Five years after the transfer of Deer Island flows offshore, 2006

Massachusetts Water Resources Authority · from the MWRA · 78 pages · 1.1 MB
open the file ↗ · original at the source ↗

5 years after transfer of Deer
Island flows offshore: an update of
water-quality improvements in 
Boston Harbor 
Massachusetts Water Resources Authority 
Environmental Quality Department 
Report ENQUAD 2006-16

5 years after transfer of Deer Island flows offshore: 
an update of water- quality improvements 
in Boston Harbor 
prepared by 
David I. Taylor 
MASSACHUSETTS WATER RESOURCES AUTHORITY 
Environmental Quality Department 
100 First Avenue 
Charlestown Navy Yard 
Boston, MA 02129 
(617) 242-6000 
September 2006 
Report No: 2006-16 
2

Citation 
Taylor D. I. 2006. 5 years after transfer of Deer Island flows offshore: an update of water-quality
improvements in Boston Harbor. Boston: Massachusetts Water Resources Authority. Report ENQUAD
2006-16. 77 p 
3

TABLE OF CONTENTS 
Page 
EXECUTIVE SUMMARY ……………………………………………… 5 
INTRODUCTION ………………………………………………………. 10 
Aspects addressed by the report...……………………..………… 11 
METHODS 
Field sampling and laboratory analytical procedures…………… 12 
Data and statistical analysis…………………………………….. 15 
RESULTS 
I. DIFFERENCES BETWEEN THE 5-YEARS AND 
BASELINE……………………………………………………….17 
Nitrogen………………...………………………………… 17 
Phosphorus …………………..…………………………... 25 
Molar ratios of N:P……………………………………….. 31 
Phytoplankton biomass (chlorophyll- a )……………..…… 36 
POC and suspended solids……………………………….. 43 
Water clarity……………………………………………… 48 
Bottom-water dissolved oxygen (DO)……………………. 52 
II. SUMMARY OF DIFFERENCES BETWEEN PERIODS…… 57 
DISCUSSION 
Comparison with the first 36-months after transfer………………. 63 
Observed versus predicted changes.……………………………… 65 
Conclusions…………………………………………….................. 68 
ACKNOWLEDGEMENTS………………………………………………. 69 
LITERATURE CITED …………………………………………………… 70 
APPENDIX A 
Salinity and water-temperature…………………………................ 74 
4

EXECUTIVE SUMMARY 
In September 2000, the Massachusetts Water Resources Authority (MWRA) transferred
the flows from the Deer Island wastewater treatment facility (WWTF) offshore, for
diffusion into the bottom waters of Massachusetts Bay. This transfer, which we refer to
here as ‘offshore transfer’, ended over a century of direct WWTF discharges to the
Harbor. 
A number of reports have documented the changes in the water-column of the Harbor
during the first two to three years after transfer. The purpose of the following report was
to update these earlier reports, and document the changes over the first 5 years after the
transfer. With an additional two years of post-transfer data, we can be more confident of
any changes that might have occurred in the Harbor. 
To identify the changes since transfer, we compare water quality in the Harbor during the
5-years with water quality during a 5- to 7-year baseline period before the transfer. This
report, like our earlier reports, focuses on eutrophication-related water quality, because it
was this aspect of the Harbor that was predicted by earlier numerical models, to be most
changed by transfer. 
MWRA collected all water-quality data used in the report. The data were collected at 10
stations, located in the four major regions of the Harbor; the Inner Harbor, North West
Harbor, Central Harbor and South East Harbor. Table I summarizes some of the
differences in the Harbor water-column between the 5-years and baseline. Only
differences significant at p < 0.05 are presented in the Table. 
‘Solid arrows’ indicate variables where the differences between the two periods were
significant for the data averaged Harbor-wide. The ‘hollow arrows’ indicate variables,
where the differences were not significant for the data averaged Harbor-wide, but were
significant at certain stations. The ‘dashes’ indicate variables where the differences were
not significant for the data averaged Harbor-wide or for any of the stations. 
5

Table |. Summary of differences in Harbor water-quality between the 5-years and baseline.

VARIABLE 5 - YEARS
[ TN (uumol I") g -10.7 (-35%)
DIN ( umol!7") ¥ -6.5 (-55%)
DIN as % TN -22 (-55%)
: ¥
Fe TP ( umolI-! ) 3 0.58 (-28%)
E DIP ( ymol I"") § -0.40 (-38%)
z
DIP as % TP & 7 (-15%)
TN:TP UY 0.7 (-5%6)
DIN:DIP a 3.4 (30%)
at ¢ TOTAL CHL-A (,ug!"") 8g “1.7 (-26%)
= = ‘ACTIVE’ CHL-A( jig!) ¥ “1.3 (-28%)
zo
a PHAEOPHYTIN (ug I") a -0.5 (-28%)
- [ POC( umol 17" ) % -12.0 (-28%)
~ 4
A
Ek Tss(mg!"') cent +0.20 (+5%)
a
L POC as %TSS ge 6,0 (-33%)
> — n
& = k (m7) Vv 4 +0.01 (+19)
= 3 a SECCHI DEPTH (m) 4 +0.1 (+496)
6 DO CONC. (mgI!) 4s $0.3 (45%)
5 \,
L DO % SAT. 4> 3 (3%)
C.] ec
ze SALINITY (ppt) - “1.2 {-4%)
cs
ew oO
ae L TEMPERATURE (°C) - 05 (-5%)
wv

Blue arrows Indicate change that might be interpreted as an ‘improvement’. Red arrows Indicate change that
might not be viewed as an Improvement. Gray symbols Indicates a change that cannot be assessed as an
improvement or non-improvement. Values adjacent to arrows indicate the difference in the Harbor-wide
averages between the 5-years and baseline. This difference expressed as a percent of the average during baseline
Is shown In parentheses.

For 18 of the 20 variables shown in Table I, the averages after transfer were significantly
different from baseline, for either the data averaged Harbor-wide, or for individual
stations. For 12 of the 18 variables, the differences were significant for the data averaged
Harbor-wide; for 6 of the 18 variables, the differences were significant only at individual
stations. 
For concentrations of most of the N and P nutrients, and especially for the dissolved
inorganic fractions, the Harbor-wide averages after transfer were significantly different
(and in all cases, lower) than baseline. The decreases were also significant at most or all
of the 10 stations, suggesting that the decreases were observed over most of the area of
the Harbor. 
For most of the N and P nutrients that showed significant decreases (excluding the
particulate fractions), the decreases were also observed during all four seasons, but were
largest during fall and winters, the seasons when concentrations in the Harbor tend to be
seasonally greatest. 
For concentrations of chlorophyll- a (chl- a ), a measure of phytoplankton biomass, Harbor-
wide average concentrations during the 5-years were also significantly lower than
baseline, presumably in response to the reduction in N loadings that followed transfer.
The decreases were confined to a localized area of the Central Harbor, and were largest
during summers, the season when concentrations in the Harbor tended to be highest. 
The Harbor also showed significantly lowered concentrations of particulate organic
carbon (POC), a measure of detrital mass, during the 5-years. As for N and P, the
decreases were observed over most of the area of the Harbor. As for chl- a , but unlike for
N and P, the decreases were most pronounced during summers, again when
concentrations were seasonally greatest. 
For TSS, the Harbor-wide averages during the 5-years were not significantly different
from baseline. (Note, the bulk of the decrease in wastewater loadings of TSS occurred 
7

before offshore transfer, and the decrease after offshore transfer was much smaller than
for N or P). Significant differences in TSS were observed between the 5-years and
baseline, at individual stations. At stations in the Inner Harbor, and perhaps along the
west margin of the Harbor, concentrations were lower than baseline. At the mouth of the
Outer Harbor, the opposite applied. 
For both attenuation coefficients ( k ) and secchi depth, the Harbor-wide averages during
the 5-years were not significantly different from baseline. Both variables indicated
elevated clarity in the Inner Harbor extending along the west margin of the Harbor. The 
k data in turn indicated significantly lowered clarity at the mouth of the Outer Harbor,
perhaps extending into the South Harbor. 
For neither dissolved oxygen (DO) concentrations nor DO % saturation, were the year-
round, Harbor-wide averages after transfer significantly different from baseline. Both
variables, however, showed significant increases for the data averaged Harbor-wide,
during the ‘mid-summers’ after transfer. For the year-round data, average DO
concentrations were significantly elevated compared to baseline, at stations in the mid- to
upper Inner Harbor. 
For only two of the 20 variables, salinity and water-temperatures, were the averages after
transfer not significantly different from baseline, for both the Harbor-wide averages and
the averages for the individual stations. The fact that these variables were not
significantly different between periods, suggests that background environmental changes
were not responsible for other differences in water quality observed between the two
periods. 
In the Discussion section of the report we compare the differences observed for the 5-
years, with the differences predicted by others before transfer, using 3-D numerical
models. For all four variables where such comparisons were possible (dissolved
inorganic nitrogen, chl- a , DO concentrations, and salinity), the observed differences were
within < 15% of the predicted differences. 
8

