsic benefit estimates were related to recreational
activity which might not capture all non-user benefits. .
— A potentially large category of benefits not captured in
this economic analysis is ecological benefits--benefits
related to preservation and restoration of the harbor and bay
habitats. The volume of pollutants controlled by the STPs is
far greater than that controlled by the CSOs (approximately
30 times greater). Therefore, from an ecological perspective
we need to be very concerned about the long term impacts that
those heavy metals, toxics and other constituents in the STP
effluents have on the harbor and bay habitats even though
they are not immediately reflected or easily captured in the
economic analysis. The CSOs are also of concern because of
their proximity to highly productive saltmarshes along the
shoreline.
— In this study we have looked at uses of the Harbor waters
which could be most directly analyzed within our economic
analysis framework. This resulted in the exclusion of the
inner Far bo r CSO Plan except for the Constitution Beach
area. The Inner Harbor CSO control plan (reducing odor,
floatables, and toxic substances) would include benefit
categories of commercial use, aesthetics and ecological, none
of which were monetizable. There are relatively few
recreational uses in this area. Given the large amount of
effluent discharged (about 11 billion gallons per year) , the
control costs are quite high and it would not appear that
this CSO plan would be as important as the others in its
overall impact.
Costs
— The costs for the CSO control options are estimates for
preferred control alternatives. However, the costs for the
Quincy Storm Sewers may not be comparable to the costs as
used in the rest of the report. The Quincy cost study is
still in the preliminary stages and not nearly as detailed as
the other CSO plans. Thus, we show in the summary tables an
upper range estimate equal to the CSO control costs for
Dorchester Bay, its neighbor to the north.
As is clear from the discussion above, the benefit estimate numbers
presented in Tables 1-5 through 1-7 should not be taken as especially
-------
1-24
important or precise in themselves. They are approximations and represent
means computed from ranges, sometimes wide ranges, that have been developed
for each benefit category; they are the result of, for the most part,
conservative assumptions; and they generally underestimate the benefit values
of the treatment options. R>r instance, as discussed above, ecological
benefits have not been included as they are considered non-monetizable (see
Section 10). Recreational boating and fishing benefits (except for Charles
River) have been computed only for the Harbor as a whole, since data was
unavailable to break the totals down by option. The totals were included,
however, to give an idea of the possible magnitude of these benefits. Despite
these shortcomings, it is apparent from the conclusions that have been drawn
that an economic analysis of the beneficial impacts of water quality
improvements is feasible and is a useful tool for providing information to
decision makers to facilitate improved policy decisions, especially where
there is a choice to be made among various alternatives and a limit to the
available funding.
1.5 Specific Benefit Btimates
Benefits accrue to households who recreate in, on or near the water, to
consumers of commercial fisheries, to consumers who benefit directly and
indirectly from the increased economic activity in the primary sector, and to
non-users of Harbor waters, who derive intrinsic benefits. Each benefit
category, estimation procedure, and benefit estimate are briefly described
below.
1.5.1 Recreation
Benefits from increased recreational opportunities are the greatest of all
the monetizable benefit categories. Benefits accrue to swimmers, boaters,
-------
1-25
anglers and those who recreate near the water. Two major components of
consumer surplus have been estimated which fully capture benefits from ^
improved water quality: (1) increase in participation, and (2) increase in
the price participants are willing to pay per visit for the improved quality
of the recreational experience. The following is a brief summary of the three
major recreation benefit categories considered in this study.
Swimming. A variety of benefit estimation methodologies were employed to
estimate swimming-related benefits. These included: (1) using recreation
studies to predict and value increases in participation; (2) applying a travel
\
cost, conditional logit model to estimates gains in consumer surplus due to
increased participation and increased satisfaction per trip; and
(3) calculating consumer losses stemming from beach closings. Results from
the travel cost model are the most accurate of all the methodologies because
of the theoretical and empirical strengths of the logit model. Benefits
associated with the CSO control options are substantial: $18-19 million
for swimmers throughout the Harbor area for a full plan of STP and CSO
controls. About $15 million of this is related to GSO controls because of the
proximity of their discharges to the shoreline beaches. (See Chapter 6.)
Fishing and Boating. Fishing and boating benefits have been calculated
only for the entire Harbor study area because of data limitations. Benefits
for both these categories are substantial: $12 to 15 million for both
activities for combined STP and CSO controls. (See Chapter 6.)
Boston Harbor Islands—Ml Recreation activities. Ohe Boston Harbor
Islands are a unique recreation resource that will benefit from improved water
quality resulting from the implementation of the STP treatment alternatives.
-------
1-26
Recreational data was used to predict increase in participation in all Boston
Harbor Island activities. Benefits total $1 to 3 million. (See Chapter 6.)
1.5.2 Health
Health benefits from water pollution abatement include willingness to
pay to avoid swimming-related illnesses and shellfish consumption-related
illnesses. Dose-response data were used to evaluate swimming-illness
benefits. NO such functions exist for consumption of shellfish, and thus
these benefits were developed by assuming that a percentage reduction in
shellfish-borne diseases is directly proportional to percentage reduction
in the concentration of the fecal coliform in the water. Total health
benefits from CSO and STP controls are about $1.5 million. They are
lowest at Constitution Beach and highest at the Wollaston/Quincy beaches,
which have the highest swimming attendance and are in close proximity to
the Quincy storm sewers. Shellfish consumption benefits can only be
linked to pollution reduction throughout the entire harbor. Benefits are
small, from $0.001 million to $0.005 million. (See Chapter 7.)
1.5.3 Commercial fisheries
Water pollution abatement in Boston Harbor would probably result in a
recla' sification of shellfish beds from grossly contaminated (closed beds) to
moderately contaminated (restricted beds), thereby allowing increased
shellfish harvesting with depuration. Moderate benefits are about $0.06
million for combined STP and CSO controls. (See Chapter 8.) These benefits
do not include the sizable commercial catches of finfish and lobster. Current
-------
1-27
annual value of these catches reaches $18 million. We were not able to
calculate incremental annual benefits for this portion of commercial fishing
bene f i t s, howeve r.
1.5.4 Intrinsic Benefits
Water pollution abatement is predicted to have an important effect on
benefits which are not specifically related to actual water use, such as
option, existence, and aesthetic values. Except for the Charles River,
t
because of the lack of appropriate willingness to pay survey data which could
be applied to the different treatment alternatives in the study area,
intrinsic values have been estimated by assuming that non-user benefits are
one-half as great as recreational user benefits. (See Chapter 9.) Moderate
estimates for intrinsic benefits total $16-17 million.
1.5.5 Ecological Impacts
Pollution abatement might positively influence ecological processes in
saltmarsh areas throughout the harbor. Although attempts have been made to
estimate the economic value of marshlands by valuing the role of the marsh as
a factor of production, and by estimating the cost of duplicating these
functions, it was not possible to apply these results to the Boston Harbor
study area. This is because the connection between the levels of pollution
control, the subseq »nt reduction of pollutant loadings to the water column
and the functioning of the marshlands is unknown for the harbor. Furthermore,
the role of pollutants already in the sediments, that could be resuspended '•
into the water as loadings are reduced, is not well understood at this time.
Therefore, these benefits have been considered non-monetizeable. (See Chapter
10.)
-------
1-28
The adverse ecological impacts believed to be caused by current and past
levels of pollutant loadings include:
— the alteration of benthic populations which may reduce the food
supply, thereby resulting in a decrease in commercially valuable
fish variety and numbers;
— the accumulation of toxics by benthic fauna and then passage up
the food chain where they pose a health risk to consumers (copper,
mercury, PCBs, silver found in tissues of lobster and winter
flounder);
— bioaccumulation which can affect species reproduction, increase
potential for disease (fin erosion in winter flounder associated
with PCB contamination), and impair predator avoidance behavior
which could result in reduced numbers and variety of fish.
Important commercial species that may be adversely impacted include lobsters,
manhaden, cod, bluefish, striped bass and eels. Efcological benefits would
accrue to the pollution control measures if the reduction in pollutant
loadings caused reductions in the aforementioned adverse impacts.
The ecological benefits of the STP options may be larger because the
volume of discharge is about 30 times as great as for the CSOs. However, the
ocean outfall option will negatively impact some of the areas in
Massachusetts Bay which include:
— commercially valuable species such as tautog, cod, pollack,
haddock, halibut, mackeral; and
— migratory and endangered species such as whales, sea turtles,
sturgeon and the leregrine falcon.
1.5.6 Secondary Effects
Improving water quality will result in secondary effects from increases in
economic activity generated in an area by direct impacts, such as commercial
fisheries or recreation activities. A range of input and output multipliers wei|
-------
1-29
applied to each benefit category to compute all secondary economic effects.
Secondary effects cannot be linked to each pollution control option for every
primary benefit category because some of the benefit categories, such as fishing
and boating, could only be developed on a harbor-wide basis. We have chosen to
refer to these values as effects, rather than benefits, because only under certain
circumstances can secondary effects be considered benefits and the labor market
analysis required for delineation and definition of these circumstances was beyond
the scope of this case study. For these reasons we have calculated the different
secondary effects, but have not included the dollar value in the summary of total
pollution control benefits. (See Chapter 11.)
1.5.7 Charles River Basin
Benefits to instream, near-stream users and non-users of the Charles River were
calculated by estimating increase in boating participation and by applying results
from a willingness to pay survey. Boating benefits are small ($0.51 million)
because all river acres in the Charles River Basin currently are used for boating
and because user day values used to value this increase are moderate. The benefits
of improving water quality along the Charles more accurately are measured by
applying results of a willingness to pay survey, which captures benefits to users
and non-users alike. Benefits calculated using this methodology are substantial:
$4.7 million. Despite the large size of these benefits, they are approximately
half of the estimated $10.43 million annual cost, of implementing the Charles River
f
Basin CSO plan. (See Chapter 12.) '
1.6 Guide to the Report
This chapter has summarized the features of the study area, the treatment
alternatives and the benefit categories. It also has presented a brief
analysis of the treatment options and a brief summary of study results and
-------
1-30
conclusions. The specific STP and (50 treatment options are discussed in
detail in Sections 2 and 3. Their effects on Harbor water quality are
included in Section 4. Section 5 presents a brief introduction to the
theoretical and methodological approaches used to measure benefits from
improving water quality, and discusses the benefit categories applicable to
this case study. The next six sections describe each benefit category and
include benefit estimation methodology, data bases used in the analysis,
benefit estimates, and limits to the analyses: Section 6, Recreation
Benefits; Section 7, Health Benefits; Section 8, Commercial Fisheries;
Section 9, Intrinsic Benefits; Section 10, Ecological Benefits; and Section
11, Secondary Effects. Section 12 presents a separate analysis of benefits
from implementing the Charles River Basin CSO Plan.
Several Appendices follow the major text. Appendix A gives a more
detailed view of STP treatment alternatives and their effects on Harbor water
quality. Appendix B presents detailed calculations for the different
methodologies used to estimate recreation benefits and includes a description
of the major recreation sources used in this analysis. Appendix C explains
how health benefits are calculated and Appendix D presents a step by step
analysis of commercial fisheries benefits calculations. Appendix E summarizes
calculations of recreation boating benefits from water quality improvement in
the Charles River Basin.
-------
Section 2
Municipal Sewage Treatment Plant Operations, Options
and Water Quality Impacts
The 43 towns and cities belonging to the Metropolitan Sewage System (see
Figure 2-1) generated approximately 167,900 million gallons of raw, mixed
domestic, commercial, and industrial wastewater in 1980. Among the
responsibilities of the Metropolitan District Commission (MDC), a public
service authority for the 43 municipalities, is the collection, treatment and
disposal of these municipal wastes. To fulfill its responsibility, the MDC
owns and operates two sewage treatment plants (STP), one at Deer Island and
the other at Nut Island, which handle the wastes from the northern and
southern member municipalities, respectively. At present, both plants are
designed to carry out primary treatment, which is essentially a screening,
sedimentation, and chlorination procedure. They then discharge both the
treated effluent and concentrated, digested sludges into the outer harbor.
Under the legal mandate of the 1972 and 1977 Clean Water Act and
Amendments, the Environmental Protection Agency (EPA) established standards
and procedures for the treatment and disposal of municipal wastes. The new
regulations call for treatment at the secondary level (in addition to primar-'
treatment) and a cessation of sludge disposal in the ocean. Ohe intent of
the regulation is to reduce the degradation of water quality that is caused
by municipal waste loadings.
-------
2-2
Figure 2-1
Area Served by the
MDC Sewerage System
MASSACHUSETTS
communities which will
possibly be added to
the system
a portion of Hingham
is presently part of
the system
-------
2-3
Prompted by the aforementioned regulation, numerous studies have been
undertaken to determine the engineering feasibility of treatment alterna-
tives, how to manage and handle residual sludges, etc. For the most part,
these studies have limited their analyses to what can be done to satisfy the
new regulatory requirements, either through direct compliance (secondary
treatment and no sludge discharges to the ocean) or with options available
through waiver opportunities (upgraded primary treatment with a deep ocean
outfall, sludge barging).
Section 3 discusses the technical, environmental, and financial options
for combined sewers in the Boston Metropolitan area. STPs are discussed here
separately. At times it will be necessary to bring combined sewers into the
following discussion since their performance can affect that of the STPs and
vice-versa.
The intent of this section is to: (1) describe current Deer and Nut
Island STP performance and pollutant loadings; (2) present the financial and
expected performance characteristics of two proposed STP options; and (3)
discuss the potential water quality impacts of these proposed STP options. A
vast amount of information was analyzed for the development of this chapter.
What is presented here is essentially the conclusions of that effort. The
background and formulation of the most important analyses are explained in
Appendix A.I.
2.1 Current STP Performance
The existing Deer Island and Nut Island STPs are designed to treat
municipal wastewaters at the primary level. As the flow diagram of Figure 2-2
illustrates, most constituents of the municipalities' wastewaters eventually
-------
Figure 2-2.
Schematic of Sources of Pollutant Loadings to Boston Harbor
stormwatersj
domestic,
commercial, &
industrial
wastewaters
disinfection
(largely of bacteria
and other pathogens)
combined
sewers
to ^
STPs "
——{influent) — *-
sewage
treatment
process
BOSTOtt
HARBOR
WATERS
V
to
-------
2-5
reach the harbor in one form or another, with the exception of a portion of
the organic constituents which are lost through disinfection in the treatment
process. Metals and other non-destructables remain relatively unchanged
while passing through the treatment system and are, therefore, discharged in
either the sludges or effluent from the STPs.-3r The STP-source loadings
have been calculated on an annual basis for comparison of their relative
magnitudes; they are presented in Table 2-1. Effluent and sludge loading
information was calculated using measurements taken of wastewaters that had
undergone complete treatment, which does not account for raw (untreated)
wastewater discharges. Therefore, loadings from STP bypasses of raw
wastewaters were calculated from influent composition and bypass volume
data.-*/
Both the treated wastewaters and the solids sludges extracted by STP
treatment are discharged through local outfalls into Presidents and close to
Nantasket Roads from the Deer Island and Nut Island STPs, respectively. These
two 'Roads* are the major deep and fast-flowing channels of the Harbor (see
Figure 2-3 for their location). Whereas much of the harbor is only 10 to 15
feet deep, the depths of President and Nantasket Roads range up to 90 feet.
The STPs discharge to these locations because of their capacity for carrying
and dispersing effluent and sludge loads. The plants' effluents are
di' -harged continuously whereas sludges ideally are released only on outgoing
tides. Since the sludges generally contain a high percentage of the original
influent's pollutants, their releases are timed for maximum removal from the
2/ Some chemical recombination and physical change of the wastewater
constituents can be expected, but essentially, mass is conserved.
—/ CSO loadings have not been calculated from this same raw wastewater
information because data regarding the frequency, duration, stormwater
dilution, and volumes of overflow events are not available.
-------
2-6
Table 2-1.
Comparison of STP loadings
for Deer and Nut Islands Combined
BOD5
TSS
Cd
Cr
Cu
Pb
Hg
Ni
Zn
STP
Effluent
(Ibs/yr)
154x10^
124x106
26,000
138,000
325,000
178,000
2,000
241,000
702,000
Exist!
1 STP
1 Sludges
1 (Ibs/yr)
17xl06
4 5x10 6
4,800
72,600
115,700
36,500
700
25,900
. 222,700
nq loadings
1 STP
1 Bypasses
1 (Ibs/yr)
10-15xl06
10-15xl06
1,000
4,700
22,400
6,300
100
22,700
. 29,000
1 Total STP
1 Discharges
1 (Ibs/yr)
181-186xl06
17 9-18 4x10 6
31,800
215,300
463,100
220,800
2,800
289,600
! 953,700
STP
Effluent
from
Ocean
Outfall
(Ibs/yr)
18 Ox 10 6
135xl06
26,000
138,000
325,000
178,000
2,000
241,000
702,000
STP
Effluent
from
Secondary
Treatment
(Ibs/yr)
46xl06
46xl06
23,000
101,000
159,000
141,000
0
172,000
. 419,000
£/ Conversions of mg/1 data to Ibs/year figures made assuming 500 million
gal/day of effluent discharged from Deer Island and Nut Island combined. See
Appendix A.I for further explanation of calculations.
Sources: US EPA (1978), Tables 3.2-6 and 3.2-7; US EPA (1983), p.2;
Metcalf & Eddy (1982), Tables 3-10 and 3-11; ERT (1978), Table 2.2-8;
Damanoski (1982).
-------
"I'llt
Charles River or
, * Presidents Roads
- tasket Roads
Ml.no-.
Figure 2-3. Location of Sewage Treatment
• Plants in Boston Harbor Study Area
A Sewage Treatment Plants
X Outfalls
HULL
K)
-------
2-8
Harbor. However, due to outfall pipe deterioration, inadequate holding
capacity, and system malfunctions, the sludge releases are not always
properly co-ordinated with the tides. F^droscience's model of sludge
transport from the outfalls predicts that 20 percent of the sludges
discharged on the outgoing tide are carried back into the Harbor on the
return tide.
The Deer and Nut Island STPs are currently operating below design
criteria. This has led to:
a. the bypassing of raw sewage directly into the
Harbor;
b. the release of sludges on currents other than
the out-going tide;
c. the backing up of sewers from the STP, causing
the combined sewers to overflow; backups can
occur if some unit of the STP malfunctions,
halting incoming flows or if incoming flows
simply exceed the capacity of the system;
d. overall, less than design-optimal treatment
performance because of tanks settling, tank
covers missing, screens in poor condition, pumps
malfunctioning, and other operational problems,
including the problem of saltwater influent into
STP due to malfunctioning tide gates.
A properly operating and properly sized sewage treatment system could
alleviate these problems. Necessary steps to correct the above
deficiencies include:
improving the capacity of the combined sewers
(particularly holding facilities) in order to
moderate heavy (storm) flows to the STPs;
repairing STPs to restore capacity (pumping,
etc.); expanding STPs to increase capacity (of
holding tanks, etc.);
-------
2-9
b. that sludges be released only on out-going tides
(sludge-release timing problems) or that another
method of sludge disposal be found;
c. that either or both of the following actions be
taken to solve the influent back-up problem:
o expand combined sewer facilities to
accommodate what the STP cannot, and/or
o increase STP ability to accept incoming
flows;
d. the repair of units to restore their design
functions and performance.
Funds recently have become available for the repair and
rehabilitation of Nut and Deer Islands' STPs, which may restore their
original design performance. Increased operation and maintenance
efforts made by the MDC can result in a change in the Harbor's water
quality prior to implementation of any of the proposed STP options.
Actual loadings to the Harbor might be more consistent with the sum of
the first two columns of Table 2-1 once the STPs are operating well,
whereas now the loads are higher since bypassing occurs.
2.2 STP Options and Costs
There are many options under consideration for STP modification.
Obey represent different combinations of primary and secondary level
treatment facilities at Deer, Nut and/or Long Islands, witi. either local
or deep ocean outfalls. Two options have been chosen for the purposes of
this analysis: (1) upgraded primary treatment with a deep ocean outfall
and (2) secondary treatment with a Presidents Roads (local) outfall.
Because of resource limitations, only two options could be included.
-------
2-10
At the time of this analysis several other options are under discussion — ,
but all are either primary treatment with a deep ocean outfall or some type
of secondary treatment. Thus, the options described here are meant to be
representative of the range of options possible under current federal
regulations.
One of the STP options considered here calls for upgrading the existing
facilities to achieve primary treatment plus the construction of a deep ocean
outfall diffuser system to discharge the combined, treated effluents from
Deer and Nut island plants into the waters of Massachusetts Bay, out of the
Inner Harbor estuary, at a depth of 32 M (105 feet) (see Figure 2-3). The
outfall system would consist of a 10 foot diameter pipeline extending
4.7 miles from Nut Island; 56.6 cubic meters per second (1.29 billion gallons
per day) capacity effluent pumping station on Deer island; and an outfall
tunnel 7.5 miles long and 19 feet in diameter, terminating at a diffuser
manifold 1.3 miles in length. The proposed deep ocean outfall would
discharge the treated effluents from the Deer Island and Nut Island
facilities. At the mouth of the outfall would be a diffuser, which is
designed to rest on the ocean floor at a depth of approximately 100 feet.
The other STP option includes expanded primary treatment at the Nut
Island facility with the waste flow sent to Deer Island where all of the
system's wastes would be treated at the secondary level. The combined local
outfall would be into President's Roads (see Figure 2-3).
£ see CE Maguire (1983). Specific options were chosen in consultation
with Region I, Environmental Protection Agency, personnel. At the time of
this analysis these were the options preferred by the HOC.
-------
2-11
The capital and operating and maintenance costs for the two STP options
are presented in Table 2-2, and the expected loadings in terms of pollutant
concentrations are compared to concentrations in the existing STP effluents
in Table 2-3. These options and their associated costs assume that the
present facilities operated by the MDC will be modified according to the
presently planned "fast-track improvements". The costs of these immediate
upgrade improvements will be $10 million at Nut Island and $40 million at
Deer Island (CE Maguire, 1983).
2.3 Areas Impacted by STP Discharges
Existing water quality in different areas in the Harbor is due to current
STP effluent discharges, bypasses and sludge discharges as well as the
natural composition of the waters, CSOs, surface runoff, long-term discharges
to the harbor (industrial, residential STPs, etc.), discharges from marine
craft, etc.
In terms of the incremental contributions to pollutant concentrations
made by STPs, some areas are impacted more than others. The affected areas
may be grouped as follows (see Figure 2-4):
o areas of heaviest loadings;
between Deer Island and Long Island
between Long Island and Love11 Island
Quincy Bay, south of Moon Island
between Nut and Peddocks Island
o areas of moderate loadings:
east of Lovell Island
western half of Hingham Bay
northwest and northeast of Deer Island STP
Quincy Bay shoreline
-------
2-12
Table 2-2. Costs of the Two STP Options
(Millions 1982$)
Wastewater 1 1
Treatment 1 Capital Cost 1
STP Options 1 (1983$)
Upgraded Primary
With Ocean Outfall 774.8
Secondary 887.4
1
(19 8 23) 2/
728.9
834.8
1
I
Annualized Costs
I Capital^ O&M
74.9 22.0
85.8 45.2
1
Total
96.9
131.0
£/ Expressed in 1982$ (ENR=3825) because benefit estimates are expressed
in 1982$ (CPI-0=289.4)
£/ Based on 8 1/8 percent interest and 20 year period.
Source: CE Maguire (Draft, 1983), Table 2. These costs are to be
considered preliminary estimates only.
-------
2-13
Table 2-3. Pollutant Concentrations in
Effluent for STP Options
Effluent
BODg
TSS
Cadmium
Chromium
Copper
Lead
Mercury
Nickel
Zinc
Pollutants 1
mg/1
% removal
mg/1
% removal
mg/1
% removal
mg/1
% removal
mg/1
% removal
mg/1
% removal
mg/1
% removal
mg/1
% removal
mg/1
% removal
1
Existing
STP
107
29
87
55
.017
' 15
.090
16
.212
39
.116
19
.0011
21
.157
72
.458
33
Ocean Outfall
Option
115
28
86
47
.017
15
.090
16
.212
39
.116
19
.0011
21
.157
72
.458
33
Secondary
Option
30
81
30
81
.015
25
.066
38
.104
70
.094
34
.001
28
.112
80
.273
60
Sources: US EPA (1978), Tables 3.2-6 and 3.2-7; US EPA (1983), p.2;
Metcalf & Eddy (1982), Tables 3-10 and 3-11.
Note: Existing values for the metals are averages of sampling done in years
1975-1977. Samples taken in 1982 show decreases in chromium, lead, and zinc
with increases in the other metals (Metcalf & Eddy, 1983). Whether this
represents a significant decreasing trend can only be ascertained through a
concerted monitoring plan.
-------
diaries River ."A .
• /*'Presidents Roads ^
v/^jv^r •. • • . /v
)[ • , % . Doftiitim Ity
_*- ,"*• Najitasket Roads
1 'I- AllvMon
, , •
/ > ' _ Tbeopien J «. A
Figure 2-4. Dispersion of Current
STP Discharges
Dilution Ratio
(Seawater:Effluent)
«50:1
^100:1
£200:1
>200 to 1000:1
Sewage Treatment
Plants
Outfalls
^ Source: Metcalf & Eddy (1979),
Figure 6-16.
fir /»-7 "\^i
C^% 'P^
'• M
HINGHAM
-------
2-15
o areas minimally influenced by STP discharges:
the Brewsters Islands
eastern half of Hingham Bay
Inner harbor
Dorchester Bay shoreline
Neponset River
The highest pollution loads are located along the incoming and outgoing
tidal paths of Presidents and mntasket Roads (the two main current channels
of the harbor, in which Deer Island and Nut Island STPs have their outfalls,
respectively), current STP discharges have a greater impact on the Outer
Harbor Islands and the eastern part of Quincy Bay than on the other shoreline
at the perimeter of the harbor.
If a deep ocean outfall option is selected, the harbor will certainly
experience a reduction in pollutant loadings. The reduction for the harbor
creates a trade-off, however, by introducing wastes to previously unpolluted
areas. Figure 2-5 identifies three zones of impact for the proposed ocean
outfall option. In terms of the areas of concern to this benefits study, the
zones which sustain degradation of water quality due to the construction of
the deep ocean outfall are:
o Massachusetts Bay (highest level of impact); and
o Nantasket Beach and the Brewsters (moderate level of impact).
The advantage of the proposed deep ocean outfall is the dilution of
effluent that is obtainable in its vicinity as compared to the dilution in
the vicinity of the local outfalls currently in use in the harbor. The
disadvantage is that total pollutant loadings are not reduced to the extent
they would be under the secondary treatment option, and the proposed location
may not provide for sufficient transport and dispersion of the diluted
wastewater.
-------
Figure 2-5. Dispersion of Proposed
Ocean Outfall Discharges
«I»M
DORCHESTER
Source: ~:alf & Eddy (1979),
re BII-12.
Dilution Ratios
(SeawaterrEf fluent;
200:1
500:1
Site of Ocean
Outfall Diffuser
Charles River ,'.(•'
-------
2-17
The secondary treatment option considered in this study will reduce total
pollutant loadings and change the location of the current STP discharges in
Nantasket and Presidents Roads to a single discharge within the harbor in
Presidents Roads. Figure 2-6 identifies the areas in the harbor which will
be affected by this discharge. The highest level of impact will be on the
Outer Harbor Islands.
None of the proposed STP options will eliminate the pollution of harbor
waters. The incremental loadings to the harbor waters can be reduced,
thereby improving water quality. Most pollutants (i.e., metals and solids)
tend to settle out of the water column and into the sediments. Therefore,
the pollutant concentrations of Boston Harbor sediments will probably
continue to rise unless the rate of pollutant loading can be supressed by
some sort of biological, chemical, or physical neutralization process within
the sediments. What happens to pollutants in sediments is not known,
however, nor are the effects on aquatic organisms of pollutant build-up in
sediments fully understood. The present and potential status of water and
sediment qualities can be quantified but the significance of such qualities
is not clear.
-------
Charles River of/-' .
Df Proposed
Treatment Discharges
QUINCY
Source: Metcalf & Eddy (1979),
Figures 6-16 and 6-20.
Dilution Ratio
(Seawater:Effluent)
t. 50:1
HULL
HINGHAM
< 200:1
>200 to 1000:1
\ Diffusers
-------
2-19
References
Dumanoski, Diane, 1982, Boston Globe, December 19, 20, and 21, Boston, MA.
Environmental Research and Technology, 1978, Draft Report for the
National Science Foundation, C-PRA77-15337.
Metcalf & Eddy, Inc., June, 1982, Nut Island Wastewater Treatment Plant
Facilities Planning Project, Phase I, Site Options Study, for the
Metropolitan District Commission, Boston, MA.
Metcalf & Eddy Inc., September 13, 1979, Application for Modification of
Secondary Treatment Requirements for Its Deer Island and Nut Island
Effluent Discharges into Marine Waters, for the Metropolitan District
Commission, Boston, MA.
Metcalf & Eddy, Inc., January 1983, Application for Modification of
Secondary Treatment Requirements for Its Deer Island and Nut Island
Effluent Discharges into Marine Waters, Executive Summary, for the
Metropolitan District Commission, Boston, MA.
Maguire, CE, Inc., December 20, 1983, Preliminary Screening Results for
Boston Harbor DEIS Supplemental, for Environmental Protection Agency*
Region I, Boston, MA.
U.S. Environmental Protection Agency, June 30, 1983, Analysis of the Section
301(h) Secondary Treatment Waiver Application for Boston Metropolitan
District Commission, Office of Marine Discharge Evaluation, Washington,
DC.
U.S. Environmental Protection Agency, August 1978, Draft Environmental
Impact Statement on the Upgrading of the Boston Metropolitan Area
Sewerage System. Boston, MA.
-------
Section 3
Combined Sewer Overflow Control in Boston Harbor
In its effort to develop a comprehensive plan for combined sewer
overflow (CSO) control in Boston Harbor, the Metropolitan District
Commission (MDC) has designated four CSO planning areas. The four areas
are defined on the basis of their existing water use and coastal use
patterns. The designated areas are: (1) Dorchester Bay, (2) Neponset
River, (3) Inner Harbor and (4) Charles River Basin (see Figure 3-1). In
addition, the City of Quincy has storm sewer outfalls into Quincy Bay
which may impact the study area in a manner similar to the CSOs. For each
of these five planning areas engineering firms have been hired to study
alternative methods of control. All information pertaining to specific
areas is drawn from the contractor reports, and these reports are
referenced at the end of this Section.
3.1 Scope of the Combined Sewer Overflow Problem
The water quality of all four planning areas is compromised by
pollution from combined sewer overflows, stormwater discharges, and dry
weather overflows (DWO) (see Figure 3-1). Storm-related combined sewer
overflows vary in duration (depending on the nature of the storm) and occur
from 50 to 100 times a year (depending on the planning area location). Dry
weather overflows may be caused by sewer blockages, regulator malfunctions
and/or tide gate failures. DWO's are continual discharges of sanitary
-------
Figure 3-1. Combined Sewer Overflow
and Storm Sewer Project
Planning Areas
CSO/Storm Sewer outlets
lrr.it.,.
Neponset
River CSO
Area
Quincy
Storm
Sewers
-------
3-3
wastewater and are considered by the HOC to be the single most important
source of pollution in Boston Harbor. Treatment of dry weather overflows
is considered in all the CSO plans.
Different parts of Boston Harbor have different standards (see Figure
3-2). Die Dorchester Bay and Neponset River Estuary areas are both
classified as SB. An SB classification implies suitability for primary
water contact sports (i.e., swimming) and shellfishing and means that the
dissolved oxygen in the water must be greater than 6.0 mg/1 and the total
coliform count must have a median level less than 700 MPN/100 ml. The
Inner Harbor is classified as SC which makes it suitable for secondary
recreation and means that dissolved oxygen must be greater than 6.0 mg/1
and the total coliform count must have a median not greater than 1000
MPN/100 ml. The Charles River is classified as "C", the fresh water
counterpart of SC, which makes the river suitable for secondary recreation.
Some areas of Quincy Bay and Hingham Bay, outside the CSO planning
areas but within this studies' boundaries, are classified SA. An SA
classification is the same as SB for DO but has stricter limits on total
coliform counts (70 MPN/lOOml) for the protection and propagation of
aquatic life and so that shellfish harvesting can take place without
depuration in approved areas (Metcalf & Eddy, 1979). The plan for the
city of Quincy's storm sewers is discussed separately below (Section 3.6).
The MDC and its contractors found the following violations of the
standards, of which all of the violations are caused jointly by combined
sewer and dry weather overflows (MDC, April 1982) and by STP loadings:
-------
Figure 3-2. Water Quality Standard Classifications
in Boston Harbor
"rn««l llvir .
*$&
Wtll Mv»i
HULL
HINGHAM
u>
I
-------
3-5
Coliform bacteria standards are intermittently or
continually violated in all areas.
Dissolved oxygen standards are frequently violated in the
Charles River Basin planning area and in the Inner Harbor;
they are less frequently violated in Dorchester Bay and the
Neponset River Estuary and Constitution Beach.
Suspended solids possibly limit the most sensitive
designated uses of receiving waters; settleable solids cause
violations in certain locations, and floating materials, oil
and grease violate the standards in all areas. The
standards for all of these parameters are non-quantitative.
Nutrients are creating enriched conditions conducive to
excessive algal growth in fresh waters in the Neponset River
Estuary.
Heavy metals are potentially significant contaminants to
finfish and shellfish in all areas.
The shellfishing standards for total coliforms are violated
in the Neponset River Estuary, in much of Dorchester Bay,
and in areas north and east of Logan International Airport.
In order to deal with these violations, the MDC concluded that its
efforts to upgrade water quality should meet the following objectives
(MDC, April 1982):
a. eliminate dry weather overflows;
b. reduce the frequency and volume of untreated CSO's
c. reduce the release of pathogenic organisms and floating
materials;
d. and reduce the release of settleable organic solids and other
oxygen demanding material, nutrients and toxics.
-------
3-6
Control Plans designed to meet these objectives were developed for
each of the planning areas by the MDC's contractors. Table 3-1 lists the
costs and water quality characteristics of the planning areas. The
Recommended Plans, and the types of benefits to the area in which they
will be implemented, are explained in more detail in the remainder of this
Section.
3.2 Neponset River Estuary
The Neponset River Estuary Planning Area contributes to pollution in
both the river estuary and Dorchester Harbor. The area is approximately
60 percent residential and five percent industrial with remaining acreage
being either open space or commercial/institutional property. Tenean
Beach, in Dorchester Harbor, and the shellfish beds in the estuary both
experience what the MDC terms "extremely high levels for both total and
fecal coliforms, The contractor's survey of the planning area determined
that combined sewer and dry weather overflows accounted for over 90
percent of the annual total coliforms. Combined sewer overflow also was a
major contributor to floating and suspended solids. The contractor's
survey of the in-place sewerage facilities revealed broken or malfunc-
tioning tide gates, malfunctioning regulators, and much solid deposition
in the conduits. The fact that downstream interceptors had reduced flow
due to sedimentation also compromised this planning area's ability to
discharge its waste through normal channels.
-------
Table 1-1.
CSO Plannlnq Area Characteristics
Plannlnq
Area
Diner
Harhor£/
Constitution
Only
1
(torches ter
BayS/
1
Kponset
Rived*
Charles
River^
1
Qu IncyS/
1
2/ Based on 8
!>/ Values for
£/ Values for
, 6.0
Castle Is.
Pleasure B.
Carson
Mallbu
Tenean
II 1 1 II
0.61 0.10 0.71 Shellfish!
Neponset SB geoo geom
Estuary mean mean
T.col & 700 6,800 38,000
DO J 6.0
Beaches :
Tenean
III II
8.87 1.S6 10.43 tt> shell- C 300-
fishlng 12.000
or T.col ~ 1000
swimming DO % 6.0
III II
0.25 -0.02 0.27 Shellfish: SA
Quincy Bay 500- 800-
Beaches; T.col £70 16,000 34,000
Wollaston DO J 6.0
Quincy
III 1 1
1/8 percent Interest and a 20 year payback period.
Neponset River are from data gathered In August 1978 (DEQE, 1982).
Dorchester Bay and Inner Harbor are from the CSO Facilities Plans (Camp Dresser
the Charles River are from data gathered by the HOC (Ferullo. 1981).
Quincy are from sampling conducted in June-August 1982.
Su3. Total
Sol Ida Turb. BOD5 Phos. DO
ng/1 NTU mg/1 mg/1 mg/1
0-27 1.0-3.1 0.7-92 0.08- 1.3-
0.86 12.2
1 1 1
3.0-
14.0 0.1 .8-79 0.3-0.6 3.6
1 1 1
45 1.5 3.2 .22 5.2
1 1
0.09- 0.2- 0.4- 0.01- 0.0-
34.0 5.0 9.6 0.61 12.6
1 I
5- 1- 6-
50 5.0 10
1 1 1
t McKee, 1981) O'Brien t O'Gere, 1980).
-------
3-8
The Recommended Plan for this area focused initially on dry weather
flow abatement, and in this vein the Plan starts with recommendations to
fix or replace faulty tide gates, clean and inspect conduits, and re-open
the blocked regulators that in their present condition contribute to DWO.
Some new conduit and storm drain construction is recommended, both of
which are intended to reduce CSOs. The planning area is divided into two
subsections, each of which is slated to receive a storage and chlorination
facility. Such facilities will store combined sewage until such time that
the downstream treatment system can handle it.
According to the contractor's report, this Plan will reduce total
coliforms loadings by 96 to 99 percent. The costs are summarized in Table
3-2. Such a reduction will have several benefits, the most calculable
being fewer days during which total coliforms exceed the water quality
standards for swimming at Tenean Beach. In 1970, the Massachusetts
Department of public Health issued a report on Dorchester Bay beaches
indicating that the fecal coliform counts at Tenean Beach were above the
guideline of 200 MPN/100 ml for bathing water in 35 to 54 percent of the
grab samples and total coliforms exceeded the 1000 MPN/100 ml guideline in
24 to 43 percent of the samples. As a result of the findings regarding
these coliform counts and the fact that sewage was clearly being
discharged into Tenean Beach, the Department of Public Health recommended
in its 1970 report that the beach be closed for 24 hours after a rainfall
of 0.25 inches or more in a 24 hour period. The Recommended CSO Plan will
alleviate much of the reported pollution, reduce the need for closing
Tenean Beach and, thus, contribute significant recreational (especially
swimming) benefits.
-------
3-9
Table 3-2. Combined Sewer Overflow Project Costs
(1979$)
Neponset River Estuary
Capital-
Improvement Costs
I. Granite Avenue Service Area
1. Rockwell St. Drain 47,000 120
2. Stockton St. Drain 48,000 120
3. Washington St. Drain 66,000 170
4. Hilltop St. Drain 29,000 77
5. Ballet St. Drain 73,000 180
6. Adams St. Sewer 72,000 180
7. Granite Ave. Truck Sewer 696,000 1,740
8. Davenport Brook and Granite Ave. 30,000 880
(Regulator upgrading)
9. Catch Basin Cleaning 25,000
10. Monitoring Program 3,000 12,000
11. Granite Ave. Storage Facility 1,341,000 4,100
12. Net Cost at Deer Island 2,290
II. Port Norfolk Service Area
1. Chickatawbut St. Pump Station 150,000 4,000
2. lawley St. Relief Sewer 207,000 520
3. Regulator Rehabilitation 8,000
4. System Inspection and Cleaning 10,000 5,000
5. Port Norfolk Storage Facility 1,752,000 15,000
6. Net Cost at Deer Island 960
Total Costs
Total Cost for Granite Avenue
Service Area 2,405,000 46,850
Total Cost for Port Norfolk
Service Area 2,127,000 25,480
Total Cost for Bitire Plan 4,532,000 72,330
a/Based on June 1979 price levels (ENR 2900) and includes an
allowance for engineering and contingencies.
k/Based on 1979 price levels.
Source: Havens and Emerson, 1980.
-------
3-10
The benefits of reducing these coliform counts in the Neponset River
Estuary are less clear because the Recommended Plan affects only the mouth of
the river. Total coliform loadings upstream of the planning areas are
considerable and are unaffected by the Recommended Plan. There are 40 acres
of soft shell clam beds in the estuary, most of which are currently classified
as grossly contaminated. At present there is not sufficient evidence to
predict, with certainty, whether or not the Recommended Plan for the Neponset
River Estuary Planning Area will allow those shellfish beds to be opened.
A final benefit of the Recommended Plan has to do with the aesthetic
upgrading of the area due to the reduction of odor and floatable solids.
In addition to quality of life benefits, such an upgrading may also result
in an increase of secondary recreational water use (such as boating and
the development of boat ramps and yacht clubs) .
3.3 Dorchester Bay
Dorchester Bay is used mainly for swimming, boating and shellfishing.
Of all the planning areas it has the highest density of beaches. There
are five beaches in the Bay, seven yacht clubs, and 75 acres of shellfish
(soft shell clam) beds. Most of the shellfish beds are currently closed
to harvesting and four of the beaches are known to exceed total coliform
standards after rain storm? in the summer.
This degradation of Dorchester Bay's water quality is in large part
caused by eleven combined sewer outlets that discharge into waters
adjacent to public beaches. Unlike the Neponset River Estuary Planning
Area, the Dorchester Bay Planning Area has a wastewater collection system
that is in good condition. The contractors discovered "no major
structural deficiencies in the regulators, tide gates or sewer manholes.
-------
3-11
Several maintenance-related problems were discovered, generally consisting
of blockages within the sewers and regulators due to excessive sediment
buildup. These maintenance problems and a small number of direct dry weather
connections to overflow conduits that were also discovered, result in
several dry weather flow dischargers to Dorchester Bay" (camp, Dresser and
McKee, 1981) .
The Recommended Plan for the Dorchester Bay Planning Area includes a
DWO abatement program and an ongoing program to maintain high operating
efficiency in the tide gates and regulators. It also calls for a one and
one-half mile consolidation conduit designed to intercept CSOs at the
outlet tide gates and to transport the waste to a storage facility. A
final part of the plan involves the construction of the two screening and
disinfection facilities to protect the Dorchester Beaches during the
bathing season. Table 3-3 presents the costs for the Recommended Plan.
According to the contractor's calculations, DWO abatement in the
Dorchester Bay Planning Area will result in the attainment of the required
water quality standards over the long run. Short-run storm-induced
episodes of elevated coliform counts can be avoided by the addition of CSO
controls. For all but the most extended heavy storms, the Recommended
Plan will reduce total coliform loadings by 98 percent.
In addition the Recommended Plan will greatly reduce the level of
floatable solids, oil and grease in Dorchester Bay. Such changes will be
particularly beneficial to a planning area that is used primarily for
recreation and shellfishing. Reduced coliforms and a reduction in floatable
solids will have the direct benefit of increasing swimming and improving
-------
3-12
Table 3-3.
Combined Sewer Overflow Project Costs
(19793)
Dorchester Bay
1. Structural Capital^ 0 &
South Boston
- Consolidation Conduit 9,620,000
- Storage/Containment Facility 18,380,000 38,000
Dorchester
- Hoyt St. Regulator Modification 90,000
- Commercial Point Screening/
Disinfection Facility 3,030,000 22,000
- Fox Point Screening/
Disinfection Facility 2,740,000 21,000
- Pine Neck Creek/
Storm Drain Relocation 2,340,000
Subtotal 36,200,000 81,000
II. Non-Structural
Dry weather flow abatement
program 100,000
Post Management Practices 25,000 200,000
Subtotal 125,000 200,000
III. additional
Cleaning of CSO Conduits 300,000
Dredging 40,000
Landfill at CSO Outlet BOS-090 20,000
Subtotal 360,000 -0-
Total 36,685,000 281,000
3/Based on June 1979 price levels (ENR 2900) and includes an
allowance for engineering and contigencies.
b/Based on 1979 price levels.
Source: Camp, Dresser and McKee, 1981.
-------
3-13
health through decreasing the number of pathogens in the water.
Shellfishing should, over time, increase as bed acreage is opened to
harvest.
3.4 Inner Harbor
The Inner Harbor Planning Area has two distinct uses. The majority of
the planning area is classified for commercial use while a very small
section, Constitution Beach, is classified for swimming. To deal with
these divergent uses Table 3-4 is divided to show the costs of attending
to the Inner Harbor proper and Constitution Beach separately. The Inner
Harbor proper is characterized by industrial, transportation, shipping and
energy production uses. The area to the north and east of Logan
international Airport has shellfishing and recreational uses similar to
the Dorchester Bay. There is swimming at Constitution Beach and there are
areas of restricted shellfish harvesting around the airport (MDC, 1982).