The similarity of the observed and predicted differences indicates that, at least for these
four variables, the transfer of the WWTF discharges offshore was largely responsible for
the differences observed between periods. A more sophisticated statistical analysis
would allow a better separation of the effects of the transfer from background
environmental changes that might have occurred over the same period. 
9

INTRODUCTION 
In September 2000, the flows from the Deer Island (DI) wastewater treatment facility
(WWTF) to Boston Harbor were transferred 16-km offshore, for diffusion into the
bottom-waters of Massachusetts Bay (Rex et al . 2000). This transfer, which we refer to
here as ‘offshore transfer’, ended over a century of direct WWTF discharges to the
Harbor. 
Before this transfer, non-oceanic loadings of materials that cause or exacerbate
‘eutrophication’, and specifically of total nitrogen (TN), were estimated to be among the
highest reported for bays or estuaries in the USA (Kelly 1997). The WWTFs that at the
time discharged to the Harbor, were in turn responsible for the bulk (>90%) of these
elevated loadings of TN (and TP) (Alber and Chan 1994). 
Numerical modeling studies conducted by others before the transfer (HydroQual and
Normandeau 1995; Signell et al . 2000), predicted that the transfer of the WWTF
discharges offshore would lead to improvements in eutrophication-related Harbor water-
quality. Some of the changes in the Harbor water-column since transfer have been
documented by Libby et al . (2006), Taylor (2004) and Rex et al . (2002). 
Most of these studies, however, focused on the first two to three years after the transfer.
The purpose of the following report was to update these reports, and document the
changes during the first 5 years after transfer. With an additional two years of post-
transfer data now available, we can be more certain of the changes that have taken place
in the Harbor. 
To identify the changes in the Harbor, and we used the same approach in some of our
earlier reports, we compared eutrophication-related conditions in the Harbor water-
column during the first 5-years after transfer, with conditions during a baseline period
before transfer. We did this at two levels; one at the level of the Harbor as a whole, using 
10

Harbor-wide averages, and the other, at the level of the individual stations sampled in the
Harbor, to determine the spatial extents of the differences. 
In this report we also took the comparison a step further, and compared water quality
before and after transfer during different seasons, to determine whether the sizes and the
nature of the changes in the Harbor were different during different seasons. The changes
in the Harbor were then compared (where possible) with the changes predicted by others
before transfer, using numerical models. 
The earlier model predictions were made by ‘numerically transferring’ the WWTF
discharges offshore, meaning that the predicted changes could be used to determine the
extent to which the changes were caused by transfer of WWTF discharges. In this study,
the 5-years after transfer extended from 7 September 2000, the day after the actual
transfer, through 31 December 2005. The ‘baseline’ period extended from August 1993
(or August 1997, depending on variable), through 6 September 2000. 
Aspects of water quality addressed 
This report, like Taylor (2004), focuses on aspects related to eutrophication, or as defined
by Nixon (1995), ‘organic over-enrichment’ of the Harbor water-column. Specific
aspects of eutrophication addressed included; concentrations of N and P (the two
nutrients most responsible for eutrophication), phytoplankton biomass (measured as
chlorophyll- a ), concentrations of total suspended solids (TSS), concentrations of
particulate organic carbon (POC), water-clarity, and bottom-water dissolved oxygen
(DO). 
Others have addressed other aspects related to eutrophication of the Harbor, including
rates of pelagic primary production (Libby et al . 2006), structure of the phytoplankton
communities (Libby et al . 2006), sediment redox characteristics (Maciolek et al . 2006),
benthic metabolism and nutrient fluxes (Tucker et al . 2005), and biomass and structure of 
11

the benthic invertebrate communities (Maciolek et al . 2006). Others have addressed the
changes to Massachusetts Bay. 
We focused the report on eutrophication of the Harbor, because of the recent scientific
interest in eutrophication of coastal aquatic ecosystems (Nixon 1995, Bricker et al . 1999),
and the numerous symptoms of eutrophication that had been documented in the Harbor
before transfer (some of these symptoms have been described in Taylor 2001). Loadings
of N and P to the Harbor had been shown to be elevated compared to other bays and
estuaries (Kelly 1997, 1998). 
The discharges from the WWTFs, that would in turn be transferred offshore, had been
shown to be responsible for the bulk of these elevated loadings (Alber and Chan 1994).
The report focused on the water-column, because prior to the transfer, the wastewater
was discharged directly to the water-column, and the water-column is often the part of an
aquatic ecosystem that responds most rapidly to changes in external inputs. 
METHODS 
Field sampling and laboratory analytical procedures 
The Massachusetts Water Resources Authority (MWRA), the agency responsible for the
transfer of the DI discharges offshore, collected all the Harbor water-quality data used in
the report. The data were collected as part of the Boston Harbor Water Quality
Monitoring (BHWQM) project. For details of the BHWQM project, see its Quality
Assurance Project Plan (Rex and Taylor 2000). 
All water-quality data were collected at 10 sampling stations (Fig. 1). The names and
coordinates of the 10 stations are listed in Table 1. Note: the 10 stations were located in 
each of the 4 major regions of the Harbor; three in the ‘Inner Harbor’ region, three in the
‘North West Harbor’, three in the ‘Central Harbor’, and one in the ‘South East Harbor’
region. 
12

At 8 of the 10 stations, sampling was initiated in August 1993; at two of the 10 stations,
specifically Station 077 and 137, sampling was initiated in June 1994 and June 1995,
respectively. Sampling at these stations was added during the course of the study, to
increase the spatial coverage of the Harbor. At all stations, measurements were
conducted weekly from May through October, and every two weeks from November
through April. 
For most variables, measurements at each station were conducted at two depths; one,
‘near-surface’ (at ca. 0.3 m below the water surface), and the other, ‘near-bottom’ (or ca. 
13

0.5 m above the Harbor bottom). The variables monitored at both depths included
dissolved inorganic nitrogen (DIN), ammonium (NH 4 ), nitrate + nitrite (NO 3+2 ),
chlorophyll- a (chl- a ), phaeophytin, total suspended solids (TSS), dissolved oxygen (DO),
salinity and temperature. 
Table 1. Locations of the stations sampled to track differences in Harbor water-
quality between the 5-years and baseline. 
Region Station name Station ID Latitude (N) Longitude (W) 
NORTH HARBOR 
Inner Harbor Mouth Mystic River 137 42 o 23.20 71 o 03.80 
New England Aquarium 138 42 o 21.59 71 o 02.82 
Mouth Inner Harbor 024 42 o 20.59 71 o 00.48 
North West Long Island 106 42 o 20.00 70 o 57.60 
Harbor Calf Island 142 42 o 20.35 70 o 55.89 
Neponset River/ 140 42 o 18.35 71 o 02.43 
Dorchester Bay 
SOUTH HARBOR 
Central Harbor Inner Quincy Bay 077 42 o 16.51 70 o 59.31 
Hangman Island 139 42 o 17.20 70 o 58.10 
Nantasket Roads 141 42 o 18.30 70 o 55.85 
South East Hingham Bay 124 42 o 16.36 70 o 53.86 
Harbor 
For all these variables, excluding DO and the variables that were measured at ‘near-
surface’ alone, all data presented in the report were averages for the two depths. For DO,
only the data from the ‘near-bottom’ depth were reported; this being the depth where any
differences in DO would be of most concern. The variables sampled at ‘near-surface’
alone included, total N (TN), particulate N (PN), non-DIN, total P (TP), particulate P
(PP), non-DIP and POC. 
14

Concentrations of TN were estimated indirectly by summing concentrations of total
dissolved nitrogen (TDN) and PN for each sample collected at the ‘near-surface’ depth. 
Similarly, concentrations of TP were estimated by summing concentrations of total
dissolved phosphorus (TDP) and PP. Note, for chl- a and for TP, data were only available
after mid-1995. 
Table 2 provides a summary of the field procedures and analytical techniques employed
in the study. All standard operating procedures for all analytical techniques are archived
at the MWRA Central Laboratory, Deer Island, Winthrop, MA 02152. All data used in
the report are archived in the EM & MS Oracle database (MWRA Environmental Quality
Department, Charlestown Navy Yard, Boston MA 02129), and are available on request. 
Data and statistical analysis 
In this report, for both the Harbor as a whole, and for the individual stations, we used the
non-parametric Mann-Whitney U test to test whether means after transfer were
significantly different from the means during baseline (SPSS 10.1, SPSS 2002). We used
the Mann-Whitney U test in preference to conventional ANOVA, because for many of
the variables, especially at individual stations, the requirement of non-homogeneity of
variance of the data could not be met (even with data transformation). 
To test for differences between periods for the Harbor as a whole, we applied the Mann-
Whitney U test to volume-weighted Harbor-wide averages. These averages were
computed as follows (after Sung 1991): 
Volume-weighted average = ( a *0.119) + ( b *0.418) + ( c *0.342) + ( d *0.12) 
where, a = average concentration for all stations in the Inner Harbor, b = average
concentration for all stations in North West Harbor, c = average concentration for all
stations in Central Harbor, and d = average concentration for all stations in South East 
15

Table 2. Summary of field and analytical methods. 
VARIABLE METHOD 
TDN a and TDP a 
PN a 
PP a 
Ammonium b 
Nitrate + nitrite b 
Phosphate b 
Chlorophyll a b , 
phaeophytin b 
Secchi depth d 
k d 
TSS b 
Dissolved oxygen c 
Salinity b and water 
temperature b 
Solarzano and Sharp (1980 b ), Whatman G/F filters 
Perkin Elmer CHN analyzer, Whatman GF/F 
Solarzano and Sharp (1980 a ), Whatman GF/F 
Fiore and O'Brien (1962), modified as in Clesceri et al.
(1998; Method 4500-NH3 H), Skalar SAN plus autoanalyzer,
Whatman GF/F filters 
Bendschneider and Robinson (1952), modified as 
in Clesceri et al. (1998; Method 4500-NO3 F), 
Skalar SAN plus autoanalyzer, Whatman GF/F 
filters 
Murphy and Riley (1962), modified as in Clesceri et al.
(1998; Method 4500-P F), Skalar SAN plus autoanalyzer,
Whatman GF/F filters 
After Holm Hansen (1965) as described in 
EPA (1992). Sequioa Turner Model 450 
fluorometer, Whatman GF/F filters 
20 cm standard (all-white) secchi disc 
Li Cor PAR sensor Model LI-193 SB 
Clesceri et al. (1998, Method 2540D), using 
nucleopore filters 
YSI 3800 through July 1997, Hydrolab 
Datasonde 4 thereafter 
YSI 3800 through July 1997, Hydrolab 
Datasonde 4 thereafter 
a = surface samples only, b = samples/measurements taken surface plus bottom, c = measurement
taken at bottom only, d = profile through water column. 
Harbor. The constants, 0.119, 0.418, 0.342 and 0.12, were the volumes of the respective
regions expressed as a proportion of 1 (volumes from Sung 1991, citing Ketchum 1951). 
16

For the year-round data, (for both the Harbor-wide averages and the averages for
individual stations), we applied the Mann-Whitney U tests to average monthly data,
rather than to data from individual surveys. This precaution was necessary to avoid
biasing the findings by the different intensities of sampling during different periods of the
year. For the data for the individual seasons, we applied the test to the averages for
individual surveys. 
In this report, unlike in Taylor (2004), none of the data were ‘de-seasonalized’ before we
applied the Mann-Whitney U tests to the data. Thus, the tests (as applied) compare the
averages before and after transfer, with each of the averages inclusive of any seasonal
‘signal’ within the period. To separate out ‘any seasonal differences in responses, we
then applied the Mann-Whitney U tests to the survey averages, for individual seasons. 
In this report we have recognized three levels of significance. When the Mann-Whitney
U test yielded p values < 0.05, we considered the difference to be ‘significant’ (at 95%
CL), and used an asterisk (*) to denote this condition. When p was > 0.05 but < 0.10, we
considered the difference ‘almost significant’ (i.e. significant at 90% CL), and used a ‘ ? ’
to denote this condition. When the p values were > 0.10, we considered the difference
‘not significant’. 
RESULTS 
I. DIFFERENCES BETWEEN THE 5-YEARS AND BASELINE 
Nitrogen 
Total nitrogen (TN). During the 5-years after offshore transfer, Harbor-wide average
concentrations of TN were significantly lower than baseline (Table 3). During the
baseline, TN concentrations averaged 30.9 + 6.4 μmol l -1 . During the 5-years, they 
17

averaged 20.2 + 2.9 μmol l -1 . The difference of -10.7-μmol l -1 was equivalent to -35% of
average baseline values, and was significant at p < 0.05 (Mann-Whitney U test). 
Table 3. Nitrogen concentrations . Comparison of volume-weighted Harbor-wide
average concentrations between the 5-years after transfer, and baseline. Values are
averages + 1 x SD of average monthly values, with the n value in parentheses. 
Variable Average values during Difference p 
between baseline 
Baseline 5-years and 5-years 
TN 30.9 + 6.4 20.2 + 2.9 -10.7 (-35%) <0.01 * 
( μ mol l -1 ) (61) (62) 
DIN 11.8 + 6.4 5.3 + 3.5 -6.5 (-55%) <0.01 * 
( μ mol l -1 ) (75) (62) 
DON 13.0 + 3.7 11.6 + 1.9 -1.5 (-11%) 0.075 ? 
( μ mol l -1 ) (61) (62) 
NH 4 6.3 + 3.4 1.2 + 0.9 -5.1 (-81%) <0.01 * 
( μ mol l -1 ) (75) (62) 
NO 3+2 5.5 + 3.8 4.1 + 3.2 -1.4 (-25%) 0.02 * 
( μ mol l -1 ) (77) (62) 
PN 6.1 + 2.4 4.1 + 1.3 -2.0 (-33%) 0.06 ? 
( μ mol l -1 ) (60) (60) 
DIN 39 + 17 17 + 14 -22 (-55%) <0.01 * 
as %TN (68) (62) 
NH 4 53 + 14 29 + 15 -29 (-51%) <0.01 * 
as %DIN (75) (62) 
The fact that TN concentrations were lowered after transfer is readily evident in the time-
series and bar-chart plots of the Harbor-wide average concentrations in Figures 2 and 3,
respectively. The time-series plots show also that the strong seasonal cycle of
concentrations seen before transfer was ‘dampened’ during the 5 years after transfer. 
18

1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005
0
10
20
30
40
50
60 
Fig. 2. Total nitrogen (TN) partioned into DIN and non-DIN components. Time-series
plots of survey average, volume-weighted Harbor-wide average concentrations. 
Offshore
transfer
AVERAGE CONCENTRATION
( UMOL L -1 ) 
DIN 
OTHER 
1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005
0
5
10
15
20
25
30
35
40 NH 4
NO 3+2
Non-DIN
Fig. 3. Annual average, Harbor-wide average concentrations of TN partitioned into the NH 4 , 
NO 3+2 , and non-DIN fractions , 1994 through 2005. 
Offshore
transfer 
ANNUAL AVERAGE CONC.
( UMOL L -1 ) 
19

As can be seen in the bar chart plot, the annual average TN-concentrations during each of
the 5 years were consistently lower than during all years during baseline. During the 5-
years after the discharges were ended, TN concentrations showed a background increase.
The increase was between one-fourth and one-third the size of the decrease that followed
transfer. 
At the individual stations, average TN concentrations were significantly lower during the
5-years than baseline at all 10 stations (Fig. 4). At the individual stations the decreases
ranged in size from -9-μmol l -1 to -14-μmol l -1 . Expressed as percent of baseline, they
ranged from -21% (Station 137), to -46% of baseline (Station 106). (Note; the shaded
areas in this and later Figures simply enclose the stations that showed differences
significant at p < 0.05. The shading is not meant to be a contour plot). 
The fact that the decreases were significant at all 10 stations suggests that after transfer
average concentrations of TN were lower than baseline over most of the area of the
Harbor. Some of the largest decreases were observed at stations located closest to the
former DI (Stations 106 and 142) and NI outfalls (Station 139), but decreases were also
observed up into the Inner Harbor and down into Hingham Bay. 
For the TN-data averaged Harbor-wide, significant decreases were also observed during
all four seasons (Table 4). Thus, concentrations after transfer were significantly lower
than baseline year-round. Both in absolute and percent terms, the decreases were largest
during fall and winter, when before transfer concentrations tended to be seasonally
greatest, and smallest during spring and summer, when concentrations tended to be
smallest. 
During three of the four seasons, and specifically during summer, fall and winter, the
decreases in TN were significant at all 10 stations (Fig. 5). During the remaining season,
spring, the decreases were significant at 6 stations extending over most of the area of the
Outer Harbor, and ‘almost’ significant at one other station located at the mouth of the 
20