According to the contractor for the Inner Harbor Planning Area, over
11 billion gallons of overflow enter the water every year. Seventy-five
percent of these discharges are attributable to dry weather overflow and
the rest can be accounted for by storm-related discharges. Dry weather
overflows are the main cause of elevated coliform counts and floatable
solids.
The contractor's report states that the water quality standards set
for the inner Harbor (they are less stringent than those set for the more
recreational Dorchester Bay and Neponset River Estuary Planning Areas) can
be met with the elimination of dry weather overflows (DWO) and the control
of combined sewer overflows (CSO).
-------
3-14
Table 3-4.
Combined Sewer Overflow Project Costs
(1979$)
Inner Harbor Planning Area
A. Inner Harbor
Capital O&M
CSO Consolidated Pipelines Costs 2/ Costs**/
Boston Waterfront $5,858,000 N/A
South Boston 5,149,000
East Boston-Southern Waterfront 6,221,000
East Boston-Western Waterfront 8,156,000
East Boston-Lexington Square 980,000
Chelsea River Waterfront 4,097,000
Reserved Channel 2,932,000
CSO Treatment Facilities
Fort Point Channel 45,542,000 816,000
Somerville 3,450,000 71,200
East Boston-Southern Waterfront 6,894,000 131,000
East Boston-Western Waterfront 6,400,000 48,000
East Boston-Lexington Square 3,575,000 31,700
Chelsea-Pearl Street 3,404,000 66,000
Chelsea Willoughby Street 2,005,000 39,100
Reserved Channel 2,032,000 50,400
Causeway Street 243,000
Commercial Street 243,000
Charles River Estuary 5,800
Management Practices
Tidegate Improvements 250,000 118,000
Regulator Improvements 500,000 118,000
107,921,000 1,495,300
B. Constitution Beach 315,000 8,700
£/ Based on June 1979 price levels (ENR 2900) and includes an
allowance for engineering and contingencies.
b/ Based on 1979 price levels.
Source: O'Brien and Gere Engineers, 1980.
-------
3-15
The main benefit of this plan is that it will clean up the Inner
Harbor by reducing floating solids and thus reduce the consequent
aesthetic problems, since the inner Harbor is not now used for contact
recreation, and there are no plans for it ever to be put to that use.
Constitution Beach currently meets swimmable standards and the
improvements at that site are to ensure that the standards will be
maintained. It is difficult to predict if the improvements in water
quality will result in more shellfish beds being opened for harvest.
3.5 Charles River Basin
The Charles River Basin includes the Back Bay Fens, the Muddy River,
Alewife Brook and the Charles River itself. The basin is mixed fresh and
salt water and is used mainly for non-contact recreation, both on the water
and at the water's edge. The River is an extremely important and much used
recreational source for local residents. Many sailing clubs maintain
marinas on the River and every area college and many high schools use the
River for rowing and sculling. The entire basin exceeds the water quality
standards set for it by the state. Those standards (a rating of "C") allow
non-contact recreational use. The main results of the basin pollution are
extremely high coliform counts (both total and fecal), odors, floatables,
debris and turbidity. The primary objectives of the MDC's efforts are to
(Metcalf & Eddy, 1982):
a. reduce excessive levels of bacteriological organisms for
public health reasons and remove floatables and turbidity for
aesthetic reasons, and
b. remove solids and organic matter to prevent build up of
benthic deposits.
-------
3-16
In order to meet these objectives, the contractor designed a plan that
involves the capture, transport and storage of most of the basin's
combined sewer overflows. The plan's costs are presented in Table 3-5.
The Recommended Plan will reduce coliforms, floatable solids and
suspended solids (and, therefore, turbidity) in the Charles River Planning
Area. Secondary recreation (boating but not swimming) can be expected to
increase because of the decrease in objectionable odors and floating
debris.
3.6 Quincy Storm Sewers
Another source of pollutant loadings to Boston Harbor is the Quincy
storm sewers. The Quincy storm sewers discharge waters with fecal
coliform, BOD and TSS concentrations that are higher than levels expected
from storm water runoff (Moore, 1980). Storm water contamination can
result from cross-connections between sanitary and storm drains. These
cross-connections can be due to broken pipes, exfiltration from sanitary
sewers in disrepair and illegal "tie-ins" to the storm sewer system,
although the latter has not been documented in Quincy. The problem in
Quincy is compounded by the fact that North Quincy is relatively flat
(especially adjacent to the beach areas) and, therefore, the drains in the
area have slopes close to, and in some cases, less than the recommended
minimums. This tends to cause blockages in the sanitary system and
surcharges and exfiltration of sewage results, especially where pipes are
cracked or have loose joints (Moore, 1977). A factor which increases the
frequency of surcharging is the excessive infiltration and inflow into the
-------
3-17
•able 3-5.
Combined Sewer Overflow Project Costs (19799)
Charles River Basin
Capital O&M
Design Package Costs 3/ Coats £/
1. Phase I In-System
Modification 510,000 5,000
2. Consolidation and Rebuilding
of Boston Gatehouses fl and 12 6,650,000 87,000
3. Grit Removal and In-System
Storage at Beacon Street and
Charlesgate East with Phase II
In-System Modification to MDC
Fens Gatehouse 4,900,000 72,000
4. Restoration of the Fens with
Phase II In-System Modifications
in the Muddy River Sub-area 2,000,000 20,000
5. Connection from Stony Brook
and Old Stony Brook Conduits
to the Boston Main Drainage
Relief Sewer 1,060,000 187,000
6. Grit and Sludge Removal from
Stony Brook and Old Stony Brook
Conduits 4,750,000
7. Stony Brook Screening Disin-
fection, and In-System Storage
Facility near Tremont and Gurney
Streets 7,500,000 360,000
8. Stony Brook In-System Storage
Facility and Base Brook Pumping
Station at Green Street 10,400,000 116,000
9. Phase II In-Line Storage
Tannery Brook 5,900,000 14,000
10. Surface Storage of Canterbury,
Bussey and Stony Brooks 1,260,000 50,000
-------
3-18
Table 3-5 (continued).
Combined Sewer Overflow Project Costs (1979$)
Charles River Basin
Capital
Design Package Costs £/
11. St. Mary's Street In-System
Storage 335,000
12. Concord, Rindge and Mass. Aves.
Industrial Sewer Separation 3,600,000
13. St. Mary's, Street Diversion
to Cottage Farm Facility 2,800,000
14. Phase III In-System Modifications 360,000
15. Brighton-Allston Phase II
In-Line Storage 10,850,000
16. Brighton-Allston Phase III
Off- Line Storage 2,500,000
Management Practices and Monitoring
Program 40,000
O&M
Costs b/
2,000
4,000
11,000
88,000
56,000
108,000
•total 65,415,000 1,180,000
£/ Based on June 1979 price levels (ENR 2900) and includes an allowance
for engineering and contingencies.
b/ Based on 1979 price levels.
Source: Metcalf and Eddy, 1982.
-------
3-19
sanitary system. Quincy is the last (i.e., downstream) city in the South
Metropolitan Sewer District so that excessive flows from as many as 20
cities are channeled through Quincy on their way to the Nut island
Treatment Plant. It has been estimated that as much as 57% of the flow
reaching Nut island during a rainstorm is due to infiltration/inflow
(Moore, 1981). Thus, the problem of correcting stormwater contamination
in Quincy involves repair and rehabilitation of both the sanitary and
storm sewer systems.
Several investigations and improvements have been undertaken in recent
years to locate sources of contamination of storm drains,^ in particular in
order to reduce total and fecal coliform levels at Wollaston Beach to
within acceptable levels, as determined by State standards (Moore, 1980).
In addition, studies of the infiltration/inflow problem are continuing
(Moore, 1981). It should be noted that other sources of contamination of
the area's beaches include the Nut Island sewage treatment plant
discharges and, in particular, recurring by-passes from both Nut Island
and Moon Island.
Although estimates of treatment costs for the Quincy storm sewers
comparable to those for the CSO planning areas are not available, recent
studies give an indication of the order of magnitude of the costs involved
(Table 3-6).
3.7 Summary of Options
The annual cost of implementing all five of the CSO and Storm Sewer
plans is about $30 million (1982$). The costs of implementing portions of
the plans or only some of the plans will, of course, be less.
-------
3-20
•Cable 3-6. Potential Storm Sewer and Infiltration/Inflow
Project Costs for City of Quincy (1981$)
Recommended Facility Costs
Sewer System Evaluation Survey
and Rehabilitation of Sewers 417,000
Construction of Relief Interceptors
North Quincy 204,000
West Quincy 844,000
Quincy Point Diversion/
Relief Interceptor 132,000
Town River Bay Interceptor 703,000
Rehabilitation of Quincy Point Pump Station 40,000
Construction of Rirnace Brook
Emergency Relief Lift Station (MDC) 180,000
•natal Capital Costs 2,520,000
Annual O & M Costs -22,000
Nate: Many problems remain and the city of Quincy has authorized a
new engineering study so that these estimates of costs are
preliminary only. They are taken from Table 8, Moore (1981),
and do not include land and easement acquisition costs. The
annual operation and maintenance costs are expected to
decrease as a result of the infiltration/inflow removal
program.
-------
3-21
In order to gain swimming benefits at all beaches and shellfishing
benefits at many of the currently closed shellfish beds in Boston Harbor,
the Constitution part of the Inner Harbor plan, the entire Neponset River
Estuary and Dorchester Bay Plans, and the Quincy plan must be implemented.
Such a treatment option would cost more than $6.3 million a year (in
1982$), and it would affect neither the Charles River Basin nor the Inner
Harbor proper.
Another option might be to implement only the Dorchester Bay and
Neponset River Estuary plans. This would cost about $6 million (in 19829)
annually, but while making swimming safe in Dorchester Bay, it might
compromise the water quality at Constitution Beach and Quincy Bay beaches
in the long run as the population of these areas increases and wastewater
discharges increase.
Table 3-7 shows the annual costs of the CSO and storm sewer options
along with the approximate percentage reduction in pollutant loadings,
including fecal coliform, floatable and suspended solids, and oil and
grease. The top part of the table presents the four CSO plans as
designated by the MDC. The bottom part shows the options used in the
benefit-cost analyses in this study. The options as defined in the lower
half of the table correspond more appropriately with the benefit estimates
associated with the uses of the Harbor. For example, all the swimming and
shellfishing uses affected by the CSOs (and therefore the corresponding
benefits estimates) can be captured by including only the Constitution
Beach portion of the Inner Harbor Plan plus the Dorchester Bay, Neponset
River, and Quincy plans. The numbers in the table reflect incremental
increases in annual costs.
-------
3-22
Table 3-7. Incremental Costs and Potential Reductions
in Pollutant Loadings for the CSO Options
(Millions 1982$)
1
1
Treatment 1
Alternative/ I
Receptor I
Inner Harbor
a) Including
Constitution
b) Constitution
only
Dorchester Bay
Neponset River
Charles River Basin
Implementation of
all MDC design-
ated CSO plans
1
Inner Harbor
Constitution
Beach only
Dorchester Bay/
Neponset River
Quincy Storm
Sewers £/
Above three plans
combined
Charles River
MDC PLANNING AREA DESIGNATION
Annualized I
Capital Cost £/ |
•
14.63
0.04
4.97
0.61
8.87
35.44
STUDY
0.04
5.59
0.27
5.90
8.87
Annual
O&M Cost
1.97
0.01
0.37
0.10
1.56
4.00
1 Total
1 Annual Cost
16.61
0.05
5.34
0.71
10.43
33.39
AREA DESIGNATION
0.01
0.47
-.02
0.46
1.56
•
0.05
6.06
0.25
6.36
10.43
1
1 Percentage
1 Reduction in
1 Pollutant
1 Loadings !>/
•
50 - 99
70 - 99
60 - 98
65 - 100
50 - 100
1
50 - 99
60 - 99
60 - 99 I/
50 - 99
65 - 100
-------
3-23
References
Metropolitan District Commission, April 1982, Combined Sewer Overflow
Project; Summary Report on Facilities Planning, Boston, MA.
Massachusetts Department of Environmental Quality Bigineering, May 1982,
Neponset River 1978 Water Quality Data, Publication No.
12808-39-35-5-82-CR, Boston, MA.
Ferullo, Alfred, Paul DiPietro, and Robert Schaughnessy, June 1981,
Oiarles River Articifial Destratification Project, Metropolitan
District Commission, Boston, MA.
Havens & 'Emerson, Inc., September 1980, Combined Sewer Overflow Facilities
Plan for the Neponset River Estuary, Boston, MA.
Massachusetts Department of Public Health, 1970, Report of the
Massachusetts Department of Public Health to the Interagency Task
Force on the Survey of the Dorchester Bay Beaches, Boston, MA.
Moore Associates, Inc., H.W., 1977, Drainage Contamination Study for North
Quincy, Quincy, Massachusetts, Boston, MA.
Moore Associates, Inc., H.W., 1981, Facility Plan for Water Pollution Control,
Volume I, Quincy, Massachusetts, Boston, MA.
Moore Associates, Inc., H.W., 1980, Wollaston Beach Exploration/Remedial
Program Regarding Storm Water Contamination, Boston, MA.
Camp Dresser & McRee, Inc., October 1981, Report on Combined Sewer
Overflows in the Dorchester Bay Area.
O'Brien & Gere Engineers, Inc., June 1980, Combined Sewer Overflow Project
Inner Harbor Area Facilities Plan.
Metcalf & Eddy, Inc., September 13, 1979, Application for Modification of
Secondary Treatment Requirements for Its Deer Island and Nut Island
Effluent Discharges into Marine Waters, for the Metropolitan District
Commission, Boston, MA.
Metcalf & Eddy, Inc., May 1982, Final Report to the Metropolitan District
Commission on Combined Sewer Overflows; Charles River Basin Facilities
Planning Area, Boston, MA.
-------
Section 4
Water Quality Impacts
To estimate the change in water quality that is expected to take place
under the various options for reducing pollutant loadings it is necessary to
take into account the change in loadings, the dispersion pattern in the
Harbor from the point of discharge to the areas where recreation and fishing
take place (receptor areas), and the current ambient water quality in these
areas. The reception areas defined for the purposes of this study are shown
in Figure 4-1. Pollutant loadings continue under all treatment options but
at rates less than the current ones. Thus, percent improvements in water
quality are related to percent reductions in pollutant loadings under the
various options. The changes experienced under any of the options are not
expected to be in the form of new dispersion patterns but rather are expected
to be concentration reductions in the water column. The changes are
incremental ones, evaluated in relation to current loadings and current
ambient quality.
4.1 Water Quality Impacts of STP Dischargers
To assess the impact of STP discharges in Boston Harbor it is important
to know how such discharges are dispersed throughout the Harbor. Since
discharges to the Harbor are subject to diverse and variable conditions, the
water quality throughout the Harbor is not uniform. A few models have been
developed to quantitatively explain some of these variations and to correlate
-------
Figure 4-1. Receptor Areas for the Boston Harbor Study
harles River .TA"
Wtll Klvif
HULL
V
r>
,r">»\* •>/>>-, C
• •• M.ii fir' A'V*
1. Constitution Beach
2. Castle Island
3. Pleasure Bay
4. City Point
5. L&M Streets Beach
6. Carson Beach
7. Malibu Beach
8. Tenean Beach
9. Wollaston Beach
10. Quincy Town Beaches
11. Weymouth Bay
12. Hingham Harbor
13. Hull Bay
14. Outer Harbor Islands
15. Brewsters Islands
16. Nantasket Beach
17. Massachusetts Bay
ro
PfcAiife
'L ^)
HINGHAM
-------
4-3
STP discharges with water quality. The DISPER model/ developed at Massachu-
setts Institute of Technology, was designed specifically to quantify the
dispersion of STP discharges into Boston Harbor. This model was used in the
assessment of a deep ocean outfall in the MDC's application for a waiver to
secondary treatment (Metcalf and Eddy, 1979). He use the dilution ratio
results to predict relative changes in water quality but use ambient water
quality data from other sources.
The DISPER model (and the associated CAFE model) relies largely on water
movement (currents) to describe dispersion.£/ It models BOD only and
predicts volumetric inflows and outflows from the Harbor. Whether pollutant
loadings move exactly as does the water is unknown because settlement and
decomposition in transport, propensities of marine organisms to assimilate
wastes, etc., are not precisely understood. Assumptions regarding settling
rates, decay rates, biological uptake, and chemical reactions are employed in
r
running DISPER. This model is useful in comparing relative dispersion
differences for the different STP options while precise, absolute values
predicted by DISPER may not be as reliable. It was with this in mind that
the maps of dilution ratios in Section 2 were developed based on the DISPER
model (Figures 2-4, 2-5, and 2-6).
In order to use the dilution ratios produced by DISPER to assess water
quality impacts, current water quality must be known. The Boston Regional
Office of the Environmental Protection Agency (Region I) has recently
undertaken to bring together all water quality sampling data collected in the
Harbor since 1968. They have stored the data in a computer system called the
*/ See Appendix A.2 for a further description of this model.
-------
4-4
Boston Harbor Data Management System and, in December 1983, could produce
computer-generated maps with statistically-averaged data for various points
throughout the Harbor and adjacent waters. The information from this system
that we used in the analysis below includes data on fecal coliform, BOD ,
and total suspended solids averaged over the years 1968 to 1983 at the
receptor sites of interest to this study.
To calculate the water quality impacts of reduced pollutant loadings
under the various STP options, the change in effluent concentrations were
multiplied by the dilution ratios at the various receptor sites (Table 4-1).
The reduction was compared to current ambient quality to calculate a
percentage change in water quality. This simplified approach is clearly not
accurate if absolute values for water quality are desired. The nature of
both the current water quality data and the limitations of the dispersion
model preclude any attempt to predict absolute values. However, for the
purposes of our analysis percentage changes in water quality with a range to
indicate the degree of uncertainty is sufficient.
4.2 Water Quality Impacts of Combined Sewer Overflows
The individual contractor reports on combined sewer overflows included
modeling for water quality impacts. In those reports the impact was
evalvited using both statistical and time-varying models. The statistical
modeling was used to produce a long-term picture of the quality of water in
different segments of the harbor. The time-variable model produced dynamic
changes in water quality over a finite period of time in order to predict the
results of discrete storm events. Total coliform counts were used in both
-------
4-5
Table 4-1. Effluent Concentrations and Dilution Ratios
Used in the Water Quality Impact Analysis
EFFLUENT CONCENTRATIONS ^/
Pollutant
1
1
1
Existing Facilities
Deer Nut
Island^ Island
1 Deep Ocean
1 Outfall
1 Option
1 Secondary
1 Treatment
1 Option
Fecal Coliform
(MPN/100 ml)
BOD5 (mg/1)
TSS (mg/1)
I
1500
127.6
121
1500
105
110
I
1500
115
86
1500
30
30
£/ Values as summarized in EPA (1983) and Metcalf & Eddy (1979).
£/ Includes sludge discharged into Presidents Roads.
DILUTION RATIOS £/
Receptor Area
Outfall Location
Presidents Roads
(Deer Island)
Nantasket Roads
(Nut Island)
Ocean Outfall
Constitution Beach
Dorchester Bay
Quincy Bay
Hingham Bay
Outer Harbor Islands
Brewsters Islands
Nantasket Beach
Massachusetts Bay
500
100-200
50
500
1000
—
50-100
100-200
50
500
500
1000
200
200
200
£/ From DISPER contour maps.
-------
4-6
the statistical and time-variable models. Although the models predict actual
total coliform counts for both existing conditions and under the recommended
plan, we state the results in terms of relative percentage changes both to
indicate the degree of uncertainty and as sufficient for our purposes.
The studies of the Quincy sewer systems did not model water quality. In
this study we have assumed the situation to be similar to the Dorchester Bay
area in this regard.
4.3 Estimated Water Quality Impacts of the STP and CSO Treatment Options
Table 4-2 presents the results of the water quality impact analyses. The
entries are ranges of predicted percentage change in water quality due to
each treatment option at each receptor site. Table 4-3 presents best-guess
point estimates for the same options and receptor sites. (Appendix A gives
details for these calculations.) These were compiled for use in several of
the benefit estimation approaches. Again it should be noted that limitations
of both data and methodology preclude estimation of absolute changes in water
quality. However, relative percentage changes are adequate for the benefit
estimation procedures to be used in the remaining sections of this report.
This report investigates pollution due to sewage treatment plant
discharges and combined sewer overflows. Other point and non-point sources
exist which were not included in the scope of this report. They include the
large amount of shipping and boating in the ffirbor, run-off from urban areas
not collected by the sewer system and potential resuspension of pollutants
from sediments in the ffcrbor. Thus, our estimates of water quality changes
do not reflect complete reduction of pollutant levels because of these other
sources whose impact is, essentially, unknown at this time.
-------
4-7
Table 4-2. Estimated Water Quality impacts of the CSO and STP
Treatment Options
1 Efercent Follution Reduction by
1 Treatment Option
Receptor Area
Constitution Beach
Dorchester Bay
Quincy Bay
Hingham Bay
Outer Harbor Islands
Brews ters Islands
Nantasket Beach
Massachusetts Bay
Charles River
1 Combined 1
1 Sewer Overflow/ 1
Storm Sewer
50 to 80
60 to 90
60 to 90
—
—
—
—
—
50 to 80
1 1
Deep Ocean
Outfall
5 to 10
10 to 25
10 to 20
15 to 40
60 to 90
-10 to -15
-5 to -10
'-35 to -45
—
1 Secondary
1 Treatment
0 to 5
5 to 15
10 to 20
15 to 40
30 to 80
30 to 40
0 to 5
15 to 20
1
Nate: Positive figures denote improved water quality. Negative figures
denote degradation in water quality.
Source: See Appendix A for details of the calculations.
-------
4-8
Table 4-3. Estimates of Pollution Reduction at Receptor
Sites in Study Area (Point Estimates)
I Percent Pollution Reduction by Treatment Option
I CSO/Storm I Deep Ocean I Secondary
I Sewer I Outfall I Treatment
Constitution 70 10 5
Dorchester/Neponset Bay
Castle Island
Pleasure Bay
Carson
Malibu
Tenean
Wollaston
Quincy
Weymouth
Hingham
Hull
Outer Harbor Islands
Brewsters Island
Nantasket Beach
Massachusetts Bay
Charles River
80
80
80
80
80
80
80
__
—
—
—
—
—
—
70
1
10
10
10
10
10
10
10
30
30
30
80
-15
-10
-40
__
1
10
10
10
10
10
10
10
30
30
30
70
40
—
20
__
1
Note: Positive figures denote improved water quality. Negative figures denote
degradation in water quality. Based on Table 4-2.
-------
4-9
References
Environmental Protection Agency, June 30, 1983, Analysis of the
Section 301 (h) Secondary Treatment Waiver Application for Boston
Metropolitan District Oommission, Office of Marine Discharge Evaluation,
Washington, DC.
Metcalf & Eddy, Inc., September 13, 1979, Application for Modification
of Secondary Treatment Requirements for Its Deer Island and Nut Island
Effluent Discharges into Marine Waters, for the Metropolitan District
Oommission, Boston, MA.
-------
Section 5
Approaches to Measuring Benefits from Water Quality Improvement
Estimates of changes due to changing ambient pollutant levels are the
basis for benefit measurements. These changes include effects on human
health^ human activities, such as recreation, and the availability of goods
and services. The economic value individuals place on the reduction of the
adverse effects due to pollutants is the measure of benefits. As will be
seen throughout this report, for some effects, such as ecological changes,
current efforts can only, at best, delineate the physical changes; for
others, either a partial or full economic evaluation is possible. This
section describes the economic theory appropriate to measuring such benefits
and the classification scheme used in this study.
5.1 Theoretical Concepts
The benefits of improved water quality resulting from implementation of
pollution control options can be classified in many ways. One way is to
divide them into benefits to users of the water resource and benefits to
non-users, or intrinsic benefits, as presented in Table 5-1. Potential
benefits from water pollution abatement accrue from current users or from
intrinsic values. Current user benefits stem from either indirect use
(near-stream activities that are enhanced by the water body such as
picnicking, jogging, hiking or viewing), direct use of water resources for
-------
Table 5-1.A Spectrum of Water Quality Benefits
Potential
Water
Quality
Benefits
Current
User
Benefits
Intrinsic
Benefits
Direct
Use
Indirect
Use
Potential
Use
No
Use
In Stream
Withdrawal—
Recreational--fishing, swimming, boating, rafting, etc.
Commercial—fishing, navigation
Municipal—drinking water, waste disposal
Agricultural—Irrigation
Industrial/Commercial—cooling, process treatment,
waste disposal, steam generation
Near Stream'
—— Recreational—hiking, picknicking, birdwatching, photography, etc.
Relaxation—viewing
— Aesthetic—enhancement of adjoining site amenities
en
I
NJ
Option
Existence —
Near-term potential use
Long-term potential use
•Stewardship—maintaining a good environment for everyone to enjoy
(including future family use—bequest)
Vicarious consumption—enjoyment from the knowledge that others
are using the resource.
Source: Adapted from RTI, 1983.
-------
5-3
instrearn purposes (recreational and commercial)/ or withdrawal purposes
(municipal, agricultural, industrial/commercial). Intrinsic benefits are
based on non-user valuation of the existence of the resource, and on the
potential future use of the resource. Since the distinction between these
types of benefits is not always clear-cut and since many of the analytical
techniques used to measure benefits cover more than one of these types of
uses, we have chosen to reclassify the water uses according to the economic
entity to which the benefits accrue (see Table 5-2). Here, benefits flow to
households as recreators in, on or near the water and as consumers, who
benefit directly or indirectly (secondary benefits) from the increased
economic activity in the primary sectors, and to producers who use the water
resources. The benefits that will accrue from pollution abatement in Boston
Harbor are noted with an asterisk in Table 5-2.
Most of the methodologies used to measure the benefits to society from
environmental improvements are based on the theory of welfare economics and
the concept of willingness to pay (WTP). This economic theory is founded on
the principle that the "demand" for water quality is the sum or aggregate of
how much individuals of a society would be willing to pay to receive
additional increments of improved water quality. The concept of willingness
to pay has been translated into other alternative theoretical measures of
willingness to pay, including consumer surplus, compensating variation, and
equivalent variation. In simple terms, consumer surplus is the difference
between what individuals are willing to pay and what they actually pay for a
good. Figure 5-1 illustrates this individual demand function which
2/ The following discussion is based on material discussed and presented
in RTI, 1983.
-------
5-4
Table 5-2. Economic Benefit Categories
(Alternative Typology)
I. Benefits to Households
A. Recreation Benefits:
1. Swimming*
2. Fishing* '
3. Boating*
4. Aesthetic
5. Near-stream recreation*
6. Option value*-
7. Existence*-
B. Consumption Benefits:
1. Commercial Fisheries*
2. Health
a. Swimming*
b. Food Consumption*
Direct Use
Indirect Use
Potential or non-use
C. Ecological*
II. Benefits to Producers:
A. Commercial Fishing*
B. Municipal drinking and wastewater
C. Agricultural
D. Industrial
£. Navigational
III. Secondary Effects*
* Benefits from pollution abatement in Boston Harbor.
-------
5-5
Figure 5-1. The Demand Function and
the Consumer Surplus Welfare Measure.
Price
($/unit)
B
Quantity/time
Source: RTI, 19E2
-------
5-6
describes, for any commodity X, the maximum amount an individual would be
willing to pay for each quantity of X.^/ The downward slope of the curve
illustrates that individuals are willing to buy more of commodity X at lower
prices than at higher prices. The simple two-dimensional diagram in Figure
5-1 assumes all other factors that might influence demand—income, the prices
of related goods, etc.—do not change. At price PQ the individual will
purchase Qg of X and make a total expenditure of PgAQgO. Because the
demand curve measures the individual's maximum willingness to pay for each
level of consumption, the total willingness to pay for QQ can be derived:
total expenditures plus the triangle PjPQA. The difference between what
individuals actually pay with a constant price per unit and the amount they
are willing to pay is defined as the consumer surplus.
As a dollar measure of individual welfare, however, consumer surplus is
not ideal. The most direct way of understanding its limitations is to
consider the measurements underlying an ordinary Marshallian demand
function. An individual's demand function describes the maximum an
individual with a given nominal income would be willing to pay for each level
of consumption of a particular good. Specifically, if the price paid
changes, it will affect not only what the individual can purchase of this
good, but also the purchases of all other commodities through its effect on
the remaining disposable incor e. Thus, movement along a conventional demand
function affects the level of satisfaction an individual will be able to
achieve with a given income. For example, suppose the price of hypothetical
good X declines to Pj^ The individual can purchase the same quantity of X
at its new price as indicated in Figure 5-1 by the area OP,BQ0 and have
-------
5-7
income remaining, as given by ^PgAE, to purchase more X or more of other
goods and services. The movement to a consumption level of OQ. describes
the increased selection of X under the new price. This change leads to a
higher utility level because more goods and services can be consumed with the
same income. For consumer surplus to provide an "ideal" dollar measure of
individual well-being, however, the appropriate area under an Hicksian
income-compensated demand curve rather than an ordinary Marshallian demand
curve, should be used. Nevertheless, ordinary Marshallian demand curves are
much easier to estimate, and Willig (1976) has shown that they provide a
reasonably close approximation to the "ideal" measure.
The four "ideal" Hicksian welfare measures are summarized below (Hicks,
1943):
o Compensating variation (CV)—the amount of compensation that
must be taken from an individual to leave him/her at the same
level of satisfaction as before the change.
o Equivalent variation (EV)—the amount of compensation that
must be given to an individual, in the absence of the change,
to enable him/her to realize the same level of satisfaction
he/she would have with the price change.
o Compensating surplus (CS)—the amount of compensation that
must be taken from an individual, leaving him/her just as
well off as before the change if he/she were constrained to
buy at the new price, the quantity of the commodity he/she
would buy in the absence of compensation.
o Equivalent surplus (ES)—the amount of compensation that must
be given to an individual, in the a sence of the change, to
make him/her as well off as he/she would be with the change
if he/she were constrained to buy at the old price the
quantity of the commodity he/she would buy in the absence of
compensation.
If commodity X in Figure 5-1 represents environmental quality, then in
order to measure environmental improvement benefits it is necessary to
-------
5-8
measure the marginal benefit curve for environmental quality, estimate the
levels of environmental quality before and after environmental changes, and
then calculate the area under the marginal benefit curve. This is difficult
to do because there exists no explicit market for environmental quality.
Therefore, a variety of alternative techniques to measuring willingness to
pay for improvements in environmental quality have been developed. These
techniques fit three major categories: (1) the specific damages approach;
(2) the implicit market approach; and (3) the hypothetical contingent
valuation approach. The specific damages approach involves monetizing a
physical measure of damage per unit receptor per pollutant and combines this
with the amount of receptor population. This measure is considered a crude,
lower-bound proxy for willingness to pay. The implicit market approach stems
from the observation that perceptions and values of environmental quality are
reflected in individuals' behavior in markets related to environmental
quality, such as property values or travel costs to recreational sites. The
contingent valuation approach relies on surveys or bidding experiments which
elicit direct measures which are contingent on the hypothetical framework
from which individual valuations are obtained.
The most fundamental approach to benefit valuation is the implicit market
approach, or supply/demand analysis because it enables the calculation of
consumer and producer surplus at an equilibrium. The demand for water
resources of a particular quality arises from a desired use activity—uses
for recreational activities, industrial water uses, withdrawals for supplies,
etc. Each of these uses requires a certain quality of water and the demand
depends on potential uses at a given geographic location. To evaluate the
-------
5-9
effects of changes in water quality, demand for a use activity must be
calculated. It is not always possible, however, to conduct-demand curve
estimation for benefit calculations. In reality, only a partial form of
demand analysis can be done. Moreover, the success (or reliability of the
estimate) of the analysis varies by benefit category.
For an in-depth discussion of these issues and methodologies that are
used to estimate economic benefits from pollution abatement see Freeman, The
Benefits of Environmental Improvement (1979), and Air and Water Pollution
Control; A Benefit-Post Assessment (1982); Feenberg and Mills, Measuring the
Benefits of Water Pollution Abatement (1980); and Research Triangle
Institute, A Comparison of Alternative Approaches for Estimating Recreation
and Related Benefits of Water Quality Improvements, (1983).
5.2 Study Methodology
Our strategy in this study is to employ methods developed by previous
researchers and to compute benefits for each category using a variety of
estimation techniques whenever possible.
The various categories of effects (or beneficial use classes) are
•*» *"
summarized in Table 5-3. The table also indicates the approach which has
been used to estimate the effect/benefit, and an evaluation of the
reliability and availability of the methodology and data.
-------
5-10
Table 5-3
Benefit Categories and Methodologies for Boston Harbor Study Area
I
I
Reliability/
I I Reliability of I Availability
Benefit/Effect I Benefit Estimation Approach I Methodology I of Data
I I I
Recreation
Swimming
Boating
Fishing
o Travel cost (logit model) excellent
o Regional participation good
o Beach closings cost savings fair
o Regional participation
o Regional participation
fair
fair
excellent
fair to good
fair to good
fair
fair
Health
Swimming
o Dose-response function
(incidence of disease)
Food consumption
o Dose-response function
(incidence of disease)
excellent
good
good
fair to good
Commercial fisheries
o Demand and supply
functions
good
poor
Intrinsic Benefits
o Contingent valuation survey
o Direct % of recreation fair
benefits good
fair
good
Ecological
o No approach available to
apply a dollar value for
benefits
Secondary Effects
o Input-output multipliers
fair
fair
-------
5-11
References
Feenberg, Daniel and Edwin Mills, 1980. Measuring the Benefits of Water
Pollution Abatement, Academic Press, New York, NY.
Freeman, A.M. 1979. The Benefits of Environmental Improvement, Johns
Hopkins University Press, Baltimore, MD.
Freeman, A.M., 1982. Air and Water Pollution Control; A Benefit-Cost
Assessment, John Wiley and Sons, Inc., New York, NY.
Hicks, John R., 1943. The Four Consumers' Surplus, Review of Economic
Studies, 11:31-41.
Research Triangle Institute, 1983. A Comparison of Alternative Approaches
for Estimating Recreation and Related Benefits of Water Quality
Improvement, Research Triangle Park, North Carolina, EPA.
Willig, Robert D., 1976. Consumer Surplus Without Apology, American Economic
Review, 66:589-597.
-------
Section 6
Recreation Benefits
The recreation benefits of improving water quality in Boston Harbor are
many. Boston Harbor is surrounded by a major metropolitan area of 2.8
million people and provides a setting for many diverse water uses including
boating, sailing, canoeing, fishing, swimming and beach activities. In
addition, in recent years the harbor has become an aesthetic focal point for
water-enhanced recreation activities such as picnicking, bicycling, camping,
hiking and sight-seeing. Figure 6-1 shows the various locations (called
receptor sites) of these water uses.
Although the CSOs and the STPs affect some of the same harbor areas of
the study, in general the receptor sites are primarily affected by one or the
other source. The CSOs affect recreation areas closest to the shore and,
thus, have the greatest impact on swimming and shore-related fishing and
boating. Of all the CSO planning areas, Dorchester Bay is influenced the
most because of the great concentration of CSOs and beaches in the bay. The
Quincy storm sewers affect water quality at local town beaches and Wollaston
Beach. The Charles River CSOs have a major impact on boating. This area is
discussed separately in Section 11 because of differences in data bases and
the nature of the water resources.
The areas primarily affected by the STP discharges are the waters and
islands surrounding the STP outfalls. Beaches in the towns of Quincy,
-------
Figure 6-1. Receptor Areas for the Boston Harbor Study
•(•lit I r
inrlcs River IT"/'
DORCHESTER
"iroMtt Mvir .
•#fi
1. Constitution Beach
2. Castle Island
3. Pleasure Bay
4. City Point
5. L&M Streets Beach
6. Carson Beach
7. Malibu Beach
8. Tenean Beach
9. Wollaston Beach
10. Quincy Town Beaches
11. Weymouth Bay
12. Hingham Harbor
13. Hull Bay
14. Outer Harbor Islands
15. Brewsters Islands
16. Nantasket Beach
17. Massachusetts Bay
HULL
f1
HINGHAM
-------
6-3
Weymouth, Hingham and Hill and the Boston Harbor Islands are the swimming
areas primarily affected by the STPs.
The Boston ffirbor Islands—Slate, Bumpkin, Qrape, (Gorges, Ioveils,
Gallups, Deer, Long, Rainsford, Moon, Thompson, Spectacle, Sheep, Peddocks,
and the Brewsters--are a unique natural resource in a metropolitan area
possessing only one-half of the recommended minimum acreage of open space per
thousand population, The Islands offer a wide range of activities such as
boating, swimming, picnicking, fishing, hiking, camping, scuba diving, and
historic sight-seeing. Many of the islands have limited recreational
facilities which restrict current and potential visits. Poor water quality,
however, is also a major factor restricting recreational activities. Effluent
from Deer and Nut Island sewage treatment plants seriously degrades water
quality around the Islands, particularly discouraging swimming and fishing.
Assuming that the planned recreational facilities were constructed, then
upgrading the plants and/or discharging the effluent into the ocean would lead
to a significant improvement in water quality, which would lead to a
corresponding increase in both frequency of participation and total number of
visitors.-/
Fishing and boating are also affected by the STPs since a large percentage
of these activities take place in the outer harbor study area rather than on
or near the shore. Participation in all boating--sailing, motor boating,
canoeing and windsurfing—and fishing activities in Boston Harbor is expected
to increase with decreases in water pollution Ievelsj2/
£/ The exception to this assumption is the Brewsters Islands and
Nantasket Beach, which are expected to be negatively influenced by the ocean
outfall option. ,
—/ The degradation of water quality in Massachusetts Bay under the ocean
outfall option is expected to primarily affect commercial fishing.
-------
6-4
6.1 Data Needs and Data Bases
The data needed to estimate recreational activity in these various areas
and to relate the uses to changes in water quality come from a variety of
sources. This section discusses the data bases used to estimate recreation
•-
benefits. It is followed by a discussion of the various methodologies which
have been applied to the Boston Harbor case to arrive at a range of benefit
estimates for each separate benefit category.
6.1.1 Swimming Attendance
Seven of the beaches managed by the Metropolitan District Commission
(MDC) are affected by CSOs and/or STPs in the study area: Constitution,
Castle Island, Pleasure Bay (including City Point), Carson,—' Malibu,
Tfenean, and Wollaston. tearby cities and towns also have small neighborhood
beaches which are affected by pollution control sources. The cities of
Quincy, Weymouth, Hingham and Hill recognize ten beaches besides Wollaston
for water quality collection purposes. In addition, swimming occurs on an
informal basis on many of the eleven Boston Iferbor Islands. Rough estimates
put recent seasonal attendance of all these affected beaches at 4.0 million
people (see Table 6-1). Unfortunately, neither the MDC, the towns, nor the
Massachusetts Department of Environmental Management (DEM) keep attendance
records or make official co- \ts during the season. In addition, people swim
at the beaches during warm weather in the spring and fall, even though they
are not officially open. Information from a 1975 recreation survey (Binkley
and Hanemann) and from the MDC indicate that some of the Boston area beaches
i L and M Street Beach, part of Carson Beach, is managed by the City of
Boston.
-------
6-5
T&ble 6-1. Seasonal Swimming Supply
Current 1
Seasonal Beach 1
Attendance 1
Constitution
325,000
Dorchester/Neponset
590,000
Castle island
15,000
Pleasure Bay
175,000
Carson
100,000
Malibu
150,000
Tfenean
150,000
Wollaston
2,750,000
Quincy
158,900
Weymouth
105,820
Hingham
22,200
Hill
66,000
Nantasket Beach
3,035,000
Seasonal £/
Capacity
582,780
5,044,878
291,390
1,548,155
1,899,774
632,449
673,110
4,595,976
320,568
763,680
355,200
532,800
£/
I Seasonal £/
1 Excess
1 Supply
257,780
4,454,878
276,390
1,373,155
1,799,774
482,449
523,110
1,845,976
161,668
657,860
333,000
466,800
£/
5/ Based on 40 ft2 per person; turnover of 3 times per day; 29.6 peak
user days per season. Except Wollaston Beach with four times per day
turnover and 39.4 peak user days per season. (Derived from US Department
of Interior, 1970.)
b/ Excess supply = (Capacity) minus (Current attendance).
£/ Not applicable since expect degraded or unchanged water quality.
Source: See Appendix B, Table B-l.
Note: Brewsters Islands are omitted because most of the recreational activity
is fishing and boating, and Massachusetts Bay is omitted because the primary
activity is commercial fishing.
-------
6-6
draw people from many parts of the Boston Metropolitan area. Other beaches
appear to be used almost exclusively by people from a nearby section of the
city, such as Carson Beach by South Boston residents and Constitution Beach
by East Boston residents.
Attendance data used for calculating swimming benefits were estimated by
MDC personnel and by recreation and park department officials in Quincy,
Weymouth, Hingham, and Hull. We also compared attendance figures reported by
the MDC in the 1975 Binkley and Pfinemann study along with attendance figures
generated from a survey used in their study. This range of values can be
found in Table B-l, Appendix B. Data on beach acreage and/or linear feet of
beach/shoreline was also supplied by the MDC and municipalities and was used
to develop a range of beach capacities for each affected area based on
national recreation standards. Estimates for beach capacity and beach
attendance-numbers are presented in Table 6-1. These attendance and capacity
figures are used in several approaches to calculating swimming-related
benefits in this report. The accuracy of these methods is linked to the
accuracy of the recreation data.
Other factors could also act to limit the increased participation
predicted as a result of water quality improvement. They include crowding
and congestion, available parking facilities, presence of jellyfish and,
particularly for Boston Iterbor, cold temperatures of the air and water.
Although these effects can be significant, the first three factors were not
considered here because of insufficient data. The effects of air and water
temperatures were incorporated in a lower bound estimate of increased
participation.
-------
6-7
As a qualitative assessment, we have assumed that crowding would not have
as severe an impact on the study area beaches as in other recreation areas
because these beaches are extremely urban and, as one municipal source noted,
visitors are used to constant crowding.
Parking facilities close to the beaches could limit visits on a given day
as these beaches are used by people throughout the area, currently, the MDC
estimates that on a normal sunny day parking is at 80 percent of capacity
although on the hottest days demand for parking greatly exceeds capacity and
substantial traffic congestion occurs. Beachgoer preference is to drive to
the beach rather than use public transportation which is available and
convenient to the cities' beaches. Thus, alternatives to parking do exist if
the increased participation should exceed the available parking supply.
With regard to jellyfish, there are practically no data available on this
form of life except for some research done in Chesapeake Bay by the
University of Maryland's Chesapeake Bay Laboratory. Most of the work has
been done in open ocean. Observations in Boston Harbor indicate the presence
of a substantial jellyfish population. The fish are present throughout the
summer months and, in 1984, have been observed as early as April. The
prevalent theory is that polluted water promotes an algae growth within the
jellyfish food chain and the population increases in accordance with the food
supply. However, scientists caution that there is no evidence to support
this theory. Jellyfish are considered to have little food value and
consequently have no predators to act as a population control mechanism.
Population levels are thought to be decreased by storms, currents and changes
in the salinity of the marine environment. The introduction of fresh water
-------
6-8
into the harbor through CSO's could result in reduced salinity which in turn
could promote or deter jellyfish population growth. However, a lack of data
makes the issue speculative. An agreed to fact is that the presence of
jellyfish in the waters generates an adverse public reaction and acts as a
deterrent to water contact activity and, possibly, increased visits to the
beaches on days when jellyfish are present. 3/
In attempting to account for the effects of air and water temperature on
swimming attendance, for an upper bound estimate the base seasonal attendance
figures are limited to the three summer months. For a lower bound estimate
the predicted increased attendance is modified according to the distribution
of air temperatures during these summer months. — On those days with
cooler temperatures not all the predicted increased participation due to
improved water quality is assumed to take place. Thus, a factor is applied
reducing the upper bound'estimate in relation to the distribution of air
temperature during the summer months (see Appendix B.3) .
6.1.2 Recreation Studies
Information on general recreational activities such as percentage of
population participating in swimming and percentage of unmet demand for
boating and fishing was drawn from a number of existing city, state and
federal reports. These include, the New tork-New Bigland Recreational Demand
Study (Abt, 1979), the 1980 National Survey of Fishing, Hunting and Wildlife
i/ Information in this section was provided by EPA, Region I, Boston,
MA.
^/ Air temperature is assumed to affect beach attendance. Air and water
temperature are assumed to affect the amount of swimming done by those who go
to the beach (and are taken into account in estimating swimming health
benefits in Chapter 7) .
-------
6-9
Associated Recreation (US DOI) , The 1982-1983 itetionwide Recreation Survey
(US DPI), Eastern Massachusetts Metropolitan Area Study (EMMA), (Metcalf &
Eddy, 1975) , Boston ffirbor Islands Comprehensive Plan (Metropolitan Planning
Council, 1977), and the Massachusetts SCORP (Massachusetts DEM, 1976). Not
all of the information in these studies is specific to Boston Harbor nor does
each study supply exactly what is needed for estimating pollution abatement
benefits, for example, there is some information about swimming and
beach-related activities, but there is very little information available
describing fishing and boating activities. In addition, much of the data in
these studies are only estimates, rather than statistically-derived results
from rigorous sampling, which compromises their use in benefit estimation
techniques. We have evaluated a number of these recreation studies for their
accuracy, sampling methods and applicabilty to the Boston Ksrbor case study,
and have used only those statistics and numbers which we believe to be
representative and unbiased. A brief discussion of each recreational source
can be found in Appendix B.7.
6.1.3 Water Quality Data for log it Model
Water quality data is needed for the application of the travel cost logit
model (see Section 6.2.2 below).—/ There is information about ambient
water quality concentrations throughout most of the harbor but it is of
limited usefulness due to the shortcomings in sampling procedures (frequency,
consistency, regularity, comprehensiveness) and in the comparability of the
measurements used to describe water quality. Recently, the MDC has started a
water quality sampling program to better identify ambient concentrations of a
variety of pollutants such as BOD5, heavy metals, oil and grease.
§/ At the time the logit model analysis was run the Boston Harbor Data
Management System was not available so that this data had to be collected
independently.
-------
6-10
Currently, the only readily available water quality data for the MDC and
town beaches are measures of fecal and total coliform concentrations.
Binkley and Itenemann (1975) collected water quality samples for a number of
water quality parameters to be used in their recreational travel cost model,
but we have chosen not to use any of their data because water quality samples
were only taken once during the summer and thus cannot be considered
statistically representative of water quality for the entire swimming
season. For this Boston Harbor case study, we collected 1974-1982 fecal and
total coliform concentrations and information on beach closings/costings,
from the seven MDC beaches and several town beaches in Quincy, Weymouth,
Hingham and Hill. In general, the MDC and towns sampled once a week,
resampling when high counts were recorded. In cases where only total
coliform concentrations were reported, we substituted fecal coliform values
based on a statistically significant regression function relating fecal
coliform concentrations to total coliform concentrations (see Appendix C) .
This water quality data, together with data from several other towns in the
Boston Metropolitan area, was used in the travel cost logit model.
6.1.4 User (Unit) Day Values
The application of user day values to estimate recreation benefits is the
most common and widely used of all the estimation techniques because of its
simple methodology and minimal data requirements. Essentially, a single
dollar value per recreation day (not per visit) is developed to represent the
market value of the recreation services. Originally, this figure per
recreation day was based on recreational costs including entrance charges and
equipment expenditures. The federal government has adopted a schedule of
values to distinguish between "general" and "specialized" recreation
-------
6-11
activities.^ A single unit value is assigned per recreation day regardless
of whether the user engages in one activity or several. This value should
reflect the quality of the activity and the degree to which opportunities to
engage in a number of activities are available (Dwyer et al., 1977). We have
reviewed a number of user day values for their applicability to Boston Harbor
and present the values and their sources in Appendix B, Table B-3.
There are many shortcomings and problems with using user day values to
estimate recreation benefits. These limitations are discussed in detail in
Dwyer, et al., 1977. The most basic problem is that most user day
values--whether based on government or private schedules--may not be developed
from empirical data on the actual willingness of participants to pay for
recreation. This lack of theoretical or empirical justification for many user
day values often leads to arbitrary and biased estimates of the value of a
recreation day.
User day values have been developed both nationally and locally. Many of
these values tend to be site-specific, reflecting regional socio-economic
biases and, more often than not, cannot capture the effects of incremental
changes in environmental quality. In addition, user days cannot capture the
increased value or utility of the individual recreator. As a result, user
day values may produce biased estimates of consumer surplus from improved
water quality.
6.1.5 Water Quality Impact
All of the above categories of data are needed to evaluate the response of
recreators to water quality changes. The remaining piece of data that is
- See Federal Register. Vol. 48, No. 48, March 10, 1983.
-------
6-12
needed is what the estimated percentage change in water quality will be, given
the implementation of a treatment option. Section 4 explained how the
percentage reductions in pollution were estimated for the various receptor
sites. T&ble 4-2 and 4-3 presented best-guess ranges and point estimates of
the water quality changes. We use these numbers in the benefit calculations.
6.2 Benefits
Reducing pollution in the harbor by upgrading STPs and improving CSOs will
lead to recreation benefits throughout the Boston Harbor area. Two major
components of consumer surplus should be estimated in order to fully represent
all benefits from improved water quality. These components are:
o increase in participation (both frequency and total numbers)
resulting from decreased time and travel costs
resulting from a higher quality recreational experience
resulting from increase in water areas available for
recreation; and
o increase in the price participants are willing to pay (WTP)
for the improved quality of the recreational experience.
A third component can be measured by calculating the value of lost participa-
tion due to severe water contamination, such as that resulting from beach
closings.
We have used a number of techniques to calculate a range of economic
recreation benefits associated with improving water quality in Boston Harbor
by upgrading the sewage treatment plants and improving the CSOs. These
include:
Measure of
Benefit Consumer Surplus Benefit Estimation Approach
Swimming o Increase in participation o Regional participation
o Travel cost (logit model)
o Increase in WTP per trip o Travel cost (logit model)
o lost participation o Beach closings
-------
6-13
Measure of
Benefit Consumer Surplus Benefit Estimation Approach
Boating/
Fishing o Increase in participation o Regional participation
All Recreation
Boston Harbor
Islands o Increase in participation o Regional participation
Each of these estimation techniques and benefits categories are discussed
separately, below. Included in this discussion is a presentation and
analysis of the range of benefit values corresponding to the pollution
abatement program, limits of the analysis, and pertinent references. A
detailed description of the benefit computations and the empirical data is
presented in Appendix B.
6.2.1 Swimming—Increase in Participation
One of the significant consumer surplus benefits associated with water
pollution abatement in the Boston arbor study area is the increased use of
the beaches by current users and new use by previous non-participants. This
is one of the more difficult benefits to measure because of the need for
reliable and accurate calculations of user and non-user response. For Boston
Harbor, we have assumed that an improvement in water quality--specifically
fecal coliform--is equivalent to an increase in total supply of the water
resource. Theoretically it is therefore possible to relate this increase in
a water resource to a corresponding increase in participation. Increased
participation, measured in total visits, should capture both increase in
frequency of visits by those already participating, as well as increased new
use by previous non-users. Once this population number is calculated, it is
possible to value this increased participation by applying user-day values.
-------
6-14
Estimating benefits accruing from increase in participation involves the
following:
a) determine which areas are affected by each pollution control
plan;
b) calculate excess seasonal beach supply;
c) estimate the range of increase in participation using
information from regional participation studies;
d) relate the increase in participation to the pollution control
plan; then
e) value increased participation by applying a range of user day
values.
The first step in estimating the benefits from increased participation
involves determining which beaches are affected by the different treatment
options. These were determined in Section 4 and presented in Table 4-3. The
next step is to calculate the excess supply of each beach, such that increased
demand will not exceed the existing supply. This will prevent overstating
swimming benefits. Excess seasonal supply of these beaches was estimated
using beach attendance data from the MDC and towns, and the capacity of each
beach was calculated using a variety of recreational standards and information
from town governments and the MDC on acreage and linear feet of shoreline.
This data was summarized in Table 6-1. Other factors could serve to limit
increased participation, as discussed in Section 6.1.1. However, these
effects were not considered here because of insufficient data.
6.2.1.1 Regional Participation Model
The most important step in this methodology involves estimating a
range of increased participation. The first approach presented here to
estimating increased participation is based on regional and local
-------
6-15
recreation participation studies. Results of these studies suggest that the
number of unmet user days (often called latent demand) in the Boston SMSA is
4.3 to 5.2 million user days. IBing this information, we can calculate unmet
demand at the beaches that will be supplied by the different pollution
control options. These calculations are summarized in Appendix B.I. It is
possible to relate this total increase in beach participation to the
pollution control plans by assuming that the percentage reduction in
pollution will supply a corresponding percentage of the excess supply in
terms of additional user days. A number of other assumptions were made in
order to calculate increase in participation:
s~
(a) water quality is the major constraint affecting unmet
demand;
(b) current facilities are adequate to fulfill the needs of
additional visitors;
(c) time available for recreation is not a constraining factor;
(d) fecal coliform is the best available measure of overall
water quality affecting participation;
(e) there is little effect of substitution of sites on
participation at individual beaches; and
(f) people use the beaches for swimming purposes.
These assumptions and calculations produce the upper bound
estimates of increased user days presented in Table 6-2. R>r the
lower bound estimates a factor based on the distribu .ion of air
temperatures during the summer months is applied. It is assumed
that on days when the air temperature is below 79° Farenheit, not
all the predicted increase in beach visits may actually occur even
with the improved water quality because of the relatively lower air
temperature (see Appendix B.3 for details of the calculations) .
-------
6-16
T&ble 6-2. increased Swimming Participation—Regional
Participation Model §/
Beach
CSO
Ocean
Outfall
Secondary
Treatment
LOWER BOUND ESTIMATES
Constitution
Dorchester
Wollaston
Qu incy
Weymouth
Hingham
Hull
TOTAL
76,099
157,884
735,900
42,522
0
0
0
1,012,485
10,871
19,736
91,988
5,315
10,619
2,228
6,623
147,380
5,436
19,736
91,988
5,315
10,619
2,228
6,623
141,945
CSO and
Ocean Outfall
(User Days)
86,970
177,620
827,888
47,837
10,619
2,228
6,623
1,159,785
UPPER BOUND ESTIMATES (User
Constitution
Dorchester
Wollaston
Qu incy
Weymouth
Hi ngham
Hull
TOTAL
113,750
236,000
1,100,000
63,560
0
0
0
1,513,310
16,250
29,500
137,500
7,945
15,873
3,330
9,900
220,298
8,125
29,500
137,500
7,945
15,873
3,330
9,900
212,173
130,000
265,500
1,237,500
71,505
15,873
3,330
9,900
1,733,608
CSO and
Secondary
Treatment
81,535
177,620
827,888
47,837
10,619
2,228
6,623
1,154,400
Days)
121,875
265,500
1,237,500
71,505
15,873
3,330
9,900
1,725,483
i/ See Appendix B for details of the calculations.
-------
6-17
/
An alternative approach to estimating increase in participation is to use
results from the logit model (described below in Section 6.2.2) which
predicts increased visits based on a percent reduction of water pollutants to
calculate unmet demand.
It is important to compare the estimates of increased participation due
to increases in water quality with the availability of excess supply, in
order not to overestimate swimming benefits. We have assumed in the case of
the torchester/teponset Bay beaches that if increased participation exceeds
capacity at any one beach, then other nearby beaches will serve as substitute
sites. This enables us to treat the Dorchester Bay beaches as a unit, rather
than individually, and simplifies the analysis.
6.2.1.2 Benefit Estimates
The final step in this methodology is to value the increased
participation by applying a range of appropriate user day values, which
represent a crude proxy for individual consumer surplus. The results of this
valuation are presented in Table 6-3.
Table 6-3. Annual Benefit of Increased Swimming
Participation for all Boston Harbor Beaches (1982 $000)
User Day
Value
$1.60
$5.80
$11.06
CSO
1,620.0 a/
7,324.8 b/
16,737.2 £/
Ocean
Outfall
235.8
1,066.3
2,436.5
Secondary
Treatment
227.1
1,026.9
2,346.6
CSO plus
Ocean Outfall
1,855.7
8,390.8
19,173.7
CSO plus
Secondary
Treatment
1,847.0
8,351.7
19,083.8
§/ Lower bound estimate of increased visits (from Table 6-2) multiplied
by user day value (from Table B-3, Appendix B).
£/ Average of lower and upper bound estimates of increased visits
multiplied by user day value.
c/ upper bound estimate of increased visits multiplied by user day value.
-------
6-18
There is a wide range of estimated benefit values for increased
participation because of the many different user day values. Benefits are
most substantial for the Dorchester/Nsponset Bay Beaches and for the
Wollaston and Quincy Beaches. Benefits are more modest for Constitution
Beach. Benefits are substantial for torchester/Neponset Bay and
Wollaston/Quincy beaches because these areas have poor water quality, a large
predicted percent cleanup, and a great number of visitors. Thus, cleaning up
these areas will attract a large number of new recreators and significantly
increase the frequency of participation of current users. Swimming benefits
from an increase in participation are small for the STP affected beaches
because of the fewer number of people who visit these beaches and because the
STP option is expected to abate pollution by only 30 percent.
6.2.1.3 Higher Valued Experience
Improved water quality may also lead to an increase in the price that
participants are willing to pay for the improved quality of the recreation
experience. This higher valued experience is often very difficult to
quantify. Other benefits studies have relied upon surveys of willingness to
pay for various improvements in recreational water quality (See for example,
Ditton and Oaodale, 1972 and Bricson, 1975). Such surveys are often locally
biased and, thus, cannot be applied to other areas because of sociological,
environmental and economic differences.
No such studies were found to be applicable to Boston Harbor because of
the previously mentioned biases. We therefore, were unable to calculate the
portion of consumer surplus attributable to a higher valued experience using
this method.
-------
6-19
6.2.1.4 Limits of analysis
An analysis of increased participation was limited by both benefit
estimation methodology and by data bases. latent or unmet demand was
difficult to measure. Estimates were based on results from regional and
local recreational studies, which may be inaccurate for a number of reasons.
(For more details see Appendix B.7.) The accuracy of our benefit estimates
is greatly influenced by recreation attendance data and capacity estimates.
Current attendance figures were based on professional estimates, rather than
actual field data, and thus must be considered "best guesses". In addition,
these estimates of attendance figures were based on seasonal summer
attendance, from Memorial Day to labor Day, and did not include the number of
people who swim before or after the "summer" season. Benefits to these
recreators are not captured and, therefore, total benefits may be
understated. Beach capacity estimates also represent our best professional
judgment. For example, Wollaston has an estimated capacity of 2.75 million
people, but the MDC has estimated seasonal attendance to be over 3.5
million. In this case we concluded that the development capacity for
Wollaston represents a lower bound and assumed a greater turnover rate than
normal and a greater than expected crowding. Other factors, including
adequate parking facilities, cold water temperatures and the presence of
jellyfish which could limit attendance in a manner similar to beach capacity
were not considered because of the lack of data.
These benefit estimates are also limited by the many assumptions which
were made, including assumptions about the appropriateness of fecal coliform
as the best available water quality indicator, time constraints, and the
effect of water quality improvement on increased participation. It was
-------
6-20
assumed that many relationships were strictly linear, such as the
relationship between percentage increase in use and the percentage reduction
in water pollution. Such an assumption seems feasible here, since the
baseline water quality level is so poor; however, in general, the
relationship between percentage reduction in pollution and percentage
increase in participation is very sensitive to the baseline water quality
level, for example, a 90 percent reduction of pollution in a water body that
has relatively good water quality may result in little or no increase in
participation. We also assumed that user day values were the best available
proxy for consumer surplus. In reality, user day values cannot capture total
consumer surplus because they cannot measure increased utility of each visit
due to improved water quality. The higher range of user day values
($5.80-$11.06) is, therefore, more appropriate to use than the lower one
($1.60-35.80) in estimating recreation benefits. All of these limitations,
shortcomings and the state-of-the-art nature of benefit estimation will be
reflected in the final range of swimming benefits and must be taken into
consideration when interpreting the values.
6.2.2 Travel Cost Model—Conditional Log it Analysis
An alternative apporach to estimating increased participation is the
logit model which incorporates the probability of visiting a beach as a
function of distance to the sites, socioeconomic factors and water quality
variables. This approach is a specialization of the so-called travel cost
approach first suggested by Harold tt>telling in 1949, then developed by
Clawson and Knetsch (1966) , and since applied by many others (se_e Binkley,
1977, for a review of the literature).
-------
6-21
6.2.2.1 Methodology
This methodology uses observed recreation travel patterns to infer the
recreationists' response to price changes. Havel costs play the role of
price in estimating a demand curve for a specific site. Other personal
characteristics of the recreationist, such as income and age, are used in the
same equation to control for tastes and preferences. Because a demand curve •
measures the marginal willingness to pay for a good, estimates of recreation
benefits can be obtained from the area under a demand curve using travel cost
data and information on socioeconomic characteristics. In the present case,
we extend this basic methodology to include water quality characteristics in
the demand function. Then we can infer the changes in price which would be
equivalent to a change in water quality, and from that information can infer
the benefit of the change in water quality.
The principal theoretical shortcoming of this approach is the use of
travel costs to simulate prices. The recreationist may not respond to prices '
(i.e., an entry fee) in the same way as he/she does to travel costs because
travel may have a special utility or disutility in itself. Part of the
disutility of travel might be related to travel time as well as travel
costs. (See below for a further discussion of the time issue.) Another
common difficulty in the application of the travel cost method is the
allocation of joint costs of travel made to several recreation sites as part
of a single trip. Because travel costs are used as a proxy for prices, to
determine the "price" of an individual site it becomes necessary to separate
the cost of travel to one site from that to other sites. Consider, for
example, a trip from Boston to the Grand Canyon, then to Yellowstone National
Park, and then back to Boston. To infer the recreational value of the Grand
-------
6-22
Canyon from this trip, we would need to know what part of the travel costs
associated with the whole trip to assign to the visit to the Grand Canyon.
The appropriate cost is probably less than the total cost, but could well be
more than just the additional cost of including the Grand Canyon in the
trip. In short, there is no unambigious way to allocate joint costs of
recreation travel. Fortunately, day trips in an urban setting are not likely
to be conducted as part of a larger recreational outing, so our analysis
probably does not suffer from this limitation.
It is important to discuss the major ways that our methodology differs
from the classic implementation of the travel cost approach. First, we
consider a system of competing recreation sites. That is, demand for
recreation at one site depends on the characteristics of other possible
recreation sites that an individual might choose. To our knowledge, aside
from the direct antecedents of this work, only three other studies (Cicchetti
et. al, 1976; Hurt and Brewer, 1971; Morey, 1981) have incorporated this
important feature of the problem.
Second, we attempt to explicitly account for travel time as well as
travel cost. It is easy to show that ignoring the cost of time spent in
recreation leads to biased estimates of the value of a recreation site. This
point is well-recogni :ed in the literature (see, for example, Wilman, 1980).
The following section on the conditional multinomial logit model acknowledges
the empirical difficulties we had in obtaining usable estimates of the value
of time and discusses this point further.
Third, we model recreational demand as a discrete choice process. That
is, over the summer the individual chooses to go to some sites, perhaps none,
but probably not to all the available sites. Consequently there are
-------
6-23
typically quite a few observations of zero visits/ and these observations
tell us very little about how he/she trades off water quality with travel
distance. Therefore, we would like a model of recreation demand which
explicitly accounts for the kind of information contained in this large
number of zero observations. The multinominal logit model, borrowed from
transportation demand analysis, is one such model. This model was first
proposed for recreation demand analysis by Binkley and Ifinemann (1975) and
subsequently has been developed by Hanemann (1978) and by Feenberg and Mills
(1980). Peterson et. a_l (1983) applied a version of this model to activity
choice at the Boundary Waters Canoe area.
•Die first three studies rely on the same basic data. In 1974, a sample
of 500 households representative of the Boston SMSA were interviewed
concerning their recreation visits to 29 fresh and saltwater beaches in the
Boston area during that summer. A total of 467 usable questionnaires
resulted from the survey. Pertinent social and economic data on these
families were collected along with information on recreation habits. To
compute travel distance and, hence, cost, each of the sample points was
located on a map as were each of the recreation sites. In the original three
studies, travel distance was computed as the straight line distance between
the two points. Also, water quality variables used in the demand equations
were derived from one single sample at each beach during July of 1974.
(Binkley and Hanemann, 1975, describe the data more fully.)
While sharing a common estimation strategy with these other three studies,
the present work employs a somewhat different data base. Recreation patterns
and socioeconomic data from the Boston survey were used, but improved
information on travel costs and water quality was incorporated. In an urban
-------
6-24
area, straight line distance is a particularly poor measure of actual travel
distance. Fortunately, in the mid-197O's the Central Transportation Planning
Staff (CTPS), a regional transportation planning agency for the Boston area,
developed a detailed origin-destination travel distance and time matrix for
the region. Our sample points and beaches were located in the CTPS
transportation zones, and the minimum travel distance and time from each
sample point to each beach was computed. Consequently, the measure of
distance used in this research reflects much more accurately the actual
distance between each individual and the various beaches. In addition, the
transportation time information derived from the CTPS study offered the
possibility of estimating the value of time in travel for recreation.
Due to large sampling errors, the 'told" (Binkley and ffinemann, 1975)
measure of water quality—a one time grab sample--might not reflect the true
water quality level. We assembled measures of coliform levels from the
records of the Metropolitan District Commission and the appropriate towns.
These were averaged over the summer, and we employed the median level of fecal
coliforms as our "new" measure of water quality. The agencies responsible for
some of the beaches did not collect information on fecal coliforms. R>r these
cases, a regression equation was developed relating the old water quality data
to the new estimate of fecal coliforms. for the sites where there was no new
information, this equation was used to estimate . .ie new data from the old data
on fecal coliform (OLD):
NEW = -53.27 + 13.22 log (OLD) N = 19
(-1.99) (3.17) R2 = 0.371
-------
6-25
6.2.2.2 The Conditional Multinomial log it Model
The multinomial logit model of multiple site demand is ideally suited for
the situation we consider here: the choice of one or more beaches from a
known universe of possible sites.—^ This section describes the model
informally and explains how we obtain estimates of the benefits of water
quality improvements from the model. Appendix B.4 presents the model and
benefit estimation procedures in more detail.
We want to model the number of visits an individual will make to one or
more of the beaches as a function of beach characteristics (including water
quality), travel costs/time, and socio-economic characteristics of the
individual. With such a model, we can alter the level of water quality at
one or more of the sites and simulate how use at all of the sites will
change. From those simulated changes in use, we can infer the value of the
change in water quality.
The conditional logit model is divided into two parts. The first part
describes the probability that an individual will choose to visit each of the
beaches given that she/he takes a trip to any of the sites. Equivalently,
t
this part of the model can be thought of as predicting thejproportion of all
beach visits which will be made to each of the individual beaches. This
first part of the model is referred to as the Usite_ choice" moov.1 in the
following discussions.
The model is called a "conditional" logit model because the choice of
sites is conditional on knowing the total number of visits that the
individual takes. Hence, the second part of the model explains the total
S/ See Domenich and McFadden (1975) for an authoritative treatment of
this model.
-------
6-26
number of visits an individual makes to any of the study beaches. This
second part of the model is referred to as the "visitation" model.
The overall structure of the model can be summarized as follows. The
number of visits by individual i to beach j is njj. The individual makes a
total of n^ beach visits during the summer. The probability of an
individual i going to beach j (or the proportion of her/his total beach
visits which are made to beach j) is pjj. Then we model the number of
visits to beach j by individual i as:
nij • ni Pij (
We estimate n^ in the visitation model and p. j in the site choice model
and predict n^. using this equation.
For the site choice model, the dependent variable is the probability of
visiting a given beach. While this variable is precisely the probability of
visiting a certain beach, it can also be considered the proportion of the
time that an individual will go to a particular beach when she/he goes to the
beach at all. The probability of visiting a certain beach (given that a trip
is taken) is assumed to be a function of the desirability of that beach. We
take desirability to depend on the characteristics of the beach (e.g., water
quality), the travel costAime associated with a visit to that beach, and the
socioeconomic characteristics of the invididual who is making the trip.
Through the procedures described in Appendix B, the probability of visiting a
beach is estimated as a linear function of these variables. The results for
the site choice model which are presented below can be interpreted much as
one would interpret an ordinary linear regression.
-------
6-27
The dependent variable for the visitation model is the number of visits
an individual made to any of the study area beaches during the summer. We
assume that the total number of beach visits an individual takes is related
to the socioeconomic characteristic of the individual and the overall
desirability of the sites available to her/him. While there are many ways
this latter variable might be measured, the details of constructing the
conditional logit model require that it be derived from the site choice model
in a specific way. This variable, called the "inclusive price", measures the
average desirability of the available sites. Here, the term desirability has
the same meaning as it did in the description of the site choice model and
includes the level of water quality at each of the beaches. Through the
inclusive price term in the visitation model, a change in water quality at
one or more beaches will not only affect the split of visits among the
various beaches, but will also affect the total number of beach visits which
will be taken.
Linking together the site choice model, the visitation model, and
Equation 6.1 permits one to model how changes in water quality at any of the
sites will affect total number of visits to each of the sites. To simulate
the effect of a change in water quality at one or more of the sites, we use
the visitation model to predict total number of visits after the change in
water quality, the site choice model to predict the fraction of the visits
which will be made to each site, and Equation 6.1 to determine the number of
visits made to each site.
In general, the benefits associated with a simulated improvement in water
quality come from two sources: an increase in the total number of visits and
an increase in the value of each of the visits. Of course, because the
V.
-------
6-28
demand model includes the interaction among beaches, a water quality improve-
ment at one beach might lead to a decrease in use at some other beach. All
of these shifts in usage are included in our benefits calculations.
Conceptually, we are interested in determining the equivalent variation.
Suppose we improve water quality at some set of beaches. The equivalent
variation is the amount of income we would have to take away from an
individual to make her/him indifferent between the situation with higher
income/lower water quality and that with lower income/higher water quality.
The equivalent variation measures this willingness of an individual to pay
for an improvement in water quality. This measure of benefit is a good
approximation to other measures of benefit (Willig, 1976 and 1978) and also
is of interest in its own right.
Because income is not explicitly incorporated in the demand model, the
equivalent variation cannot be estimated directly. We use a modification of
a procedure developed by Small and Rosen (1982) and adapted to this problem
by Jeenberg and Mills (1980) to determine the equivalent variation associated
with a change in water quality. The details of the procedure are presented
in Appendix B.4, but the method can be outlined as follows. Demand is a
function of travel distance and water quality. In the estimated demand
model, then, we know how an individual trades off travel distance and water
quality. We can estimate the value of a simulated improvement in water
quality by asking how much further could the individual travel given the
water quality improvement and be no worse off than she/he was before the
water quality improvement took place. Benefits are measured in units of
distance. Therefore, in order to put benefits in dollar units, we need to
know the cost per unit distance.
-------
6-29
Here we take cost to have two components: a money cost and a time cost.
It is important to discuss how time should enter the model. Because time is
a scarce resource which the recreationist must allocate, the amount of time
spent in travel and on the site itself should be included in the model.
Failure to do so will lead to an underestimate of the value of the site.
Unfortunately, the data available for this study does not permit usable
estimates of the effect of these two time variables. The survey data on time
spent on the site contained numerous missing observations. Further, it is
not conceptually clear how to measure the amount of time which would be spent
on sites not visited. Thought of in another way, we need to estimate a three
part model--site choice, visitation and time spent on site—and the data are
not adequate to do so. Attempts to include travel time along with distance
in the model failed because of the high collinearity between the two
variables.
An alternative procedure was employed to partially account for the value
of time. Ossario (1976) suggested that the value of travel time for
recreation is about one third the wage rate. Consequently, our estimates of
welfare change were converted to money values of the basis of $0.12/4nile (the
national average in 1974) plus travel time valued at one-third the
individual's wage rate.
The wage rate was computed from information on income and the number of
days worked per year. From the household survey, we know the number of days
taken off per week, the number of holidays per year and the annual amount of
v
vacation time. By subtracting the sum of these figures from 365 days, we
know the number of days worked per year. Annual income is divided by the
number of working days to determine the average daily wage. Daily wage is
converted to an hourly wage assuming eight hours per work day.
-------
6-30
6.2.2.3 Model Results
The recreation demand analysis provides several kinds of results. First,
we present the estimates of the site choice and visitation models. These
results are compared with those of Feenberg and Mills (1980) to show the
effect of our different and, in our view, better measures of travel costs and
water quality. Second, we use the procedures, outlined above and detailed in
Appendix B.4, to simulate the effect of changes in water quality on
recreation patterns and to estimate the recreation benefits of several
specific water quality improvement scenarios for the Boston Harbor study
area. These results depict total benefit curves for each of the water
quality improvement scenarios.
Table 6-4 presents the estimates of the model parameters. The model,
using all 467 cases, predicts the site choice correctly in 15.9 percent of
the cases compared with 34.7 percent for the Feenberg-Mills model. We
attribute this difference primarily to the fact that Eeenberg and Mills
grouped individuals according to residential (origin) location, which we did
not. In addition, our specification of the site choice model omits several
interaction terms (age x distance, income x distance). We felt that there
was no good a priori rationale for including these interaction terms. The
dis'ance coefficient for the Feenberg-Mills model is about 0.33 expressed in
one-way miles and evaluated at the mean of the interaction terms. This is
more than three times higher than the value we obtained indicating the
magnitude of the error from using straight line distance to proxy for actual
travel distance in an urban area.
There are several other interesting differences in the two models which
can be seen in the simulation results. A 10 percent reduction in coliform
-------
6-31
Table 6-4. conditional log it Model Estimates
Site Choice:
Distance (miles one way)
Water Temperature (°F)
Fresh water (dummy)
Fecal Col if or m (median)
log likelihood (xlflS)
percent correctly predicted
coefficient
-0.1003
-0.4088
-1.607
-0.01275
At
Convergence
-0.1443
15.9
t
-50.71
-41.17
-27.79
-18.47
At
Zero
-0.1658
3.5
Visitation
Intercept
Inclusive Price
Age (years)
Education (years)
Income ($1974 x 10 3)
coefficient
172.7
5.757
-0.3095
-0.5758
0.2550
R2 = 0.078
f (4,462) = 9.79
t
—
3.26
4.12
1.68
2.31
Nate: Parameter estimates for the site choice model were obtained using
QUAIL Version 3.5.
Source: Model developed and run by Clark Binkley, Yale university, School of
Forestry and Environmental Sciences.
-------
6-32
levels can be accompanied by an increase in two way travel distance of 0.254
miles and leave the individual's utility level unchanged. In the Feenberg-
Mills model evaluated at the mean value of all the interaction terms/ a 10
percent reduction in all water quality variables (total bacteria, oil, color)
offsets an 0.5 mile increase in travel distance. It is curious that we find
a negative value for the fresh water dummy variable, indicating Bostonians
prefer salt water to fresh water beaches, where feenberg and Mills report a
positive value. In sum, our model, using better travel cost and water
quality data for a larger sample of individuals, seems to be more sensitive
to water quality and less sensitive to distance than is the Feenberg-Mills
model.
6.2.2.4 Benefit Estimates
The model presented above can be used to obtain estimates of the benefits
of water quality improvement. Recall that the benefit measure we use is the
equivalent variation measured in units of distance. These units are
converted to units of dollars at the rate of $0.12/tnile for travel costs plus
an amount which reflects the time cost of travel: travel time valued at
one-third the individual's wage rate. Travel time was determined from the
CTPS study mentioned above. The wage rate was computed from information on
income and the number of days wr rked per year as was described above. These
per mile figures were doubled to reflect the fact that the demand model was
estimated on one-way rather than two-way distance.
Four sets of simulations were performed. In each case the fecal coliform
level at a single beach or group of beaches in the Boston Harbor Study area
was decreased in increments of 10 percent up to a 90 percent improvement in
water quality. These simulations map out the total benefit curve for water
-------
6-33
pollution abatement in the various segments of Boston Harbor. Sites 7
(Constitution Beach) and 15 (Wollaston Beach) were examined separately
because of their importance to Boston Harbor-based recreation and because of
their location within the harbor. Sites 8 - 14, the beaches in the
torchester/Neponset Bay CSO planning areas, (south from Castle Island to
Tenean Beach), were treated as a group in a third simulation. Finally, a
simulation including all of the sites 7-15 was performed, with 10 percent
less water quality improvement at site 7 than the others. This simulation
shows the effect of a full water pollution abatement program for the Boston
Harbor Study area.'
The summary results from these simulations are given in Table 6-5. The
entries in the table are benefits per person per year and the corresponding
» _
change in visits per person for a given pollution reduction. Thus, to get a
value per visitor day for the site the per capita benefit is divided by the
change in per capita visits. The value per visitor day for the different '
sites and pollution reduction levels ranges from $5.60 to $5.70 (in 1974
dollars) and is within the range of user-day values found in the literature
(see Table B-3, Appendix B). Total benefits rise steadily with increasing
levels of water quality improvement, and the curve continues to climb even as
high levels of abatement are achieved. This results in a marginal benefit
curve which curves upward rather than down wan. as is commonly assumed. This
unusual result might stem from the fact that the demand model was estimated
using data from beaches generally having water quality levels much less than
the 80 to 90 percent levels imply.
Table 6-6 summarizes the change in per capita visits for each of the
control options. Then, the increases in number of visits are derived by
multiplying change in per capita visits by the 1980 Census Boston
-------
6-34
Table 6-5. Per Capita Ainual Benefit Estimates from
Conditional Logit Model S/
($19 7 4/capita/year)
Value per visitor day£/
Equivalent 1982 dollars
% Reduction in Water
Pollution
10
20
30
40
50
60
70
80
90
1
1 7
Constitution
5.62
11.00
.0340
(.006054)
.0687
(.01222)
.1040
(.01851)
.1400
(.02491)
.1766
( .0 314 3)
.2140
(.03807)
.2521
(.04481)
.2908
(.05174)
—
SITES
1 8-14 |
Dorchester
5.62
11.00
.1562
(.02779)
.3240
(.05765)
.5055
(.08995)
.7030
(.1251)
.9192
(.1636)
1.158
(.2060)
1.422
(.2530)
1.718
(.3056)
2.050
(.3646)
15 I
Wollaston
5.69
11.14
.1176
(.02069)
.2469
(.0434)
.3889
(.06837)
.5446
(.09575)
.7155
(.1258)
.9027
(.1587)
1.108
(.1947)
1.332
(.2342)
1.577
(.2773)
7-15 b/
All
5.65
11.06
.2731
( .04835)
.6014
(.1065)
.9539
(.1689)
1.334
(.2361)
1.744
(.3087)
2.189
(.3874)
2.672
(.4729)
3.199
(.5661)
3.774
(.6680)
*/ Change in per capita visits for given change in pollution is in
parentheses.
£/ Reduction at site 7 is 10 percent less than reduction at site 8-15 (i.e.,
the first entry is a 10 percent reduction at 8-15 and no reduction at 7) .
—' Calculated by dividing $/capita/year by change in per capita visits and
averaged over all percent pollution reduction simulations.
Nate: for location of sites see map (Figure 6-1) .
-------
6-35
Table 6-6. increased Participation Estimates from Conditional log it Model
Site
No.
Beach
Ocean Secondary
CSO Outfall Treatment
CSO plus
CSO plus Secondary
Ocean Outfall Treatment
7 Constitution
8-14 Dorchester/
Neponset
15 Wollaston
7-15 All Sites
7 Constitution
8-14 Dorchester/
Neponset
15 Wollaston
7-15 All Sites
70
Percent Pollution Abatement §/
10 5 80
80
80
70/80
10
10
10
10 90
10 90
5/10 80/90
Increase in Per Capita Visits
.0448
.3056
.2342
.5661
.0278 .0278
.0207 .0207
.0484 .0484
.3646
.2773
.6680
75
90
90
75/90
.048
.3646
.2773
.6171
Lower Bound Increase in Number of Visits £/
7 Constitution
8-14 Dorchester/
Neponset
15 Wollaston
7-15 All Sites
82,821 11,277 5,546
95,577
88,737
564,958
432,962
1,046,541
51,393 51,393
38,267 38,267
89,477 89,477
674,031
512,641
1,234,922
7 Constitution 123,798
8-14 Dorchester/
Neponset 844,482
15 Wollaston 647,178
7-15 All Sites 1,564,336
674,031
512,641
1,140,824
Upper Bound Increase in Number of Visits
-------
6-36
SMSA population. The value of increased visits can be calculated by
multiplying increased visits by the consumer surplus per visit. These are
presented in Table 6-7. Nat surprisingly/ these annual benefits are high.
This is the result of both the large number of beach users and the large
estimated percentage reduction in pollution.
6.2.2.5 Limits of Analysis
The principle theoretical shortcoming of this conditional logit approach
is the link between objective water quality parameters and the subjective
perception by recreationists of water quality. Some water quality parameters
(e.g., dissolved oxygen) are not easily perceived by recreationists and/
consequently, an improvement in water quality (i.e., an increase in DO levels
in the water) may not be perceived by recreationists, and their behavior
(i.e., frequency of visits to the site) may not change.
This is not likely to be the case for the beaches in the Boston Harbor
study area. Dornbusch's study (1975) found that floating debris and oil and
grease were the most frequently perceived water quality indicators applicable
to the logit, travel cost model as applied here. The presence of high fecal
coliform counts, the water quality parameter used in this study, is highly
correlated to oil and grease measures (Hanemann, 1978), parameters which are
perceived by recreationists. Thus, the concern that recreation behavior is
governed by perception and, ideally, the predicted changes in behavior be
linked to water quality parameters that are perceived by recreationists has
been addressed in this application of the logit model by using fecal
coliform, instead of dissolved oxygen, as the water quality variable.
-------
6-37
Table 6-7. Annual Benefit Estimates from
Conditional Logit Model (1982 $000)
Site
No.
7
8-14
15
7-15
7
8-14
15
7-15
Beach
Constitution
Dorchester/
Neponset
Wollaston
All Sites
Constitution
Dorchester/
Neponset
Wollaston
All Sites
CSO
911
6,214
4,823
11,574
1,361
9,289
7,209
17,301
.0
.5
.2
.7
.8
.3
.6
.6
Ocean Secondary CSO plus
Outfall Treatment Ocean Outfall
124
565
426
989
185
845
637
1,479
LOWER
.0
.3
.3
.6
UPPER
.4
.0
.2
.2 1
BOUND
61.0
565.3
426.3
989.6
BOUND
91.2
845.0
637.2
,479.2
ESTIMATES
1.
7,
5,
13,
ESTIMATES
1,
11,
8,
20,
051.
414.
710.
658.
571.
082.
536.
451.
3
3
8
2
2
7
3
9
CSO plus
Secondary
Treatment
976
7,414
5,710
12,617
1,459
11,082
8,536
18,860
.1
.3
.8
.5
.1
.7
.3
.3
Source: Derived by multiplying $1982 value per visitor day from Table 6-6
by increase in number of visits.
-------
6-38
An additional shortcoming of this conditional logit approach is the use
of travel costs to simulate prices. Travel costs may be difficult to specify
because travel may have a special utility or disutility in itself/ based on
aesthetics of the travel route and travel time, in addition to travel costs.
Hie improved water quality data/ the incorporation of travel
time, and the estimation of travel distance, and the estimation of consumer
surplus, however, make the logit model the most empirically and theoretically
sound of all the methodologies used to estimate swimming benefits from
improving water quality in Boston terbor.
Despite these limitations, the benefit estimates resulting from the logit
model are instructive in two ways. The difference in the estimates of
increase in demand as measured by user days using the logit technique (Table
6-6) as opposed to the increased participation technique (Table 6-2) depend
on the treatment option and the beach location. The logit model predicts
greater participation for the STP options (ocean outfall and secondary
treatment) and less participation under the CSO and CSO and STP combined
options than does the increased participation approach, for the individual
beaches the logit model predicts greater participation at Dorchester/Nsponset
and Constitution while less participation at Wollaston. The predicted
increased days for the logit model are within the bounds of seasonal capacity
as estimated above (see Section 6.1.1) . The o<
-------
6-39
6.2.3 Swimming--Beach Closings
An alternative method for calculating swimming benefits from increased
participation because of improved water quality is to determine the value of
lost participation if beaches are closed because of fecal contamination.
Essentially, this technique estimates the dollar value of the number of daily
beach closings by multiplying the average consumer surplus per daytrip (in
dollars per unit) by the daily attendance at each beach and by the number of
daily beach closings due to water pollution.
The information needed to calculate these benefits using this technique is
usually more readily available than detailed information required for benefit
estimation using the previously described increased participaton technique,
and thus this method has often formed the basis for calculating total swimming
benefits. In the case of Boston ffcrbor beaches, different health standards
are applied according to beach ownership. The MDC does not actually close
beaches when fecal coliform measures are high enough to represent a health
hazard, but they do post signs that the beaches are unsafe for swimming.
Signs are posted at an MDC beach when fecal coliform counts exceed 500 MPN/100
ml. A few towns use a standard of 1,000 MPN/100 ml total coliform. Federal
standards are the most strict, suggesting closure when fecal coliform counts
exceed 200 MPN/100 ml.
The first step in this technique is to decide which health standard to
apply. We have chosen tc use the strict federal standard of 200 MPN/100 ml to
establish an upper bound and the MDC standard of 500 MPN/100 ml as a lower
bound. We did not choose the 1,000 MPN/100 ml as a lower bound because few of
the affected town beaches use this level, and there are few times during the
-------
6-40
season when coliform concentrations reach this high a level. We have also
assumed that there is limited or no attendance at the beaches during the days
when fecal coliform counts exceed the 200 MPN/100 ml and 500 MPN/100 ml levels.
The next step is to relate bacteriological contamination with daily
attendance figures so that we can arrive at a number of lost recreation days.
Unfortunately, as previously described, the only attendance figures available
are seasona1 (Memorial Day to labor Day) data, making it difficult to assess
the exact number of swimmers affected by daily beach closings. There is also
the added complication that weekend attendance at beaches is usually greater
than weekday attendance and, therefore, weekend violations of water quality
standards have a greater impact on potential losses than weekday violations.
Data limitations prevented us from considering this effect. Instead we have
assumed a direct proportional relationship between total seasonal attendance
figures and percentage of times during the season that water quality levels
exceed 200 MPN/100 ml and 500 MPN/100 ml. for example, if a beach has water
quality levels which exceed 200 MPN/100 ml during five percent of the season,
then we assume that five percent of total attendance will be affected and will
not go to the beach (see Appendix B.5 for details) . This assumption probably
understates the case since water quality problems tend to be the worst during
the hottest times of the year, when beach attendance is the highest.
6.2.3.1 Boston Harbor Beaches
In order to arrive at savings according to the CSO and STP options, it is
necessary to multiply these base visits by the predicted percent cleanup.
These base-case lost visits and their corresponding averted lost visits due
to pollution control programs are presented at the top of Tables 6-8
-------
6-41
and 6-9. The final step in this methodology is to value these averted lost
attendance days by applying a range of user-day dollar values. These values
represent the savings due to averted beach closings due to pollution
abatement in Boston Harbor and are presented at the bottom of Tables 6-8 and
6-9.
6.2.3.2 Nantasket Beach
The only other swimming beach in our study area is Nantasket Beach. It
is expected to be adversely affected by the deep ocean outfall option (see
Table 4-3) . We have used only the beach closing method to estimate the
effects on swimming at Nantasket Beach because of the limitations of
available data and methodology for measuring effects of increases in
pollutant levels.
Seasonal population at Nantasket Beach is estimated to be 3,035,000,
based on information from Binkley and ffinemann (1975) and the MDC.
Currently, Nantasket Beach has water quality levels which exceed 200 MPN/100
ml approximately 2.3 percent of the season. Water quality is expected to
decrease by 10 percent from current levels if a deep ocean outfall is
constructed. It is difficult to predict the relationship between this
percentage decrease in water quality and the corresponding percentage changes
in pollutant concentrations exceeding 200 MPN/100 ml and 500 MPN/100 ml. We
have chosen to conservatively assume that the water quality level at
Nantasket will exceed 500 MPN/100 ml at least as frequently as it was
exceeded at the 200 MPN/100 ml level, or 2.3 percent of the season. By
multiplying the seasonal attendance estimates by this percentage, we arrive
at a number of lost visits totalling 69,805. These lost visits can be valued
by applying a range of user day values from $1.60 to $11.06. Thus, we arrive
-------
6-42
Table 6-8. Ainu a 1 Benefit of Averted Beach Closings
at 200 MPN/lOOml (1982 $000)
Beach Number of
CSO
Ocean
Outfall
CSO plus CSO plus
Secondary Ocean Secondary
Treatment Outfall Treatment
Averted Lost Visits &/
Lost Visits a/
Constitution
Dorchester
Castle Island
Pleasure Bay
Carson
Malibu
Tenean
Wollaston
Quincy
Weymouth
Hingham
Hull
TOTAL
User Day
Value
3 1.60
$ 5.80
$11.06
29,019
1,010
11,779
6,604
14,423
41,519
518,870
13,687
11,966
—
3,505
652,382
20,313
808
9,423
5,283
11,539
33,215
415,096
10,950
-
-
—
506,627
2,902
101
1,178
660
1,442
4,152
51,887
1,369
3,590
-
1,052
68,333
Annual Benefit of
for All
810.6
2,938.4
5,603.3
1,451
101
1,178
660
1,442
4,152
51,887
1,369
3,590
-
1,052
66,882
23,215
909
10,601
5,943
12,981
37,367
466,983
12,319
3,590
-
1,052
574,960
21,764
909
10,601
5,943
12,981
37,367
466,983
12,319
3,590
-
1,052
573,54
Averted Beach Closings £/
Boston Harbor Beaches
109.3
396.3
755.7
107.0
387.9
739.7
(1982 $000)
876.7
3,178.2
6,060.4
860.0
3,117.5
5,994.8
a/ See Appendix B.5.
b/ Number of lost visits multiplied by percent pollution abatement (in Table
4-3) .