™N

Legend

Former outfalls

Decrease

Increase

Nadel

significant
p<0.05

LA
+ +
~ & Ye

‘almost’ significant
p<0.10> 0.051

Fig. 4. Total nitrogen (TN) and dissolved inorganic nitrogen (DIN). Differences in average
concentrations (mol I-! ) between the first 5-years after offshore transfer and baseline, at the 10 stations.
‘Down-facing triangles’ indicate stations at which subtraction of averages after transfer from averages before transfer
yielded negative values; ‘up-facing triangles’ ndicate stations at which the opposite applied. ‘Blue triangles’ indicate
differences might be viewed as beneficial; ‘red triangle; differences might not be viewed as beneficial. ‘Asterisks’ are
used to identify the stations at which the differences between the two periods were significant (p < 0.05). ‘?' indicates
stations at which the differences were significant at p > 0.05, but < 0.10. a

21

Table 4. Differences in average N concentrations between the 5-years and baseline, 
by season. ‘Spring’ = April, May, ‘Summer’ = June, July August, September, ‘Fall’ =
October, November, ‘Winter’ = December, January, February, March. 
Fraction Difference (μmol l -1 , % baseline) 
Spring Summer Fall Winter 
TN -7.8 (-27%) * -7.1 (-28%) * -12 (-36%) * -15.4 (-42%) * 
DIN -5.1 (-52%) * -3.1 (-59%) * -8.7 (-63%) * -10.4 (-54%) * 
NH 4 -3.8 (-81%) * -2.5 (-75%) * -7.0 (-78%) * -7.8 (-85%) * 
NO 3+2 -1.2 (-24%) * -0.4 (-23%) ? -1.7 (-35%) ns -2.6 (-26%) * 
PN -2.0 (-31%) * -3.0 (-38%) * -1.7 (-27%) * -1.6 (-35%) * 
DON -1.3 (-10%) ns -1.3 (-10%) ns 1.25 (-9%) ns -1.25 (-9%) ns 
Inner Harbor. (Note, at the three stations located closest to river mouths, the decreases
during spring were not significant.) 
Fractions of TN. For three of the 5 fractions of TN that we monitored, Harbor-wide
average concentrations during the 5-years were also significantly lower than baseline (at 
p < 0.05). The three fractions that showed the significant decreases were dissolved
inorganic nitrogen (DIN), and the two fractions of DIN; specifically ammonium (NH 4 )
and nitrate + nitrite (NO 3+2 ). 
For DIN, the concentrations after transfer were -6.5-μmol l -1 (or -55%) lower than
baseline. This decrease of -6.5-μmol l -1, in turn accounted for most (and 60%) of the 
22

decrease in TN of -10.7 -μmol l -1 . Note, DIN too contributed most, and in this case, 73%
of the TN in the secondary-treated wastewater discharged to the Harbor during baseline
(MWRA unpublished data). 
Average NH 4 concentrations during the 5-years were -5.1-μmol l -1 (or 81%) lower than
baseline, and accounted for 78% of the decrease in DIN. NO 3+2 accounted for the
remaining -1.4-μmol l -1 or 22% of the decrease in DIN. Note, NH 4 too accounted for 
23

most (and in this case >95%) of the DIN in the secondary-treated wastewater discharged
to the Harbor during baseline (MWRA, unpublished data). 
For both particulate nitrogen (PN) and dissolved organic nitrogen (DON), subtraction of
the Harbor-wide averages after transfer from the averages before transfer yielded
negative values, suggestive of decreases. For both variables, the differences were
‘almost’ significant ( p > 0.05, but < 0.10; 90% C.L.). With longer-term monitoring, these
differences may or may not be shown to be significant. 
For DIN, as for TN, the decreases were significant at all 10 stations (Fig. 4). (The same
applied for NH 4 and NO 3+2 , MWRA unpublished data). At the individual stations, the
decreases in DIN ranged in size from -6-μmol l -1 to -8-μmol l -1 . As for TN, the stations
located closest to the former DI outfalls (Stations 106 and 142) showed some of the
largest decreases. 
For three of the fractions of TN, specifically DIN, NH 4 and PN, the decreases in Harbor-
wide average concentrations were significant during all four seasons (Table 4). For DIN
and NH 4 , the decreases were largest during winter; for PN, they were largest during
summer. (For the respective fractions, as for TN, the decreases were largest during the
seasons when concentrations before transfer were seasonally greatest). 
For NO 3+2 , the Harbor-wide average concentrations showed significant decreases during
winter and spring, but not during summer or fall. For DON, the Harbor-wide averages
before and after transfer were not significantly different during any of the four seasons. 
Nature of the TN pool. Figure 6 compares the nature of the TN-pool of the Harbor
water-column before and after transfer. Before transfer, DIN and DON contributed most
and similar proportions of the TN (38% and 42%, respectively). After transfer, the
percent contribution of DON was increased to more than one-half (57%), the percent
contribution of DIN was decreased to about one-fourth (26%), and the percent
contribution of PN remained unchanged at 20%. 
24

0
5
10
15
20
25
30
35
40 
NO 2+3
NH 4
DON
PN 
Figure 6. Nitrogen pool . Comparison of the nature of the TN pool of the Harbor water column
between the baseline period, and the first 5-years after wastewater discharges to the
Harbor were ended. Values in bars are the percent contributions of the different
constituents to the total pool. 
PN
NO 2+3
NH 4
DON 
36-MONTHS
POST-TRANSFER
BASELINE 
6%
20%
20%
57%
18%
20%
20%
42% 
42 
AVERAGE CONC. ( UMOL L -1 ) 
The nature of the DIN fraction of the TN-pool was also changed. Before transfer, NH 4 
and NO 3+2 contributed similar proportions (and 53% and 47%) of the TN. After transfer,
the percent contribution of NO 3+2 was increased to three-fourths (77%); and NH 4 
decreased to about one-fourth (and 23%). The background increase in NO 3+2 during the
5 years (Fig. 4), will also have contributed to the increase in the percent contribution of
NO 3+2 . 
Phosphorus 
Total phosphorus (TP). During the 5-years, Harbor-wide average concentrations of TP
were also significantly lower than baseline (Table 5). During baseline, TP concentrations
averaged 2.06 + 0.32 μmol l -1 ; during the 5-years they averaged 1.48 + 0.31 μmol l -1 . 
25

The difference of -0.58-μmol l -1 was equivalent to -28% of baseline, and was significant.
The -28% decrease was slightly smaller than the -34% decrease for TN. 
Table 5. Phosphorus concentrations . Comparison of volume-weighted Harbor-wide
average concentrations, between the 5-years after transfer and baseline. 
Variable Average values during Difference p 
between baseline 
Baseline 5-years and 5-years 
TP 2.06 + 0.32 1.48 + 0.31 -0.58 (-28%) <0.01 * 
( μ mol l -1 ) (67) (63) 
DIP 1.05 + 0.37 0.65 + 0.27 -0.40 (-38%) <0.01 * 
( μ mol l -1 ) (68) (63) 
DOP 0.39 + 0.24 0.40 + 0.22 +0.01 (+3%) 0.85 ns 
( μ mol l -1 ) (61) (63) 
PP 0.58 + 0.18 0.42 + 0.12 -0.15 (-26%) <0.01 * 
( μ mol l -1 ) (61) (63) 
DIP 51 + 16 43 + 16 -7 (-15%) <0.01 * 
as % TP (61) (63) 
The fact that the concentrations of TP were lowered after transfer is evident in the time-
series and bar chart plots of Harbor-wide average concentrations of TP in Figures 7 and
8. The seasonal cycle of TP was also ‘dampened’ after transfer, but less so than for TN.
Also unlike for TN, TP showed no evidence of a background increase through the 5 years
after transfer. 
The scatter plot of annual average concentrations of TN versus TP (Fig. 9) shows that
during the years after transfer when the TP concentrations were lowest, the
concentrations of TN were also lowest. The shallower slope of the observed data relative
to the diagonal line (which represents the 1:10 TP:TN ratio), suggests that the decrease
for TP was smaller relative to the decrease for TN. 
26