£/ Total averted lost visits multiplied by user day value (in Table B-3,
Appendix B).
-------
6-43
T&ble 6-9. Annual Benefit of Averted Beach Closings
at 500 MPN/lOOml (1982 $000)
Beach
Constitution
Dorchester
Castle Island
Pleasure Bay
Carson
Malibu
Tenean
Wollaston
Quincy
Weymouth
Hi ngham
Hull
TOTAL
User Day
Value
3 1.60
$ 5.80
$11.06
Number of
Lost Visits a/
11,606
433
5,049
4,714
4,328
24,107
259,435
6,537
-
-
3,505
319,714
CSO
8,124
346
4,039
3,771
3,462
19,286
207,548
5,230
-
-
—
251,806
Ocean Secondary
Outfall Treatment
Averted Lost Visits
1,161 580
43 43
505 505
471 471
433 433
2,411 2,411
25,944 25,944
654 654
-
-
1,052 1,052
32,674 32,093
Annual Benefit of Averted Beach
for All
402.9
1,460.5
2,785.0
CSO plus
Ocean
Outfall
b/
9,285
389
4,544
4,242
3,895
21,697
233,492
5,884
-
-
1,052
284,480
Closings £/
CSO plus
Secondary
Treatment
8,704
389
4,544
4,242
3,895
21,697
233,492
5,844
—
-
1,052
283,899
Boston Harbor Beaches (1982 $000)
52.3 51.3
189.5 186.1
361.4 354.9
455.2
1,650.0
3,146.3
454.2
1,646.6
3,139.9
§/ See Appendix B.5.
b/ Number of lost visits multiplied by percent pollution abatement (in Table 4-3).
£/ Total averted lost visits multiplied by user day value (in Table B-3, Appendix
B).
-------
6-44
at a range of 3111,688 to $772,043, which represents a conservative estimate
of swimming-related pollution costs at Nantasket Beach attributable to
implementation of the deep ocean outfall option.
6.2.3.3 Benefit Estimates
It is clear that the greatest benefits will derive from cleaning up the
Dorchester Bay and Wollaston Beaches because these are the areas with the
greatest and most frequent water quality violations, and they are the most
popular beaches. lenean and Wollaston Beaches, especially, have the greatest
number of averted lost visits. Based on the strict 200 MPN/100 ml standard,
Wollaston has nearly 520,000 lost visits while Tfenean has over 41,500.
Benefits to the STP-affected beaches of Weymouth, Hingham and Hull
are extremely low for both the upper bound and lower bound case for a number
of reasons. These include the fairly good quality of shoreline water, the
fact that the STP pollution control programs are expected to reduce fecal
coliform concentration and, thus, reduce beach closings, by only 30 percent,
and the fact that attendance is low at these beaches.
6.2.3.4 Limits of Analysis
These dollar benefits are significantly lower than the values calculated
fo.. swimming benefits using the increased participation methodology,
previously described. The reasons for this difference are many and only
serve to emphasize the many limitations and shortcomings of using this
methodology to estimate recreation benefits. Normally, beach closings are
calculated by relating the intensity of rain events to CSO overflow and the
corresponding effect on ambient water quality and beach attendance. This
methodology was not utilized, however, because of data limitations and
-------
6-45
because a substantial portion of ambient water quality problems in beach
areas in Boston Harbor stems from problems with dry weather overflow (DWO).
The beach closing methodology attempts to capture the general seasonal
relationship between CSO/DWO events and beach participation based on seasonal
averages of ambient water quality and estimates of seasonal beach
attendance. It underestimates total swimming benefits because it cannot
capture the dollar value of increased number of visits due to cleaner and
more attractive beaches, nor can it capture the increase in willingness to
pay for safer and cleaner bathing areas. In addition, these estimates for
Boston Harbor are based on the assumption that there is a direct correlation
between percent fecal contamination and percent beach closings. In reality,
this relationship may not be directly proportional and, in fact, there may
not be a significant relationship between the two parameters. We can only
conclude that this methodology seriously underestimates swimming-related
benefits, and that this range of values is a less appropriate measure of
water pollution abatement benefits than values derived from previously
described techniques.
6.3 Recreational Boating
One of the significant consumer surplus benefits associated with water
pollution abatement in Boston Harbor is the increased use and utility of
harbor waters by boaters, and the savings in dollars spent on these
activities. Uifortunately, unlike the previously described swimming-related
benefits, there is little available information upon which to base these
benefits, instead, we make only very general estimates of consumer surplus
using a number of assumptions about increased participation and the
corresponding value of these increases and applying aggregated information
from regional and federal recreation studies.
-------
6-46
6.3.1 Increased Participation
It has been well documented that improved water quality can have an
important effect on the level of recreational boating (Davidson, Adams and
Seneca, 1966). Participation in all boating activities in Boston Harbor—
sailing, motor boating, canoeing and windsurfing--is expected to increase
with corresponding decreases in water pollutant levels. Benefits from this
improvement stem from an increase in frequency of participation by previous
users, willingness to pay a higher price for the boating experience because
of improved water quality, and new participation by previous non-users. Much
of this increased participation is likely to come from increases in the
aesthetic boating experience due to the decreased offensiveness of presently
polluted areas, especially those areas directly surrounding the sewage
treatment plants and near CSO outfalls. Improvements to CSOs in Dorchester
Bay and the Deer and Nut Island STPs will most definitely improve water
quality and, thereby, encourage increased recreational boating in these
areas. Unfortunately, there are few boating participation studies which link
a change in water quality to a change in boater use of water resources which
are applicable to Boston Harbor and, thus, recreation participation data on
present use, along with data on unmet demand, was used to estimate boating
benefits from improvements in water quality.
We have used a benefit estimation methodology which is similar to the
increased participation technique described for swimming related benefits.
Using data from a variety of recreational sources we have estimated the
number of user days per year for two categories of boating—motor boating and
sailing. Although there are no quantitative measures of predicted percentage
increases in boating that are expected to occur under the various CSO and STP
-------
6-47
options, we can estimate this increased participation by assuming that
cleaner waters will supply a portion of unmet (latent) demand. Two of the
recreational studies have estimated a 45-69 percent unmet demand in the
Boston Metropolitan area for boating. This translates into a need of 1.8 to
2.8 million days for motor boating and 0.8 to 1.3 million days for sailing.
We can assume that some of this demand will be met by cleaning up harbor
waters, although it is not immediately clear what percentage will actually be
met. Because fishing and boating take place throughout the harbor and are
not restricted to certain areas we have calculated these benefits on a
harbor-wide basis for the two combined options, CSO plus Ocean Outfall and
CSO plus Secondary Treatment. We have assumed that abating pollution from
CSO and Ocean Outfall controls will lead to a 2 to 10 percent reduction in
unmet demand. We assumed the CSO plus Secondary Treatment option would meet
5 to 12 percent of unmet demand. The lower figures for the deep ocean
outfall option reflect the adverse impact this option is expected to have on
the area around the Brewsters Islands.
Although these figures might appear to be overly conservative, we have
chosen them for two reasons. First, we believe that the latent demand of
45-69 percent reported in the recreational studies is probably an
overestimate (and have chosen to use 50% in ' ur calculations) . Second, even
though more boaters might increase their use of Boston ffirbor when pollution
is decreased, there is a limited supply of available marinas, boatyards and
docks. Thus, for every ten new boaters who might want to use the harbor,
only one might actually be able to because of limited facilities. In other
words, we have assumed that the binding constraint on increases in boating
-------
6-48
participation is not only poor water quality, but the supply of boating
facilities as well. This has been demonstrated by Davidson et al. (1966) who
determined that the supply of beatable water is affected by the depth, width,
access, and quality of a water resource. In this study the upper bound
benefit estimate is determined by the facility availability constraint.
6.3.2 Benefits Estimates
Using these assumed recreational figures, it is possible to calculate the
number of increased boating days. By applying a lower bound user day value
of $18.14 and an upper bound value of $45.19 (see Table B-3, Appendix B) to
the range of increased boating days, we arrive at the estimated value of
benefits for boating activities (see Table 6-10) .
6.3.3 Limits of Analysis
Calculation of boating-related benefits is limited by both methodology
and data base. Statistics on use and participation were inconsistent among
all sources, requiring us to judge which statistics were the most appropriate
for a given step in the estimation process. There was scant information on
latent demand, requiring us to use a possibly overstated estimate from a
Boston-based study. Benefit estimation was further compromised by having to
assume what percentage of latent demand was met by cleanin'- up harbor waters,
a prediction based on professional judgment rather than quantitative
information. All of these shortcomings are reflected in the final benefit
values. In addition, this benefit methodology does not capture total consumer
surplus in that only the benefits of water quality improvement to new
participants, and not increased utility and increased participation of
-------
6-49
Table 6-10. Annual Saltwater Boating Benefits
(1982 $)
Motor
Boating Sailing Total
LATENT DEMAND
% Of SMSA 22 15
ft of recreators 607,938 414,504 1,022,442
User Days per Participant 6.7 4.5
# of User Days 4,073,185 1,865,268 5,938,453
Latent Demand (50%) 2,036,593 932,634 2,969,226
LOWER BOUND ESTIMATES
% latent Demand met by
CSOs and Ocean Outfall 222
CSOs and Secondary Treatment 5 55
Days of Latent Demand met by
CSOs and Ocean Outfall 40,732 18,653 59,385
CSOs and Secondary Treatment 101,830 46,632 148,462
Annual Benefits (User Day Value = $18.14g/)
CSOs and Ocean Outfall 3,694,000 1,692,000 5,386,000
CSOs and Secondary Treatment 4,433,000 2,030,000 6,463,000
UPPER BOUND ESTIMATES
% latent Demand met by
CSOs and Ocean Outfall 10 10 10
CSOs and Secondary Treatment 12 12 12
Days of Latent Demand met by
CSCs and Ocean Outfall 203,659 93,263 296,922
CSOs and Secondary 244,391 111,916 356,307
Annual Benefits (User Day Value = $40.89a/)
CSOs and Ocean Outfall 8,328,000 3,814,000 12,129,000
CSOs and Secondary Treatment 9,993,000 4,576,000 14,569,000
£/See Table B-3, Appendix B.
-------
6-50
previous users, is measured. These benefit values also understate total
boating-related benefits because other boating activities, such as canoeing
and windsurfing, have not been considered and because reductions in the
amount of fouling of boats and equipment have not been considered. Finally,
although boating benefits are substantial when estimated for the entire study
area in the Boston Harbor, data limitations prevented disaggregating these
benefits to the level of the areas specifically affected by the pollution
abatement options. Thus, these benefit estimates can only be used to
emphasize the relative importance of the effect of improved water quality on
recreational boating and to underscore the conclusion that these effects are
both monetizable and significant.
6.4 Recreational Fishing
The benefits to recreational fishing of improving water quality in Boston
Harbor has two components. First, cleaner water will affect the availability
of fish, both species and numbers. Second, this change in fish availability
will affect fishing participation rates. In addition, there may be a
"perception" effect on fishing activity which is independent of this
availability, implying a more positive response towards fishing in cleaner
water.—' The consumer surplus from improving water quality should, thus,
be measured by calculating increases in participation stemming from changes
in fish species and numbers and the increased utility or willingness to pay a
higher price to fish in cleaner water.
i/ An informal survey by Metcalf and Eddy (1982) reported that, although
in general it did not appear that fishers avoided discharge areas, one bait
shop owner had reported that the Nut Island discharge made the area
unattractive for his clients. In another, larger, survey conducted by the
Massachusetts Division of Marine Fisheries (1982), concern was expressed over
the effects of pollution by toxic chemicals and sewage waste (65-60 percent
felt these were serious problems) , loss of fish habitat (57 percent),
adequate stocks of fish to catch (43 percent).
-------
6-51
6.4.1 Components of Recreational Fishing
Calculating this fishing-related consumer surplus is difficult however,
because it involves assessing the technical effects/impacts of the pollution
control actions, including changes in ecological habitat, as well as
determining the behavioral effect of these actions. These steps are
summarized in Figure 6-2.
A number of studies have attempted to model and analyze the effects and
responses of fish and anglers to changes in water quality from pollution
control programs. Bell and Canterbery (1976) modeled biological production
functions of important recreational fish and applied them to recreational
fisheries data to arrive at estimates of recreational fishing benefits for
each state in the Uiion. We have chosen not to apply their results to the
study area because of methodological and data limitations. One other study
(Russell and vaughan, 1982) developed a model to estimate the probability of
being an angler, the probability of spending time to fish, and the average
length of time for each type of fishing. Their model estimates the effects of
water quality changes on number of fishing sites, types of supportable fish
population, and change in aesthetic experience. This model can only be used
for freshwater fishing areas and, thus, cannot be applied to the Boston Harbor
Study area.
It was not possible to calculate many of the effects and responses listed
in Figure 6-2, which is a prerequisite to calculating measures of consumer
surplus. It was particularly difficult to determine how pollution control
plan effluents would precisely affect or change the ecological habitats of
important recreational fish. The preferred summer recreational fish in the
harbor is winter flounder (Pseudopleuronectes americanus) although other
-------
6-52
Figure 6-2. Effects and Responses to STP, CSO and
Sewer Controls
STP, CSO
Storm Sewer
Treatments
STP Treatment
Storm Sewer Controls
CSO Controls
Technical Effects
of STP, CSO and
Storm Sewer Treatments
Changes in Effluents
Changes in Water Quality
Change in Ecological
Habitat
Effects on Economic
Agents
Behavioral Effects
of Water Quality
Standards Action(s)
-c
Behavioral Responses
of Economic Agents
-------
6-53
desirable species include striped bass (Morone saxatilis) , bluefish
(Pomatomus saltatrix), and cod (Gadus morhua). Winter flounder appears to be
the only species definitely affected by Harbor pollution, preferring the more
organically polluted areas to the cleaner ones. Despite this attraction to
polluted areas it was not possible to link these changes with the specific~
pollution control options. In general, productivity throughout the Boston
Harbor Study area is expected to increase with corresponding decreases in
water pollutants, although we were not able to quantitatively determine the
increase in productivity. These data limitations required us to apply a
general participation approach to estimate fishing benefits, similar to the
method previously described under boating benefits.
/
Recreation studies provided information on percentage participation,
value of user days, and total user days per year for marine fishing. We were
unable to find direct, reliable figures on latent demand and, thus, we
assumed a rate identical to that used for boating. We applied a user day
value of from $12.90 to $28.46 per user-day derived from a number of studies '>
presented in Table B-3, Appendix B. The results are presented in Table 6-11.
/
6.4.2 Benefits Estimates
Fishing benefits can only be estimated for the entire Boston Harbor Study
area, rather than for each distinct geographical area. The possibility of
double counting some boaters who primarily fish from their boats exists.
However, no information was available to suggest how prevalent this kind of
behavior might be. For this reason, these benefit figures should be
interpreted with caution.
-------
6-54
•able 6-11.
Ainual Recreational Fishing Benefits
(1982 $)
lower
Bound
Upper
Bound
latent Demand
% of SMSA
# of recreators
User Days per Participant
# of User Days
Latent Demand (50%)
% of latent Demand met by
CSOs and Ocean Outfall
CSOs and Secondary Treatment
Days of Latent Demand met by
CSOs and Ocean Outfall
CSOs and Secondary Treatment
User Day Value */
Annual Benefit
CSOs and Ocean Outfall
CSOs and Secondary Treatment
7
193,435
12
2,321,220
1,160,610
2
5
23,212
58,030
$12.89
299,000
749,000
14
386,810
12
4,642,440
2,321,220
10
12
232,122
278,546
$34.08
7,911,000
9,493,000
a/ See Table B-3, Appendix B
-------
6-55
6.4.3 Limits of Analysis
Estimation of recreational fishing benefits is limited by methodology and
data base in ways similar to those described under boating benefits. A major
limitation of this analysis is the lack of information linking changes in
water quality to corresponding changes in both biological habit and fish
population. This lack of data prevented a precise estimation of the effects
of availability and number of fish species on fishing participation. Another
problem was that the available recreation fishing statistics on participation
and unmet demand were often inconsistent, requiring us to judge which were
the most appropriate for a given step in the estimation process. Another
limitation of the analysis is that the methodology used here does not capture
all components of consumer surplus. Benefit values reflect only benefits to
new participants, and not the value of increased utility or increase in
participation by previous users. The last limitation of this analysis is the
possibility of some double counting of fishing and boating benefits. Thus,
these estimates can only be used to emphasize the importance of the effect of
improved water quality on recreational fishing.
6.5 Boston Harbor Islands
The Boston Harbor Islands are a unique natural resource in a metropolitan
area which possesses only lalf of the recommended minimum acreage of open
space per thousand population. The Islands are predominantly open, natural
areas which offer a wide range of activities such as swimming, boating,
fishing, hiking, picknicking, camping and historic sight-seeing. Most of the
islands have limited recreational facilities, which restrict current and
potential visits. However, effluent from the two sewage treatment plants
-------
6-56
seriously degrades water quality around the islands, also discouraging
recreation. Assuming that the planned recreational facilites were
constructed, then improving water quality around the islands would lead to a
corresponding increase in both frequency of participation and total number of
visitors. It is possible to roughly estimate this increased participation,
despite scarce recreational data.
6.5.1 Increased participation
Recreational data from the Boston Harbor Islands Comprehensive Plan,
(Metropolitan Planning Council, 1972) suggests that current attendance at all
the Islands for all recreational activities is 265,000 per season and that
total capacity, assuming the planned structural improvements and additions
are implemented, is 560,000 per season. This results in an excess supply of
295,000 visits per season. Given the unique nature of the Harbor Islands, we
have assumed that some of the latent demand for recreation in the
harbor--especially swimming, boating and fishing--could be met largely by
improving water quality around the Islands. Implementation of either of the
STP options is expected to improve the water quality around the nearest
Harbor Islands. However, implementation of the deep ocean outfall option is
expected to have adverse effects on the Brewsters Islands, which are the
outermost islands of Boston Harbor. Thf Brewsters include Great Brewster,
Middle Brewster, Outer Brewster, Calf, Little Calf and Green Islands, Shag
Rocks, and the Graves. These islands constitute one of the most unique
marine environments on the Massachusetts coast, providing a highly accessible
marine habitat, conservation areas, and excellent sites for recreational
diving. Water quality is expected to decrease by 10 to 15 percent in the
area surrounding these islands because they are so close to the ocean outfall
-------
6-57
diffuser. Consequently, many of the recreational activities such as diving,
swimming, boating and hiking will be affected by this degradation of water
quality.
To develop benefit estimates for recreational activities at the Harbor
islands we have assumed a percentage increase (decrease) in visits and
applied a range of previously utilized user day values. The assumptions and
calculations of these benefit values are presented in Table 6-12.
6.5.2 Limits of Analysis
The previously described methodology is limited by both its data bases
and its assumptions. There is little available information on latent demand
for the Boston Harbor Islands and, thus, we had to assume an upper and lower
bound participation rate. Although there are accurate estimates for current
Harbor Island attendance, capacity estimates should be interpreted and used
with caution. The derived benefit estimates probably underestimate
STP-related benefits for the Islands because the applied methodology cannot,
theoretically, capture either the dollar value of increased utility or the
value of increases in frequency of participation. These benefit values
should also be viewed as rough estimates because of the possibility of
double-counting from other benefit categories such as boating and fishing for
the entire harbor and because costs of upgrading recreational facilities,
which are a necesary prerequisite to increased participation, have not been
included.
6.6 Summary of Recreation Benefits
Reducing water pollution in the Boston Harbor Study area by implementing
the different pollution control options will result in many recreation
-------
6-58
T&ble 6-12. Annual Benefits for Recreation on Boston ffirbor Islands
(1982 $000)
Oar rent Attendance
Capacity
Excess Supply (latent demand)
% Change in Water Quality
Ocean Outfall Option
Secondary Treatment Option
% of Latent Demand met by
Ocean Outfall Option
Secondary Treatment Option
Outer Harbor
Islands
258,000
546,000
288,000
60 to 90
30 to 80
50 to 90
50 to 75
Brews ters
Islands
7,000
14,000
7,000
-10 to -15
30 to 40
-20 to -30
50 to 75
Change in Visitor Days due to §/
Ocean Outfall Option
Secondary Treatment Option
User Day Values &/
Annual Benefit values (1982 $000)
Ocean Outfall Option
Secondary Treatment Option
144,000 to 259,200 -1,400 to -2,100
144,000 to 216,000 3,500 to 5,250
$5.80 to $11.06 $5.80 to $11.06
835 to 2,867
835 to 2,389
-8.1 to -23.2
20.3 to 58.1
§/ Change in Visitor Days calculated by multiplying latent demand by the
percentage of latent demand met by the different treatment options.
b/ See Table B-3, Appendix B.
-------
6-59
benefits (see Table 6-13) . A variety of methodologies have been used to
calculate the range of these benefits. These include: (1) swimming—
increased participation; (2) swimming—travel cost with conditional logit
model; (3) swimming—beach closings; (4) boating and fishing—increased
participation; (5) all recreation activities for Boston Harbor
Islands—increased participation.
Recreation benefits as calculated by the travel cost method, are greatest
in the category of swimming. Benefits associated with the CSO options are
substantial while STP-related swimming benefits are minor, because the
majority of swimming in the harbor study area takes place along shorelines,
which are not as adversely affected by STPs. Fishing and boating benefits
have been calculated only for the entire harbor and not for each treatment
alternative, because of data limitations. Benefits for both these categories
are also substantial while the greatest STP-related recreation benefits are
from water quality improvements near the Boston Harbor Islands.
-------
6-60
Table 6-13. Ainual Recreation Benefits
(Thousands of 19823)
Benefit
A.
SWIMMING
1. Increased participation
a. Recreation studies £
High:
low:
Moderate:
2. Increased Participation
a. Log it model: — '
High:
CSO
/
16,737
1,620
7,325
Ocean
Outfall
2,436
236
1,066
Secondary
Treatment
2,347
227
1,027
CSO plus
Ocean
Outfall
19,174
1,856
8,391
CSO plus
Secondary
Treatment
19,084
1,847
8,352
and Increased Utility of Visit
17,302
low: 11,575
B.
C.
Moderate:
3. Beach Closings
a. Strict £/ 200 MPN
High:
low:
Moderate:
b. Lenient £/ 500 MPN
High:
low:
Moderate:
c. Nantasket Beach £/
High:
low:
Moderate:
14,439
f .c.
5,603
811
2,938
f .c.
2,785
403
1,461
0
0
0
1,479
990
1,235
756
109
396
351
52
189
(772)
(112)
(405)
1,479
990
1,235
740
107
388
355
51
186
0
0
0
20,416
13,658
17,037
6,060
877
3,178
3,146
455
1,650
(772)
(112)
(405)
18,860
12,618
15,739
5,945
860
3,118
3,140
454
1,647
0
0
0
increased Participation
High:
low:
Moderate:
FISHING 2/
NA
NA
NA
NA
NA
NA
NA
NA
NA
12,129
5,386
8,758
14,569
6,463
10,516
Increased Participation
D.
High:
low:
Moderate:
BOSTON HARBOR ISLANDS
NA
NA
NA
NA
NA
NA
NA
NA
NA
7,911
299
4,105
9,493
749
5,121
Increased Participation h/
High:
low:
Moderate:
0
0
0
2,844
827
1,835
2,447
855
1,651
2,844
827
1,835
2,447
855
1,651
2/ From Table 6-3.
!•>/ From Table 6-7, does not include.
Quincy town beaches.
£/ From Table 6-8.
£/ From Section 6.2.3.2;
costs not benefits.
±/ From Table 6-10.
£/ From Table 6-11.
-------
6-61
References
Abt Associates, 1979. New York-New England Recreational Demand Study, Vol. I
and II, Cambridge, MA
Bell, P.W. and E.R. Canterbery (1975) . An Assessment of the Economic
Benefits Which Will Accrue to Cbmmercial and Recreational Fisheries from
Incremental Improvements in the Quality of Coastal Waters, Florida State
University, Tallahassee, FL.
Ben-Akiva, J., 1973. The Structure of Passenger Travel Demand. MIT PhD.
dissertation, Department of Civil Engineering.
Binkley, Clark S. and W. Michael Hanemann, 1975. The Recreation Benefits
of Water Quality improvement. U.S. Environmental Protection Agency,
Washington, DC, NTIS PB257719.
Binkley, Clark S., 1977. Estimating Recreation Benefits: Critical Review
and Bibliography. CPL Exchange Bibliography 1219.
Burt, Oscar R. and David Brewer, 1971. Estimation of net social benefits
from outdoor recreation. Bsonometrica. pp 613-827 in Abel, Tihansky and
Walsh, 1976, National Benefits of Water Pollution Control, Draft U.S. EPA
Office of Research and Development, Washington, DC.
Cesario, F. J. , 1976. The Value of Time in Recreation Benefit Studies. Land
Economics, 52: 32-41.
Charbonneau, J. and J. Hay, 1978. "Determinants and Economic Values of
Hunting and Fishing." Paper presented at the 43rd north flnerican
Wildlife and Natural Resource Conference. Phoenix, Arizona.
Cicchetti, C., A. Fisher and V. K. Smith, 1976. An Econometric Evaluation of
a Generalized Consumer's Surplus Measure: the Mineral King Controversy.
Econometrica, 44: 1253-1276.
Clawson, M. and J. H. Knetsch, 1966. The Economics of Outdoor Recreation.
Johns Hopkins University Press: Baltimore.
Davidson, P., G. Adams and J. Seneca, 1966. The social value of water
recreational facilities from an improvement in water quality: The
Delaware Estuary, Water Research, Allen Rneese and Stephen C. Smith,
eds., Baltimore: Johns Hopkins university Press for Resources for the
Future, 1966.
Development Planning and Research Associates, Inc., 1976. National Benefits
of Achieving the 1977, 1983 and 1985 Water Quality Goals, U.S. EPA,
Office of Research and Development, Washington, DC.
Dornbusch, David M., 1975. The Impact of Water Quality improvements on
Residential Property Prices, National Commission on Water Quality,
Washington, DC.
-------
6-62
References
Ditton, R. and T. Goodale, 1972. Marine Recreation Uses of Green Bay; A
Study of Hainan Behavior and Attitude Patterns, Technical Report No. 17,
Sea Grant Program, university of Wisconsin, Madison.
Domenich, T. and Daniel Mcfadden, 1975. Urban Travel Demand. North Bslland,
Amsterdam.
Dwyer, John, John R. Kelly and Michael D. Bowes, 1977. Improved Procedures
for Valuation of the Contribution of Recreation to National Bconomic
Development.University of Illinois, Water Resources Center, Urbana, IL.
Bricson, Raymond, 1975. Valuation of Water Quality in Outdoor Recreation,
PhD. Dissertation, Department of Economics, Colorado State University,
fort Collins, 00.
Feenberg, Daniel and Edwin Mills, 1980. Measuring the Benefits of Water
Pollution abatement, academic Press, New »rk, NY.
Federal Register, Vol. 48, No. 48, March 10, 1983.
Banemann, W. M. , 1978. A Methodological and Bnpirical Study of the
Recreation Benefits from Water Quality Improvement. PhD dissertation,
Harvard University, Cambridge, MA
Heintz, H.T. , A. Hershaft, and G.C. Horak, 1976. National Damage of Air and
Water Pollution, U.S. Environmental Agency, Office of Research and
Development, Washington, DC.
Massachusetts Department of Environmental Management, December, 1976.
Massachusetts Outdoors; Statewide Comprehensive Outdoor Recreation Plan
(SCORP) , Boston, MA.
Massachusetts Division of Marine Fisheries, 1982. Massachusetts Marine
Fisheries Management Report, Boston, MA.
McFadden, D., 1973. Conditional Choice Logit Analysis of Qualitative Choice
Behavior in P. Zarembka, ed. Frontiers in Econometrics, Academic Press,
New York.
Metcalf and Eddy, 1975. Eastern Massaschusetts Metropolitan Area Study
(EMMA) . Technical Data (volume 13B) , Socio-Economic Impact Analysis,
Boston, MA.
Metcalf and Efldy, Inc., 1982. Application for Modification of Secondary
Treatment Requirements for its Deer Island and Nut Island Effluent
Discharges into Marine Waters, for Metropolitan District Commission,
Boston, MA.
Metropolitan Planning Council, October 1972. Boston Harbor Islands
Comprehensive Plan, for Massachusetts Department of Natural Resources,
Boston, MA.
-------
6-63
References
Morey, E.R., 1981. The Demand for Site-Specific Recreational Activities: A
Characteristics Approach. J. Ehv. Eton, and Management, 8:345-371.
National Planning Association (NPA), 1975. Water-Related Recreation Benefits
Resulting from Public law 92-500. rational Commission on Water Quality,
Washington, DC.
Peterson, G. , D. H. Anderson and D. W. Lime, 1983. Multiple-Use Site Demand
Analysis: An Application to the Boundary Waters Canoe Area. J. liesure
Res. 14: 27-36.
Russell, C.S., and W.T. Vaughan, 1982. The national fishing benefits of
water pollution control, Jburnal of Environmental Bionomics and
Management, 9:328-353.
Small, K. and H. Rosen, 1981. Applied Welfare Economics with Discrete Choice
Models. Econometric a 40: 105-130.
U.S. Department of Commerce, Bureau of the Census, 1982. 1980 Census of
Population and Reusing, Massachusetts. Washington, OC.
U.S. Department of Interior, March 1970. Outdoor Recreation Space Standards,
Washington, DC.
U.S. Department of the Interior, 1982. 1980 Survey of Fishing, Hunting, and
Wildlife Associated Recreation, Fish and Wildlife Service and U.S.
Department of Commerce, Bureau of the Census, Washington, DC.
U.S. Department of the Ulterior, April 1984. The 1982-1983 Nationwide
Recreation Survey, National Park Service, Washington, DC.
Willig, R., 1976. Consumer's Surplus without Apology. AER 66: 589-597.
Willig, R., 1978. Incremental Consumer's Surplus and Hedonic Price
Adjustment. J. Hson. Theory 17:227-253.
Wilman, E. , 1980. The Value of Time in Recreation Benefit Studies. J. Env.
Ebon, and Management 7: 272-286.
-------
Section 7
Health Benefits
In order to assess the health benefits of reducing the level of pollution
in Boston Harbor, it is first necessary to understand the adverse effects
that such a level of pollution might have on users of Boston Harbor waters.
Until recently, most health effects associated with water have been estimated
for withdrawal uses for drinking water supplies rather than for instream
uses, such as swimming, or other withdrawal uses, such as fish consumption.
This focus, in part, has bee'n due to what the public views as the more
serious nature of ingesting sewage contaminated water, but it has also been
affected by the relative ease of determining causal relationships between
water ingestion and illness as opposed to water contact and illness or the
less direct link of water pollutants to the food chain. Attempts to quantify
morbidity values and the corresponding benefits of decreasing the incidence
of illnesses contracted while swimming in polluted waters or consumption of
contaminated food have been made difficult by the lack of data on dose-
response and the corresponding population at risk.
This section focuses on two types of health benefits: swimming-related
illness and illness r' 'ated to bacterial contamination of shellfish. Other
health risks, such as those due to the accumulation in the food chain of
heavy metals and toxics (e.g., copper, mercury, PCBs and silver found in the
tissues of lobsters and winter flounder), cannot be estimated because little
is known about how the accumulation takes place, the effects of consumption
-------
7-2
or the dose response. Consequently, the benefits described in this section
must be viewed as a partial analysis of the possible health benefits of
improving water quality in the Boston Harbor.
7.1 Swimming-related Health Benefits
The method used to estimate swimming-related health benefits defines the
population at risk and then applies a dose-response relationship. A
discussion of the dose-response relationship used in this analysis is
included below because this approach is a fairly recent development.
7.1.1 Benefit Measurement Approach
One of the dose-response data problems for water contact and disease is
related to the indicators used to predict and quantify illness in the
population. The conventional wisdom regarding public health and water borne
disease assumes that since sewage contains fecal material and fecal material
may contain pathogens, then the level of fecal material is an adequate
measure of the potential for pathogens in the water. The parameter most
commonly used as an indicator of the potential for pathogens is the fecal
coliform bacterial count in the water column. Fecal coliforms are, in fact,
an excellent indicator of the presence of domestic sewage, but they do not
supply the kind of information needed to develop a dose-response relationship
for swimming-related illnesses.
Recently, it has been established that the presence of another bacterial
indicator, Qiterococci, is a more accurate measure of water quality than
fecal coliforms (Cabelli et. aj^_, 1980, 1982; Meiscier e_t ajk_, 1982). This is
principally due to the fact that Biterococci better mimic the aquatic
behavior of the viruses responsible for the potentially most serious
-------
7-3
(infectious hepatitis) and common (gastroenteritis) water-related enteric
diseases. In his 1980 and 1982 articles, Cabelli developed a dose-response
relationship between Biterococci density and the number of cases of
gastrointestinal symptoms per 1000 swimmers.
In order to apply this dose-response data to Boston Harbor beaches it was
necessary to perform some preliminary calculations and transformations of the
water quality data. All of the water quality data for Boston area beaches is
recorded in terms of concentrations of fecal and total coliforms, as required
by local, state and federal health standards, rather than in concentrations
of Enterococci. Using Enterococci data gathered from local Boston beaches we
developed a statistical relationship between the more available indicator,
fecal coliform, and the more accurate indicator, Enterococci. (See Appendix
C for more details.)
Given the correspondence between fecal coliform and Enterococci and the
dose-response relationship between Enterococci and gastrointestinal symptoms,
it was possible to correlate water quality at affected beaches with potential
swimming-related illness. Water quality data from 1974-1982 were collected
and averaged for all Boston area beaches and a percentage range of fecal
coliform concentrations was established. As described under the swimming/
beach closings section, population at risk was calculated by assuming
proportional relationships between seasonal attendance figures and percent of
time during the season that water quality levels fell into various ranges.
For example, if fecal coliform standards fell between 30 and 50 MPN/lOOml for
two percent of the entire season at a beach, we assumed that two percent of
the seasonal swimming population would be affected by this level of fecal
-------
7-4
coliform. in addition, we assumed that there were no swimmers among the
visitors on days when fecal coliform counts were above 500 MPN/100 ml since
this is the standard most of the towns and municipalities use for closing
beaches or posting them as unsafe for swimming.
Given these different water quality levels and number of bathers at risk,
we estimated the number of potential cases of gastrointestinal illness.
These are presented in Table 7-1. (See Appendix C for details of the
calculation.) For a lower bound estimate of number of cases of illness,
population at risk can be changed to reflect visitors to the beach who
actually go swimming. If not all visitors to a beach go swimming, then not
all visitors would be exposed to water pollution. The lower bound estimates
of numbers of cases of illness reflect an estimate of 49% of all beach
visitors actually go swimming. In addition, even with the improved water
quality not all of the predicted increased visitors may go swimming because
of air and water temperatures. During the 1982 and 1983 summer season, for
example, over half of the days had water temperature below 65° F or air
temperature below 75° F. For such days, some beach visitors may not go
swimming. To take into account these relatively colder temperatures in the
Boston Harbor area a factor based on the distribution of air and water
temperatures is applied to reduce population at risk and, thus, the number of
cases of illness. (See Appendix C.3 for derivation of population- \t-risk.)
The final stage in estimating swimming-related health benefits was to
value these illnesses. Based on information from Cabelli et al. (1980), we
have assumed that each case lasts from one to two days and requires sick
leave from work but does not require medical treatment. We have applied a
-------
7-5
Table 7-1. ftinual Reduction in Cases of Gastrointestinal Illnesses
Beach
Constitution
Dorchester Bay
Castle Island
Pleasure Bay
Carson
Malibu
Tenean
Wollaston
Quincy
Weymouth
Hingham
Hull
Nantasket
(SO
Option
161-596
21-77
242-896
134-497
198-735
65-239
2419-8961
238-881
0
0
0
0
Total 3478-12882
Ocean
Outfall
Option
21-79
2-7
21-79
12-45
18-68
15-57
293-1085
19-70
45-168
9-35
27-100
(352) -(1302) *
133-491
Secondary
Treatment
Option
11-39
2-7
21-79
12-45
18-68
15-57
293-1085
19-70
45-168
9-35
27-100
0
473-1753
CSO Plus
Ocean
Outfall
Option
248-919
28-103
325-1203
182-675
- 285-1056
175-647
4144-15348
344-1275
45-168
9-35
27-100
(352) -(1302)*
5461-20227
CSO Plus
Secondary
Treatmen t
Option
200-741
28-103
325-1203
182-675
285-1056
175-647
4144-15348
344-1275
45-168
9-35
27-100
0
5765-21351
* Increased cases of illness
See Appendix C for details of the calculations.
-------
7-6
full wage rate of $8.10/hour for two days to arrive at an upper bound value of
$129.56 per case and one-half the wage rate of $8.10/hour for one day to arrive
at a lower bound value of $32.40 per case (19823). These results are presented
in Table 7-2. Since the cost of illness is not the same as the willingness to
pay to avoid illness, these lost earnings represent a conservative proxy for
the value of good health. Other factors might include a value for discomfort
avoided and expenditures on medical care.
7.1.2 Benefit Estimates
The health benefits that are derived from cleaning up harbor waters are
substantial for some parts of the Boston Harbor Study area and insignificant
for others. The Wollaston and Quincy beaches show the greatest benefit
because of the great number of beach visitors, the poor level of water
quality, and the large percentage of predicted cleanup. Benefits for the
(institution and Dorchester Bay Beaches are not as great because, although
water quality is often poor at the beaches, the water is not consistently
dirty and, therefore, the greater number of cases of swimming-related
gastroenteritis occur only sporadically. The benefits at Weymouth, Hingham,
and Hull beaches are low because the water is relatively clean during most of
the season, percent predicted cleanup is only 30 percent, and attendance
figures are low compared to other Boston Harbor beaches.
7.1.3 Limits of Analysis
The key difficulties in accurately calculating health benefits are the
water quality and population-at-risk data limitations, as well as the
problems associated with valuing morbidity. Although we were able to develop
-------
7-7
Table 7-2. Swimming Health Benefits3./
(1982 $000)
Constitution
Dorchester Bay
Castle Island
Pleasure Bay
Carson
Malibu
Tenean
Wollaston
Cuincy
pmouth
Hinghain
Hull
Nantasket£/
TOTAL
CSO Option
$32.40-$129.56
5.2-77.2
21.3-316.7
0.7-10.0
7.8-116.1
4.3-64.4
6.4-95.2
2.1-31.0
78.4-1161.0
7.7-114.1
0
0
0
0
112.7-1,669.0
Ocean
Outfall
Option
$32.40-$129.56
0.7-10.2
2.3-33.1
0.1-0.9
0.7-10.2
0.4-5.8
0.6-8.8
0.5-7.4
9.5-140.6
0.6-9.1
1.5-21.8
0.3-4.5
0.9-13.0
(11.3)- (168.7)
4.3-63.6
Second a ry
Treatment
Option
CSO Plus
Ocean Outfall
Option
$32.40-$129.56 $3 2. 4 0-$129. 56
0.3-5.1
2.3-33.1
0.1-0.9
0.7-10.2
0.4-5.8
0.6-8.8
0.5-7.4
9.5-140.6
0.6-9.1
1.5-21.8
0.3-4.5
0.9-13.0
0
15.3-227.2
8.0-119.1
32.2-477.3
0.9-13.3
10.5-155.9
5.9-87.5
9.-2-136.8
5.7-83.8
134.3-1988.5
11.2-165.2
1.5-21.8
0.3-4.5
0.9-13.0
(11.3)- (168.7)
176.9-2,620.7
CSO Plus
Secondary
Treatment
Option
$32.40-3129.56
6.5-96.0
32.2-477.3
0.9-13.3
10.5-155.9
5.9-87.5
9.2-136.8
5.7-83.8
134.3-1988.5
11.2-165.2
1.5-21.8
0.3-4.5
0.9-13.0
0
186.8-2,766.3
2/Value per case of illness times number of cases from Table 7-1.
b_/$32.40 represents one day lost work at one-half wage rate and $129.56 represents
two days lost work at full wage rate.
°/Increased costs rather than savings.
-------
7-8
a good statistical relationship between fecal coliform and Enterococci because
of available Boston data, in general such relationships are difficult if not
impossible to determine because of variability in water quality conditions,
which affect the survival patterns and relationships between various bacterial
indicators in marine waters. Benefit estimates are also subject to bias
because of assumptions made about water quality levels and swimming
participation, because attendance figures only measure seasonal, and not
yearly, beach visits because beach attendance may not reflect actual time
spent in the water, and because the costs of illness do not include any
measure of medical treatment.
In addition, estimating health benefits from swimming may be subject to
double counting since swimmers may perceive most of the health effects
associated with water pollution. These benefits would thus be captured in
whole or in part by the logit estimation, described in the previous Section
of this report. More important than these limitations, however, is the fact
that previously unavailable dose-response information can now be used to
predict the number of swimming-related illnesses, provided towns and cities
measure the appropriate indicator of bacterial contamination.
A note of caution is warranted in using the Cabelli et al. dose response
function. This study is based on limited testing and the results have not
been duplicated or verified by other studies.
7.2 Siellfish Consumption
Theoretically, health benefits resulting from improved water quality can
be estimated by relating the reduction in frequency of water-related diseases
to the reduced contamination of shellfish attributed to various levels of
pollution abatement. Quantifying these benefits is difficult because of the
-------
7-9
/
unavailability of a dose-response function for shellfish-borne diseases such
as gastroenteritis, infectious hepatitis, and salmonellosis. Additional
difficulties are caused by the lack of information on the magnitude of
shellfish contamination and corresponding estimates of the population at
risk. Benefit estimation is further complicated by the difficulty in valuing
morbidity effects. Despite these methodological shortcomings, it is important
to attempt to estimate some of the shellfish-related benefits, if only to
illustrate that such techniques can be applied, given appropriate data.
It is possible to calculate benefits from reduction in incidence of
disease by applying assumed, rather than scientifically-derived, relationships
between water quality levels and incidence of disease. Assuming that disease
rates are proportional to the level of contamination, it is possible to
calculate a percentage reduction in the number of shellfish-borne cases of
disease based on a corresponding percentage cleanup. Almost one-half of the
shellfish acreage in Boston Harbor is classified as "grossly" contaminated and
is closed to harvesting because of potential health threats. It has been
estimated that, despite this closure, hundreds of bushels of contaminated
clams are being illegally harvested ("bootlegged") from these closed beds, and
sold on the open market. It is difficult to estimate the number of
contaminated clams that are reaching consumer tables, and even more difficult
to estimate what proportion of these clams can be linked to occurrence of
diseases. The only available indicator of shellfish-related diseases are the
actual reported outbreaks of gastroenteritis, hepatitis and other diseases.
In Boston, there have been few reported outbreaks of gastroenteritis or
other shellfish-related diseases. The Commonwealth of Massachusetts recorded
one outbreak of 30 cases of shellfish-related gastroenteritis in 1980. This
low disease, rate does not necessarily indicate that there is little risk of
-------
7-10
contracting shellfish-borne diseases or that shellfish contamination, due to
polluted waters, does not exist. Rather, it suggests that a high proportion
of cases are unreported, especially for the more common gastroenteritis
cases. One study (Singley, et al., 1975) suggested that the ratio of actual
to reported cases of foodborne diseases is 12:1. If this ratio were applied
to the data from Boston, then we would expect a minimum of 360 cases per year
of gastroenteritis due to shellfish contamination. Assuming a similar
scenario as described under swimming effects, these cases could be valued at
a low of $32.40 and a high of $129.56. lotential damages would then range
from $11,664 to $46,642.
It is not possible to relate reduction in water pollution, resulting from
implementation of different pollution control plans, to corresponding
reductions in incidence rate of these diseases and corresponding reductions
in morbidity values because of the inadequate information relating a specific
case to a specific shellfish area. It is important to note, however, that
provided adequate data, the above technique can be applied, and corresponding
benefits can be valued.
-------
7-11
References
Cabelli, Victor J. , e_t a_l., 1980. Health Effects Quality Criteria for Marine
Recreational Waters, Environmental Protection Agency, EPA-600/1-80-031.
Cabelli, V.J., A.P. Dufour, L.J. McCabe, and M.A. Levin. 1982. Swimming
Associated (Sstroenteritis and Water Quality. American Journal of
Epidemiology, 115:606-616.
Meiscier, John J. and Victor J. Cabelli, 1982. Enterococci and Other
Microbial Indicators in Municipal Wastewater Effluents, Jburnal of the Water
Pollution Control Federation, Vol. 54, No. 12.
Singley, J. Edward e_t a_l. , 1975, A Benefit/Cost Evaluation of Drinking Water
Hygiene Programs, U.S. Environmental Protection Agency.
-------
Section 8
Commercial Fisheries Benefits
Commercial fishing within Boston Harbor and the perimeter of
Massachusetts Bay includes shellfishing, lobstering and finfishing. It is
difficult to predict the precise impact of the various pollution abatement
options because of lack of data on both productivity changes in relation to
pollutant levels and current yields from the study area, especially for
lobstering and finfishing. Because of differences in the available data,
this section presents a general view of the potential impacts on lobsters and
finfish and more detailed calculations for shellfishing.
As will be seen, the near-term benefits from reducing water pollution are
modest. The most important factor affecting this lack of improvement is the
problem of sediment contamination, which is affected by all sources of
pollution (STPs, CSOs, non-point runoff, unauthorized site dumping, illegal •
discharges, and town sewers). The sediment throughout Boston Harbor is a
sink for a number of toxic pollutants, particularly for heavy metals such as
mercury, copper, nickel and silver, for PCBs, and for a number of pesticides,
all of which are potentially detrimental to fish productivity and consumer
health. There is scarce information about the precise levels of these
contaminants in the sediment and even less information about their turnover
and flushing rates. Added to this dilemma of ediment contamination is the
problem of bacterial pollution from illegal dischargers, non-point sources
and town sewers, all of which are difficult to locate, making it nearly
impossible to precisely define their corresponding receptors. For these
reasons, we have had to apply quite restrictive assumptions to the benefit
calculations.
-------
8-2
8.1 lobster ing and Finfishing
/ Lobstering is the most valuable fishery conducted within Massachusetts
v/
state waters. Tbtal 1981 lobster landings were 9.5 million pounds and, at a
a/
value of $2.09 per pound, were worth $19.8 million. Most of the
lobstering activity occurs in Essex and Plymouth counties, along shoreline
areas. Prior to 1979, the Massachusetts Division of Marine Fisheries did not
keep data in a form which made it possible to determine amounts which were
ff
harvested in any particular area of the Harbor, 'Metcalf & Eddy (1982) have
estimated that Dorchester Bay is the most productive area of the Harbor,
b/
followed in productivity by Quincy and Hingham Bays. In 1979, however,
the Division expanded the boundaries of the statistical catch area for lobster
to include the entire Boston Harbor and portions of Massachusetts Bay out to a
depth of 120 feet. Within this area, stretching from Lynn to Scituate and east
past the Brewsters Islands, the total 1981 lobster catch was 2.6 million pound
worth $5.4 million if valued at $2.09 per pound, accounting for about 27
percent of total Massachusetts lobster supply.
Finfishing is also a commercial activity in Boston Harbor and the
immediate Massachusetts Bay area. Boston is one of 51 commercial fishing
harbors in Massachusetts, and in 1979 ranked third in Massachusetts in pounds
of finfish landed. The approximately 57 gilt net line trawl vessels operating
in and around the Harbor fish primarily for winter flounder, cod, and pollock,
mostly during the summer months. There are also 29 draggers registered in
Boston of which a small percentage fish within the Harbor area for menhaden
and, just outside Boston Harbor, for winter flounder, yellow tail flounder, and
cod. In addition, there are four seine boats which are known to fish the
\ a/
— Massachusetts Division of Marine Fisheries estimates.
b/
— Lobster harvest was approximately 140,000 kg (308,000 Ibs.) in
Dorchester Bay in 1967 and 80,000 kg (176,000 Ibs.) in Hingham Bay in 1970.
-------
8-3
waters at the perimeter of Boston Harbor and Massachusetts Bay for sea
herring. The National Marine Fisheries service records finfish landings in -
Boston Harbor but, unfortunately, these records do not include where the fish
species are caught. For the year 1981, 28.4 million pounds of fish were
landed in the port of Boston for a value of $12.4 million (National Marine
Fisheries Service, 1983).
It is expected that reducing pollutant levels from the CSOs and the STPs
,will increase the productivity of lobstering and finfishing within the study
I
1 areas but it is not possible to say by how much. On the other hand one
{treatment alternative, the deep ocean outfall option, will increase pollutant
levels immediately surrounding the ocean diffuser in Masschusetts Bay. This
option is expected to have an adverse impact on lobstering and finfishing
activities in that area.
It is difficult to predict the precise impact that effluent from the ocean
outfall discharge—which includes BOD, suspended solids, heavy metals and
( ^ toxic chemicals—will have on the productivity of lobstering and finfishing
because of insufficient dose-response data at sublethal concentrations and
s\
because of deficiencies in current knowledge of variations in ambient
concentrations of water pollutants, which vary according to depth, current
patterns, temperature conditions, tidal influences and estuarine influences.
We must assume that pollution from ocean outfall effluent will have similai.
environmental effects as those reported for Boston Harbor, despite their
biological, chemical and physical differences. Some information does exist,
however, which enables us to predict the range of transport of some of the
pollutants and the corresponding qualitative predicted impact of discharge on
benthic fauna and commercial fisheries productivity.
-------
8-4
Circulation in Massachusetts Bay (location of the ocean outfall) is not as
efficient in terms of dispersion as are other area coastal locations, because
the Bay is partially enclosed. Circulation is further restricted because of
the depressed topographic features. The predicted ocean outfall discharge of
494,200 Ibs/day of BOD and 369,000 Ibs/day of suspended solids (including
associated toxic pollutants such as PCBs, pesticides, and heavy metals) is
expected to have an adverse effect on the biological population within the
immediate discharge area and beyond the zone of initial dilution, although
exact quantification of these effects is currently not possible. The
discharge from the ocean diffuser is not expected to violate the
Massachusetts' dissolved oxygen standard at the boundary of initial dilution,
but it could be expected to violate the far-field and steady state benthic
oxygen demand criteria due to abrupt resuspension.—
As stated in the waiver denial (US EPA, 1983) the proposed deep ocean
outfall is expected to contribute nutrient stimulation of phytoplankton
resulting in an adverse increase of pollution-tolerant phytoplankton and an
increase in the amount of phytoplankton propagated at the existing
b/
site. No measurable effects are expected for zooplankton-populations.
The dilution dynamics at the proposed discharge site, the differences in the
community structure of some of the populations, and the numerous near-shore
pollution sources make it difficult to predict precisely the nature of the
impact on biological community dynamics. In general, the proposed discharge
is predicted to result in moderate, and possibly major, adverse impacts on the
benthos. Major benthic alterations resulting from a sedimentation rate of 486
<3/m2/yr would be expected to cover an area about 37 times the area of the
— For a complete discussion of discharge and projected qualitative
impacts, see US EPA, 1983.
-' Based on observed impacts at present discharge areas in Boston Harbor,
and a calculated deposition rate of sewage particles resulting in organic
enrichment.
-------
8-5
zone of initial dilution (2.4 mi ) whereas moderate impacts resulting from a
sedimentation rate of 92 g/m /yr would extend over an area about 2,500 times
that of the zone of initial dilution (166 mi2) (US EPA, 1983; Tetra Tech,
s
1980) .
The effects of these benthic changes on commercial fisheries are not
immediately clear. In general, the reduction and changes in benthic fauna are
expected to result in a decrease in available foods for finfish, crabs, and,
to a lesser extent, lobsters over a 166 mi area of Massachusetts Bay.
Unfortunately, it is extremely difficult to quantify the exact magnitude of
these effects on finfish and lobster productivity.
The part of this study area which is most likely to be affected by the
proposed ocean outfall, and which also supports lobster populations, is the
area of the Brewsters Islands on the perimeter of Boston Harbor and
Massachusetts Bay. It is possible that an area of lobster exclusion may be
formed around the Brewsters based on observed exclusions at the existing Lynn
Wastewater discharge (Tetra Tech, 1982). This exclusion would result,
however, in only a small reduction in total lobster catch. This is because
the amount of lobster caught in the Brewster Islands area represents only a
fraction of the over 2 million pounds of lobster harvested in the entire area
(which extends from Lynn to Scituate, and includes inner Boston Harbor).
Insufficient data on the number of pounds of lobster caught in this area
prevents precise quantification of these effects.
Estimates of costs to commercial finfishery are equally difficult to
determine. As was the case for lobsters, increased concentrations of
pollutants are expected to detrimentally affect many of the fish populations.
Fin erosion, particularly in winter flounder, is one of the few impacts which
-------
8-6
are directly observable. Fin erosion has been detected in winter flounder
taken from inshore Harbor locations, although the exact cause of fin erosion
is not known. There is some evidence that fish develop the disease when
maintained in contact with contaminated sediments. There is also additional
evidence that PCBs may be involved in the development of the disease (US EPA,
1983; Sherwood, 1982). Based on this information, it is predicted that
finfish (particularly the winter flounder, which will be attracted to the
sediments because of their organic enrichment) will be affected by this
disease. Given the lack of information on how this disease specifically
alters species productivity and recruitment, however, it is currently not
possible to quantitatively estimate these effects on the economics of
commercial finfishing in the study area.
One final concern is the problem of toxic pollutants. Tbxic pollutants
and pesticides can exert a number of adverse effects on marine organisms. The
ocean outfall option is expected to increase the concentrations of a number of
toxic pollutants in the ambient waters and sediments surrounding the ocean
outfall diffuser. Based on analysis by Tetra Tech (1980) and US EPA (1983),
it is predicted that copper, mercury, silver, and PCBs may exceed EPA water
quality criteria after initial dilution, unless alleviated by a toxic control
program. Although an initial dilution of 133:1 will help assure that metals
concentrations will fall below EPA water quality criteria, the unusually large
predicted volume of particulate matter and its associated toxic substances are
likely to result in high sediment concentrations of particulate-iassociated
toxicants which will adversely affect marine biota (US EPA, 1983). Lobsters
are particularly sensitive to copper concentrations; however, there is
uncertainty about the sublethal, chronic effects of this heavy metal on
-------
8-7
lobster population dynamics. Even less is known about synergystic pollutant
a/
effects on both finfish and lobster.
Although toxic materials may be bioaccumulating in lobster and finfish
tissue and adversely affecting the dynamics of these populations, we must
>
conclude that because of insufficient biological, chemical and economic data,
the economic effects on these commercial fisheries must remain unquantified.
8.2 Commercial Shellfishing Industry
The shellfishing industry is the sector of commercial fishing to which the
greatest value could accrue from CSO or STP pollution abatement in Boston
Harbor. The soft shelled clam (Mya arenaria) is the most abundant
commercially valuable shellfish species found in Boston Harbor. Blue mussels
(Mytilus edulis) are also found but are not commercially valuable. The Boston
Harbor fishery is an important part of the Massachusetts shellfishing
industry; approximately twelve percent of the 1981 soft shelled clam harvest
came from the area. There are fifty-six shellfish areas in Boston Harbor
defined by the Massachusetts Division of Marine Fisheries, ranging in size
from one that is three acres in Weymouth to one of 400 acres in Hingham (see /
Figure 8-1.). Total shellfish acreage is about 4,700 acres (see Table 8-1).
,' Almost one-half of this acreage (2,273) is classified as grossly contaminated
I
| and, therefore, closed to harvesting. Slightly over one-half is classified as
moderately contaminated and is open to harvesting only by licensed master
£/ Despite the fact that toxic pollutants are expected to adversely
affect the marine biota, bioaccuraulation of these toxic chemicals are not
expected to exceed the FDA tolerance level for finfish and lobster (US EPA,
1983).
-------
Figure 8-1. Commercial Finfishing and
Shellfishing Resources in Boston Harbor
rhnrlc.s River.?/
Source: Metcalf & Eddy (1982),
'gure 2-5.
Will dlv.l
HULL
HINGHAM
* I Areas closed to
I Shellfishing
Areas restricted to
Master Diggers
\| Approximate location
of commercial
fisheries
00
CO
-------
8-9
Table 8-1
Characteristics of Boston Harbor Shellfish Areas!/
Name of Adjacent 1 1
City or Town or 1 Number of 1
Land Area 1 Shellfish Areas 1
Constitution Beach Area
Winthrop
East Boston
Dorchester Bay Area
South Boston
Dorchester
Quincy
Weyraouth
Hingham
Hull
Boston Harbor Islands:
Slate
Grape
Bumpkin
Georges
Lo veils
Gallups
Deer
Long
Spectacle
• Thompson
Rains ford
Sheep
Peddocks
TOTAL FLAT AREA
Estimated Productive Tidal Ar. a
10
3
7
4
2
2
11
7
3
8
13
1
1
1
1
1
1
1
1
1
1
1
1
1
56
2,300 acres
Acreage by Classification^/
1
Closed 1 Restricted
470
38
432
425
125
300
581
129
37
172
689
28
106
20
18
106
46
180
37
18
130
2,503
1,150
426
316
110
70
40
30
777
272
464
344
105
30
55
20
2,458
£/ Department of Environmental Quality Engineering estimates.
b/ These acreages represent total flat area as opposed to tidal area.
Productive acreage may be much smaller.
-------
8-10
diggers and their employees. None of this area is open to unrestricted
digging. Special requirements such as the posting of a surety bond are placed
upon those who are issued master digger licenses by the state. Shellfish
from moderately contaminated areas must undergo depuration at the Shellfish
Purification Plant in Newburyport, Massachusetts, before being sold. The
Massachusetts shellfish sanitation program classifies shellfish areas by
standards developed by the U.S. Public Health Service and member states of the
Cooperative Program for Certification of Interstate Shellfish Shippers. Among
other criteria, areas are classified according to the MPN (mean probability
number) of total coliform bacteria per 100 ml of the overlying waters. Zero
to seventy MPN is defined as clean, seventy-one to seven hundred MPN is
defined as moderately contaminated (restricted) and above 700 is defined as
grossly contaminated (closed). Although bacterial quality of the water is one
criteria, the guidelines contain other requirements so that any potential
sources of pollution, direct or indirect, may be sufficient to declare an area
unfit even though bacterial limits were met.
8.2.1 Pollution Abatement Impacts
The implementation of CSO controls or STP improvements can be expected to
reduce the fecal and total coliform counts in the waters overlying the
shellfish areas in Boston Harbor, as discussed in the previous chapters.
Table 8-2 illustrates the changes that might occur in the classification of
shellfish bed acreage if the CSO and/or STP controls were implemented. The
anticipated changes would mean reclassification from grossly contaminated
(closed) to moderately contaminated (restricted), thereby allowing harvesting
-------
Table 8-2. Estimated Potential Impacts of Pollution
Abatement Options on Boston Harbor Shellfish Areas */
Adjacent
Land Area
Winthrop
East Boston
South Boston
Dorchester
Quincy «!/
Weymouth
Hlnghan
Hull
Potential Additional Acres Open to 1
Restricted Harvesting due to Control Option -/ 1
Option 1
CSO 1 STP 1
Const. I Dorch/Nep. I Quincy 1 Ocean Outfall or I
1 1 1 Secondary Trmt. 1
5 ~ — 14
55 — — 161
16
75
80 6
20 1
6
— — — — — — 7
9
Optimum I Increased Yield Due
Annua 1 1
Yield 1
For Each 1 CSO
Area I Const I Dorch./Nep.
-------
8-12
with depuration. It is not likely that areas now classified as restricted
could be opened to unrestricted harvesting, due to such factors as sediment
contamination which are unaffected by CSO controls or STP upgrading.
It should be noted that, while this analysis specifically looks at two
main factors affecting the Boston Harbor shellfisheries1 soft-shelled clams
(CSOs and STP discharges), other factors will also have an impact (e.g.,
winter-kills on the clam beds and harbor maintenance through channel
dredging). Also, as mentioned above, criteria other than bacterial levels are
used to classify shellfish harvesting areas.
Based on information from the Massachusetts Department of Environmental
Quality Engineering, about 725 acres could be reclassified if all pollution
abatement options were implemented. This represents about 30 percent of the
estimated total productive tidal area (as opposed to total flat area, see
Tables 8-1 and 8-2) in the harbor and about 60 percent of the closed
productive tidal area. The reclassification of acreage presented in Table 8-2
must be considered as only a general estimate. Areas would have to be
surveyed and sampled extensively after implementation of any of the options
before any reclassification could take place.
In order to determine the impact of the pollution abatement options on the
shellfishing industry, it is necessary to translate the potential additional
acreage open to restricted digging into an increased harvest which can be
valued economically. To do this, an estimated optimum yield factor is used
(see Table 8-2). The optimum yield is an estimate of the ideal annual level
of harvest of a particular area which will maximize both present and future
economic revenues derived from the fishery. It is based on the maximum
-------
8-13
sustainable yield (MSY), which is a biologically determined level indicating
the annual harvest rate at which the productivity of the resource is
maximized. Any change from this level of fish catch, more or less, would
result in a decrease in the equilibrium population of fish. Optimum yield
differs from HSY in that it also accounts for fishing industry effort levels
and benefits to society at large (see Pierce and Hughes, 1979). The optimum
annual yield of a fishery is a function of costs and expected returns as well
as the natural rate of growth of the fish population. It may be a different
number than the MSY and, theoretically, allows for a profit-maximizing firm to
deplete the resource. It is not expected that the pollution controls in
question would lower the growth rate of shellfish in affected areas, so
current optimum yields have been used here.
The production and yield of a shellfish resource is generally determined
from a population density study of the area which place clams into class sizes
seed, juveniles, intermediates and mature in the order of size groupings.
These results afford information on the generation of yearly stock and of
succeeding crop families. Data also is produced on the health of the
shellfish, predation and a general distribution pattern of the shellfish in
the area. The information on optimum yield in Table 7-2 was provided by the
Massachusetts Department of Environmental Quality Engineering. Where no
studies have been made an average figure of 50 bushels per acre was used.3/
a/ From Harrington (no date). Also, the Maine Department of Marine
Resources rates acreage productivity for less than 25 bu/acres as poor, for
25-50 bu/acre as fair, for 50-75 bu/acre as good and for greater than 75
bu/acre as excellent (provided by E. Wong, Environmental Protection Agency,
Region I, Boston, MA).
-------
8-14
Multiplying the optimum annual yield by the acreage potentially
reclassified due to each abatement option gives the increased annual yield
that could be realized, as shown in the last four columns of Table 8-2. •'The
economic benefits associated with these increased yields depend upon the
economics of the industry and the supply and demand for soft shelled clams, as
discussed below. It should be noted that compared with an estimated current
16,000 bushels annual yield in Boston Harbor, the maximum estimated increase
of 34,000 bushels from all pollution abatement options amounts to twice the
current annual yield. This potential increase would impact on the depuration
plant, patrol surveillances, and laboratory and water quality monitoring.
These factors could act to limit actual acreages opened to increased
harvesting.
8.2.2 Benefit Assessment Methodology
Two types of benefits—change in producer surplus and change in consumer
surplus—may be associated with an increased shellfish harvest resulting from
pollution abatement. Producer surplus is a measure of the well-being of a
firm and is defined as the excess of revenues over costs. Figure 8-2
illustrates typical, simplified demand (D) and supply (S.) curves for the
shellfish industry. In the figure, producer surplus is the area below the
price line (Pn) and above the supply curve (Sg); it is equal to the area
labeled "B" plus the area labeled "F". Consumer surplus is a measure of the
satisfaction a consumer derives from the purchase of goods and services and is
defined as the difference between what the individual is willing to pay and
what is actually paid. In Figure 8-2, consumer surplus is the area above the
price line (PQ) and below the demand curve (D) (i.e., the area labeled "A").
-------
8-15
Figure 8-2.
Typical Demand and Supply Curves for the Shellfish Industry
Market
Price
($/unit)
Quantity of
Shellfish
-------
8-16
If the fishery is regulated and managed so that free entry by new firms is
restricted, then a change in producer surplus may occur. If the increase in
harvest is accompanied by either an unchanging price level or by a decrease in
per unit harvest costs greater than the decrease in price, then increased
profits will accrue to those firms in the restricted fishery throughout the
time frame of the analysis. If entry is unrestricted, however, then the
increased profits or rents to existing firms would be dissipated (after
several years duration at best) as new firms are attracted to the industry,
resulting in no long-run producer surplus.
A change in consumer surplus would depend upon a change in market price.
If the increase in harvest is large relative to the total local market, then
the market price could decrease, resulting in an increase in consumer
surplus. If the increase in harvest is relatively small, or if the industry
is oligopolistic (i.e., composed of only a few firms so that each can affect
the whole industry) and the firms influence market price, then the price might
not decline and no increase in consumer surplus would accrue.
Whether changes in either producer or consumer surpluses would result from
the increased shellfish harvest estimated in the previous subsection for the
pollution abatement options depends upon the shapes of the demand and supply
curves for the industry. As mentioned above, in Figure 8-2 for price equals
PO and quantity equals QQ, consumer surplus is defined as the area A and
producer surplus as the sum of the areas B + F. In the case illustrated, an
increase in quantity to Q^ along with a downward shift in the supply curve
from Sg to S^, representing a decrease in per unit harvest costs
(resulting from pollution abatement), results in a new lower equilibrium
price, PI. in this hypothetical example, both consumer and producer
-------
8-17
surpluses are increased and these changes can be valued as economic benefits
associated with the pollution abatement, as follows:
Change in consumer surplus (CS) = New CS - Old CS
= (A + B + C + E)-A
» B + C + E
Change in producer surplus (PS) = New PS - Old PS
(P + G + H) - (B + F)
G + H - B.
These supply and demand curves must be estimated empirically for the
relevant benefits to be determined. For example, if the demand curve is very
elastic (i.e., flat) in the region of interest, then we can expect no
significant consumer surplus benefits to accompany an increase in quantity
produced. Broadly speaking, demand is elastic if quantity demanded is highly
responsive to price changes and is inelastic if it is not. A very elastic
demand curve would be one that is approaching a horizontal line and,
therefore, the change in consumer surplus (B + C + E in the above example)
would be very small. Or if, for instance, the supply curve for the industry
is not upward sloping in the region of concern, then no producer surplus would
be associated wth the production increase. Benefits estimated for a
particular fishery could include either consumer surplus benefits only or
producer surplus benefits only, or both types together, or no long-term
benefits, depending upon the shapes of the empirically estimated curves and
whether or n"»t the fishery is regulated (i.e., entry restricted).
8.2.3 Benefit Estimates
Although the theory for estimating commercial fishing benefits is well
developed and straightforward, the application of that theory is difficult.
There are no readily available studies which define consumer demand or supply
curves for the soft shelled clam industry in Massachusetts or elsewhere.
-------
8-18
landings data (data on the quantity of shellfish harvested) are collected by
the state but are felt to be reasonably accurate only for recent years.
Exvessel price (price to the digger or firm) data are not available. The
Boston area, however, is a major market for the industry. In 1980 consumption
was estimated at approximately 625,000 bushels.£/ Only 20 percent of that
quantity was harvested in Massachusetts, about 125,000 bushels. About 20 to
25 percent was harvested in Maryland and the remainder in Maine. Maine and
Maryland collect more extensive price and landings data than does
Massachusetts.
A study was done in Maryland in the mid-1970s for various fisheries in the
Chesapeake Bay, including the soft shelled clam fishery (Marasco, 1975). This
study developed the following demand function for the soft shelled clam
fishery, calibrated to late 1960s landings and price data in Maryland:
log Q = 2.4606 - 2.3588 log (P/CPI) + .6067 log (I/CPI) R2 = .91
(-9.5022)£/ (.9463)
where,
Q = landings in 1,000 Ibs.
P = exvessel price in £/lb.
I = per capita income
CPI = consumer price index.
Price elasticity of demand is defined as the ratio of the relative change
in quantity to the relative change in price, i.e., (*Q/Q)/(AP/P). The price
elasticity for clams in t! above equation is -2.3588. Price elasticities for
other species included in this study ranged from -.1 to -2. (See Appendix D.I
for a discussion of other demand curves investigated.)
S/ Based on Division of Marine Fisheries estimates.
£/ Significant at the .01 level.
-------
8-19
Unfortunately, the above demand function and other demand curves
considered represent the total demand faced by the fishermen for their product
which is shipped to more than one consumer market and not all consumed in
Maryland. So the estimated price elasticity (-2.3588) cannot be automatically
applied to develop a demand curve for Massachusetts consumers, even if the
markets were assumed comparable. The price elasticity for Massachusetts
consumers might be higher than the one in the above equation because many x
other fish species might be considered close substitutes. On the other hand,
it has been said that demand for soft shelled clams in Massachusetts in the
summer is unlimited; any that can be dug can be sold because of the high
tourist demand for this well-known local specialty.
f
To account for the lack of data, consumer demand functions have been
estimated for Massachusetts for a particular year (1981) for a range of price
elasticities, from more elastic (-3) to less elastic (-.5) than the number in
the above equation. Given the changes in yield estimated in the previous
subsection for each pollution abatement option and given an estimated average
price for that year ($31.41/bu^/), new prices were estimated for each
assumed price elasticity. The demand equation used is of the following form:
Q82= Ax P82 or,
log Q82 = Log * + cf*x Log Pfl2
£/ Based on Resources for Cape Ann, 1982, price for 1980 ($28.00) updated
to 1982 price using soft shelled clams price index from National Marine
Fisheries Service, NOAA, 1983.
-------
8-20
where,
Q82 = quantity consumed in the Boston market in 1982
A = constant
oc " assumed price elasticity
P82 = average 1982 exvessel price for soft shelled clams in
Massachusetts.
Table 8-3 displays the results of these estimates. The table shows that
as price elasticity increases (from -.5 to -3) and the demand curve becomes
flatter, the price changes resulting from the increases in clam harvest due to
the abatement options, decrease. The price decrease is greatest for the
combined CSO and STP upgrade option with an inelastic demand curve assumed
(* = -.5). The price change is least for the CSO options (taken separately)
s,
with an elastic demand curve assumed ( o( = -3) .
For reasons which are described below, it is likely that the primary
source of commercial fisheries benefits that would be associated with the
pollution abatement options would result from changes in consumer surplus
rather than producer surplus. If no producer surplus changes occur (see
below) , then total commercial fisheries benefits (equal to change in consumer
surplus) would be as shown in Table 8-4, following the same price elasticity
assumptions that were made for Table 8-3.
Consumer surplus benefits (Table 8-4) are estimated from the price changes
shown in Table 8-3 and from the changes in yields previously estimated for
each abatement option (see Table 8-2) . These changes in consumer surplus were
calculated from the following equation:
ACS - AP x Qo + 1/2 (AP x AQ)
where,
& CS = change in consumer surplus ($)
A p = change in price (3)
QO = initial consumption (bushels)
&Q = change in consumption (bushels) .
-------
8-21
Table 8-3. Estimated Changes in Price of Soft Shelled
Clans Associated with Alternative Abatement Options
and with Assumed Price Elasticities of Demand (19829)
Abatement Option
CSO
Constitution
Dorchester/Neponset
Qu incy
Combined CSO3/
STP: Ocean Outfall
or Secondary Treatment
Combined CSO and STpS/
Price
AP
Price
AP
Price
AP
Price
AP
Price
AP
Price
AP
1 E
1 -.5
31.11
-.30
30.94
-.47
31.18
-.23
30.42
-.99
28.76
-2.65
27.89
1 -3.52
last
1 -1
31.26
-.15
31.17
-.24
31.30
-.11
30.91
-.50
30.05
-1.36
29.60
1 -1'81
i c i t y
1 -2
31.33
-.08
31.29
-.12
31.35
-.06
31.16
-.25
30.72
-.69
30.49
1 -92
( <*)
1 -3
31.36
-.05
31.33
-.08
31.37
-.04
31.24
-.17
30.95
-.46
30.79
1 "62
—' All CSO options are combined in this row. Price changes are greater for
the combined plans than for the sum of the separate plans, because the demand
equation is not linear.
-------
8-22
Table 8-4. Estimated Total Benefits
Associated with Alternative Abatement Options
and with Assumed Price Elasticities of Demand (19823)
Abatement Option
CSO
Constitution
Dorchester/Neponset
Quincy
Combined CSO
1 E
1 -.5
5,239
8,674
3,936
20,727
last!
1 -1
2,626
4,353
1,971
10,446
city
1 -2
1,314
2,181
987
5,243
(*)
1 -3
877
1,455
658
3,501
STP: Ocean Outfall
or Secondary Treatment
Combined CSO and STP
79,847 40,804 20,627 13,812
123,537 63,602 32,273 21,622
-------
8-23
It was assumed that the harvest from Boston Harbor shellfish areas is consumed
in the Boston area market. In addition, 16,000 bushels was used as a
reasonable estimate of the annual harvest from Boston Harbor restricted areas
before pollution abatement and, therefore, as the initial consumption estimate
(Qo)—' . For a more detailed discussion of the computation methods used to
obtain the new prices, price changes and consumer surplus benefits, see
Appendix D.2.
As shown in Table 8-4, the total benefit levels vary in roughly the same
way as the price changes shown in Table 8-3. This is because as the price
decreases, the difference between price and willingness to pay increases, so
that consumer surplus increases, and is shown by positive numbers in the
table. The greatest benefits are obtained from the options with the greatest,
increase in yield and the most inelastic demand. Total benefits are larger
for the combined options than for the sum of the separate options, because the
demand equation is not linear.
It could also be legitimately argued that the change in consumer surplus
could be zero. If all the pollution abatement options were implemented, then
the increased harvest (34,000 bushels) would represent about six percent of
the total market (625,000 bushels). Since it appears that none of the firms
included in the Boston area market can influence price and since only a small
percentage of them would be affected by the pollution abatement, it could be
reasonably agreed that there would not be a change in consumer surplus given
the small percentage increases in harvest just mentioned. Not enough is known
about the consumer demand curve, however, to make a definitive judgment.
a/ Division of Marine Fisheries
-------
8-24
Thus, from the considerations just discussed, we can conclude that the
range of commercial fisheries benefits resulting from implementation of the
pollution abatement options in Boston Harbor would be from zero to the
higheststimates.levels presented in Table 8-4. The benefits estimates shown
in Table 8-4, column 2 (price elasticity = -1) represent moderate levels
between the upper and lower bounds just described.
As indicated above, no definitive estimates concerning producer surplus
changes could be made due to lack of data. Attempts were made to develop a
supply curve but were unsuccessful; these are described in Appendix D.3 along
with an example showing how to compute change in producer surplus, if such
benefits exist.
A reasonable argument can be made that the change in producer surplus
would be zero for commercial shellfishing in Boston Harbor. This argument is
that the supply curve is flat in the range of interest. If there is unlimited
entry of firms into the fishery, then the additional profits or rents which
would accrue to the master diggers currently operating in Boston Harbor
restricted areas would be dissipated over the long run, leaving no long-term
producer surplus benefits. There do exist institutional constraints on entry
to the fishery; the State of Massachusetts places some restrictions upon
master diggers allowed to operate in moderately contaminated areas: they must
have a special license, post a surety bond, utilize specially licensed
employees, meet certain transport requirements, keep certain records' and are
not allowed to concurrently harvest in areas classified as closed. There are
no absolute restrictions to entry, however; as long as a firm meets the
requirements, it may participate.^/
S/ For a discussion of various options for entry or effort regulation of
New Bigland fisheries, see Smith and Peterson, 1977.
-------
8-25
In addition to the question of official restrictions on entry into the
Boston Harbor shellfishing industry there is also evidence, as mentioned in
the Section on Health Benefits, that thousands of bushels of contaminated
clans are being bootlegged (illegally harvested) from the shellfish areas that
are classified as closed by the state.3/ This evidence shows that the
official restrictions on Boston Harbor shellfishing are often ignored and that
in practice there are few barriers to entry. It is, therefore, probable that
the change in producer surplus that would result from the control options
would only extend over a limited number of years until new firms attracted by
the increased profits are able to meet the entry requirements. It is
impossible to say how long these impediments would prevent new entries, but
over the long term they may not keep the additional profits generated by the
pollution abatement options from being reduced to zero.
8.2.4 Limits of Analysis
The major limitation of this analysis of commercial fisheries benefits is
the lack of well-developed consumer demand and supply curves for the soft
shelled clam industry. This makes application of the theory for estimating
commercial benefits difficult. However, it is unlikely that a producer
surplus exists and the true demand elasticity probably falls within the
estimated demand elasticity range used in this study. Thus, the analysis was
able to put bounds around the uncertainty.
Other data deficiencies include no good historical data for Massachusetts
on harvest of soft shelled clams, numbers depurated and price to the digger.
Little information also exists on the Boston consumer market and its sources
and changes over time. Furthermore, there is only a small amount of data on
Discussions with Division of Marine Fisheries staff and others.
-------
8-26
costs of the firms in the industry, particularly those with special licenses
to operate in restricted areas. The impacts of pollution abatement and of
theresulting increase in yields on these costs are hard to judge, especially
the changes in numbers of employees and income to the master diggers. This
lack of data thus prevented a more precise estimation of shellfishing benefits.
-------
8-27
References
Altobello, Marilyn A., David A. Storey and Jon M. Conrad, January 1977.
Hie Atlantic Sea Scallop Fishery; A Descriptive and Econometric Analysis,
Research Bulletin No. 643, Massachusetts Agricultural Experiment Station,
University of Massachusetts, Boston, MA.
Arnold, David, January 31, 1983. Clammers Can't Work, Can't Get Benefits, The
Boston Globe, Boston, MA.
Cape Cod Planning and Economic Development Commission, 1978. An Economic
Profile of the Cape and Islands Fisheries.
Commonwealth of Massachusetts, Division of Marine Fisheries, 1981. Abstracts
of Massachusetts Marine Fisheries Law, Boston, MA.
Commonwealth of Massachusetts, Division of Marine Fisheries, March 1982.
Massachusetts Marine Fisheries Management Policy Report, Boston, MA.
Commonwealth of Massachusetts, Division of Marine Fisheries, 1982. Rules and
Regulations Pertaining to the Issuance and Use of Master and Subordinate
Digger Permits, Boston, MA.
Commonwealth of Massachusetts, Division of Marine Fisheries, 1982. Shellfish
Areas Approved for Restricted Harvest with Depuration and Corresponding
Routes to Newburyport Purification Plant, Boston, MA.
Crutchfield, Stephen R., February 1983. Soft Clam Exvessel Demand Functions
and Clam Fishery Data, unpublished data. Department of Resource Economics,
University of Rhode Island, Kingston, RI.
Dumanoski, Dianne, December 19 to 21, 1982. Boston's Open Sewer, three part
series, The Boston Globe, Boston, MA.
Harrington, Peter, (no date). Shellfish Resource Estimates, unpublished memo,
Massachusetts Department of Environmental Quality Engineering, Boston, MA.
Marasco, Richard J., May 1975. An Analysis of Future Demands, Supplies,
Prices and Needs for Fishery Resources of the Chesapeake Bay, MP 868,
Agricultural Exper_- ent Station, University of Maryland, College Park,
Maryland.
Marchesseault, Guy, Joseph Mueller, Lars Vidaeus and W.. Gail Willette, 1981.
"Bio-Economic Simulation of the Atlantic Sea Scallop Fishery: A
Preliminary Report", in K. Brian Haley (ed.), Applied Operations Research
in Fishing, Plenum Publishing Corp., New York, NY.
Metcalf and Eddy, Inc., June 1982. Nut Island Wastewater Treatment Plant
Facilities Planning Project, Phase I, Site Options Study, for the
Metropolitan District Commission, Boston, MA.
-------
8-28
J
National Marine Fisheries Service, NOAA, 1983. Fisheries of the U.S,
Washington, DC.
National Marine Fisheries Service, NOAA, 1976. Fishery Statistics of the U.S,
Washington, DC.
Pierce, David E. and Patricia E. Hughes, January 1979. Insight into the
Methodology and Logic Behind National Marine Fisheries Services Fish
Stock Assessments, Massachusetts Division of Marine Fisheries and Coastal
Zone Management Office, Boston, MA.
Resources for Cape Ann, April 1982. The Costs of Pollution! The Shellfish
Industry and the Effects of Coastal Water Pollution, Massachusetts Audubon
Society.
Sherwood, M.J. 1982. Fin Erosion, Liver Condition, and Trace Contaminant
Exposure in Fishes from Three Coastal Regions. In: "Ecological Stress
and the New York Bight," Science and Management, pp. 359-377. Estuarine
Research Federation, Columbia, South Carolina.
Smith, Leah J. and Susan B. Peterson, August 1977. The New England Fishing
Industryi A Basis for Management, Technical Report WHOI-77-51, Woods Hole
Oceanographic Institution, Woods Hole, MA.
Tetra Tech, Inc. 1980. Technical Evaluation of Deer Island and Nut Island
Treatment Plants Section 301(h) Application for Modification of Secondary
Treatment Requirements for Discharge into Marine Waters for U.S.
Environmental Protection Agency.
Tetra Tech, Inc. 1981. Technical Evaluation of Lynn Wastewater Treatment
Plant Section 301(h) Application for Modification of Secondary Treatment
Requirements for Discharge into Marine Waters, for U.S. Environmental
Protection Agency.
Townsend, Ralph and Hugh Briggs, September 1980. Some Estimates of Harvesting
and Processing Costs for Maine's Marine Industries, technical report,
Department of Economics, University of Maine.
U.S. Environmental Protection Agency, Economic Analysis Division, 1982. The
Handbook of Benefit-Cost Assessm*nt for Water Programs, Draft, Washington
D.C.
U.S. Environmental Protection Agency, Office of Marine Discharge Evaluation,
1983. Analysis of the Section 301(h) Secondary Treatment Waiver
Application for Boston Metropolitan District Commission, Washington, D.C.
Wang, Der-Hsiung, Joel B. Dirlam and Virgil J. Norton, March 1978. Demand
Analysis of Atlantic Groundfish, (preliminary report), Staff Paper No. 7,
Agricultural Experiment Station, University of Rhode Island, Kingston, RI.
Williams, Doug, (no date). Data and Procedures Used to Estimate Technical
Coefficients for the Clam/Worm Sector, unpublished paper, Department of
Agricultural and Resource Economics, University of Maine.
-------
Section 9
Intrinsic Benefits
Intrinsic benefits are all benefits that are associated with a resource,
which are not specifically related to current direct use of that resource.
Although these non-user benefits are not directly observable, it is important
to emphasize that they are as real and economically important as the more
easily measured user benefits.
Briefly, intrinsic benefits can be categorized as the sum of option
(bequest) values, existence value, and aesthetics.-^/ Option value is
defined as the amount of money, beyond user values, that individuals are
willing to pay to insure access to the resource (or a level of environmental
quality) in the future when there is uncertainty in resource availability
and/or individual use (demand), regardless of whether the individual is a
current user. Option benefits reflect the value of reducing uncertainties
and of avoiding irreversibilities. When option values reflect
intergenerational concerns they are referred to as bequest motives. Bequest
values are defined as the willingness to pay (HTP) for the satisfaction
associated with endowing future generations with the resource. Existence
value is defined as the willingness to pay for the knowledge that the
resource is available and ecosystems are being protected, independent of any
§/ For an in depth discussion of intrinsic benefits and their
estimation, see RTI, 1983; Freeman, 1979; Fisher and Raucher, 1982; Mitchell
and Carson, 1981.
-------
9-2
anticipated use by the individual. These values are distinct_from aesthetic
benefits and concerns over retaining the option of future use. Aesthetic
values pertain to enhanced appreciation of water-related (instream vs. near
stream) experiences. Given that improved water quality could enhance the
aesthetic values of users as well as non-users of the resource/ there could
be an aesthetic component in both use benefits and intrinsic benefits.
Definitions of bequest values tend to obscure the distinction between
existence and option values in the literature. Sometimes bequest values are
placed in a separate category of intrinsic values; sometimes they are treated
as part of existence values and at other times they are considered as option
values. For example. Freeman (1979) considers the utility of the expectation
of future use by descendants as a bequest form of vicarious existence
benefits. Yet, bequest values can be considered for long term potential use
where there may be uncertainties associated with future demand and supply.
Hence, this concept may be treated as part of option value. Mitchell and
Carson (1981) , for example, separate option value into current and bequest
categories.
Although the distinction between user and intrinsic benefits is often
unclear, there is substantial agreement that these intrinsic benefits may
account for a large portion of all pollution abatement benefits (see Fisher
and Raucher, 1982). Intrinsic benefits are usually derived from lemand
functions. Data for these functions are most frequently obtained from
surveys, questionnaires, and voting referenda. Assuming that people are
willing to pay for these values, these techniques are intended to yield
information on the prices that consumers are willing to pay for cleaner water
even though they do not intend to use the resource directly. This generated
-------
9-3
price information is used to construct demand equations from which the
welfare changes associated with cleaner water can be measured. Despite the
criticisms leveled at this contingent valuation approach/ due to several
potential biases, the survey method represents the best available technique
to quantify all these benefits.
Property value data may also be used to infer estimates of intrinsic
benefits. The property value approach is based on the hedonic valuation
method, which relates the price or value of a property to a variety of
discrete characteristics. These characteristics include site and
neighborhood characteristics, socio-economic factors, and environmental
quality variables such as degree of water pollution. A major limitation of
the property value technique is that it neglects the benefits to those who do
not own property near the affected water body. The approach also records the
response of property owners to an actual change in water quality, a change
which may not necessarily reflect what property owners would be willing to
pay for potential improvements in water quality, or for improved water
quality at other locations. As a result, a significant fraction of value, in
the form of consumer surplus, may be omitted when applying this technique.
In addition, the hedonic approach may produce biased benefit estimates
because of the difficulty in disaggregating the benefits between use
(recreation, for example) and nonuse. There have been several attempts to
model this relationship despite the extensive data required for this
technique. One such effort, described in Feenberg and Hills (1980), uses
property values derived from a study by Harrison and Rubenfeld (1978).
9.1 Methodology
Intrinsic benefits are difficult to measure and value. A number of
studies have attempted to measure intrinsic values using the NTP survey
-------
9-4
approach. We know of no specific study that can be applied directly to the
entire Boston Harbor or that can be associated with the range of pollution
abatement options which accurately relates either dichotomous or incremental
changes in water quality to corresponding changes in intrinsic values. The
most recent willingness to pay surveys measure benefits to users and
non-users of rivers (RTI, 1983; Cronin, 1982) and are inappropriate to apply
to a marine resource such as Boston Jferbor. The Qcamlich (1977) study, which
measures willingness to pay for improving water to a swimmable level in the
Charles River/ cannot be applied to Boston Harbor because Gramlich's bids are
averages across both users and nonusers, representing total values, and
because the Charles River is not a marine resource.
Other researchers have attempted to establish a relationship between
intrinsic values and user values (see Fisher and Raucher, 1982, for a
critical review). Results from this approach suggest that intrinsic values
are substantial: they generally are at least one-half as great as
recreational user benefits. Because of the lack of appropriate WTP survey
data which can be applied to the different control options in the study area,
estimates of intrinsic benefits were made by assuming that these non-user
benefits are one-half as great as recreational user benefits.
9.2 Benefits Estimates
Intrinsic benefits for the CSO and STP pollution control options
are accordingly based on one-half the benefit estimates derived from
the recreational benefits estimated in Section 6.$r These benefit values
a/ Includes swimming participation (logit model plus Quincy, Weymouth,
Hingham, Hall and fentasket estimates), boating, fishing, and Boston Harbor
Islands recreation. For swimming the user day value ($11.06) derived in the
logit model is applied to increased user day figures (see text in Section 6
for user day values for other recreational activities).
-------
9-5
/
incorporate both current and future benefits from water quality improvements
and are presented in Table 9-1. The range of values represents a very rough
approximation of non-user benefits.