1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005
0
1
2
3
4 
Fig. 7. Total phosphorus (TP) partioned into DIP and non-DIP components. Time-series
plots of survey volume-weighted Harbor-wide average concentrations. 
Offshore
transfer
AVERAGE CONCENTRATION
( UMOL L -1 ) 
DIP 
NONDIP 
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005
0
1
2
3 
Figure 8. Annual Harbor-wide average TP concentrations partitioned into the DIP and
non-DIP fractions, 1995 through 2005.
Offshore
transfer 
ANNUAL AVERAGE P CONC
( UMOL L -1 ) 
DIP 
NONDIP 
27

10 15 20 25 30 35 40 45
0.5
1.0
1.5
2.0
2.5
3.0
3.5 1:10 
Fig. 9. TP versus TN . Data are annual, volume-weighted Harbor-wide averages. Note:
diagonal line represents a 1 : 10 ratio of TP:TN.
ANNUAL AVERAGE TP CONC
( UMOL L -1 )
ANNUAL AVERAGE TN CONC.
( UMOL L -1 ) 
BASELINE 
POST 
The decreases in TP were also significant at all 10 stations (Fig. 10). The sizes of the
decreases ranged from -0.4-μmol l -1 to -0.8-μmol l -1 . The decreases were largest in the
outer North East Harbor and the mid- Central Harbor, the locations of the former WWTF
outfalls. Thus, the decreases were also observed over most of the area of the Harbor, as
they were for TN. 
For the data averaged Harbor-wide, and as for TN, the decreases were also significant
during all four seasons (Table 6). Again as for TN, the decreases were largest during
winter and smallest during summer. The differences in the sizes of the decreases during
winter and summer were smaller than for TN, accounting for the smaller ‘dampening’ of
the seasonal cycle after transfer for TP than TN. 
28

Other P fractions. For two of the three fractions of TP that we monitored, DIP and PP,
Harbor-wide averages during the 5-years were also significantly lower than baseline.
For DIP, average concentrations decreased by -0.40-μmol l -1 (or -38%) between the two 
29

periods; for PP, they decreased by 0.15-μmol l -1 (or -26%). For DOP, as for DON, the
averages before and after transfer were not significantly different. 
Table 6. Seasonal differences in P concentrations between the 5-years and baseline . 
Fraction Difference (μmol l -1 , % baseline) 
Spring Summer Fall Winter 
TP -0.58 (-32%) * -0.53 (-26%) * -0.61 (-25%) * -0.65 (-33%) * 
DIP -0.24 (-39%) * -0.34 (-37%) * -0.61 (-40%) * -0.46 (-40%) * 
PP -0.24 (-36%) * -0.21 (-29%) * -0.1 (-21%) * -0.08 (-19%) ? 
DOP -0.08 (-17%) ns -0.02 (-4%) ns -0.16 (-47%) ns -0.05 (-16%) ns 
As for N, the dissolved inorganic fraction (DIP) accounted for most (69%) of the
decrease in TP. In fact, the 69% contribution by DIP was almost identical to the 68%
that DIN contributed to the TN-decrease. As for DIN, the decrease in DIP was also
significant at all 10 stations (lower panel, Fig. 10). 
Differences in the TP-pool . The nature of the TP-pool of the Harbor water-column was
also changed after transfer (Fig. 11). The percent contribution of DIP decreased from
51% before transfer, to 44% after. (Note, both before and after transfer, DIP contributed
a greater proportion of the TP pool than did DIN of TN.) Conversely, the percent
contribution of DOP was increased from 19% to 27%. The percent contribution of DOP
remained unchanged, and 28%. 
30

0.0
0.4
0.8
1.2
1.6
2.0
2.4 
Figure 11. Total phosphorus pool . Differences in the nature of the TP pool of the
Harbor water-column, before and 5-years after offshore transfer. 
PP
DIP
DOP
PP
DIP 
5-YEARS
POST-TRANSFER
BASELINE
AVERAGE CONC. ( UMOL L -1 ) 
DOP 27% 19%
44%
51%
28%
28% 
Molar ratios of N:P 
Molar TN:TP . Unlike for TN and TP, the Harbor-wide average ratios of TN:TP during
the 5-years were not significantly different from baseline (Table 7). During baseline,
TN:TP ratios averaged 15.2 + 3.5:1; during the 5-years, they averaged 14.4 + 3.5:1.
Subtraction yielded a negative value of -0.7:1 (or -5%), but the difference was not
significant. 
Both before and after transfer, and especially during winters, average TN:TP ratios in the
Harbor varied widely between years (Fig. 12 and 13). This variability was such that the
annual averages during the 5 years spanned the range of averages during baseline. The
fact that the annual averages were not consistently lower than baseline during all 5 years
after transfer was unlike for TN and TP. 
31

Table 7. Molar ratios of N:P . Comparison of volume-weighted, Harbor-wide
average ratios between the 5-years after transfer, and baseline. 
Variable Average values during Difference p 
between baseline 
Baseline 5-years and 5-years 
TN:TP 15.2 + 3.2 14.4 + 3.5 -0.7 (-5%) 0.17 ns 
(61) (61) 
DIN:DIP 11.4 + 5.2 8.0 + 5.5 -3.4 (-30%) <0.01 * 
(68) (63) 
DON:DOP 43 + 33 35 + 15 -8 (-18%) 0.15 ns 
(60) (61) 
PN:PP 10.0 + 2.3 9.6 + 1.8 -0.5 (-5%) 0.27 ns 
(61) (62) 
At the individual stations, average TN:TP ratios during the 5-years were significantly
different from, and in both cases significantly lower than baseline, at two of the 10
stations (Fig. 14). The two stations were Stations 106 and 139, the stations located
closest to the former DI and NI WWTF outfalls, respectively. At two other stations, also
located in the Outer Harbor, the decreases were ‘almost’ significant. 
During the 5-years, the Harbor-wide average TN:TP ratios were significantly different
from baseline during one of the four seasons, specifically winter (Table 8). Average
TN:TP ratios during the winters after transfer were -2.2:1 (or -12%) lower than baseline.
During the other three seasons, the averages after transfer were not significantly different
from baseline. 
Thus, the Harbor did show significant decreases in ratios of TN:TP, as it did for the
concentrations of TN and TP individually, but the decreases were smaller, and more
spatially localized and seasonally restricted than for TN and TP. 
32

1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005
0
10
20
30 
Fig. 12. Molar ratios of TN:TP and DIN:DIP. Time-series plots of survey volume-weighted
Harbor-wide averages. Also shown is the Redfield Ratio of 16:1.
Offshore
transfer DIN : DIP 
TN : TP 
Redfield
Ratio 
MOLAR N:P RATIO 
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005
0
5
10
15
20
25 
Fig. 13. Molar N:P ratios . Annual Harbor-wide average molar N:P ratios, 1995 through 2005.
Redfield
ratio 
Offshore
transfer
ANNUAL AVERAGE N : P RATIOS TNTP 
DINDIP 
33

Other molar N:P ratios . For one of the other three N:P ratios that we monitored,
specifically DIN:DIP, the Harbor-wide averages during the 5-years were significantly
different from, and in this case lower than baseline (Table 7). For the other two N:P 
34

Table 8. Seasonal differences in N:P ratios . Differences in average molar N:P ratios
between the 5-years and baseline, by season. 
Fraction Difference (% baseline) 
Spring Summer Fall Winter 
TN:TP -1.4 (-9%) ns -0.1 (-1%) ns -0.9 (-6%) ns -2.2 (-12%) * 
DIN:DIP -3.8 (-33%) ns -3.3 (-48%) * -3.8 (-31%) * -3.2 (-20%) * 
PN:PP -1.0 (-10%) ns -1.4 (-13%) * +0.7 (+8%) ns +0.1 (<+1%) ns 
DON:DOP +3.8 (+12%) ns -5.4 (-13%) ns -16.6 (-36%) ns -11.1 (-23%) ns 
ratios, PN:PP and DON:DOP, subtraction yielded negative values, but as for TN:TP, the
differences were not significant. 
For DIN:DIP, the Harbor-wide averages decreased from 11.4 + 5.2:1 during baseline, to
8.0 + 5.5:1 during the 5-years. The difference of -3.4:1 was equivalent to -30% of
baseline, and was significant. This -30% decrease was smaller than the -55% and -40%
decreases for concentrations of DIN and DIP, respectively. Thus, the decrease for
DIN:DIP was greater than for TN:TP, but smaller than for DIN or DIP. 
Average DIN:DIP ratios in the Harbor also showed significant decreases at 8 of the 10
stations; c.f. two stations for TN:TP (Fig. 14). The 8 stations that showed the significant
decreases were located in the main basin of the Outer Harbor extending up into the Inner
Harbor. Only at Stations 137 and 140, the two stations located at river mouths, were the
decreases not significant. 
35