Table 9-1
Annual Intrinsic Benefits
(Millions 1982$)
Pollution Control
Option
1 CSO I
1 plus I
1 Ocean 1
1 Outfall I
50% of Recreation
Benefits
i
1 High: 21.8
1 Low: 10.1
1 Moderate: 15.9
CSO
plus
Seconda ry
Treatment
23.2
10.7
17.0
9.3 Limits of Analysis
Non-user benefits are especially difficult to measure and project, and
estimation of these benefits is limited by both methodology and data.
Appropriate willingness to pay surveys and property studies were not
available to estimate benefits from the variety of pollution control
options. As a result, these benefits may be biased because they might be
capturing benefits calculated under other categories such as fishing,
swimming, or boating (i.e., double counting).
-------
9-6
References
Cronin, Francis, 1982. Valuing Nonmarket Goods Through Contingent Markets,
Richland, Washington: Pacific Northwest laboratory, PNL-4255.
Feenberg, Daniel and E.S. Mills, 1980. Measuring the Benefits of Water
Pollution Abatement, teademic Press, New York, NY, 187 pp.
Fisher, Ann and Robert Raucher, 1982, Comparison of Alternative Methods of
Evaluating the Intrinsic Benefits of Improved Water Quality, presented at
the American Economics Association proceedings, New York.
Freeman, A. Myrick, 1979. The Benefits of Environmental Improvement. Johns
Hopkins University Press, Baltimore, Maryland.
Gramlich, Frederick, 1977. The Demand for dean Water: The Case of the
Charles River, National Tax Journal 2:183.
Harrison, David and Daniel Rubenfeld, 1982. Hedonic Housing Prices and the
Demand for Clean Air. Journal of Environmental Economics and Management
5: 81-102.
Mitchell, Robert and Richard Carson, 1981. An Experiment in Determining
Willingness to Pay for National Water Quality Improvements. U.S.
Environmental Protection Agency.
Research Triangle Institute, 1983. A Comparison of Alternative Approaches
for Estimating Recreation and Related Benefits of Water Quality
Improvement, Research Triangle Park, North Carolina.
-------
Section 10
Biological Effects
Several of the pollution abatement options considered are expected to
have a positive influence on the ecological processes in the estuarine areas
of Boston Harbor because of significant reductions in pollutant loadings and
corresponding reductions in concentrations of fecal coliform, suspended
solids, organic toxics, heavy metals, and increases in the level of dissolved
oxygen. Implementation of the ocean outfall option is also expected, on the
one hand, to beneficially impact the ecological processes in Boston Harbor
while, on the other hand, to detrimentally affect the ecological processes in
Massachusetts Bay because of removal of pollutants from the Harbor to the Bay.
It is not easy to capture the ecological costs and benefits of these
pollution control options because of the lack of information linking
pollutant transport and dispersion to specific dose-response relationships,
and the difficulty in expressing these changes and effects in monetary
units. Therefore, the following discussion of the ecological effects of the
different treatment options will be presented qualitatively, as opposed to
the quantitative benefits and costs described in previous chapters.
10.1 CSO and Secondary Treatment Options
It is likely that the G50 and STP pollution abatement options will
positively influence the biological ecosystem within Boston Harbor,
-------
10-2
particularly the highly productive saltmarsh habitats. Phytoplankton,
benthic organisms and the communities of shellfish, finfish and lobster will
be specifically affected. This positive effect will occur because both
treatment options will reduce loadings of BOD, suspended solids and fecal
coliform to the Harbor area, as well as reducing concentrations of heavy
metals (see Table 2-3) and possibly organic toxics such as pesticides and
PCBs.a-' Although none of Massachusetts' major saltmarshes are located in
Boston Harbor, it does contain a significant amount of marsh acreage. Quincy
Bay has 209 areas of saltmarsh, Dorchester Bay 363 acres, Hingham Bay 644
acres and there is also Belle Isle Marsh along the inlet in Winthrop. These
marshlands play an important role in the biological productivity of the
adjacent coastal waters as well as performing other useful functions. It is
well documented (Odum, 1961; Teal, 1962) that these areas are the most
efficient primary producing environments on earth and provide natural
spawning, nursery and feeding habitat for many species of fish and
invertebrates. The sheltered waters and grasses provide food and cover for
furbearing animals, shorebirds, and waterfowl. From two-thirds to three-
quarters of the commercially or recreationally important finfish, such as
herring, striped bass and flounder, and shellfish spend part of their
lifecycle in saltmarshes.
Marshlands tran form carbon dioxide water into oxygen and food. They are
highly productive of organic matter; because of the tides, wastewater
a/ In general, the STP secondary option will reduce conventional and
non-conventional pollutant loadings to a greater extent than the CSO option,
although the greatest difference in reduction are changes in BOD and
suspended solids.
-------
10-3
products are regularly removed and organic material and nutrients are added.
It has been estimated that a saltmarsh produces 10,000 pounds of organic
matter per acre per year (Odum, 1961). These lands concentrate and recycle
carbon, nitrogen and phosphorus and are important to the global cycles of
nitrogen and sulfur. Marsh areas have a very high value as providers of
tertiary sewage treatment since they remove and recycle inorganic nutrients.
Saltmarshes are also important for stabilizing the shoreline. They
provide a buffer zone which limits coastal erosion by flood, wave, and wind
action. Marshes act as reservoirs during flooding and absorb sediments and
wave energy during storms which aids in keeping harbors open and in
preserving beaches.
Attempts have been made to estimate the economic value of saltmarshes by
valuing the productivity of the marsh, by valuing the role of the marsh as a
factor of production, and by estimating the cost of duplicating the functions
of a marsh, such as providing tertiary wastewater treatment. Annual values
ranging from $100 to $4,000 per acre were developed in one study (Gosselink,
Odum and Pope, 1973). These types of values have been criticized as
representing total value rather than net benefits and much smaller values
(3.25-3.30 per acre) were estimated for marsh areas as factors of production
(Lynn, Conroy and Prochaska, 1981). Another study points out the many
functions of the marsh are not !:• luded when only the productivity of the
marsh is valued (Westraore, 1977). In any case, if, for illustration
purposes, such a range of values is applied to the total marsh acreage of
Boston Harbor (1216+ acres), an economic value ranging from $121,600 to
$4,864,000 per year is estimated.
-------
10-4
Whatever value of marshland is selected, the problem for this case study
is determining the impact of the pollution abatement option on the marsh.
For the most part, the studies cited above and others are concerned with
development that will destroy the marsh by dredging or filling. Here, the
concern is with the impact of pollutants (and their abatement) on the
functioning of the marsh. It is known that large amounts of untreated
organic materials greatly stress marshes and reduce dissolved oxygen to
undesirable levels. However, smaller amounts of these materials may enhance
marsh productivity. Oilorinated hydrocarbons, and organophosphorous
pesticides have been measured in the Harbor in sufficient concentrations to
have sublethal or lethal effects on adult crustaceans, larval mollusks and
embryonic and larval forms of finfish. Other effects on saltmarsh flora and
fauna are unknown.
The proposed pollution abatement options under consideration in this
study will control coliform bacteria, pesticides and some heavy metals in
Harbor marshlands. The connection between the levels of control and the
effect on the functioning of the marshlands, however, is unknown. Since we
are unable to measure the extent of the impacts, these marshland benefits
must be considered nonmonetizable.
The effects on the plankton and benthic communities throughout the rest
of the Harbor generally will be the opposite of those described below for the
ocean outfall option. Reduction in conventional loadings may increase
species diversity and there will be a shift whereby pollution
sensitive-species will replace many of the pollution-tolerant species now
dominating the Harbor. These community changes will influence the abundance
-------
10-5
and diversity of species who feed on these organisms in the lower portion of
the food chain, leading to a shift towards pollution-intolerant species. For
example, yellow tail flounder may replace winter flounder who prefer
organically enriched sites.
Reductions in metals and possibly organic toxicants will have a positive
effect on many species in the Harbor, particularly the shellfish and finfish
who tend to bioaccumulate toxic substances such as PCBs and organically
complexed metals such as mercury and lead.— These effects may include a
reduction in disease (such as finfish erosion), increases in juvenile
survival and increases in productivity and community stability.
10.2 Ocean Outfall Option
The ocean outfall plan is expected to have negative effects on the
biological ecosystem of a portion of Massachusetts Bay. As discussed in
Section 2 of this report, the pollution abatement plan calls for an ocean
outfall diffuser system to discharge the combined, treated effluent from Deer
and Nut Island plants into Massachusetts Bay, 7.5 miles (12.1 km) northeast
of Deer Island. This discharge area will not provide for sufficient
a/ It is important to note however, that although the pollution abatement
options under consideration will eliminate some of the toxic substances and
metals in the Harbor waters, significant concentrations of the- .• pollutants
reside in the harbor sediment and are constantly being re-suspended. It is
not known what the flushing rate is for Boston Harbor but the rate is
probably considerably reduced because of the very shallow depths of all the
harbor waters. Thus, many of these pollutants will remain in the sediment
and water columns for many years to come and continue to negatively affect
the ecological communities.
-------
10-6
transport and dispersion of the diluted wastewater and particulates because
it is topographically depressed. This, in turn, will restrict circulation
and dilution and will lead to an accumulation of BOD and suspended solids,
and several toxic pollutants. In addition, the proposed discharge of
suspended solids is expected to violate the Commonwealth's dissolved oxygen
standard.
Discharge from the proposed outfall is expected to negatively affect the
structure and function of many of the components of the marine ecosystem in
this area including phytoplankton, benthic invertebrates, and communities of
lobster, crab and finfish. It is also possible that several species of
whales, including the endangered Right whale, will be influenced by discharge
of pollutants into Massachusetts Bay.
10.2.1 Plankton
The proposed ocean discharge of BOD and suspended solids (which include
toxic pollutants) is predicted to significantly enrich the waters within the
immediate 2.4 square miles surrounding the diffuser and extend to a much
larger zone of 166 mi and thus greatly increase the levels of available
nutrients such as nitrogen (the most limiting nutrient in marine waters) and
phosphorous. Increased amounts of these nutrients will consequently
stimulate phytoplankton productivity and lead to increases in phytoplankton
biomass, as well as resulting in an adverse shift from pollutant-intolerant
phytoplankton to pollutant-tolerant species. The composition and
distribution of the zooplankton populations are not expected to be
significantly affected because of the increased limited dilution and because
the zooplankton community is inherently able to quickly recover from
-------
10-7
pollutant stress. As discussed in the waiver documents (Tetra Tech. 1980;
US EPA, 1983) the most polluted of waters appear to depress numbers of
zooplankton without measurably altering species composition or distribution.
The only effects from these increased pollutant loadings would be a
proportional decrease in actual numbers of individuals of all species.
10.2.2 Benthos
The benthic community in the proposed ocean outfall area is currently
dominated by high densities of surface-deposit feeders, to the exclusion of
other more pollution-intolerant species. The structure and density of this
existing benthic community suggests that the site is already organically
enriched. The effect of the large amounts of discharge on the benthic
community is predicted to be significant. The additional nutrient levels and
decreasing oxygen levels would exceed the assimilative capacity of the
community and would result in major structural and functional alterations in
the macrobenthos. These include major reductions in total density, species
richness, diversity and eveness. Pollution-sensitive species would be
greatly reduced or eliminated resulting in a shift to highly
pollution-tolerant species. Major effects are likely to appear in the
immediate 2.4 square mile area surrounding the diffuser, and moderate effects
would extend over a much larger area (166 square miles) of Massachusetts Bay.
10.2.3 Finfish/Lobsters
The proposed ocean outfall option is expected to negatively affect local
populations of finfish and lobster for a number of reasons. The anticipated
changes in the benthic community are expected to have a negative impact on
the finfish and lobster who feed on these benthic organisms. The resulting
-------
10-8
alterations in diversity and structure of the benthos will reduce the amount
of food which is available to the finfish and lobsters (Ennis, 1973) and thus
will reduce finfish and lobster population within the immediate zone of
intial dilution. This effect may extend over a much larger area of
Massachusetts Bay. Lobsters may be more negatively influenced than the
finfish by the increased organic loading from the discharger as was observed
near another wastewater discharge north of Boston Harbor (Tetra Tech, 1981).
A slight shift in the distribution and abundance of the finfish community
may also occur because of the increased amounts of organic loading. The
settling of these effluent solids is predicted to alter the substrate
composition of the site to one preferred by winter flounder. As a result, it
is expected that the winter flounder will replace other finfish species,
particularly the now-dominant yellow tail flounder.
The discharge into Massachusetts Bay will also contain toxic materials
including some heavy metals and PCBs. These toxic pollutants can affect
marine organisms in a number of ways. Acute exposure can lead to death,
while exposure to lower concentrations can induce sublethal effects such as
reduced survival of young, lowered resistance to disease and deleterious
changes in behavior. These sublethal, chronic concentrations can, in turn,
reduce species distribution and abundance.
The toxicity of certain heavy metals is influenced, however, by the
chemical form taken by the metal. Acute, short-term effects are more likely
to occur when the metals are in ionic form while chronic, long-term effects
are most likely to occur when metals are complexed in organic form and are
-------
10-9
relatively non-ionic. It is in this chemical state that the metals will
accumulate within body tissues and can be transferred to other organisms
through the food chain.
Bioaccumulation of toxic substances is even more likely to occur with
organic toxicants, such as certain types of pesticides and PCBs, because
their neutrally charged organic form allow a much easier passage across
cellular membranes. In addition, many of these organic compounds are very
resistant to degradation. As a result, these long-lasting residues will pass
through the food web, ending up in commercially and recreationally important
species of fish, and will be transferred to humans when these fish are
consumed.
The proposed ocean outfall option will remove about the same percentage
of metals, pesticides, PCBs and other toxic materials as does the existing
STP (see Table 2-3 in Section 2). This means that metals such as cadmium,
chromium, copper, lead, mercury and zinc will, at most, be reduced by 40
percent from their influent concentrations. Based on data collected near the
current Deer Island and Nut Island outfalls (US EPA, 1983) annual average
concentrations of three metals, copper mercury and silver, were found to
exceed EPA water quality criteria.— It was also found that PCBs were
appearing in the effluent at 19 to 320 times the EPA criterion. A study of
the toxic chemical concentrations in the tissues of lobster and winter
flounder near the discharges indicated that PCBs are bioaccuraulating in the
edible tissues of these species. It was shown, however, that the other
chemicals sampled—DDT, mercury, silver, cadmium, copper and lead--were not
a/ (See US EPA, 1983 and 45 Fed. Reg. 79318, November 38, 1980.)
-------
10-10
bioaccumulating in fish and lobster tissues, although this does not mean that
these organisms are otherwise not being negatively influenced by
concentrations in the water column.
Finally, the discharge from the ocean outfall is expected to contribute
to the problem of fin erosion in demersal fish. Although the exact cause of
fin erosion is not known, there is evidence to suggest that fish develop the
disease when they come into constant contact with contaminated sediments,
particularly those contaminated with PCBs (Sherwood, 1979; OS EPA, 1983).
There is evidence that current HOC discharges into Boston Harbor are
contributing to fin erosion, particularly in winter flounder, and thus it is
likely that the proposed discharge of effluent into Massachusetts will have a
similar negative effect in local fish populations.
10.2.4 Bidangered or Threatened Species
The ocean outfall option may adversely affect transient threatened or
endangered species which appear in or obtain nutrients from the waters of
Massachusetts Bay. The affected organisms include several species of whale,
and are listed below:
Blue Whale Balaenoptera musculus
Finback Whale B. physalus
Sei Whale B. borealis
linke Whale B. acutorostrata
Humpback Whale Megaptera noveanglias
Right Whale Bibalaena glacialis
Loggerhead Sea Turtle Caretta caretta
Leatherback Sea Turtle Dermochelys coriacea
Shortness Sturgeon Acipenser brevirostrum
American Peregrine Falcon Palco peregrinus anatum
-------
10-11
All of these species are migratory, particularly the whales who travel
from the Gulf of Maine down the coast to Delaware Bay and southward to
Georgia and Florida. Two endangered species, the Right and Humpback Whale,
and the threatened Fin Whale are known to feed in summer along the shoreline
areas of Massachusetts and Cape Cod Bay on their migration along the East
coast. Their food sources include fish, krill or related crustaceans, and
zooplankton, which, as discussed previously, are likely to be negatively
affected by the conventional pollutants or by toxic pollutants discharged
into Massachusetts Bay. Although it is impossible to quantify these effects
on these species of whale and on the other species, it is likely that heavy
metals and the organic toxics will have the most deleterious impacts on these
endangered/threatened organisms.
-------
10-12
References
Ennis, G.P., 1973. Pood, feeding and condition of lobsters, Homarus
Americanus, throughout the seasonal cycle in Bonavista Bay, Newfoundland,
Journal of Fisheries Research Board of Canada, 30:1905-1909.
Gosselink, J.G., E.P. Odum, R.M. Pope, 1973. The Value of the Tidal Marsh,
Center for Wetlands Research, Louisiana State University, Baton Rouge,
Louisiana.
Lynne, G.O., P. Conroy, F.J. Prochaska, 1981. Economic Valuation of Marsh
Areas for Marine Production Processes, Journal of Environmental Economics
and Management, Vol. 8, No. 2.
Metcalf and Eddy, September 13, 1979. Application for Modification of
Secondary Treatment Requirements for Its Deer Island and Nut Island
Effluent Discharges into Marine Waters, for the Metropolitan District
Commission, Boston, MA.
Odum, E.P., 1961. Fundamentals of Ecology, W.B. Saunders Co.
Sherwood, M.J., 1979. The Fish Erosion Syndrome, Coastal Water Research
Project Annual Report for 1978, W. Bascom (ed.) SCCWRP, El Segundo,
California.
Teal, J.M., 1962. Energy Flow in the Saltmarsh Ecosystem of Georgia,
Ecology 43:614.
Tetra Tech, Inc., 1980. Technical Evaluation of Deer Island and Nut Island
Treatment Plants Section 301(h) Application for Modification of Secondary
Treatment Requirements for Discharge into Marine Waters, Bellevue,
Washington.
Tetra Tech, Inc., 1981. Technical Evaluation of Lynn Wastewater Treatment
Plant Section 301 (h) Application for Modification of Requirements for
Discharge into Marine Waters, Bellevue, Washington.
U.S. Environmental Protection Agency, 1983. Analysis of the Section
301(h) Secondary Treatment Waiver Application for Boston Metropolitan
District Commission, Office of Marine Discharge Evaluation,
Washington, DC.
Welch, E.B., 1980. Ecological Effects of Wastewater, Cambridge University
Press, New York, New York.
Westman, Walter E., 1977. Row Much Are Nature's Services Worth,
Science 2: 960-964.
-------
Section 11
Secondary Effects
The benefits associated with the previously discussed pollution abatement
options which accrue from increases in recreational activity/ commercial
fishing and other activities/ are all primary benefits; that is/ they are
direct impacts of the proposed projects. Another type of benefit—secondary
benefits—measures the net increase in economic activity generated by the
direct impacts and indirectly attributable to the treatment alternatives.
Secondary benefits are added to the primary benefits of a pollution abatement
project only if there is widespread unemployment nationally or regionally and
only if it is expected that these unemployed resources would be used in the
economic activity thus generated. Otherwise/ it can be assumed that any
increased economic activity stimulated by the project would represent only a
transfer of productive resources from one use to another and would not be a
net benefit. The rules and procedures governing the inclusion of secondary
benefits are found in Section XI 2.11 of Water Resources Council/ "Economic
and Environmental Principles and Guidelines for Water and Related Land
Resources Implementation Studies" (1983).
These Principles and Guidelines state th t conceptually any employment of
otherwise unemployed resources that results from a project represents a
benefit but that difficulties in identification and measurement may preclude
any but those labor resources employed onsite in the construction of the
project be counted. For this case study/ the construction options have not
been sufficiently developed to categorize types of labor resources required.
Instead/ some of the other indirect employment categories are discussed.
-------
11-2
Since unemployment is often cyclical, secondary benefits may not accrue
to the proposed projects over the long-run unless structural unemployment
(unemployment unaffected by normal cyclical upturns in the economy) is
alleviated. A detailed labor market analysis is required to determine the
types of unemployed resources that exist and whether the mix of skills
required for the economic activity generated by the pollution abatement
options would use those resources. Even in a less than full-employment
economy, as is currently the case, some resources that would be employed to
meet the increase in economic activity would be transferred from other
productive uses either within the region or outside the region (e.g. outside
Massachusetts or New England). If this were the case, these effects,
although they might be very important to the region, would not represent net
benefits from a national perspective (unless structural unemployment was
affected, as mentioned before). This section, therefore, refers to the
indirect impacts attributable to the treatment alternatives as secondary
effects and presents a method for their valuation. Under certain conditions
these effects may be considered benefits but the labor market analysis
required for this determination is beyond the scope of this case study.
11.1 Methodology
Secondary effects can accrue to a region from increased activity in any
local industry. For example, additional wages are spent on food, clothes,
rent, etc. and increased business production requires additional purchases of
materials used in production. These purchases stimulate increased economic
activity. For every additional dollar of direct income or of total output
(sales) from the industry, a certain dollar amount of associated economic
activity is generated; these amounts are known as multipliers for that
-------
11-3
Industry and provide a way to estimate the economic value of secondary
effects. Multipliers for estimating increased economic activity in an area
usually cover three kinds of effects: direct, indirect and induced. Direct
effects are the changes in income to households resulting directly from the
changes in output of the industries of interest. Indirect effects are
additional economic activities stimulated by the direct impacts of the
project, i.e., changes in activity in all industries which supply goods and
services to the primary impact industries. Induced effects are those that
result when consumers adjust their consumption patterns in response to
changes in income. All three effects may be of interest in this case.
Two types of multipliers are used to estimate increased economic activity
generated by an industry. The output multiplier is used to compute the total
value of economic activity generated. Not all of this value remains in a
community or region, however (and, as discussed before, much of it may
represent a diversion of resources rather than a net gain). Some goods and
services purchased by businesses or by employees are produced locally and
*
others are produced outside the area. The income multiplier measures only
the portion of the economic activity generated which remains in an_area as
income to residents. For the purpose of measuring secondary benefits from
pollution abatement options, the best measure would be the output multiplier
a? we are interested in national welfare rather than regional effects.
To estimate the secondary effects which would accrue to the Boston Harbor
pollution abatement options, multipliers are used that have been estimated
from economic input-output analyses. Input-output models represent the
economy of an area and the transactions which occur among industries located
there. From such a model it is possible to estimate the effects of a change
-------
11-4
in one industry on all the other industries. The advantage of input-output
analysis over other methods of estimating multipliers is that it provides
both comprehensive and detailed coverage of the industries of interest.9/
The disadvantage of this and other methods is that only gross changes are
estimated; net effects exclusive of transfers of resources are not measured.
11.2 Benefit Estimates
The multipliers used to estimate secondary effects should correspond to
the type of data available on the impact of the pollution abatement options.
In this case, it is easier to estimate the impact on the output (sales) of an
of an affected industry (such as shellfishing or boating) than to estimate
the impact on direct income (wages). Thus, the multipliers shown in Table
11-1 estimate the total direct, indirect and induced effects of a one dollar
change in the sales of each impacted industry.^/
A range of multipliers has been included in Table 11-1. The multipliers
for the shellfishing and related industries come from three studies, one of
Cape Cod, one of the Southern Mew England Marine Region (SNEMR), including
Rhode Island, Cape Cod and parts of Southeastern Massachusetts and
Connecticut, and one of the State of Maine (Cape Cod Planning and Economic
— Other types of multipliers have been developed. For example, E.
Wong (1969) has estimated a multiplier for shellfish which computes the value
added by harvesters, wholesalers and retailers both inside and outside the
community. This kind of multiplier would not capture the indirect or induced
effects of the shellfish industry the way an input-output derived multiplier
would.
k-/ They could be converted for use with direct income impact data by
dividing by factors which show the effect on direct income of a one dollar
change in output for each industry.
-------
11-5
Table 11-1
Multipliers Showing Direct, Indirect and Induced Effects
Per $1 Change in Output
Cape Cod Study
SNEMR Study
Income
Income Output
Industry Multipliers Multipliers Multipliers Multipliers Multipliers
Wisconsin
Study
Output
Maine Study
Income
Commercial
Shellfishing 1.1749
Fish Processing
Clam and Worm
Processing
Shellfish Whole-
saling 1.0772
Seafood, Whole-
saling and
retail
Eating and
Drinking
Establishments .5158
Marinas and
Boatyards .6829
Charter Sport-
fishing .9038
TouristS/
3.0010
3.6444
2.0179
2.4971
2.8200
1.1441
.7027
.7781
.7997
.7037
.7982
1.54
1.65
2.2705
2.1741
Weighted average of impacts of tourist expenditures on all industries.
Sources: Briggs, Townsend and Wilson, 1982; Cape Cod Planning and Economic
Development Commission, 1978; Grigalunas and Ascari, 1982; Strang, 1971.
-------
11-6
Development Commission, 1978, Grigalunas and Ascari, 1982, and Briggs et al.,
1982) &
Both output and income multipliers are available from the Cape Cod study
while only income multipliers are available from the SNEMR and Maine
studies. As can be seen from the table, the Cape Cod output multipliers are
about three times greater than the income multipliers for the same study.
Although it was not possible to calculate output multipliers for the SNEMR or
Maine studies because of lack of data, the difference between income and
output multipliers would be less for these studies than for the Cape Cod
study. The reason for this is that both the State of Maine and the SNEMR
region are larger and more self-sufficient and would therefore retain more
earnings and import fewer goods and services.
There were no input-output analyses available for marine activities in
the Boston area. Since the structure of harvesters, wholesalers and
retailers of soft shelled clams in the Boston area is probably similar to
those of Maine, Cape Cod and the SNEMR, the multipliers presented in Table
11-1 can be used to provide a range of secondary effects estimates for the
pollution abatement options as shown in Table 11-2.
Although, as mentioned earlier, income multipliers measure only income
remaining in an area and, therefore, understate the total national welfare
impacts of the pollution abatement options, they are included as part of the
range in Table 11-2 for two reasons. First, Boston area output multipliers
£/ Multipliers from two other input-output analyses, an earlier SNEMR
study and a Rhode Island study, were presented in Grigalunas and Ascari.
Unfortunately, they were of the form that is multiplied by direct income
rather than by sales, and data were not available to convert them to the form
useable here. In the form that they were available, however, these
multipliers fell between the Cape Cod and SNEMR figures, and so would
probably lie within the range shown in Table 11-1.
-------
Table 11-2. Secondary Effects Estimates
(Thousands S1982)
Btlmated Change In Sales for Bich Multiplier
Industry Pollution Abatement Option (Thousands 19623) Ranqe
(SO .1 STP (3)
Dorchester, I Ocean Outfall
Constitution Nenoniet Oulncv 1 or Secondary
Cbranerclal
Shellflahlng
Harvesting 94.2 149.2 72.1 B8S.6 1.14-3.00
Distribution
and Pro-
cess Inq »/ 74.0 117.2 56.7 695. 8 0.70-3.64
Restaurants i/ 104.3 165.2 79.8 980.3 0.80-2.02
Subtotal 272.5 431.6 208.6 2,561.7
Recreation
Swimming £/ 103.3 704.7 540.1 111.6 0.80-2.27
Othec6/ 201.6
Subtotal!!/ 103.3 704.7 540.1 313.2
TOTAld/ 375.8 1,136.3 855.6 2.874.9
Secondary Effects Range
Pollution Abatement Option
CSO
Neponaet/
Constitution Dorchester
107.4- 282.6 170.1- 447.6
51.8- 269.4 82.0- 426.6
83.4- 210.7 132.2- 333.7
242.6- 762.7 384.3-1207.9
82.6- 234.5 563.8-1.600
for Each
(Thousand-)
1
1
Oulncv 1
82.2-226
39.7-206
63.8-161
353.8-583
432.1-
1,226.0
19823)
STP
Ocean Outfall
or Secoqrlarv
.3 1,009.6-
2.656.8
.4 487.1-
2.532.7
.2 784.2-
1.980.2
.9 2.280.9-
7,169.7
89.2-253.3
161.3-457.6
377.0-3642.5
82.6-234.5 563.8-1,600
352.2-997.2 948.1-
2,807.9
432.1-
1,226.0
785.9-
1,809.9
250.5-
710.9
2,531.4-
7.880.6
2/ Sales per bushel for Distribution and Processing and Restaurants assumed to maintain the same relation to harvest sales per bushel
for Boston Harbor as for Resources for Cape Ann study.
!>/ $1 per visitor-day assumed spent on food and beverages. Visitor days are average of upper and lower bounds for swimming from Table
6-6 and Cor 'other* from Table 6-12.
£/ Ten percent of boating and fishing benefits (see Section 6) assumed as sales for marinas and boatyards and for charter sportfIshlng,
respectively. Based on Table 6-10 (boating) and 6-11 (fishing).
H/ Not including fishing and boating sales and secondary effects.
-------
11-8
would probably be closer to Boston area income multipliers for the same
reasons as mentioned above for Maine and the SNEMR. Second, as discussed
above, even in a less than full-employment economy, some resources that would
be employed to meet the increases in economic activity generated by the
pollution abatement options would be transferred from other productive uses
and thus would not represent net benefits. A multiplier which underestimates
secondary effects is therefore appropriate.
Besides secondary effects generated from increased shellfish harvesting, a
certain level of economic activity may also be stimulated in the distribution
and processing and restaurant sectors for each additional bushel harvested.
To estimate these effects it was assumed that the level of sales generated in
the distribution and processing and restaurant industries as compared to the
harvesting industry would be the same for Boston Harbor as for the Cape Ann
area (see Resources for Cape Ann, 1982) and that this relationship would be
maintained across price changes.—'"-' Since Boston is a major market area
for shellfish, this is a conservative assumption. Secondary effects can
therefore be estimated for these two industries as well as for harvesting.
Recreation multipliers in Table 11-1 come from the Cape Cbd and SNEMR
studies and, for comparison purposes, from a study done for a county in
Wisconsin which has a significant tourist industry (Strang, 1971). This study
was included because there is no data available on sales generated by swimmers
-f Thus, at a price of $31.41 to the digger, for example, each bushel of
clams harvested would generate $90.86 of sales (total sales divided by number
of bushels harvested). Of this, $31.41 would be harvesting sales, $24.68
distribution and processing sales, and $34.77 restaurant sales. These per
bushel sales figures are multiplied by the increased harvest to estimate the
changes in sales shown in Table 11-2.
b/
~* The $31.41 per bushel harvest price and the other per bushel sales
figures given in footnote §./ are prices for 1980 from Resources for Cape
Ann, 1982, updated to 1982 prices using the soft shelled clams price index
from National Marine Fisheries Service, NOAA, 1982.
-------
11-9
at Bos tor. area beaches, while data from the study includes tourist receipts
in the input-output analysis. This study also developed a "tourist
multiplier" which is a weighted average of the impacts of tourist
expenditures on all industries.
The multipliers for Eating and Drinking Establishments, Marinas and
Boatyards and Charter Sportfishing from Table 11-1 were used to develop a
range of multiplers to estimate recreation secondary effects for the Boston
Harbor pollution abatement options, as shown in Table 11-2. The same
considerations discussed above for the shellfish multipliers, concerning the
inclusion of income multipliers and the assumption of comparability of
multipliers across study areas, hold for the recreation multipliers.
In order to compute secondary effects for recreation activity associated
with the pollution abatement options, a judgment was made that a maximum of
ten percent of boating sales could be applied to marinas and boatyards and
that ten percent of fishing sales could be applied to charter fishing.
Boating and fishing sales data come from Tables 6-10 and 6-11. Since there
were no data on expenditures by swimmers, it was assumed that one dollar
would be spent for each person-day and that it would most likely be spent for
food or beverages. Swimming visitor day data come from Table 6-6. Thus, the
Eating and Drinking Establishments multipliers from Table 11-1 were used to
develop a multiplier range in Table 11-2.
Table 11-3 compares multipliers which measure only indirect and induced
effects with those that also include direct income effects. These data were
only available for the SNEMR study, not for the Cape Ood study. It could be
-------
11-10
Table 11-3.
Comparison of Multipliers With and Without Direct
Effects per $1 Change in Output
I
SNEMR Study Income Multipliers
Industry
I Direct, Indirect
and Induced Effects
Indirect and
Induced Effects
I Indirect and
(Induced Effects as
I a Percentage of
I of Total Effects &
Commercial
Shellfishing
Fish Processing
1.1441
.7027
.4754
.5725
42
81
Seafood, Wholesale
and Retail
Eating and Drinking
Establishments
Marinas and
Boatyards
Charter Sport-
fishing
.7781
.7997
.7037
.7982
.6876
.4518
.3847
.4321
88
56
55
54
(Column 2 divided by column 1) x 100.
Source: Grigalunas and Ascari, 1982.
-------
11-11
argued that secondary effects should not include direct income effects. If
this were the case, then the shellfish harvesting secondary effects estimates
shown in liable 11-2 would be reduced by about 60 percent, the related shell-
fish industries by approximately 20 percent and the recreation activities by
around 50 percent. However, it does not appear that the direct income
effects would be double counting either the willingness to pay for improved
recreation experiences or the changes in producer or consumer surplus due to
increased shellfish harvest.
In evaluating the range of secondary effects estimated in Table 11-2, and
in addressing the question of whether and how much of the secondary effects
should be added to the primary benefits to derive the total benefits
associated with each pollution abatement option, the important consideration
is the level and type of unemployed resources assumed. If there is
widespread, long-term unemployment, then the full amount of the secondary
r
effects could be counted and the upper bounds in Table 11-2 used. If there
is a full employment economy, then secondary benefits would be either zero or
the difference between the value that the resources currently earn compared
to what they would earn if they were employed in activities stimulated by the
abatement option, if these values are different. As mentioned above, the
kind of detailed labor market analysis that would be required to estimate
this difference is beyond the scope of this study. If some unemployment
exists as is the present situation and if a labor market analysis showed that
it was likely to be long-term and composed of the skill levels required by
the economic activity generated, then the lower bounds in Table 11-2 may be
the best estimates to use and would represent a moderate benefit level.
-------
11-12
11.3 Limits of Analysis
The major problem in carrying out this analysis is determining whether
the secondary effects that can be estimated should be counted as benefits and
added to the primary benefits of the pollution abatement options. The data
are lacking to estimate the degree to which resources required for the
increased economic activity generated by the pollution abatement options
would be otherwise productively employed in the long run. Since we are
interested in estimating net benefits, transfers of resources already
occupied to activity stimulated by the pollution abatement options should not
be counted. Even given high unemployment as is the case in the current
recession, it is difficult to appropriately handle this problem.
Another limitation of this analysis of secondary effects is the lack of
an input-output model of marine related activities for the Boston area. A
related problem was the lack of data to compute output multipliers for the
SNEMR. The availability of these data would have produced a better range of
estimates of secondary effects.
-------
11-13
References
Briggs, Hugh, Ralph Ibwnsend and James Wilson, January 1982. An
Input-Output Analysis of Marine Fisheries in Marine Fisheries Review,
Vol. 44, No. 1.
Cape Cod Planning and Economic Development Commission, 1978. An Economic
Profile of the Cape and Islands Fisheries, Barnstable, Massachusetts.
Grigalunas, Thomas A. and Craig A. Ascari, Spring 1982. Estimation of
Income and Employment Multipliers for Marine-Related Activity in the
Southern New England Marine Region, Journal of the Northeastern
Agricultural Economics Council, Vol. XI, No. 1, pp. 25-34.
Resources for Cape Ann, April 1982. The Costs of Pollution; The Shellfish
Industry and the Effects of Coastal Water Pollution, Massachusetts
Audubon Society.
Strang, William A., 1971. Recreation and the Local Economy; Implications
for Economic and Resource Planning, Marine Technology Society,
Washington, D.C.
U.S. EPA, 1982. Benefit-Cost Assessment Handbook for Water Quality Programs,
Draft, Economic Analysis Division, Washington, D.C.
U.S. Water Resources Council, December 14, 1979. NED Benefit Evaluation
Procedure: Unemployed or Underemployed Labor Resources, Section 713.1201
of Procedures for Evaluation of National Economic Development Benefits
and Costs in Water Resources Planning (44 PR 72892).
Wong, Edward P.M., October 1969. A Multiplier for Computing the Value of
Shellfish, U.S. Department of the Interior, Federal Water Pollution
Control Administration, Needham, Massachusetts.
-------
Section 12
Charles River Basin Benefits
The Charles River Basin has been designated by the MDC as one of the four
CSO planning areas. The Charles River Basin includes the Back Bay Fens, the
Muddy River, Alewife Brook and the Charles River itself. The basin is mixed
fresh and salt water and is used primarily for non-contact recreation, both
on the water and at the water's edge. There is little or no fishing in the
Charles River Basin. The Charles River is the major water resource in the
Charles River Basin and draws the greatest number of recreators. For this
reason, as well as data limitations, we have chosen to estimate benefits only
for the Charles River.
12.1 The Charles River
The Charles River is 80 miles long/ with a watershed of 300 square
miles. The portion of the Charles that is contained in the Charles River
Basin CSO planning area runs from the Watertown Dam to the Charles River Dam
near the mouth of Boston Harbor (see Figure 12-1). This section of the River
has an average annual level of 2.38 feet, and contains approximately 675
surface acres of water. The length of the River within this stretch is 8.6
miles. The Charles River is an important water-based recreation resource,
especially to the towns through which it flows. Although there is currently
little swimming in the river (and none predicted with the proposed CSO
plans), the river plays host to a variety of boaters. Sailing and
motorboating are extremely popular, especially at the wider portions of the
river, near the Harbor. There is also a significant number of people who
-------
*'./ OHtLSEA (f -x'~»
Figure 12-1. Map of
Charles River Basin
Charles River
Basin Planning
Area
t-)
i
1 !
-------
12-3
scuii on che Charles. Every major college and university in the Boston
area has a boat house along the river; their crew members practice almost
daily during the spring and fall months. The river is also an aesthetic
focal point for other recreation-based activities. An MDC bikeway
follows the course of the river and doubles as a running path.
Picnickers, sunbathers, and strollers also take advantage of the open
space provided by the river. Major cultural events such as crew
regattas, formal and informal concerts, and city festivals take place
along the river's edge and attract thousands of residents and
sight-seers.
The Charles River violates the state water quality standards. Those
standards (a rating of "C") allow non-contact recreational use. The
river is polluted with extremely high levels of coliform counts, odors,
floatables, debris, and turbidity. The recommended CSO plan (see Section
3.5) includes capturing, transporting, and storing overflow from the CSOs
and is predicted to result in 50 to 80 percent removal of suspended and
floatable solids, coliforms and BOD_. water quality will improve
greatly although swimming will still not be permitted.
It is difficult to quantify the instream and near-stream user
benefits to be gained from improving the water quality in the Charles
because of data and methodological limitations. Unlike swimming
benefits, there are no good travel models or data available to predict
how user participation and utility will increase. There are also few
intrinsic value studies which are applicable to the Charles River area.
He have chosen two techniques and two studies to evaluate user and
-------
12-4
non-user benefits from abating pollution along the Charles River. User
benefits to boaters are estimated using a boating participation model
developed by Davidson et al. (1966) while both intrinsic and user
benefits are developed by applying results from a contingent valuation
survey (RTI, 1983).
12.2 Boating
The effect of water quality on the level of recreational boating has
been studied. The results of the Davidson et al. study (1966) show that
the number of participants within a given population as well as the
number of days of boating participation per year show significant
increases with improvement in the quantity and quality of available
waters. Davidson's approach to estimating boating-related benefits
includes calculating (a) the change in the probability of boating
participation among the general population as a result of improvement in
water quality and availability and (b) the change in number of days of
participation per year. The Davidson model attributes most of the
benefits of water quality improvement to new participants. It does not
capture any benefits accruing to current boaters. The Davidson model
estimated, in a study of the Delaware Estuary, that each increase in
recreational boating water of one acre per capita resulted in a 38
percent increase in participa- tion rates (i.e., the probability of an
individual participating in boating increased by 38 percent). The
portion of the function describing boating participation, which is
applicable to this study, can be expressed in the following reduced form:
-------
12-5
BP = 0.38485(AW) + 0.03142( * FPS)
where: BP is the probability of boating participation
w is the per capita acreage of recreational water available
FP5 is the recreational facility rating.
Hie PPS variable represents an index of the quality of boating facilities. A
rating of '!' implies "no facilities," while a rating of '5' suggests "very
good facilities. — Socioeconomic variables were included in the regres-
sion, including education, income, occupation, age, and race, but were not
well correlated with boating participation. Davidson et al. also assumed
that elimination of pollution discharges into the Delaware Estuary would
produce a minimal one point improvement (from 2.0 to 3.0) in the PPS rating.
12.2.1 Methodology
It is possible to apply this model to the Charles River. Estimation of
boating-related benefits involves the following steps:
a. Estimate the increase in recreational boating water and boating
facilities from improving water quality in the Charles River as a
result of implementation of the CSO plan;
b. Estimate the change in the probability of boating participation in
the general population as a result of improvement in water quality
and availability of boating facilities;
c. Estimate change in total participation attributable to water quality
improvements;
d. Estimate the value of the additional boating days.
ttie first step involves estimating the increase in recreation boating
water ( &W) and facilities (AFPS) as a result of improving water quality.
Although AW is the key explanatory variable in the Davidson equation,
I Davidson et al. used fishing facilities rather than boating facilities
because the former were not available for their sample area.
-------
12-6
the value of the variable is quite small for the Charles River. The Charles
River has only 675 acres of water available for boating of all kinds.
Although the Charles is polluted, there appear to be few portions of the
river which are unboatable because of pollution. Therefore, the change in
acreage of recreational water available per capita following water quality
improvements is essentially zero. Although this assumption of zero change in
water acreage might appear to be too conservative, even if we were to assume
that all 675 acres of the river were previously unboatable, a AW of 675
acres would only lead to a very small per capita acreage increase of from
0.0317 to 0.0318.-^ it is, therefore, apparent that the variable FPS will
have the greatest effect on predicting the change in boating participation.
Davidson e_t^ a_l^ assumed that eliminating pollutant discharges into the
Delaware estuary would produce a minimal one point improvement in the
recreational facilities from a rating of *2m to a rating of '3.' The same
assumption was used for the Charles River, that &FPS is 1.
Calculating the total additional boating days requires information on
current boating use of the Charles River. As described in the swimming sec-
tion, recreation statistics on attendance and days per participant are not
officially recorded by the MDC. We have, therefore, used a number of sources
to estimate a range of boating participation on the Charles. Information
from a study by Binkley and Banemann (1975) indicates that 850 visits were
made to two sites along the Charles River during the summer season, and that
5.6 percent of the visits were boating-related. Results from the study
a Depending on a range of 183,000 - 1,680,800, boating participants as
described in Appendix E.
-------
12-7
suggest that it is correct to assume that the survey sample was statistically
representative of the entire Boston SMSA. These 850 survey visits can be
extrapolated ip.to 63,000 family visitor days and approximately 183,000
visitor days (see Appendix E) . This is probably an understated estimate
because only two sections of the entire length of the Charles River were
sampled.
An alternative method is to apply the approach used in the previously-
described swimming section which is based on regional recreation studies.
This method assumes that (1) 40 percent of the population goes boating, (2) a
user population of 764,000 (see Appendix E for details), and (3) users go
boating an average of 5.5 days per year. The resulting boating days are
1,680,800. The Binkley-based estimate of 183,000 vistor days is used as a
lower bound, and the recreation study-based estimate of 1,680,800 is used as
an upper bound. The lower bound estimate appears to be the more reasonable.
Additional boating days can be estimated multiplying the previously
derived AW and AFPS value by the estimated number of general population
boaters (see Appendix E for details). The increase in visitor days ranges
from 5,750 to 52,810.
12.2.2 Benefit Estimates
Boating benefits from improved water quality resulting from implementa-
tion of CSO plans can be estimated by valuing the increase in visitor days
developed and described above. The range of user day values that have been
-------
12-8
developed for boating are presented in Appendix B (Table B-l)^. By
applying this range of values ($9.27-$18.14) to the projected increase in
boating days, we can arrive at an estimate of boating benefits, presented in
'Sable 12-1.
Table 12-1. Annual Recreational Boating Benefits
I I Total Annual
(1982$) I Number of Additional I Boating Benefits
$/Boating Day I Boating Days I (Thousands 19823)
High 18.14
low 9.27
52,810
5,750
958
53
Boating-related benefits from improving water quality on the Charles
River are modest because the estimated increase in number of boating days is
small and because boating day values which are applicable to this study
represent the lower, rather than upper, end of the range of user-day values.
12.2.3 Limits of Analysis
Calculation of boating-related benefits is limited by the methodology
employed, the data base, and the numerous assumptions made. The application
of the Davidson e_t al. boating model may lead to biased benefit estimates.
First, the model only measures benefits which accrue to new participants and
does not capture benefits of increased participation or increase in utility
*/ We have chosen not to use the boating value of $45.19 derived from the
MPA in conjunction with Charbonneau and Hay because we believe that it over-
states the particular value of boating on the Charles River. This is because
the greater portion of boaters who use the river do so in small-powered craft
(such as sculling shells, kayaks, small sailboats, canoes, and low horsepower
motor boats), rather than large-powered craft.
-------
12-9
to existing users. Second, the model may not, for a number of reasons,
be easily applied to an urban area. The key explanatory variable in the
model is the supply of boatsble water that is expected to increase
following water quality improvement. In the case of the Charles River,
the value of this variable is extremely small because virtually all 675
acres of river are currently used for boating. Even assuming that all
acres were previously unbeatable, the increase of 675 acres would only
lead to an increase of 0.0317 acres per capita and, therefore, would
account for only an 0.012 change in boating participation. The second
variable in the model—change in recreational facility rating—then
becomes the key explanatory variable of the increase in boating
participation. There are few places along the urbanized riverfront of
the diaries available for development or expansion of marinas and, thus,
we have assumed that the one point change in facility rating reflects the
improvement in boating facilities. This assumption, however, is
difficult to verify.
Other problems with estimating boating benefits from CSO pollution
control plans along the Charles lie in the available recreational data.
There is scant information about days of boating participation along the
Charles and the percentage of the entire population in the Boston
Metropolitr \ area who boat there. The use of user-day values is also
likely to bias the benefits estimates. The lower range of available user
day boating values ($9-318/day) was used to calculate benefits because of
the nature of boating (in non-motorized and small-powered craft) on the
river.
-------
12-10
12.3 Intrinsic (Non-User) and User Benefits
An alternative method for computing the benefits from CSO pollution control
plans on the Charles River is to apply the results of a contingent valuation
survey, which captures the amount users and non-users are willing to pay for
improved water quality. As mentioned previously, the Charles River is a major
aesthetic focal point for recreation-based activities. It is difficult to
estimate the exact number of people who are not direct users of the River but
who, instead, ride, picnic, run, or stroll along the Charles' shores. It is
safe to assume, however, that there are probably few families or individuals in
the towns through which the river runs who have not enjoyed the river at least
once. Calculating benefits which accrue to these "non-users" is a necessary
part of developing total benefits.^/
We have chosen the results of a contingent valuation survey described in
detail in RTI (1983) to capture instream, near-stream and intrinsic benefits
from improving water quality through upgrading CSO's in the Charles River
Basin area. A study conducted by Gramlich (1974) to determine the
willingness to pay for improving water quality in the entire 80 mile length
of the Charles River was not considered applicable here, because the survey
only recorded results for willingness to pay for obtaining a swimmable level
of water quality (classification "B"). The CSO plans and their costs have
been developed only for improving the river to a level "C," or boatable use.
Also, the results from the Gramlich study cannot be disaggregated by user and
non-user.
£/ For a discussion of non-user (Intrinsic) values and estimation
methodology, see Section 9.
-------
12-11
12.3.1 Benefit Methodology and Estimates
Estimates of willingness to pay for improving Charles River water quality
can be derived by applying the results of a study conducted by RTI, along the
Monongahela River in Western Pennsylvania. The RTI study used a contingent
valuation approach to measure willingness to pay for improved water quality.
Results from the RTI study suggest that user and non-user households are
willing to pay $18.68 (1982$) for water to go from boatable to fishable
conditions.£/ In order to calculate total benefits, it is necessary to
multiply this dollar WTP value per household times the regional household
population.
For the Charles River area, an upper bound was established by including
residents of towns bordering or very close to the Charles River: Cambridge
(95,000), Somerville (77,000), Watertown (34,000), Newton (83,000), Brookline
(55,000), Boston (560,0000) or a total of 905,000.=/ Assuming an average
household size of 2.69,-' an upper bound household population figure is
calculated to be 336,000. A lower bound can be developed by assuming that
only one half the populations of these towns benefit from CSC—based water
quality improvements, or 452,000 people, which translates to a lower bound of
168,000 households. Multiplying the RTI-derived WTP values of $18.68 by the
range of applicable households results in significant benefits, presented in
Table 12-2.
£/ This is based on a direct question framework, users and non-users.
See page 4-32, RTI, 1983.
£/ Based on data from 1980 Census.
-------
12-12
Table 12-2. Annual Estimated Willingness to Pay
for Fishable Charles River (1982$)
High
Low
1
1
Population I
905,000
452,000
Persons
per
Household
2.69
2.69
1 Willingness 1
1 to Pay 1
1 Value 1
18.68
18.68
Annual
Willingness
to Pay Value
6.28 million
3.14 million
12.3.2 Limits of Analysis
Benefits to instream users, near-stream users and non-users of the
Charles River are substantial. These results should be interpreted wih cau-
tion, however, for a number of reasons. The accuracy of benefit values is
constrained by use of off-the-shelf models. The willingness to pay values
used here are derived from a study area which may be sociologically, econo-
mically and educationally different from the population within the Charles
River Basin planning area. People in the northeast, for example, recreate
more often than those in the central regions of the east (1979 Survey of
Recreation). The Charles River population is also more highly educated and
has higher income on average than that in the Monongahela study area. The
geographical nature of the two areas is also different. The Monongahela
River, and the region surrounding it, are larger and much more rural than the
Charles River and its study area. The urban settinq of the Charles, the
relative scarcity of other closeby recreational rivers, and the previously
mentioned socio-economic differences suggest that the Charles River popula-
tion in Cambridge and other towns might be willing to pay a higher price for
river cleanup. Benefits are also understated because consumer surplus was
estimated only for the Charles River portion of the Charles River Basin CSO
plan; the 'methodology therefore does not capture benefits accruing to recrea-
tionists in the Back Back Fens, the Muddy River or Alewife Brook. The upper
-------
12-13
bound figure of $6.3 million is probably the more reliable estimate of total
benefits.
12.4 Summa ry
The benefits of improving the water in the Charles River in the CSO
Charles River Basin Planning Area are many. Benefits accrue to instream
users (boaters) and near-stream users (picnickers, strollers, bikers/ etc.)
alike. Best annual boating benefit estimates total $958,000 and probably
understate all boating benefits. Results from a contingent valuation survey
capture both user and non-user benefits by applying willingness to pay values
derived from a study of the Monongahela River. The upper value of $6.3
million is probably the more reliable estimate of total benefits from
improving water in the Charles River, although this figure may also under-
state all benefits.
-------
12-14
References
Binkley, C. and W. Hanemann, 1975. The Recreation Benefits of Water Quality
Improvement. U.S. Environmental Protection Agency, NTIS PB 257719.
Charbonneau, J. and J. Hay, 1978. Determinants and Economic Values of
Bunting and Fishing. Paper presented at the 43rd North American Wildlife
and Natural Resources Conference, Phoenix, Arizona, Washington, D.C.
Davidson, P., G. Adams, and J. Seneca, 1966. The Social Value of Water
Recreational Facilities from an Improvement in Water Quality: the
Delaware Estuary. Water Research, Allen Kneese and Stephen C. Smith,
eds. Baltimore: Johns Hopkins University Press for Resources for the
Future.
Gramlich, F.W., 1977. The Demand for Clean Water: the Case of the Charles
River. National Tax Journal 30:183-184.
National Planning Association, 1975. Water-Related Recreation Benefits
Resulting from Public Law 92-500, three volumes, Washington, D.C.
National Committee on Water Quality.
Research Triangle Institute, 1983. A Comparison of Alternative Approaches
for Estimating Recreation and Related Benefits of Water Quality
Improvements, Research Triangle Park, North Carolina.
United States Department of Commerce, Bureau of the Census, 1982. 1980
Census of Population and Housing, Masachusetts, USGPO, Washington, D.C.
U.S. Department of the Interior, Fish and Wildlife Service and U.S.
Department of Commerce, Bureau of the Census, 1982. 1980 National Survey
of Fishing, Hunting and Wildlife Associated Recreation. USGPO,
Washington, D.C.
U.S. Department of the Interior and Heritage Conservation and Recreation
Service, 1979. The Third Nationwide Outdoor Recreation Plan, USGPO,
Washington, DC.
-------
APPENDICES
-------
Appendix A
Gbrrelating STP Performance and Operation
to Boston Harbor Water Quality
One of the tasks en route to a cost-benefit analysis is to design and
cost the technical options capable of modifying, maintaining, or raising the
quality of the environment in question. The complement to this task is to
collect data on the biological, physical, and chemical parameters of the
current environment so that the changes to the environment made possible by
the different technical options can be quantified. Neither of these tasks
was performed by Meta Systems. Instead, technical and environmental data
were collected from existing sources; no new information research (i.e.,
engineering analysis or environmental monitoring) was undertaken.
Correlating STP operations and performance with the water quality of the
harbor is a complicated task. The problem is not so much that the necessary
data does not exist at all but rather that the available information may not
be collected in forms or manners suited to particular analysis goals. Water
quality data shortcomings are the result of less than optimal sampling
procedures such as:
o infrequent monitoring;
o parameter selections not consistent from one sampling to
next;
o same locations not repeatedly sampled; and
o sampling not co-ordinated with seasonal, weather-related,
tidal, STP, etc. events.
In the available reports the performance information presented for
STP options differs from that presented for CSO options. The
performance of the STPs under the various options is measured in terms^
of effluent constituent concentrations. CSO plans, on the other hand,
-------
A-2
estimate the water qualities achievable under various CSO designs as
well as reduced loadings. In order to establish potential water
qualities achievable under the different STP options it was necessary
to describe the dispersion of STP effluents throughout the harbor.
A.I Influent, Effluent, and Sludge Characteristics
Periodically, the MDC takes samples of STP influent and effluent and
conducts tests to determine the composition of raw and treated municipal
wastewaters. (See Table A-l.) Using the concentration information from this
testing, along with values for total flow volume, the pollutant loadings to
the harbor due to the Deer and Nut Islands' STP options, have been
calculated. The combined loadings are presented in Section 2, Table 2-1.
The knowledge of influent composition enables calculation of the loadings
from raw sewage discharges due to STP bypasses.
To calculate annual loadings from influent and effluent concentrations
and flow volume data:
1) milligrams per liter was converted to pounds per gallon using
(8.4 x 10~6)(mg/1) = Ibs/gal
2) the combined effluent discharge volume of Deer and Nut Islands
was assumed equal to 500 million gallons per day (350 and 150
mgd, respectively)
3) concentrations for the individual STPs were weighted by volume
of flow for a combined average concentration equal to
(0.3) x (cone, at Nut Island) + (0.7) x (cone, at Deer Island)
4) annual loading: (365 days) x (500 mgd) x (combined average
concentration)
Bypass loadings were calculated from:
1) influent (i.e., raw wastewater) composition; use of this data
probably results in an overstatement of pollutant loadings
-------
A-3
Table A-l.
MDC Treatment Facilities Current Pollutant
Removals for Wastewater Effluents
1
1
Pollutant 1
1
BOD5
TSS
Cadmium
Chromium
Copper
Lead
Mercury
Nickel
Zinc .
NUT ISLAND
Influent 1 Effluent 1
(mg/1) 1 (mg/1) 1
136.6
178.3
0.0176
0.051
0.618
0.104
0.00199
0.889
0.431 .
97.0
82.0
0.0119
0.041
0.292
0.074
0.00120
0.291
0.376 .
1
Remova 1 1
(Percent)
29.0
54.0
32.4
19.6
52.8
28.8
39.7
67.3
12.8
I DEER ISLAND
1
1 Influent 1 Effluent 1 Removal
1 (mg/1) 1 (mg/1) 1 (Percent)
150.0
155.5
0.021
0.147
0.246
0.157
0.00124
0.115
1 " 1
108.0
70.0
0.019
0.108
0.357
0.131
0.0011
0.131
0.488 .
28.0
55.0
9.5
26.5
-45.1 -/
16.6
11.3
-13.9 a/
37.2
Source: The BOD5 and TSS values are from Metcalf and Eddy, June 1982. The toxic
metals data are from US EPA (1983) Table 3.2-6 and are for the period December 1975
through September 1977.
—/ The negative value of this removal percentage may be due to 1) random sampling
error or 2) a propensity of the Deer Island's treatment process to concentrate this
metal in the effluent rather than in the sludge.
-------
A-4
since bypasses are often associated with storm events, thereby
diluting the raw wastewater.
2) bypass volume estimates:
o for Nut Is land: a./
Recorded untreated discharges to Boston Harbor
January-August, 1982—2.1 billion gallons over 50 days (0.042
billion gallons/day);
Spills of unknown amounts January-August, 1982—4 spills
over 8 days estimated at 0.34 billion gallons (8.0 x 0.042);
Total for January-August, 1982 = 2.44 billion gallons
(0.305 billion gallons/month);
Estimated annual loading = 3.66 billion gallons.
o for Deer Island:^./
Recorded untreated discharges to Boston Harbor
January-October, 1982—2.2 billion gallons
(0.22 billion gallons/month);
No spills of unknown amounts;
Estimated annual loading = 2.64 billion gallons.
o for Moon Island:^/
Estimated annual loading = .258 billion gallons.
Heavy metal loadings to the harbor from STP sludges were available from
the draft report by Environmental Research and Technology (1978).
A.2 Pollutant Transport from STP Outfalls
To assess the impact of STP discharges in Boston Harbor, it is important
to know how STP discharges are dispersed throughout the Harbor. Since
discharges to the Harbor are subject to diverse and variable conditions, the
water quality throughout the harbor is not uniform. Variations in quality
t
—' Calculations based on bypass data from Dumanowski (1982).
£/ Moore (1980).
-------
A-5
may be attributed to bottom topography, currents (directions and magnitudes)/
wind, and the location and means by which pollutants enter the Harbor. STP
discharge dispersion is not easily correlated with the water quality of the
Harbor. In order to understand the environmental consequences of STP
discharges, information is needed about:
o transport of STP loadings via water movements (current speeds,
volumes of flow, flow patterns, etc.);
o physical and chemical interactions of STP effluent and sludge with
the Harbor's waters (decay rates, settling rates, chemical reactions
which might neutralize toxics, chemical recombinations, how
pollutants get cycled through the aquatic environment, rates of
stabilization) .
o biological aspects of loadings (tolerance of aquatic organisms to
loadings, pollutant uptake by aquatic organisms).
One form of water quality information available for Boston Harbor can be
called "static data," which refers to measurements of ambient water quality
at a specific time and location. Water quality information which describes
changes in quality over time and the interactions between various elements of
the harbor (physical, chemical, biological) contributes to a dynamic
understanding of the Harbor's water quality. The problems with the static
data available for Boston Harbor could be alleviated with more regular,
extensive data collection and water quality measurement procedures. For
dynamic information, however, the complexity of the harbor environment makes
it extremely difficult to understand all interactions and interrelationships
among its elements.
Static measurements ("grab samples") of pollutant parameters represent
the contemporary environmental status of the harbor but do not clearly
reflect the impacts of STP discharges in particular. Not all of the
pollutant deposits in the Harbor are from the Deer and Nut Island's STPs.
-------
A-6
Tests of harbor waters and sediments cannot distinguish among pollutants
whose source is STP discharges, those deposited prior to STP operations, or
those that were overflowed from a combined sewer. Not enough data has been
collected to make definitive conclusions regarding discharges and their
ultimate destinations. Such conclusions require a more rigorous sampling
endeavor (periodic sampling, extensive coverage of harbor) and that water
quality sampling be scheduled in conjunction with sampling of STP and CSO
effluent to the harbor in order to correlate variations in discharges with
variations in measured qualities throughout the Harbor.
Without historical information to demonstrate dispersion phenomenon of
STP discharges within Boston Harbor, a predictive model of dispersion
dynamics is of interest to this case study because it can help to describe
what the future impacts of a number of STP options might be. Models of
dispersion dynamics are perhaps the best means of determining what will
happen to the effluent once it is discharged from an STP since available
empirical information is insufficient for this task. A few models have been
developed to quantitatively explain some aspect of the Harbor which, due to
physical or economic constraints, cannot be adequately analyzed with static
measurements. One model designed specifically to quantify the dispersion of
STP discharges into Boston Harbor is the DISPER model, developed at MIT. It
largely relies on water movement (currents) infc mation to describe
dispersion. DISPER itself is based on CAFE, another MIT-developed program
which models these water movements. DISPER has several positive qualities
which suggest that it be used as a reference. Most important is that it was
designed specially for Boston Harbor. Its output also seems to correlate
with the relative pollution concentrations measured throughout the Harbor.
t
However, this may only mean that the developers of the model fit it to the
existing situation, and thus it is descriptive but not necessarily predictive.
-------
A-7
DISFER's greatest strength lies in its ability to predict volumetric
inflows and outflows from the harbor area (across a specified, but imaginary,
boundary). The model's next strongest capacity is its ability to predict
water movement patterns (directions and magnitudes of flow). CAFE is largely
responsible for these phases of the modeling effort. How STP effluent
disperses through the harbor is the task addressed by DISPER. Whether
pollutant loadings move exactly as does the water is unknown because
settlement and decomposition in transport, propensities of marine organisms
to assimilate wastes, etc., are not precisely understood.
The impact of, for example, the ocean outfall diffuser is assessed using
a conservative solute and BOD, a substance which decays at a first order
rate. For the conservative substance, decay and settling rates and
concentrations along the ocean boundary are assumed to be zero. The source
loading is input continuously and steady-state concentrations are computed.
No other sources or sinks are modeled. The results of this modeling effort
included contour lines of constant dilution and concentrations of ultimate
BOD as incremental additions from the treatment option being modeled.
Model results available to Meta Systems for review were run by Metcalf
and Eddy. (A sample of Metcalf and Eddy's DISPER output is shown in Figure
A-l). Metcalf and Eddy suggest that their assumptions tend to be
conservative (i.e., decay rate = zero, settling rate = zero).
The predicted water quality impacts due to the various STP treatment
options presented in Section 4 of the main report were made through
comparisons of the following types of information, often in the form of
mappings:
-------
A-8
Figure A-l. Example of DISPER Output
'>v-- -.:...:
«
5000 0
L_ __. ._..-.
SCALE IN
" .-"./?•'
!-^ y
iv; /
5000 ^ S '
- . -1 ,V. / • •
FEET yS
• CONCENTRATIONS OF ULTIMATE BOD FROM
TREATMENT PLANT EFFLUENT EXISTING CONDITIONS
Source: Metcalf & Eddy (1982), Figure 3-17.
-------
A-9
c effluent pollutant concentrations;
o dispersion model output (DISPER); and
o water quality at various receptors (beaches, recreational
areas, shellfish beds).
The receptor site, Brewsters Islands, is provided as an example of the
way the calculations of percentage pollution reduction in Tables 4-2 and 4-3
of the main report were done.
(1) Current Ambient Water Quality at Brewsters Islands
Fecal coliform (MPN/100 ml) 10
BOD5 (mg/1) 1
TSS (mg/1) 10-20
Source: Maps from Region I, EPA, Boston Harbor Data Management
System, December 1983.
(2) Existing Concentration of Effluent
Deer Island Nut Island
Fecal coliform 1500 1500
BOD5 127.6 105
TSS 121 110
Source: See Table 4-1, Section 4 of main report
(3) Existing Outfall Dilution Ratio 500 500 (at Brewsters Island)
Source: See Table 4-1, Section 4 of main report
(4) Existing STP Incremental Contribution (Deer and Nut combined)
at Brewster Islands
Fecal coliform 6
BOD5 .47
TSS .46
Source: Effluent concentrations (2) divided by dilution ratios (3)
summed for both Deer and Nut Islands.
(5) Portion of Ambient Water Quality not Due to Existing STP
Fecal coliform 4
BOD5 .53
TSS 9.5-19.5
Source: Current ambient water quality (1) minus STP contribution (4).
-------
A-10
(6) Effluent concentrations
Ocean Outfall Secondary Treatment
Fecal coliform 1500 1500
BODg 115 30
TSS 86 30
Source: See Table 4-1, Section 4 of main report.
(7) Dilution Ratio at Brewsters Islands
200 500
Source: See Table 4-1, Section 4 of main report. Obtained from
DISPER contour maps.
(8) Incremental Contribution from Treatment Option (at Brewsters Islands)
Fecal coliform 7.5 3
BODs .57 .06
TSS .43 .06
Source: Effluent concentration (6) divided by dilution ratio (7) .
(9) ftnbient Water Quality with Treatment Option (at Brewsters Islands)
Fecal coliform 11.5 7
BODs 1.11 .6
TSS 10-20 9.6-19.6
Source: Portion of ambient quality not due to existing STP (5) plus
incremental contribution (8).
(10) Percentage Change in Water Quality (+ improvement / - degradation)
Eecal coliform -15 +30
BOI^ -11 +40
TSS 0 +2 to +4
Source: Current ambient quality (1) minus ambient quality with
treatment option (9) divided by (1).
-------
A-ll
References
Dumanowski, Diane, 1982. The Boston Globe, December 19, 20, and 21,
Boston, MA.
EcolSciences, March 1979. Environmental Impact Statement; MDC Proposed
Sludge Management Plan, Metropolitan District Commission, Boston, MA, for
Environmental Protection Agency, Region I, Boston, MA.
Environmental Research and Technology, 1978. Draft Report for the National
Science Foundation. C-PRA77-15337.
Metcalf & Eddy, Inc., September 13, 1979. Application for Modification of
Secondary Treatment Requirements for its Deer Island and Nut Island
Effluent Discharges into Marine Waters, for Metropolitan District
Commission, Boston, MA.
Metcalf & Eddy, Inc., June 1982. Nut Island Wastewater Treatment Plant
Facilities Planning Project, Phase I, Site Options Study, for
Metropolitan District Commission, Boston, MA.
Moore and Associates, inc., H.E., September 1980. wollaston Beach
Exploration/Remedial Program Regarding Storm Water Contamination,
Boston, MA.
-------
Appendix B
Recreation Benefit computations
B.I Seasonal Swimming—Increased participation
Increased participation in swimming due to pollution abatement control was
calculated from current swimming participation and estimated unmmet demand.
The example below is for one pollution control option (CSO controls) for the
swimming beaches in the study area.
Increase in
Participation
from Pollution =
Control
Beach
Constitution
Dorchester
Wollaston
Quincy
Weymouth
Hingham
113,750
236,000
1,100,000
63,560
0
0
0
(A)
% Pollution
Abatement
(CSO)
70
80
80
80
0
0
0
(B)
Increased
x Demand
(User days)
162,500
295,000
1,375,000
79,450
52,910
11,100
33,000
(A) Source: Section 4 of main report.
(1) (2)
entire SMS A x Unmet Demand i
User Days
(B)
Beach
Constitution
Dorchester
Wollaston
Qu incy
Weymouth
Hi ngham
Hull
Increase Demand
(user days)
146,250-178,750
avg = 162,500
265,500-324,500
avg = 295,000
1,237,500-1,512,
avg = 1,375,000
71,505-87,395
avg - 79,450
47,619-58,201
avg = 52,910
9,990-12,210
avg = 11, 100
29,700-36,300
= Proportion of entire SMS A
swimming usage supplied by beach
.034
.062
500 .291
.017
.011
.002
.007
avg = 33,000
4,258,801-5,199,090
4,258,801-5,199,090
4,253,801-5,199,090
4,253,801-5,199,090
4,253,801-5,199,090
4,253,801-5,199,090
4,253,801-5,199,090
-------
B-2
(1) Calculation of Proportion of Entire SMSA Swimming usage Supplied by Beach
Constitution Beach is used as an example. Figures for all other beaches
calculated in identical fashion.
(a) (b)
Proportion of m Total
SMSA Swimming Use = Beach - Seasonal SMSA
Supplied by Beach Attendance • Attendance
Constitution .034 = 325,000 9,452,892
(a) Source: Metropolitan District Commission and municipalities.
(b) Source: See 2b below.
(2) Unmet Demand in User Days
(a) (b)
Unmet user = Percent unmet x participation in swimming
days demand for days per year
swimming in SMSA
4,253,801-5,199,090 45-55 9,454,892
(a) Source: Department of Environmental Management, Massachusetts Outdoors (SCORPj
1976 and discussions with cities and towns.
(b) (i) (ii) (iii)
Swimming = population x proportion x average i day
Participants SMSA participating in swimming trips
9,452,892 2,763,357 .32 10.69
(i) Source: 1980 Census
(ii) Source: Department of Interior, April 1984 (figure is for all U.S.).
(iii) Source: Abt Associates, New »rk-New England Study, 1979.
-------
B-3
iB. 2 Seasonal Beach capacity and Current Attendance
The calculations above, show estimated increased number of user days due
to pollution control. It is necessary to compare the predicted increased use
with the overall beach capacity so that the estimates doe not exceed the
known capacity. The example beach capacity calculation is given for
Constitution Beach. Table 6-1, Section 6 of the main report, presents the
capacity figures for the beaches in the study area.
. . ..
Beach:
seasonal
beach
capacity
468,864
(A)
square
feet of
beach
264,000
(B)
square feet of
beach per person
50
(C)
(D)
persons per day x peak days
turnover rate per season
3 29.6
(A) Source: Metropolitan District Commission, Boston, MA.
(B) , (C) Source: Department of Interior, Outdoor Recreation Standards, 1970.
(D) Source: Department of Environmental Management, Massachusetts Outdoors
(9CORP) , 1976.
Capacities for all other beaches were calculated in a similar manner except for
Wollaston Beach. The different assumptions used for Wollaston Beach were 40 square feet
of beach per person and four persons per day turnover rate.
The predicted increased use is added to the current attendance figures before
comparison with seasonal capacity. Table B-l shows the current seasonal figures for the
study area.
-------
B-4
Table B-l.
Current Seasonal Attendance Figures
Beaches
Constitution
Dorchester Bay
Castle Island
Pleasure Bay
Carson
Malibu
Tenean
Wollaston
Quincy
Weymouth
Hingham
Hull
1 MDC and I HOC and I Binkley/ 1
1 Municipal I Municipal 1 Hanemann I
1 Estimates I Estimates I Estimate- 1
1 1982 1 1974 1 Log it Model 1
150,000 500,000 1,258,571
15,000
175,000
100,000
150,000
150,000
2,000,000 -
3,500,000 750,000 2,325,714
140,194 -
177,600
103,600 -
108,040
17,650 -
26,640
66,000
1
1
1
Range |
150,000 - 1,258,571
15,000
175,000
100,000
150,000
150,000
2,000,000 - 3,500,000
140,194 - 177,600
103,600 - 108,040
17,760 - 26,640
66,000
Best
Guess
325,000
15,000
175,000
100,000
150,000
1*0, OOC
2,750,OOC
158,927
105, 82C
22,20C
66,000
-------
B-5
B .3 lower Bound Estimate for increased Participation
Not all the projected increased participation might occur because of
relatively cold air temperatures at the beach, which might discourage
increased beach visits even with improved water quality. The predicted
increase in beach visits is reduced by a factor to take into account air
temperature. It is derived as follows in order to obtain a lower bound
estimates of increased participation.
(a) Each day of the summer season is categorized as
o poor (air temperature £ 75° Farenheit)
o good (air temperature >75° and < 79°)
o excellent (air temperature ^ 79°)
Air temperature data is available for sampled days during the
months of June, July and August, 1982 and 1983.
Source: Approach suggested and data supplied by Dr. Richard
Burns, Region 1, Environmental Protection Agency, Boston, MA
Categories based on "Weather conditions that lure People to the
Beach" by P. Rosenson and J. Havens in Maritimes, University of
Fhode Island, Graduate School of Oceanography, August 1977. Air
temperature for Boston Harbor area from NOAA, National Ocean
Survey data file.
(b) The percentage of days in each category is calculated based on a
total of 85 days sampled during the summers of 1982 and 1983.
(c) For each category of day a proportion of the predicted increased
participation due to improved water quality is assumed to take
place. For excellent days all the predicted increase is
included. However, the assumption is made that on good and poor
days only two-thirds and one-third (respectively) of the
predicted increase is retained because the cooler air
temperatures would tend to limit the increase predicted from
improved water quality.
Source: Based on graph of attendance versus daytime temperature
for a fliode island beach in "Weather Conditions that lure People
to the Beach" by P. Rosenson and J. Havens in Maritimes,
university of ftiode Island, Graduate School of Oceanography,
August 1977.
(d) Multiplying the proportion of days in each category (b) by the
proportion of the predicted increased participation (c) gives
the factor by which the upper bound estimate is reduced in order
to obtain a lower bound estimate which takes into account air
temperature.
-------
B-6
The following table presents these calculations:
(a) No. of sampled days
June, July, August,
1982 and 1983
(b) Proportion of
days in 1982
and 1983
(c) Proportion of
projected increase
in participation not
limited by air
temperature
Poor
v<75°
36
.424
.33
(d) "Reduction factor" for
lower bound : (b) x (c) .140
Good
> 75° and < 79°
12
.141
Excellent Total
.67
.094
37
.435
1.00
.435
85
1.00
.669
Ttie total predicted increased participation is mulitiplied by the sum of the reduction
factors to obtain the lower bound estimate of increased participation. For example, the
upper bound predicted increase in participation for Constitution Beach is 113,750.
The lower bound estimate is, therefore, .669 x 113,750 = 76,099.
-------
B-7
B.4 The Conditional Multinomial logit Model, in Brief
This section describes the conditional multinomial logit model of
multiple site demand. The model works from the indirect utility function
for an individual. The utility u^. individual i receives from visiting
beach j is
uij • f(dij,S.,Ii) (B.I)
where
= travel cost (perhaps time and distance) for individual i
to reach beach j
= characteristics of beach j (perhaps a vector of
characteristics) .
= characteristics of individual i (perhaps a vector of
characteristics).
Individual i will choose beach j if and only if
ui;j > uik kX j (B.2)
Suppose we recognize that the choice process is not perfect, either
because the individual has imperfect information, makes "mistakes" in beach
choice, or perhaps we do not recognize all the relevant factors in her
utility function. Then we might model the indirect utility functions as
uij = vij + ej (B>3)
where e • is an error term capturing the error in the choice process and
v^j represents the measurable, nonstochastic part of the indirect utility
function.
-------
B-8
NDW the probability of individual i choosing beach j is
Pi:j = prob JUij>uik)= prob ^vi:j + ej > vik + e
= prob vi;j - vik> 6 - ek k^j (B.4)
McPadden (1973) proved that if 6j and ek are independent with a
Weibull distribution, then
pij • exp(vij)/lexp(vik) (B.5)
K
If the nonstochastic part of the utility function, v, is specified to be
linear in parameters then (B.5) can be estimated using maximum likelihood
methods and hypotheses can be tested in that framework as well.
Our model predicts the total number of visits by individual i to site j,
n^ = n^ Pij (B.6)
where n^ = the total number of visits by individual i.
In essence (B.6) factors a joint probability model into a conditional
probability model. The underlying joint probability model predicts the
probability of making a beach visit (instead of, say, going to a movie) and
the probability of visiting a specific beach. Ben-Akiva (1973) showed the
joint model can be factored with the inclusion of a particular term in the
total visit model. The so-called "inclusive price" (IP) term reflects the
service characteristics of the set of beaches:
(B'7)
Then the total visit model can be specified as
ni - g(XPi, Ii). (B.8)
-------
B-9
Together (B.6) , (B.5) and (B.8) permit one to model how changes in site
characteristics S.. will effect the total quantity of visits and the split
of visits among the various beaches. Ohat is, we estimate the parameters of
these equations by using the data described above. To simulate the effect of
a change in the characteristic of one or more of the sites, use (B.8) to find
the total number of visits, (B.5) to find the fraction of the visits which
will be made to each site and (B.6) to determine the number of visits made to
each site.
The benefits of the simulated change in water quality at one or more
sites can be estimated using a modification of a procedure developed by Small
and Rosen (1982) and adapted to this problem by Feenberg and Mills (1980) .
The outline of this procedure is as follows, include the minimum level of
expenditure necessary to achieve a given utility level in v. Differentiate v
with respect to expenditures to obtain an expression for the change in
expenditures as a function of a change in site characteristics. This is a
compensated demand function for the site characteristic. Then integrate this
expression over a change in site characteristics to obtain an estimate of the
welfare change associated with the change in site characteristics. The
following makes this argument more specific.
Vi:j = vfdi^Sj^i^i) (B.9)
where E is the minimum expenditure for individual i to obtain utility level v
given all the other parameters.
Then _
V
S. j . .
where ^i is the marginal utility of income.
From Roy's identity X= ^V /ni' T"6"' in expectation,
-------
B-10
- ZL P, n. ±V_ _VV
We know p^ from (3.5) . Further, specify (B.8) in power function form
so that
Substituting into (B.ll) gives
f I
>•
1 exp V.. v
expvkj 2 expv../^v_ (B>12)
/ ^
To find the welfare change associated with a change in site
characteristics Sj to Sj where the characteristics might change in
more than one site we integrate this expression between those limits.
That is:
y*'?1
V j r **
EV. = / _o / ~i ds.
1 J * i "^7 D
«*
!
(I exp V )
3 L k J S°
3y
^d " 2
-------
B-ll
Table B-2.
Sites Included in Logit Model —
Site 1
Number 1
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
. 16.
' 17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
1
1
Site Name/ location 1
Kings Beach (Swampscott)
Lynn Beach (Lynn)
Nahant Beach (tenant)
Revere Beach (Revere)
Short Beach (Revere)
Winthrop Beach (Winthrop)
Constitution Beach/Orient Heights (Boston)
Castle Island (Boston)
Pleasure Bay (Boston)
City Point (Boston)
L & M Street Beaches (Boston)
Carson Beach (Boston)
Malibu Beach/Savin Hill (Boston)
Tenean Beach (Boston)
Wollaston Beach (Quincy)
Nantasket Beach (Hull)
Wingaersheek Beach (Gloucester)
Crane's Beach (Ipswich)
Plum island (Nswbury)
Duxbury Beach (Duxbury)
White Horse Beach (Plymouth)
Breakheart Reservation (Saugus)
Sandy Beach/Upper Mystic lake (Winchester)
Houghton's Pond/Blue Hills Reservation (Milton)
Wright's Pond (Medford)
Walden Pond (Concord)
Stearns Pond/Harold Parker State Fares t (Aidover)
Cochituate State Park (Natick)
Hopkinton State Park (Hapkinton)
1
Site
Ownership
MDC
MDC
MOC
MDC
MDC
MDC
MDC
MDC
MDC
MDC
Boston
MDC
MDC
MDC
MDC
MDC
Gloucester
Private
Private
Private
MDC
MDC
MDC
MDC
DNR
DNR
DNR
DNR
DNR
-* Based on Data collected by Binkley and Haneraann, 1975.
-------
B-12
B .5 Beach Closings
Beach closings were calculated using seasonal attendance and water
quality data. They were calculated for water quality levels greater than
200 and 500 MPN/100 ml fecal coliform and, in certain cases, for 700
MPN/100 ml total coliform.
Tenean Beach, at water quality level > 500 MPN/100 ml and for the CSO
control option is used as an example. Beach closings for all other
affected beaches were similarly calculated.
Number of Beach
Closings Averted
Beach (Visitor Days)
Tenean 19,286
(1)
timber of Beach
Closings Under
Present Conditions
(Visitor Days)
24,107
(2)
% Pollution Abatement
From Control Options
80
(1) Current Beach Closings
Number Beach
Closings
(Visitor Days)
24,107
(a)
% of Season
Water Quality
>500 MPN
.1607
(b)
Seasonal
Attendance
150,000
(a) Source: Meta Systems calculations based on data from Metropolitan
District Commission and towns of Quincy, Weymouth, Hingham,
and Hull.
(b) Source: See Table B-l (above).
(2) Source: See Table 4-3, Section 4 of the main report.
-------
B-13
B.6 user Day \telues
Many of the recreation benefit estimation approaches calculate the value
of benefits accruing from changes in the use of a resource by applying a
specific dollar value to an incremental change in quantity of recreation.
These user day values (also called unit day values) have been calculated
using a variety of techniques including cost of travel and survey-derived
estimates of willingness to pay. Generally an average figure is given which
may not reflect the effects of incremental changes in environmental quality.
They should be applied with care especially when user day values derived in
one area of the country are applied to a different region. Table B-3
presents the (wide) range of values to be found in the literature and which
are potentially applicable to this case study.
-------
B-14
Table B-3. user Day values
Source
User Day Value
in Study
User Day Value*'
in 1982 $
Values Chosen
for use in
Boston Harbor Study
General Recreation or Swimming
Heintz e_t a_l. 2.67 (1973$)
DPRA 2.54 (1975 3)
Binkley Log it
ModelJ/ 5.65 (1974 $)
Federal
Register 1.60 to 4.80 (1982 $)
Boating
Heintz e_t al. 8.96 (1973$)
DPRA 5.17 (1975 3)
Charbonneau
and HayS/ 22.80 (1975 $)
NPA 12.26 (1978 3)
Fishing
Heintz e_t a_l. 8.74 (1973$)
DPRA 5.15 (1975 3)
Charbonneau and Hay
Trout 21.00 (1975 $)
Bass 19.00 (1975 3)
Catfish 15.00 (1975 $)
Russell and Vaughanl/
Trout 11.10-24.10 (1979 $)
Bass
Catfish
Survey of
Fishing
Federal
Register
General
9.70-21.40 (1979 $)
7.00-16.00 (1979 $)
11.00 (1980 $)
2.30-4.80 (1982 $)
Specialized 11.20-19.00 (1982 $)
5.80
4.56
11.06
1.60 to 4.80
19.46
9.27
40.89
18.14
18.98
9.24
37.66
34.08
26.90
14.76-32.05
12.90-28.46
9.31-21.28
12.89
2.30-4.80
11.20-19.00
Harbor (moderate)
Harbor (high)
Harbor (low)
Charles River (low)
Harbor (high)
Charles River (high) ,
Harbor (low)
Harbor (high)
Harbor (low)
§/ Updated using Consumer Price Index, U.S. City Average,
All Urban Consumers, average for 1982 (CPI-U=289.1) .
*>/ As presented in Appendix B.3 and Section 6 of main report.
£/ Assuming a ratio of boating to fishing (bass) of 1:2.
£/ Lower figure assumes fees reflect real resource costs and value of
travel time is zero (net consumer surplus) . Higher figure assumes fees
are pure transfers and value of travel time is average wage rate (total
willingness to pay).
-------
B-15
References for Table B-3.
Charbonneau, J. and J. Hay, 1978, "Determinants and Economic Values of
Hjnting and Fishing,11 Paper presented at the 43rd itorth
American Wildlife and Natural Resource Conference, Phoenix,
Arizona.
Development Planning and Research Associates, Inc. (DPRA), 1976, National
Benefits of Achieving the 1977, 1983 and 1985 Water Quality
Goals, Environmental Protection Agency, Office of Research
and Development, Washington, DC.
Federal Register, Volume 48, Number 48, March 10, 1983, "Economic and
Environmental Principles and Guidelines for Water and Related
Land Resources Implementation Studies", U.S. Water Resources
Council, Washington, DC.
Heintz, H.T. , A. Hershaft and G.C. Horak, 1976, National Damages of Air and
Water Pollution, Environmental Protection Agency, Office of
Research and Development, Washington, DC.
rational Planning Association (NPA) , 1975, Water-Related Recreation
Benefits Resulting from Public Law 92-500, National
Commission on Water Quality, Washington, DC.
Russell, C. S. and W.T. Vaughan, 1982, "The National Fishing Benefits of
Water Pollution Control," J&urnal of Environmental Economics
and Management, 9:328-353.
U.S. Department of the interior, 1982, 1980 Survey of Fishing, Hunting and
Wildlife Associated Recreation, Fish and Wildlife Service and
U.S. Department of Commerce, Bureau of the Census,
Washington, DC.
-------
B-16
B .7 Sources of Recreation Data
Information pertaining to recreation participation and the corresponding
economic values were drawn from a number of existing reports. Nat all of the
information is specific to Boston Harbor, nor does each address exactly what
is needed for the case study at hand. However, it is information that can be
used to define ranges of values for both participation rates in and economic
values derived from the water resources of Boston Harbor. In order to
ascertain how the figures proposed by each source relates to this case study,
the method of their derivation and the populations from which they were
derived must be examined and compared to the objectives of this study and to
the population using (or potentially using) Boston arbor's water resources.
1. Abt Associates, 1979. New York-New England Recreational Demand Study,
vol. I and II. Cambridge, MA.
The focal point of this study was a survey designed to (1) quantify
current recreational demands in the New %rk-New England region and, then,
(2) to use that demand to develop a model of supply/demand interactions of
recreational resource availability and needs of forecasting recreational
demands.
The current demand figures from this study can be applied to the Boston
Harbor case study because the statistical techniques used were thorough
(including the breakdown of information by useful characteristics) and
because the sample size was large. The forecasted recreational data is not
applicable to Boston Harbor. One of the criticisms of the study is that
demand forecasts are a dependent variable of supply. To accurately assess
the particular effects of increasing the water quality of Boston Harbor, it
-------
B-17
would be preferable to use Boston Harbor-specific supply information in the
model. The results forecasted by this study's model are based upon much
larger geographic areas of recreational resources and thus do not directly
help in pin-pointing the benefits accrued (real or potential) from improved
harbor water quality.
2. U.S. Department of the Interior, November 1982. 1980 National Survey of
Fishing, Hinting, and Wildlife associated Recreation, Pish and Wildlife
Service and U.S. Department of Commerce, Bureau of the Census,
Washington, DC.
Every five years, since 1955, the Fish and Wildlife Service (in
cooperation with the Bureau of Census) has conducted a nationwide survey of
U.S. fishing and hunting activities. For the 1980 survey, questions about
non-consumptive wildlife associated recreation (e.g., bird watching) were
asked for the first time. Much of the information is of use to the Boston
Harbor case study, including participation rates, level of participation
intensity, and expenditures per activity. Unfortunately, there are no
willingness to pay or latent demand analyses.
The survey's strongest recommendation is its large sample size, which
lends confidence to statistical analyses derived from its data base. Over
116,000 households were sampled nationwide to determine participation rates
in various wildlife-related activities. Of particular interest and
application to the Boston Harbor case study are the statistics obtained for
saltwater fishing. Fishing participants identified in the screening phase of
the survey were re-interviewed, with attention to more details about:
o their intensities of participation Dumber of trips and days per
year);
o location of activity (fresh or saltwater, in-state or out-of-
state);
o mode of participation (boat, surf, shore, pier, etc.);
-------
B-18
o expenditures for participating in the activity; and
o demographic characteristics of the participants.
For this second phase of data collection, "sample sizes were designed to
provide statistically reliable results at the state level for fishing and
hunting and at the Census geographic division level for non-consumptive
activities".£r In Massachusetts, 700 fishermen and women were
interviewed. Of those interviewed, 272 participated in saltwater fishing
only (39 percent of Massachusetts anglers) , and 219 engaged in both fresh and
saltwater fishing (31 percent).
Since the statistics above are for Massachusetts overall, it is necessary
to consider how Boston area anglers differ from the "average" Massachusetts
anglers. Given fishing as an activity of participation, participation rate
differences between Massachusetts residents state-wide and Boston SMSA
residents are considered. The proximity of saltwater resources to Boston
suggests that the salt and freshwater fishing participation ratio might be
even higher for the Boston area. Assuming that the greatest use of Boston
Harbor is made by the local population, this is an important consideration
and it suggests that the survey's results are a lower bound estimate of
saltwater fishing participation. What might cause the survey's estimates to
be overstatements for the Boston SMSA are the characteristics of Cape Cod and
the shoreline communities to the north and south of Boston. These three
areas are apt to have higher than average fishing participation rates
assuming that individuals who like to engage is this activity are prone to
reside in these areas. A statistically equivalent sampling of these areas
could skew state-wide participation rates upward.
i/Page viii of the survey.
-------
B-19
The survey also presents participation rates by geographic area and place
of residence. for New England, the saltwater fishing geographic/demographic
distinctions are made for big cities, small cities and rural areas. Boston,
however, is rather uniquely situated with respect to most other cities of New
Qigland because it is on the Atlantic Coast. Again, if proximity of the
resource does have bearing on participation, then the study estimates are
probably underestimates of Boston SMSA fishing rates. The days of
participation figures generated by the survey are consistent with the same
measure from other studies. A final recommendation of this survey is that it
was completed quite recently (1980-1982).
3. McConnell, K.E., Smith, T.P., and Farrell, J.F., 1981. Marine
Sportfishing in Riode Island 1978. NOAA/Sea Grant, University of Hiode
Island Technical Report 83, Narragansett, Rhode Island.