During the individual seasons, the Harbor-wide average ratios of DIN:DIP were
significantly lower than baseline during summer, fall and winter; only during spring,
were the averages after transfer not significantly different from baseline. For PN:PP, the
averages after transfer were significantly lower than baseline during summer, but not the
other three seasons. For DON:DOP, we were unable to detect significant differences
between periods, during any of the four seasons. 
Phytoplankton biomass (chlorophyll- a ) 
For all three fractions of chlorophyll- a (chl- a ) that we monitored (total chl- a , acid-
corrected chl- a , and phaeophytin), the Harbor-wide averages during the 5-years were
significantly different, and lower than baseline (Table 9). For total chl- a (the sum of
acid-corrected chl- a plus phaeophytin), concentrations decreased from 6.5 + 3.9 μg l -1 
during baseline, to 4.8 + 2.4 μg l -1 during the 5-years. The difference of -1.7-μg l -1 was
equivalent to -26% of baseline, and was significant. 
Table 9. Chlorophyll-a and phaeophytin . Comparison of volume-weighted Harbor-
wide average concentrations between the 5-years after transfer, and baseline. 
Variable Average values during Difference p 
between baseline 
Baseline 5-years and 5-years 
Total chl- a 6.5 + 3.9 4.8 + 2.4 -1.7 (-26%) 0.03 * 
( μ g l -1 ) (61) (63) 
Acid-corrected 4.7 + 3.1 3.3 + 2.2 -1.3 (-28%) 0.05 * 
chl- a ( μ g l -1 ) (61) (63) 
Phaeophytin 1.9 + 1.3 1.4 + 0.5 -0.5 (-28%) <0.01 * 
( μ g l -1 ) (61) (63) 
36

For acid-corrected chl- a , the fraction most often used to measure phytoplankton biomass,
the Harbor-wide averages during the 5-years were -1.3 μg l -1 , or -28% lower than
baseline. For phaeophytin, or ‘degraded’ chl- a , the average concentrations were -0.5 μg
l -1 , or -28%. For all three fractions, the percent decreases were in the same order as for
both TN (-34%) and TP (-28%). 
The fact that the concentrations of chl- a after transfer were lower than baseline is evident
in the time-series and bar chart plots of the three fractions (Figures 15 and 16). As is
evident in the time-series plots, for all three fractions the decreases were largest during
summers, the season when concentrations in the Harbor tended to be seasonally greatest.
The fact that the decreases were largest during summers was unlike for TN and TP. 
During the 5 years, for all three chl- a fractions, the annual averages were lower than
baseline during four of the 5-years (2001, 2003, 2004 and 2005). During the remaining
years (2002), average concentrations were higher than during the other four years and
within the range of the elevated averages during baseline. (The fact that average chl- a 
concentrations during all 5 years were not lower than baseline was unlike for TN or TP). 
As can be seen in the scatter plot of annual average chl- a concentrations versus TN (Fig.
17), the years that the Harbor showed lowered chl- a , were, with the exception of 2002,
the years after transfer that the Harbor showed the lowered TN concentrations. During
2002 concentrations of chl- a in the Harbor were elevated, despite the concentrations of
TN being lowered. 
At the individual stations, the average concentrations of acid-corrected chl- a after transfer
were significantly lower than baseline, at two stations (Stations 077 and 139), both
located in the Central Harbor region (Fig. 18). At one other station, Station 124, in the
South East Harbor, and as for the other two stations, also located in the South Harbor, the
decrease was ‘almost’ significant. 
37

95 96 97 98 99 00 01 02 03 04 05
0
5
10
15
20
25 
Fig. 15. Total chl-a partitioned into the acid-corrected and phaeophytin fractions. 
Values are survey volume-weighted, Harbor-wide averages. 
Offshore
transfer
AVERAGE CONCENTRATIONS
(UG L -1 ) 
CHLA 
PHAEO 
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005
0
2
4
6
8
10
12 
Offshore
transfer
Fig. 16. Annual average total chl-a partitioned into the acid-corrected and phaeophytin fractions.
ANNUAL AVERAGE CHL-A CONC
( UG L -1 ) 
CHLA 
PHAEO 
38

10 15 20 25 30 35 40 45
0
1
2
3
4
5 
02 
1:10
Fig. 17. Acid-corrected chl-a versus TN . Data are annual, volume-weighted
Harbor-wide averages. Diagonal line represents ratios of concentrations of
chl-a :TN of 1:10.
ANNUAL AVERAGE CHL-A CONC.
( UG L -1 )
ANNUAL AVERAGE TN CONC.
( UMOL L -1 ) 
BASELINE 
POST 
Thus, as for TN and TP, the Harbor showed a significant decrease in year-round average
chl- a after transfer, but the decrease was more localized than for TN or TP. The fact that
subtraction yielded negative values at all 10 stations suggests that with longer-term
monitoring could well show the decreases in chl- a to have been significant over a larger
area of the Harbor than indicated by the two stations alone. 
For phaeophytin, average concentrations during the 5-years were significantly lower than
baseline at three stations, and ‘almost’ significantly lower at two others (lower panel,
Figure 18). Two of the three stations at which the decreases were significant were the
two stations that showed significant decreases for acid-corrected chl- a . As for acid-
corrected chl- a , subtraction yielded negative values at all 10 stations. 
39

For all three fractions, the Harbor-wide averages during the 5-years were significantly
different, and in all cases lower than baseline, during summers, but not during spring, fall 
40

or winter (Table 10). During summers after transfer, average total chl- a concentrations
were -3.7-μgl -1 (or -39%) lower than during baseline. For acid-corrected chl- a , they were
-2.7 μgl -1 (or -39%) lower. and for phaeophytin, -1.07 μgl -1 (or -37%) lower. 
Table 10. Seasonal differences in average concentrations of total chl-a, acid- 
corrected chl-a and phaeophytin between the 5-years and baseline. 
Fraction Difference (μg l -1 , % baseline) 
Spring Summer Fall Winter 
Total chl- a -4.2 (-49%) ns -3.7 (-39%) * -0.2 (-4%) ns +0.8 (+25%) ns 
Acid-corrected chl- a -2.9 (-49%) ns -2.7 (-39%) * -0.4 (-14%) ns +0.6 (+27%) ns 
Phaeophytin -1.4 (-50%) ns -1.0 (-37%) * <0.01- (<-1%) ns +0.1 (+7%) ns 
For all three chl-a fractions, the nature and the spatial extents of the changes after
transfer, varied widely between seasons. Figure 19 the seasonal differences for acid-
corrected chl- a (Fig. 19). For this fraction, average concentrations showed significant
decreases at 6 stations during spring, and all 10 stations during summer. During fall, at
none of the stations were the differences between periods significant. 
During winters, subtraction yielded positive values at all except one of the stations, and at
two of the stations that showed positive values, the averages after transfer were
significantly greater than baseline. Both stations were located in the outermost Outer
Harbor (Stations 142 and 141). At two others (Stations 24 and 106), both located in the
North Harbor, the increases were ‘almost’ significant. 
Thus, the decreases in concentrations of chl- a that followed transfer, were more
seasonally restricted than for TN or TP. The decreases in chl-a observed for the year- 
41

round data were driven by decreases during summer and spring, when the concentrations
were lowered over most of the area of the Harbor. The decreases during spring and
summer were, in turn, ‘dampened’ by the smaller, but still significant increases in chl- a 
observed at the mouth of Harbor, during the winters after transfer. 
42

POC and suspended solids 
Particulate organic carbon (POC). During the 5-years, Harbor-wide average
concentrations of POC were also significantly lower than baseline (Table 11). Average
concentrations decreased from 42.9 + 16.1-μmol l -1 during baseline, to 30.9 + 10.2 μmol
l -1 during the 5-years, or by -12 μmol l -1 (or -28%). The -28% decrease was similar to the
-26% to -35% decreases for TN, TP and chl- a . 
Table 11. POC and total suspended solids . Comparison of volume-weighted Harbor-
wide average values during baseline and the first 5-years after offshore transfer. 
Variable Average values during Difference p 
between baseline 
Baseline 5-years and 5-years 
POC 42.9 + 16.1 30.9 + 10.2 -12 (-28%) <0.01 * 
( μ mol l -1 ) (61) (60) 
TSS 3.6 + 1.2 3.8 + 1.1 +0.2 (+5%) 0.14 
(mg l -1 ) (55) (63) 
POC as %TSS 18 + 6.5 12 + 3 -6 (-33%) <0.01 * 
(by weight) (55) (60) 
POC: Chl- a 226 + 234 145 + 68 -81 (-36%) 0.32 
(by weight) (62) (62) 
As for chl- a , the decreases in POC were largest during the summers after transfer;
summer being the season when POC concentrations in the Harbor tend to be seasonally
greatest (Fig. 15). During each of the 5 years after transfer, the annual Harbor-wide
average concentrations of POC were consistently lower than in all years during baseline
(Fig. 16). 
The fact that the annual averages during all 5 years were lower than baseline was as for
TN and TP, but unlike for chl- a . Unlike for TN and TP, however, and this is evident in 
43

the bar-chart plot of the annual average values, the decrease in POC started before
offshore transfer, and in 1997 or 1998. 
95 96 97 98 99 00 01 02 03 04 05
0
20
40
60
80
100 
Fig. 20. Particulate organic carbon (POC) . Time-series plot of survey volume-weighted, Harbor
-wide average concentrations.
Offshore
transfer
AVERAGE CONCENTRATION
( UMOL L -1 ) 
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005
0
10
20
30
40
50
60 
No
data 
Offshore
transfer
Fig. 21. Annual average POC concentrations, 1996 through 2005.
ANNUAL AVERAGE POC CONC
( UMOL L -1 ) 
44

At the individual stations, averages POC concentrations after transfer were significantly
lower than baseline at all 10 stations; c.f. two stations for chl- a (Fig. 22). The decreases 
45

at the individual stations ranged in size from -5.5 μmol l -1 (at Station 137), to -15.0 μmol
l -1 (at Station 077). In percent terms, the decreases were largest at the stations located
closest the two sets of former WWTF outfalls, as was the case for TN and TP. 
Among the four seasons, the Harbor wide averages during the 5-years were significantly
different from, and in all cases lower than baseline, during spring, summer and fall, but
not winter (Table 12). The decreases ranged in size from -15.6 μmol l -1 during fall, to -
21.1μmol l -1 during summer. In percent terms, they ranged from -34% of baseline during
spring, to -40% of baseline during fall. 
Table 12. Seasonal differences in average TSS and POC concentrations between the 
5-years and baseline. 
Fraction Difference ( % baseline) 
Spring Summer Fall Winter 
POC ( μmol l -1 ) -17.3 (-34%) * -21.1 (-38%) * -15.6 (-40%) * +2.3 (+8%) ns 
TSS ( mg l -1 ) +0.32 (+9%) ns -0.31 (-7%) ns +0.56 (+17%) ns +0.48 (+18%) ns 
Total suspended solids (TSS). Unlike for POC, the Harbor-wide average concentrations
of TSS for the 5-years were not significantly different from baseline (Table 11). During
baseline, TSS concentrations averaged 3.6 + 1.2 mg l -1 . During the 5-years, they
averaged 3.8 + 1.1 mg l -1 . The difference of +0.2 mg l -1 was equivalent to +5% of
baseline, and was not significant (at p < 0.05). 
The fact that average TSS concentrations during the 5-years were not significantly
different from baseline is evident in both the time-series and bar chart plots of the
Harbor-wide average concentrations (Figures 23 and 24, respectively). The seasonal 
46

1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005
0
2
4
6
8 
Figure 23. Total suspended solids (TSS) partitioned into the particulate organic carbon (POC)
and non-POC fractions.
Offshore
transfer
AVERAGE CONCENTRATIONS
(MG L -1 ) 
POC 
NONPOC 
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005
0
1
2
3
4
5
6 
No
data 
Offshore
transfer
Fig. 24. Annual average TSS concentrations partitioned into the POC and non-POC fractions.
ANNUAL AVERAGE CONC
( MG L -1 ) 
POC 
NONPOC 
47

pattern of TSS after transfer was also not consistently different from the pattern during
baseline. (Note, both before and after transfer, non-POC or inorganic suspended material
contributed the bulk of the TSS). 
At the individual stations, average TSS concentrations after transfer were significantly
different from baseline at four stations (Fig. 22). At two stations, both located in the
Inner Harbor (Stations 137 and 138), average concentrations after transfer were
significantly lower than baseline. At two stations at the mouth of the Outer Harbor
(Stations 141 and 142), the opposite applied. 
For TSS averaged Harbor-wide, we were unable to detect significant differences between
the two periods, for any of the four seasons (Table 12). During three of the four seasons
(fall, winter and spring), subtraction yielded positive (and in all cases, non-significant)
values; during summer, subtraction yield a negative value, but again this was not
significant. 
POC as % TSS . During the 5-years after transfer, the percent contribution of POC to
TSS was also significantly lower than baseline (Table 11). During baseline, POC
contributed 18 + 6.5% of TSS; during the 5-years, it contributed 12 + 3 %. The
difference of -6% was equivalent to -33% of baseline, and was significant. Thus, the
Harbor showed no decrease in TSS after transfer, but it did show a decrease in the %
organic content of the TSS (measured as POC). Unlike for POC as % TSS, Harbor-wide
average ratios of POC : chl- a , a measure of detrital:phytoplankton biomass, were not
changed. 
Water clarity 
Vertical attenuation coefficients (k) and secchi depth . For both measures of water
clarity that we conducted ( k and secchi depth), the Harbor-wide averages during the 5-
years were not significantly different from baseline (Table 13). For both variables, 
48

49 
subtraction yielded values that were +1% to +2% of baseline, and the differences were
not significant. This was as for TSS, but unlike for chl- a and POC. 
Table 13. Water clarity . Comparison of volume-weighted Harbor-wide average
values during baseline and the first 5-years after offshore transfer. 
Variable Average values during Difference p 
between baseline 
Baseline 5-years and 5-years 
k 0.52 + 0.12 0.53 + 0.12 +0.01 (+1%) 0.85 
(m -1 ) (73) (64) 
Secchi depth 2.65 + 0.60 2.70 + 0.70 +0.05 (+2%) 0.56 
(m) (85) (64) 
The fact that the water clarity in the Harbor after transfer was not different from baseline
is evident in both the time-series (Fig. 25) and bar chart plots (Fig. 26) of the two
variables. For both variables, the Harbor-wide averages varied widely between
successive surveys, perhaps masking any changes in clarity that might have followed
transfer. 
For both variables, as for TSS, the differences between the two periods were significant
only at certain stations (Figure 27). For k , the averages after transfer were significantly
lower than baseline at one station (Station 137 in the upper Inner Harbor), and
significantly greater than baseline at one other (Station 141 in the Central Harbor). For
secchi depth, at two stations in the Inner Harbor (Stations 137 and 138), average values
after transfer were significantly greater than baseline. 
The grouping of the stations that showed positive and negative values, and the fact that
they were similar for the two variables, suggests that the Harbor may have experienced
an increase in clarity in the Inner Harbor and along the west margin of the Harbor, and a
decrease in clarity over much of the main body of the Outer Harbor. (Note, the k data

1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4 
1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005
1
2
3
4
5
6
7 Offshore
transfer
Offshore
transfer
AVERAGE K ( M -1 ) 
ATTENUATION COEFFICIENT 
Fig. 25. Vertical attenuation coefficients (k) and secchi depths. 
Values are survey-average, volume-weighted Harbor-wide averages. 
SECCHI DEPTH 
AVERAGE SECCHI DEPTH
( M ) 
1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005
0
1
2
3
4
5 
ANNUAL AVERAGE SECCHI
DEPTH ( M )
1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005
0.0
0.2
0.4
0.6
0.8
1.0 
Offshore
transfer 
Fig. 26. k and secchi depth . Annual, volume-weighted, Harbor-wide average values,
1994 through 2005. 
Offshore
transfer
ANNUAL AVERAGE k 
( M -1 ) 
50

RECIPROCAL k

Legend

Former outfalls

Decrease

Increase

significant
p<0.05

‘almost’ significant
p<0.10>0051

Legend

e Former outfalls

Decrease

Increase

significant
p<0.05

? ‘almost' significant
p<0.10>0.051

Fig. 27. k_and secchi depth. Differences in average values (m) between the first 5-years after offshore
transfer and baseline, at the 10 Harbor stations.