This study is recommended for a number of reasons, including:
o The data was collected recently, from February 1978 to
January 1979;
o The sample size is large, implying statistical confidence
(5,000 interviews were conducted at the sites of the fishing
experiences and 9,000 phone interviews were conducted
state-wide);
o The information collected pertains specifically to saltwater
fishing;
o The geographic proximity of Rhode Island to Boston Harbor
makes for similar fishing experiences in terms of the types
of fish caught and the general environmental experience
(weather, topography, vegetation, seasons); and
o The nearness of Rhode Island to the case study area captures
similar population characteristics such as attitudes,
lifestyles, economic activities, etc.
There are a few obvious differences between the two study areas. Che
difference is that the vast majority of fishing in Rhode Island does not take
place near urbanized areas. Another is that public transportation is used
-------
B-20
less often in Fhode Island than in the Boston SMSA, suggesting that travel
mode arguments are not identical for the areas. Travel time is comparable
however, because of Fhode Island's small size. R>r instance, the travel time
from Rhode Island's population centers in the northern part of the state
(including Providence, the capitol) to the southern shores (popular fishing
spots) is usually an hour or less by car; using Boston public transportation
to visit a fishing site in and around the Harbor requires a comparable amount
of time.
In addition to participation rate and intensity information, estimates of
economic expenditures for participation are also available from this study.
Average expenditures are based on "out-of-pocket" costs per trip which may or
may not include some travel costs (for instance, if gas was bought on the
trip, then it would partially account for travel costs) . An examination of
costs per trip and one-way mileage figures suggests that travel costs are not
extensively covered by the "out-of-pocket" cost data; even at the
conservative cost of 3.10 per mile, the expenditure data barely accounts for
travel costs.
By using the expenditure information available for the various modes of
fishing (shore, fixed structure, boat) together with travel cost information
specific to Boston Harbor, a range of plausible current trip expenditures for
fishing in the Harbor can be calculated. Such a range represents
demonstrated economic worth of the fishing resources but does not indicate
consumer (participant) surplus of fishing activity.
The interview questionnaire used for this study did include willingness
to pay questions, but that data has not yet been tabulated and analyzed. In
the absence of willingness-to-pay measures, the demonstrated expenditures
-------
B-21
will be taken as lower bound estimates of the economic value of Boston
Harbor's fishing resources.
4. Metcalf and Eddy, 1975. Eastern Massachusetts Metropolitan Area Study
(EMMA) . Technical Data (Molurae 13B) . Socio-Economic Dnpact Analysis.
The area of study for the EMMA series of reports roughly corresponds to the
area of this case study, so the information presented is directly relevant to
the case study at hand. The socio-economic impact analysis includes a section
on recreation in the area. It examines actual and potential recreational
activity there. Actual, or current, activity is defined as demand; potential
activity, or un-oiet demand, is defined as need. (teed is translated as latent
demand for application to this case study.)
Much of the information presented in EMMA regarding recreational
opportunities is drawn from the Eastern Massachusetts supplement to the 1972
Massachusetts Outdoor Recreation Plan. Based on information drawn from the
Outdoor Recreation and Open-Space Inventory and from census data, the
supplement provides a data baseline on recreational opportunity in the area.
Although the inventory and census were conducted in 1970, the recreational
opportunity and activity calculations are still valid since the current
population and recreational resources of the area are not much changed from
that time, if recreationa.1 habits are also alike.
The assessments of demand and latent demand were performed according to
population density groups within the MAPC area. The highest density groups
had the lowest ratios of recreation and open space acreage to population. It
appears that the analyses for latent demand were performed within each
density group; that is, if the recreational resources within a density group
area were not sufficient to meet the total potential demand for the
-------
B-22
population within that group, the availability of such resources in other
areas was not considered for satisfaction of those recreation needs. The
high density areas exhibit latent demand of water-based recreational
activities, even though the majority of municipal recreational sites is
within the very dense and dense categories. Still, the extremely dense
category has five percent of the recreational areas and 35 percent of the
population within the study area.
The quality of the available recreational sites was not a factor in
calculating recreational opportunity.
5. Metropolitan Area Planning Council, October, 1972. Boston Harbor Islands
Comprehensive Plan, for Massachusetts Department of tatural Resources.
This report describes a plan for all phases and aspects of maintaining
and developing the islands of Boston Harbor, which are considered a unique
natural resource of significance to the New England Region. The islands are
predominantly open, natural areas; some have historic sites or limited public
facilities. The Plan contains descriptions of the islands and the current
and planned activities for them. Many of the islands do not yet have the
facilities or the water quality necessary for some of the activities;
therefore, activity days figures most nearly reflect potential use of the
Islands.
The islands offer a range of activities: swimming, boating, fishing,
hiking, picnicking, group and primitive camping, play, and historic fort
visitation. Only the first three activities mentioned are of concern to this
case study because they are most directly affected by water quality.
(However, water quality can affect the experiences of other activities such
as camping and hiking.) This report is particularly useful because it
-------
B-23
provides data on the recreational potential (activity days) of the Harbor
Islands.
The economic values per day for each Boston Harbor Island activity day
were based upon the federal Water Resources Council's "Standards for Planning
Water and Land Resources" (July 1970) . These values are nationwide
estimates. Because the values in the Harbor Plan are in 1970 dollars it was
necessary to inflate them to 1980 dollars using the Consumer Price Index for
urban consumers, furthermore, the round trip ferry fee to George's Island of
$3.00 has been added to the value in order to account for a portion of the
travel costs incurred in visiting the islands. The Department of Environ-
mental Management provides a free taxi service to reach other islands from
George's Island. The travel costs incurred by private boaters to the islands
are probably at least $3.00 considering the costs of gas and/or costs of
upkeep.
6. Bureau of Outdoor Recreation, September, 1977. National Urban Recreation
Study; Boston/lowel 1/lawrence/Haverhill, Northeast Regional Office;
National Park Service and Forest Service.
This particular study offers qualitative insights into and justifications
for recreational resource preservation in its study area. (Some of the ideas
are presented here.) A basic premise of the study is that open space which
is close to home is desirable. At present, Boston ha& only 5.4 acres of open
space per thousand population, whereas the recommended minimum by the
National Recreation and Park Association and the Urban Land Institute is 10
acres per thousand population. Most of Boston's land is already developed.
Once it has been developed, it is economically and physically difficult to
reclaim as open space. Of the open spaces that do remain, there is
considerable competition for their use.
-------
B-24
Cnly about one-sixth of New England's coastline is accessible to the
general public. The recreational potential that Boston Harbor offers is
substantial by comparison since approximately 40 percent of the harbor
shoreline remains relatively undeveloped; portions of this undeveloped area
are used for recreational activities. In addition, the islands are within a
25 mile radius of 2.7 million people. The 1977 Coastal Zone Management Plan
lists three types of recreational facilities as being in greatest demand for
Boston Harbor. They are: (1) large scale beaches and waterfront parks;
(2) smaller scale beaches and parks for local use; (3) walkways. Certainly/
water quality is critical to swimming activity and can enhance the enjoyment
of parks and walkways.
Whereas the waterfront was once largely an area of warehousing and
industrial activity, new development and redevelopment styles are leading to
different interactions with the Harbor, particularly in the downtown areas
along the Inner Harbor. More people are living, shopping, and staying in
hotels near the water—their relationship to the Harbor is becoming more
intimate so the aesthetic quality and sense of open space it can offer is
becoming more important. Furthermore, as more white-collar businesses move
into the waterfront commercial spaces, perceptions and expectations of the
working environment change (visits by clientele, visual appearances of
surroundings, etc.).
7. Massachusetts Department of Environmental Management, December 1976.
Massachusetts Outdoors; Statewide Comprehensive Outdoor Recreation Plan (SCORP)
The information on recreation participation rates and latent demand in
this report is of interest to the Boston Harbor case study. However, the
methodology employed to obtain that information has a number of limitations.
The primary problem is the sample size of the data collection effort.
-------
B-25
A telephone survey was conducted of 400 households/persons throughout
Massachusetts and this survey is the data source for all subsequent analyses.
The Boston SMSA is contained within a region extending west to Worcester,
north to the state border/ and south to Bridgewater. This region is one of
seven equally sampled areas within the state, meaning that the Boston SMSA
recreational demand is calculated from only 57 (or fewer) interviews.
Some of the results of the data analysis are counter-intuitive. Che such
result suggests that power boating participation rates are more strongly
associated with low income groups than with higher income groups, although
power boat operation and maintenance can be quite expensive. Information
from the "Boston Marinas and Live-Aboards Study" indicates a high proportion
of large boats in the Boston area, thus countering the explanation that the
power boat population is dominated by small boats with outboard motors (i.e.,
less expensive power boats, affordable to low income groups).
The results of the SOORP study are more meaningful if they are
interpreted qualitatively, rather than quantitatively. The shortcomings of
the empirical findings are often mentioned by the authors throughout the
study, suggesting that SCORP results should be applied with caution.
8. Department of Interior, April 1984. The 1982-1983 Nationwide Recreation
Survey, Itetional Park Service, Washington, DC.
The most recent nationwide survey of recreation activities was designed
for comparability with certain portions of the national recreation surveys
conducted in 1960 and 1965. It includes data on participation rates,'
expenditures, reasons for recreating, and reasons for constraints on
-------
B-26
recreating. At the time of this report only nationwide figures were
available. Regional (but not as detailed as the SMSA level) figures are
expected to be published later.
-------
B-27
References
Abt Associates, 1979. New York-New England Recreational Demand Study,
Vblumes I and II, Cambridge, MA.
Ben-Akiva, J. , 1973. The Structure of Passenger Travel Demand, Ph.D.
dissertation, Massachusetts institute of Technology, Department of Civil
Engineering, Cambridge, MA.
Binkley, Clark S. and W. Michael Hanemann, 1975. The Recreation Benefits of
Water Quality improvement, NTIS PB257719, for the U.S. Environmental
Protection Agency, Washington, DC.
Department of Interior, March 1970. Outdoor Recreation Standards, Bureau of
Outdoor Recreation, Washington, DC.
Department of Interior, September 1977. National Urban Recreation Study;
Boston/lowell/lawrence/ffiverhill, Northeast Regional Office, Bureau of
Outdoor Recreation, National Park Service and Forest Service, Boston, MA.
Department of Interior, November 1982. 1980 National Survey of Fishing,
Hunting and Wildlife Association Recreation, Fish and Wildlife Service
and U.S. Department of Qammerce, Bureau of the Census, Washington, DC.
Department of Interior, April 1984. The 1982-1983 Nationwide Recreation
Survey; Summary of Selected Findings, National Park Service,
Washington, DC
Feenberg, Daniel and Edwin Mills, 1980. Measuring the Benefits of Water
Pollution flaatement, Academic Press, New York, NY.
Massachusetts Department of Environmental Management, December 1976.
Massachusetts Outdoors; Statewide Commprehensive Outdoor Recreation
Plan (SCORP) , Boston, MA.
McConnell, K.E., Smith, T.P. and Farrell, J.F., 1981. Marine Sportfishing
in Fhode Island 1978, NOAA/Sea Grant, university of Fhode island
Technical Report 83, Narragansett, RI.
Metcalf and Bidy, 1975. Eastern Massachusetts Metropolitan Area Study
(EMMA), Boston, MA.
Metropolitan Area Planning Council, October 1982. Boston Harbor Islands
Comprehensive Plan, for Massachusetts Department of Natural Resources,
Boston, MA.
Small, K and H. Rosen, 1981. Applied Welfare Economics with Discrete Choice
Models, Efconometrica 40: 105-130.
-------
Appendix C
Swimming Health Benefit Calculations
Health benefits for recreational swimming are derived using dose-response
functions and beach attendance data. The distribution of water quality
levels throughout the swimming season for each beach was used as the basis
for estimating the exposure of the swimming population. The first section of
this appendix shows how the number of highly credible gastroenteritis cases
was calculated for each water quality level at each beach. Tenean beach, at
water quality level 7 (60 MPN fecal coliforms per 100 ml), is used as a
representative example. The second section of this appendix shows the
calculations of reduced number of cases of illnesses for each treatment
option for each beach.
C.I Number of Cases of Gastrointestinal Illness
(Itenean Beach, water quality level 7)
Number of Cases
of Highly Credible
Gastroenteritis
At Water Quality
Level 7
190
(A) Number of cases
per 1000
18.09
(A) (B)
Number of Cases Population at
of HC gastroenteritis x Risk, up to Water
per 1000 Quality Level 7
10,500
18.09
(1)
= 0.2 + 12.2 log Ehterococci
= 0.2 + 12.2 x 1.47
0.75
1000
Source: Cabelli et. al, 1982.
-------
C-2
(a)
(1) Log Ehterococoi = 0.825 log Fecal coliform R = 0.82
1.47 = 0.825 x log(60)
Source: Meta derived statistical relationship using averaged MDC and
other municipal water quality data, 1974-1982.
(a) When total coliform concentrations were measured instead of fecal
coliform concentrations, total coliform concentrations were substituted
using the following relationship:
log Fecal coliform = 0.65 log total coliform R = 0.89
Source: Meta derived function, based on averaged MDC and municipal
water quality data, 1974-1982.
(1) (2)
(B) Population
at Risk at Percentage of Season
Water Quality = Seasonal Beach x Water Quality
Level 7 Attendance At Level 7
10,500 = 150,000 .07
(1) Source: MDC, Towns of Quincy, Weymouth, Hingham and Hull.
(2) The frequency, per season, of thirteen water quality levels was measured
for fecal coliform concentrations, MPN per 100 ml (see Table C-l).
C.2 Reduced Cases of Gastrointestinal Illness
The above calculations are done for each water quality level to
establish the base case for each beach. This gives the estimated number of
cases of illness occurring under current conditions. Similarly, the
calculations can be carried out assuming a certain percentage of pollution
reduction. This is done by reducing the average fecal coliform count for the
water quality level by the percentage pollutant reduction. For example, in
the base case water quality level 7 has a fecal coliform count of
60 MPN/100 ml.
-------
C-3
Table C-l. Water Quality Fecal Ooliform Levels
Level 1
1
2
3
4
5
6
7
8
9
10
11
12
13
Water
Quality Range
Fecal Col i form
0
1-5
6-10
11-20
21-30
31-50
51-70
71-130
131-170
171-330
331-470
471-730
£ 731
Median
Value Fecal
1 Col i form Used
0
3
8
15
25
40
60
100
150
250
400
600
731
% During Season
1 for Tenean Beach
0
10
13
9
1
12
7
9
7
9
6
9
10
-------
04
Under the CSO control option with 80 percent reduction the same water quality
level 7 would be assigned a fecal coliform count of 12 MPN/100 ml. Then, the
string of calculations listed in Section C.I above are repeated to estimate
the number of cases of illness under these new water quality conditions. The
number of cases for each of the water quality levels are summed to give a
total incidence of illness at that beach. Levels for which the fecal
coliform counts exceed 500 MPN/100 ml, however, are not included because we
assume the beach is closed to swimming at counts above 500 MPN/100 ml. These
calculations are shown for Tenean Beach in Table C-2.
C.3 Population at Risk
The studies of swimmers and related health effects divide the population
of visitors to a beach into swimmers and non-swimmers. TWO available studies
have this information for Boston area beaches. Their results are shown below.
Study Total § of Visitors % of Swimmers who go swimming
43 Boston area beaches
(Hanemann, 1978) 2507 32 %
2 Boston area beaches
(Cabelli e_t a_l., 1980) 4153 49%
6 Ooastal beaches in U.S.
(Cabelli e_t a_l., 1980) 16182 63%
In this study we use the figure of 4--A for a lower bound estimate of the
population at risk. In addition, a reduction factor tied to the distribution
of air and water temperature during the summer season is used. This factor
is calculated by first categorizing the summer days as follows:
-------
Table C-2. Calculation of Number of Highly Credible Gastroenteritis
Cases for Tenean Beach
Level
1
2
3
4
5
6
7
8
9
10
11
12
13
Total
Fecal
Col i form
Count
(average)^/
0
3
8
15
25
40
60
100
150
250
400
600
731
Cases
% of
Season
Hater Quality
at Given Level—
0
10
13
9
1
12
7
9
7
9
6
9
10
Total Cases below
500 MPN/100 ml
With 10%
# of Reduction
Base f.c.
Cases*/ Count2/
0
73
181
163
22
296
192
277
235
332
240
385
441
2837
2011
(a)
0
2.7
7.2
13.5
22.5
36
54
90
135
225
360
540
657.9
* of
Cases*/
0
66
172
157
21
288
187
271
230
326
236
379
434
2767
1954
(b)
Calculations for Each Treatment
Treatment
Option
CSO only
%
Ocean Outfall
Pollution
Reduction
80
10
Number of
With 80%
Reduction
f.c.
Count2'
0
0.6
1.6
3
5
8
12
20
30
50
80
120'
146.2
Option
ft of
Cases—
0
0
41
66
11
167
116
180
159
235
176
288
333
1772
1772
(c)
•
With 90%
Reduction
f.c.
Count5/
0
0.3
0.8
1.5
2.5
4
6
10
15
25
40
60
73.1
ft of
b/
Cases-'
0
0
0
24
6
111
84
137
127
194
148
246
287
1364
(
1364 (
(d)
Reduced Cases Calculation
of Illness
239
57
(a)
(a)
Method
- (c)
- (b)
Seconda ry
Treatment 10
CSO and Ocean
Outfall 90
CSO and Secondary
Treatment 90
57
647
647
(a)
(a)
(a)
(b)
(d)
(d)
a/From Table C-l.
^/Calculated using Cabelli et a_l. (1982) equation.
-------
C-6
o Poor (air temperature ^75° Fahrenheit and/or water
temperature < 65° Fahrenheit)
o good (air temperature > 75° and < 79° and water temperature £65°)
o excellent (air temperature > 79° and water temperature £ 65°)
Then, the distribution of days in each category is estimated from data on
air and surface water temperature for the months of June, July and August for
the years 1982 and 1983.. Ebr "poor" days it is assumed that only one-third of
the predicted increased population at risk will actually go swimming. For
"good" days it is assumed that two-thirds of the predicted increase due to
improved water quality will go swimming but not all of the predicted increase
because of the relatively lower air and water temperatures. Ebr "excellent"
days, all of the predicted increased population at risk is assumed to go
swimming.
Thus, the lower bound estimate of increased population at risk is 49% of
the predicted increased beach visitors times the reduction factor (.551) for
the air and water temperature constraints. We used 100% of beach visitors as
an upper bound estimate because the question in the studies is often phrased
"what is your primary beach activity" rather than "did you go swimming".
Thus, visitors may go swimming even for a limited amount of time where their
primary beach activity was something else.
-------
C-7
The following table presents the calculations for the lower bound "reduction factor": £/
Poor
Good
Excellent
Air ^ 75° and/or Air > 75° and < 79° Air £ 79° and
Water ^65° and water ^ 65° Water ^65° Total
(a) No. of sampled days
June, July and August,
1982 and 1983
(b) Proportion of days in
1982 and 1983
(c) Proportion of predicted
increase in population at
risk not limited by air
and water temperatures
55
26
.647
.33
.047
.67
.306
1.00
85
1.00
(d) "Reduction factor" for
lower bound estimate:
(b) x (c)
.214
.031
.306
.551
5/ Approach suggested and data supplied by Dr. Richard Burns, Region 1,
Environmental Protection Agency, Boston, MA. categories and proportions used
in (c) based on "Weather Conditions that Lure People to the Beach" by P.
Rosenson and J. Ravens in Maritimes, University of Rhode Island, Graduate
School of Oceanography, August 1977, and "Adapted Aquatics" by The American
National Red Cross, 1977, Washington, DC. Air and surface water temperature
for Boston Harbor Area from NOAA, National Ocean Survey data file.
-------
C-8
References
Cabelli, Victor J. , e_t al^, 1980. Health Effects Quality Criteria for Marine
Recreational Waters, Environmental Protection Agency, EPA-600/1-80-031.
Cabelli, V.J., A.P. Dufour, L.J. McCabe, and M.A. Levin, 1982. Swimming
Associated Gastroenteritis and Water Quality. American Journal of
Epidemiology, 115:606-616.
Hanemann, W.M., 1978. A Methodological and Empirical Study of the Recreation
Benefits from Water Quality improvement. PhD dissertation, Harvard
University, Cambridge, MA.
-------
Appendix D
Commercial Fisheries Benefit Computations
D.I Demand Function Estimation
Other than the one in the study done in Maryland to predict future
fisheries' supply,—' which was discussed in the main body of the report, no
other soft shelled clam demand functions were found in the literature. At
present, research is being conducted at the University of Rhode Island
Department of Resource Economics on developing such information about various
fisheries based on National Marine Fisheries Service data. Dr. Stephen
Crutchfield ran some regressions using this data to produce a range of soft
shelled clam demand functions for us.-/ One of these will be described
below for illustrative purposes. Because of the lack of information
available to calibrate these functions properly for Massachusetts, and
because these functions do not represent consumer demand in a particular
market area (as discussed in the main report concerning the Maryland demand
function), it was not possible to use them to compute the impacts of
pollution abatement in Boston Harbor. However, since this information may be
useful to others, one of these demand functions will be presented here.
The best six variable logarithmic linear model, as indicated by the
maximum improvement in the R-squared statistic, found using the stepwise
regression technique is as follows:
P = 1.876 - .076Q + .450W + .117C + .7511 + .087S + .029F
(R2 = .96)
Marasco, 1975.
Crutchfield, 1983.
-------
D-2
where,
dependent variable: P = exvessel soft shelled clam prices
(Maryland)
independent variables: Q = soft shelled clam landings (Maine)
W = wholesale prices of soft shelled clams
(New York)
C = exvessel prices of quahogs (Rhode Island)
I = per capita income
S,F = seasonal dummy variables, summer and fall.
The stepwise regressions were run using monthly data from 1960 through
1982, where available. The regressions were set up so that Q was always
included as an independent variable. Price data from Maryland and landings
(harvest) data from Maine had to be used because of insufficient time series
data elsewhere; extensive price and landings data were not available for
Massachusetts nor did the data base used have both price and landings data
for the same state. The wholesale price in New York was included as a demand
shifter since New York is a large market for soft shelled clams. Quahog
prices were added to represent demand for a competitive product. Per capita
income is used to reflect derived demand. Seasonal dummy variables were
included to account for the wide seasonal variations in demand caused by the
summer tourist season. This equation produces extremely high price and
income elasticities of demand. For this and the reasons mentioned above and
in the main report, it was not used to compute pollution abatement benefits.
D.2 Demand Function Computations
Computations to determine the constants for the demand functions for
alternative price elasticities were carried out as shown below. The
-------
D-3
following demand function was used:
Q = Ax P*
where, Q = consumption (bu.)
A = constant
P = price ($)
<* = price elasticity
and transformed to log form:
log Q = log A +
-------
D-4
log (P - AP) = 1.4340
P - i P = 27.16
P - 27.16 = A?
AP = 28.45 - 27.16 = 1.29.
Total benefits for each abatement option were calculated as shown below.
The change in consumer surplus is equal to the following:—'
A CS = A P • Q + 1/2 (A P x&Q)
where, ACS = change in consumer surplus (S)
A P = change in price ($)
Q = initial consumption (bu.)
& Q = change in consumption (bu.).
Referring back to Figure 7-2 in the main body of the report, it can be seen
thatAPxQ computes the area B + C and 1/2(A? xAQ) , the area E, and that their
sum in the above equation represents ACS equal to area B + C + E.
As an example, using the I P and AQ associated with the STP option from
the above calculations, and using 16,000 bu. as a reasonable estimate of the
initial consumption from Boston Harbor shellfish areas, total benefits (equal
to change in consumer surplus) were estimated as follows:
ACS = AP x Q + 1/2(&P x£Q)
= (1.29) (16,000) + 1/2[(1.29) (29,603)]
= (20,,. 40) + 1/2(38,188)
= (20,640) + (19,094)
= $39,734.
£/ Note that simple geometric calculations are used here rather than
integration under the curve. Even though the latter method is more accurate and
correctly assumes a non-straight-line demand curve, the former is simpler, and
given the magnitude of the possible error in the assumptions already made, will
not adversely affect the outcome.
-------
D-5
D.3 Supply Cost Data and Computations and Producer Surplus Computation Example
No estimates concerning producer surplus changes due to pollution
abatement in Boston Harbor could be made due to lack of data. Attempts were
made to develop a supply curve but were unsuccessful; these will be described
below. As mentioned in the main body of the report, it is likely that change
in producer surplus due to pollution abatement would be zero because the
fishery is unregulated and there are no limits to prevent new firms from
eventually entering and bidding away any short-run excess profits; i.e., the
supply curve is probably flat in the area of interest. Despite an extensive
search, no supply curves for the fishery were found in the literature. There
is general agreement that is would be very hard to produce such a curve due
to the extreme difficulty of modeling the biological processes affecting
shellfish supply. Thus, supply for a fishery like the soft shelled clam
industry is usually held to be exogenously determined.^/ This approach was
taken here.
As discussed in the main report, the Boston area market for clams is
supplied by Maine and Maryland as well as Massachusetts fisheries.
Harvesting cost data is available for Maine (Townsend and Briggs, 1980).
Costs for the typical Massachusetts digger are very similar to those for
Maine.—/ Costs to diggers in restricted areas in Massachusetts, however,
are higher than to others becau:. • of the special licensing requirements,
depuration costs and additional transportation necessary to get the clams to
3/ From discussions with individuals at the Maryland Department of
Natural Resources, the Maine Department of Marine Resources, and the
Universities of Maine, Maryland and Rhode Island.
b/ Massachusetts Division of Fisheries and Wildlife estimates.
-------
D-6
the purification plant. Prices to Maine diggers are lower than prices to
Massachusetts diggers.— From this information, it was assumed that the
supply curve for the Boston area soft shelled clam market could be
represented by the curve displayed in Figure D-l. This is a stepwise supply
curve in which the quantity Q^t and price P^, represent the quantity
supplied by Maine diggers at their lower cost level. Similarly, the quantity
from Q^ to Q2 represents the amount supplied by Massachusetts diggers
from unrestricted areas and from Q2 to Q3 that supplied from Boston
Harbor restricted areas at a higher cost. The dashed line at Q4 and P<
shows the decreased costs and increased quantity to the diggers that operate
in Boston Harbor as a result of pollution abatement. Maryland quantities and
costs are not included because the fishery there is highly mechanized and has
a totally different cost structure.
Initially, it was thought that, given the available Maine cost data,
costs for Massachusetts firms could be developed for both restricted and
unrestricted areas. However, with the limits on time and resources and the
lack of data, it was not possible to solve two main problems. The first was
to account for the fact that the firms that operate in the restricted areas
are composed of a master digger and subordinate diggers unlike typical other
Massachusetts and Maine firms which are single-person operations.
Information was not readily available on wager and numbers of employees. The
second problem, the really major one, was to determine what impact pollution
abatement and the potential increased supply available in Boston Harbor would
have on the harvest costs. Reasonable assumptions could be made concerning
non-labor costs such as assuming decreased per unit transportation costs
§/ Maine Department of Natural Resources and Massachusetts Division of
Fisheries and Wildlife data.
-------
D-7
Figure D-l.
Assumed Shape of Supply Curve for
Boston Area Soft Shelled Clam Market.
Price
(S/bu.)
Quantity of
Shellfish (bu.)
-------
D-8
since more clams could be hauled per daily trip to the purification plant.
However, it was very difficult to estimate the impacts on return to the
master digger or on numbers of subordinates that would be hired. Therefore,
it was not possible to complete this representation of the supply curve for
the Boston Harbor market so that it could be combined with the previously
estimated demand curves to compute changes in producer surplus. It was
thought, however, that the preliminary computations that were completed might
be useful to others and should be presented in an appendix. The following
tables and discussion show the data used and computations that were made in
order to estimate soft shelled clam harvest costs for both unrestricted areas
in Maine and restricted and unrestricted areas in Massachusetts.
Table D-l shows annual 1978 costs for a typical Maine clam digging firm,
a one-person operation, developed by Townsend and Briggs (1980). In Table
D-2, these costs are updated to 1980 dollars for Massachusetts diggers who
operate in unrestricted areas. Updated costs for Maine firms are also shown
at the bottom of this table.
Tables D-3 through D-6 show the computation of nonlabor costs for
Massachusetts shellfishing firms operating in Boston Harbor restricted
areas. Because it was not possible to develop costs for a typical firm
operating in Boston Harbor restricted areas due to the lack of information
regarding numbers of subordinate diggers employed and their wage rates, i.
was decided that costs should be developed on a per bushel basis. Table D-3
shows per bushel costs divided into four categories for computation
purposes. Nonspecialized items are those for restricted firms that
correspond to the items included in the single-person unrestricted firms
shown in Table D-2. Specialized items are those that are required by either
-------
D-9
Table D-l. Cost Data for a lypical Maine Clam Digging Firm, 1978 $
Capital Costs:
Items
Car
(1/2 cost of new car)
Boat
Trailer
Outboard Motor
SUBTOTAL:
Direct Expenses;
Items
Fuel, Car
Fuel, Boat
Auto Maintenance
Boat Maintenance
License
Insurance
Boots & Gloves
Hods
Clam Hoe
SUBTOTAL:
TOTAL:
Owner Operator Income:
1978
Cost
2500
1200
600
1000
1978
Unit Cost
.80/gal
.80/gal
12
Annual
Life Depreciation
4 625
10 120
10 60
4 250
1055
No. of
Units Annual Cost
55.6 44
7.5 6
200
200
10
100
28
2 24
1 15
627
1682
2234
Source: Ibwnsend and Briggs, 1980.
Notes;
Volumes: 210 bushels/year @ $18.65.
Gross Revenue: $3916.50.
Employment: one.
Operates: 5 months per year.
-------
Table D-2.
D-10
Costs for a Typical Massachusetts Shellfishing Firm
Operating in Unrestricted Areas, 1980S
Capital Costs:
Items
Vehicle
1978
Cost
2500
Adjustment
Factor—^
1.31
1980
Cost
3275
Life
4
Annual
Depreciation —
818.75
(1/2 cost of new car;
50% devoted to
Boat
Trailer
Outboard Motor
SUBTOTAL:
Direct Expenses
„
Items
Fuel, Car
Fuel, Boat
Auto Maint.
Boat Maint.
License
Insurance
Boots & Gloves
Hods
Clam Hoe
clamming)
1200
600
1000
•
1978
Price
.80/gal
.80/gal
200
200
-
100
28
12
15
1.31
1.31
1.31
Adjustment
Factor^
1.31
1.31
1.31
1.31
-
1.31
1.31
1.31
1.31
1572
786
1310
1980
Price
1.05
1.05
262
262
30
131
36.68
15.72
19.65
10
10
4
Quantity
55.6
(1,000 mi/yr
@ 18 mi/gal)
7.5
(300 mi/yr
@ 40 mi/gal)
1
1
1
1
1
2
1
157.2
78.6
327.5
1382.05
Total
58.4
7.9
262
262
30
131
36.68
31.44
19.65
SUBTOTAL:
839.07
ANNUAL CAPITAL COSTS PLUS DIRECT EXPENSES: 2221.12
[Similarly Updated Annual Costs for Maine Firms (1980 $) = 2203.05]
Source: Meta Systems estimates based on Townsend and Briggs, 1980 and Williams,
(no date) .
S/ CPI Boston.
k/ Assumes straight-line depreciation.
Notes;
210 bushels/yr.; average harvest.
Operates 5 mo./yr.; 100 days/yr.; 5 days/wk.
120 tides per year; 1.75 bu./tide/digger.
-------
D-ll
liable D-3. Per Bushel Nonlabor Harvest Costs for Boston
Harbor Restricted Areas
Cost Categories Cost/Bushel 1980 $
Nbnspecialized Items 5.01
Specialized Items - Subordinate Diggers 3.47
Specialized Items - Master Diggers 6.18
Depuration Costs 2.00
TOTAL: 16.66 .
Notes;
Depuration Costs: $1.00/rack; 2 rack/bu.; $2/bu.
-------
D-12
Table D-4. Per Bushel Costs for Nonspecialized Items§/
Capital Costs:
Items
Boat
Trailer
Outboard Motor
TOTAL:
Direct Expenses:
Items
Fuel, Boat
Boat Maint.
Insurance
Boots & Gloves
Hods
Clam Hoe
1978
Cost
1200
600
1000
1978
Price
.80/gal
200
100
28
12
15
Adjustment
Factor^
1.31
1.31
1.31
Adjustment
Factor^
1.31
1.31
1.31
1.31
1.31
1.31
1980
Cost
1572
786
1310
1980
Price
1.05
262
131
36.68
15.72
19.65
Annual
Life Depreciation2
10
10
4
Quantity
7.5
(300 mi/yr
@ 40 mi/gal)
1
1
1
2
1
157.2
78.6
327.5
563.3
Total
7.9
262
131
36.68
31.44
19.65
TOTAL:
488.67
ANNUAL CAPITAL COSTS PLUS DIRECT EXPENSES: 1051.97
= $5.01/bu. @ 210 bu./yr. (from Maine cost data)
1980.
Based on costs estimated for Maine diggers for 1978, Townsend and Briggs,
CPI Boston.
£/ Straight-line depreciation assumed.
-------
D-13
Table D-5. Per Bushel Specialized Costs for Subordinate Diggers
Capital Costs:
1978
Items Cost
Car (50%) 2500
Direct Expenses:
1978
Items Price
Fuel, Car .80/gal
Auto Ma int. 200
License -
TOTAL:
Adjustment 1980
Factor!/ Cost
1.31 3275
Adjustment 1980
Factor £/ Price
1.31 1.05
1.31 262
30
ANNUAL CAPITAL COSTS PLUS DIRECT EXPENSES:
= $1169.15/subordinate
digger x 49 diggers^/ 7 16,
Annual
Life Depreciation
4 818.75
Quantity Total
55.6 58.4
1 262
1 30
350.4
1169.15
500 bu. = $3.47/bu.
a/ CPI Boston
!i/ Estimated average annual number of subordinate diggers = 16,500
bu./yr. total harvest f 210 bu./digger/yr. ° 79 diggers f 30 master diggers
49 subordinate diggers. This number may be an overestimate because
restricted flats may tend to have more clams/acre and therefore the harvest
may be greater per person than indicated in the Maine data. However,
personnel must be used to transport clams to the purification plant which
would increase the employee/bushel ratio.
-------
D-14
Table 0-6. Per Bushel Specialized Costs for Master Diggers
Capital Costs:
Items
Truck
Racks
Surety Bond
SUBTOTAL:
Direct Expenses;
Items
1978
Cost
5500
1978
Price
Fuel, Truck . 80/gal
Truck Maint. 500
License -
SUBTOTAL:
ANNUAL CAPITAL COSTS PLUS
= $3401.51/master digger
Adjustment
Factor
1.31
$10 x
Adjustment
Factor
1980
Cost Life
7205 4
33 = 330 3
500 20
1980
Price Quantity
1.31 1.05 611. 2b/
1.31 " 655 1
100 1
DIRECT EXPENSES:
x 30 master diggers f 16,500 bu. = $6
Annual
Depreciation
1801.25
110
93. 5a/
2004.75
Total
641.76
655
100
1396.76
3401.51
. 18/bu.
a/ used capital recovery factor = .187 (20 yr. life, 8% interest).
£/ 611.2 = 55.6 (1000 mi/yr @ 18 mi/gal) + 555.6(10,000 mi/yr @ 18 mi/gal)
Notes;
Operates 5 mo./yr.; 5 days/week; 100 days/yr.
Approximately 16,500 bu./yr. depurated from Boston Harbor; 30 master diggers
operate in Boston Harbor; 550 bushels/master digger/yr.; 5.5 bushels/day/
master digger.
2 racks/bushel; 11 racks/day x 3 days = 33 racks/master digger.
Approximately 50 mi. from harvest area to depuration plant;
100 mi./day x 100 days/yr. = 10,000 mi./yr. to depuration plant.
-------
D-15
the master or subordinate digger because they operate in restricted areas.
Depuration costs are the per bushel costs for the clams to be handled by the
purification plant. The development of nonspecialized costs is shown in
Table D-4. Specialized costs are computed for subordinate diggers in Table
D-5 and for master diggers in Table D-6. These computations assume that the
annual harvest from Boston Harbor restricted areas is 16,500 bushels,8/
that there are 30 master diggers3/ operating in the harbor and that each
digger harvests approximately 210 bushels annually.-/
Changes in per bushel costs due to pollution abatement are shown in Table
D-7. It is assumed, for illustration purposes, that the fishery is
restricted and therefore no additional firms (master diggers) can enter.
More subordinate diggers would be hired, however. The additional yield from
the restricted areas was a preliminary figure later changed in the main body
of the report (see Table 7-2) . To compute total number of diggers, the same
annual harvest rate was assumed as for Table D-3. The main impact of the
pollution abatement was assumed to be an increased annual harvest which would
allow master diggers to transport approximately four times as many bushels
per daily trip to the purification plant as without abatement. The
purification plant is currently undergoing expansion which will allow it to
handle larger numbers of shellfish per day.
Table D-8 compares available price data with the nonlabor cost data
computed for Maine and Massachusetts. Theoretically, the difference between
the price and the nonlabor cost should reflect the income to the firm owner
I/ Division of Marine Fisheries estimates. The 16,500 was later revised
to 16,000 in the main report.
£_/ Townsend and Briggs, 1980.
-------
D-16
Table D-7. Changes in Per Bushel Nonlabor Costs for Boston Harbor
Restricted Areas Due to Pollution Abatement
Cost Per Bushel, 1980 $
Cost Categories Without Abatement With Abatement
Nonspecialized Items 5.01 5.01
Specialized Items - Subordinate Diggers 3.47 5.03
- Master Diggers 6.18 1.69
Depuration Costs 2.00 2.00
TOTAIS: 16.66 13.73
Change in Per Bushel Cost -2.93
Notes;
Annual yield: 16,500 + 49,928 -/ = 66,428 bu./yr.
66,428 bu./yr. f 30 master diggers —/ = 2,214 bu./master digger/yr.;
22.1 bu./day (4 times as many as
before abatement)
66,428 bu./yr. f 210 bu./digger —' = 316 diggers
f 30 master diggers = 286 subordinate diggers
Costs for nonspecialized items - no change.
Specialized costs - subordinate diggers:
$1169.15/subordinate digger x 286 diggers r 66,428 bu. = $5.03/bu.
Specialized costs - master diggers:
Racks: 2 racks/bu.; 44.2 racks/day x 3 days - 132.6 racks/master digger;
132.6 racks x $10 = $1326 f 3 yr. life = $442.
Cost per master digger = $3733.51 x 30 master diggers f 66,428 bu.
= 1.69/bu.
Depuration costs - no change.
I/ Assuming additional yield of 49,928 bu./yr., revised in main report.
£/ Assuming restricted fishery - no change in number of master diggers.
£/ Townsend and Briggs, 1980.
-------
D-17
Tfeble D-8. Comparison of Nonlabor Costs and Prices
location and Year Nonlabor Cost/Bu.
Maine, 1978 $
Maine, 1980 $
Massachusetts, 1980 $
Unrestricted Areas
Restricted Areas
Before Abatement
After Abatement
8.0Lb/
10.49
10.57
16.66
13.73
Inflated
Price/Bu.^ Price/Bu.
n.a. 18. 6 5k/
24.43 22.65£/
24.43 28.002/
n.a. 28.0Q5L/
n.a. 28.00^/
3/ CPI used to inflate 1978 Maine price to 1980 $.
£/ Townsend and Briggs, 1980.
£/ Maine Department of Marine Resources, Clam Production and Value,
1887-1982.
s7Resources for Cape Ann, 1982.
n.a. = Not applicable.
-------
D-18
and employees. However, there is not enough cost and price information
available to address this question adequately.
If the cost computations discussed above formed a reasonable basis on
which to estimate shifts in the supply curve, then they could be used to
calculate change in producer surplus due to pollution abatement. This is
simply not the case because of data inadequacies. For illustration purposes,
however, we could assume that they are acceptable and that the change in per
bushel cost shown in Table D-7 is a reasonable estimate of per unit supply
cost changes due to pollution abatement. Change in producer surplus would
then be computed as follows:
APS = Profits^ - ProfitSQ
= (PlQi - CiQx) - (P0Q0 - C0Q0)
• QI (PI - G!) - Q0 (PO-CO)
where,
Profitsg = initial profits = PoQo -
Profits! = new profits => P^Q^
APS = change in producer surplus (3)
P0 = initial price {$)
P! = new price ($)
QQ = initial quantity harvested (bu.)
Q! = new quantity harvested (bu.)
C0 = initial cost ($)
GI = new cost ($)
As an example, if the preliminary change in yield and initial quantity
harvested (later revised) used in Table D-7 and the initial price of
$28.00/bu. (also revised) and cost of $16.66/bu. used in Table D-8 were
assumed and if a price change of -$1.99 was also assumed (this is also a
preliminary estimate that was made using the preliminary change in yield and
one of the initial demand functions considered, later revised in the main
report), then the change in producer surplus would be computed as follows:
-------
D-19
APS = Profits^ - Profits0
= Q! (?! - Cx) - Q0 (P0-C0)
= (66,428) (26.01 - 13.73) - (16,500) (25.00 - 16.66)
= (66,428)(12.28) - (16,500)(11.34)
= 815,736 - 187,110
= $628,626.
It should be emphasized that this number is only hypothetical. As discussed
earlier, it was thought best to omit computation of producer surplus changes
in the main report because of lack of information to specify supply curve
shifts and because of the likelihood that these changes would be zero due to
the lack of regulation of the fishery.
-------
D-20
References
Crutchfield, Stephen R., February 1983. "Soft Clam Exvessel Demand Functions
and Clam Fishery Data," unpublished data, Department of Resource
Economics, University of Rhode Island, Kingston, RI.
Marasco, Richard J., May 1975. An Analysis of Future Demands, Supplies,
Prices and Needs for Fishery Resources of the Chesapeake Bay, MP 868,
Agricultural Experiment Station, University of Maryland, College Park,
Maryland.
Resources for Cape Ann, April 1982. The Costs of Pollution; The Shellfish
Industry and the Effects of Coastal Water Pollution, Massachusetts
Audubon Society.
Townsend, Ralph and Hugh Briggs, September 1980. Some Estimates of
Harvesting and Processing Costs for Maine's Marine Industries, technical
report, Department of Economics, University of Maine.
Williams, Doug, (no date). "Data and Procedures Used to Estimate Technical
Coefficients for the Clam/Worm Sector," unpublished paper, Department of
Agricultural and Resource Economics, University of Maine.
-------
Appendix E
Charles River Boating Benefits
Additional Boating Days per Year on the Charles River
(1) (2)
Current
Boating Days = ABP x Boating Days
52,810
5,750
= 0.03142
= 0.03142
1,680,800 High
183,000 low
(a) (b)
(1) BP = 0.38485 (A«) + 0.03142 ( AFPS)
0.03142 = (0.3845) (0) + 0.03142 (1)
Source: Davidson P., G. Adams and J. Seneca, 1966. The Social \felue of
Water Recreational Facilities from an Improvement in Water Quality:
the Delaware Estuary. Water Research, Allen Rieese and Stephen C.
Smith, eds. Baltimore: Johns Hopkins University Press for Resources
for the future.
(a) AW = acreage of recreational water available per capita.
= 0, because currently all 675 acres of the Charles River in the
Basin planning area are boatable.
(b) &FPS = change in recreational facility rating.
= 1 (assumed).
(2) current Boating Days
(a) (b) (c)
Boating = Portion of Population x No. days x Boating
Days Boating on Charles per Boater Population
High = 1,680,800 = .40 5.5 764,000
(a) , (b) Source: Recreation studies (see Appendix B).
-------
E-2
(c) Boating population equals population of towns bordering or
very near to the Charles River in the planning area.
Cambridge 95,000
Watertown 34,000
Newton 83,000
Brookline 55,000
3/4 Boston 420,000
Somerville 77,000
Tfotal 764,000
Source: 1980 D.S. Census.
(i) (ii)
Low a Boating = Family visitor x Family
Days days per number
season ' •
Low = 183,000 = 68,000 x 2.69
(i) Source: Calculations based on information in Binkley and Hanemann,
1975, The Recreation Benefits of Water Quality Improvement,
prepared for Environmental Protection Agency, Washington, DC.
5.6 percent of all reported 850 visits for the summer season
were boating-related activities. Sample was statistically
representative of 0.07 percent of the SMSA population.
Therefore 850 = 1,214,286 family visits, of which 5.6 percent,
or 68,000 are family visitor days.
(ii) Source: 1980 U.S. Census.
------- |