51

52 
have been reported as reciprocal values, so negative values for k correspond with positive
values for secchi depth). 
For both k and secchi depth, and again as for TSS, the Harbor-wide averages after
transfer were not significantly different from baseline during any of the four seasons
(MWRA unpublished data). 
Bottom-water dissolved oxygen (DO) concentrations and % saturation 
For both DO variables, as for TSS and water clarity, we were unable to detect significant
differences between the Harbor-wide averages for the 5-years, and baseline (Table 14).
For both variables, subtraction yielded differences that were equivalent to + 3% of
baseline, and the differences were not significant 
Table 14. Bottom-water dissolved oxygen (DO) . Comparison of volume-weighted
Harbor-wide average values between the 5-years after transfer, and baseline. 
Variable Average values during Difference p 
between baseline 
Baseline a 5-years and 5-years 
DO 8.6 + 1.2 8.9 + 1.3 +0.3 (+3%) 0.34 
(mg l -1 ) (40) (64) 
DO 95 + 7 92 + 6 -3 (-3%) 0.18 
(% saturation) (40) (64) 
a only data collected after May 1997 were used to compute baseline averages and to test for 
differences between periods. 
The time-series (Fig. 28) and bar chart plots (Fig. 29) confirm the fact that bottom-water
DO values in the Harbor for the full 5-years after transfer were not significantly different
from baseline. As is evident in the time-series plots, the Harbor showed increased

Std
Std 
1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005
40
60
80
100
120 
ANNUAL AVERAGE DO
% SAT.
1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005
4
6
8
10
12 
DO % SATURATION
DO CONCENTRATION 
Offshore
transfer
Fig. 29. Bottom-water DO concentrations and % saturation . Annual,
volume-weighted, Harbor-wide average values, 1994 through 2005.
Offshore
transfer
ANNUAL AVERAGE DO
CONC ( MG L -1 ) 
53 
1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005
4
6
8
10
12
14 
1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005
40
60
80
100
120 
Offshore
transfer
Offshore
transfer
AVERAGE DO CONC.
( MG L -1 ) 
Std.
Std. 
Fig. 28. Bottom-water DO concentrations and DO % saturation . Values are
survey-average, volume-weighted Harbor-wide averages. 
DO % SATURATION
DO CONCENTRATION 
AVERAGE DO % SAT.

DO concentrations during the summers after transfer (top panel), but the same increase
was not seen for DO % saturation (bottom panel). 
During each of the summers before transfer, bottom-water DO concentrations in the
Harbor dropped down to levels only just above the State Standard of 6 mg l -1 . During
each of the summers after transfer, concentrations fell to higher levels, well above the
State standard (discussed further below). (It is worth noting that average DO% saturation
values have fallen to lower seasonal lows during the last three years than in all previous
years). 
At the individual stations, for both DO variables, the averages after transfer were
significantly different from, and in all cases greater than baseline, at a single station,
Station 137 in the upper Inner Harbor (Fig. 30). At this one station, average DO
concentrations during the 5-years were +0.8 mg l -1 (or +11%) greater than baseline; for
DO % saturation, the increase was +6.6 % (or +9%). 
For DO concentrations, subtraction yielded positive (but non-significant) values at all 9
other stations; for DO % saturation, subtraction yielded positive (but again non-
significant) values at 5 other stations, extending from the mouth of the Inner Harbor over
the main body of the Outer Harbor. Again, longer-term monitoring may (or may not)
show any, likely small, increases in these other areas to be significant. 
For both DO variables, the Harbor-wide averages for the 5-years were not significantly
different from baseline during any of the four seasons (Table 15). If the ‘mid-summer’
data were separated out from the ‘summer’ data, however, the average DO concentrations
during ‘mid-summers’ during the 5-years were +0.4 mg l -1 (or +6%), and significantly
greater than during ‘mid-summers’ during baseline. ‘Mid-summer’, as used here, refers
to August + September. 
54

Fig. 30.

DO concentration

~O’

Va
‘

DO % saturation

*

Former outfalls
Increase
Decrease

Significant at
ps0.05

Possibly significant
P.<0.10 2 0.051

Bottom-water DO concentrations and DO % saturation. _ Differences

in average values between the first 5-years after transfer, and baseline at the 10
stations. Units = mg !-! for concentrations, and % for saturation.

55

Table 15. Seasonal differences in average bottom-water DO concentrations and DO 
% saturation values between the two periods. 
Fraction Difference ( % baseline) 
Spring Summer Mid- Fall Winter 
summer 
DO ( mg l -1 ) -0.15 (-2%) ns -0.16 (-2%) ns +0.39 (+6%) * +0.24 (+3%) ns +0.59 (+6%) ns 
DO ( % ) -5.8 (-6%) ns -1.5 (-2%) ns -0.2 (<-1%) ns -0.8 (<-1%) ns +0.3 (<+1%) ns 
For bottom-water DO concentrations, the stations that showed significant differences
varied widely among seasons (Fig. 31). Only during two of the seasons, ‘mid-summer’
and fall, were increases in bottom-water DO concentrations observed at individual
stations. During mid-summers, significant increases in bottom-water DO concentrations
were observed at 6 stations, extending from the upper Inner Harbor to Deer Island, and
then into the Central Harbor. 
During fall, subtraction yielded positive values at all 6 stations in the North Harbor and
one station in the Central Harbor, but at only one, Station 137 in the upper Inner Harbor,
were the increases significant. During winters, we were unable to detect significant
differences (increases or decreases) in DO between the two periods, at any of the 10
stations. 
During spring, average DO concentrations during the 5-years were significantly lower
than baseline at three of the stations all located in the South Harbor (Stations 077, 124 
56

and 139). At one other, Station 140 in the Neponset River area of Dorchester Bay, the
decrease was ‘almost’ significant. These localized decreases in DO during the spring
periods after transfer, will likely have ‘dampened’ the increases in bottom-water DO
expected to follow the ending of wastewater discharges to the Harbor. 
57

II. SUMMARY OF DIFFERENCES BETWEEN PERIODS 
Year-round data. Table 16 summarizes, for 20 water-quality variables, the differences
between the 5-years and baseline. The data used for this Figure are year-round data; the
differences observed at different seasons are not represented. The symbols are as in the
previous Figures, except that ‘arrows’ are used in place of ‘triangles’. Note, also that
only differences significant at p < 0.05 are indicated in the Table; differences that were
‘almost’ significant are not included. 
The ‘solid arrows show variables where the differences between periods were significant
for the data averaged Harbor-wide; the ‘hollow’ arrows indicate variables where the
differences were significant only at individual stations. The ‘dashes’ indicate variables
where the differences were not significant for the data averaged Harbor-wide or for
individual stations. 
The differences observed among the 20 variables can be partitioned into three categories.
The first category, which included 12 variables, showed the largest differences, and in all
cases decreases. The 12 variables included concentrations of the total and dissolved
inorganic fractions of N and P, the percent contributions of DIN and DIP to TN and TP
respectively, molar ratios of DIN:DIP, and concentrations of chl- a , phaeophytin and
POC. 
For all these variables, the decreases were significant for the data averaged Harbor-wide.
For most of the 12 variables, chl- a and phaeophytin excluded, the decreases were also
significant at all (or almost all) of the 10 stations. For chl- a and phaeophytin, the
decreases were confined to localized areas of the Central Harbor extending up into the
west North West Harbor region. 
58

Table 16 Summary of differences in Harbor water-quality between the 5-years and baseline.

VARIABLE 5 - YEARS
TN ( umol 17!) ¥ -10.7 (-35%) *
DIN (mol 17!) ¥ ~6.5 (-55%) *
DIN as % TN § -22 (-55%) *
TP ( mol I-1 ) co -0.58 (-28%) *
DIP ( umol 17" ) av 0.40 (-38%) *
DIP as % TP ob -7 (-15%)*
TNAP 7 07 (-5%)
DIN:DIP 5 -3.4 (30%)
TOTAL CHL-A (ug 17!) g 1.7 (-26%) *
‘ACTIVE CHL-A( gl!) B -1.3 (28%) *
PHAEOPHYTIN (ug I") e -0.5 (-28%) *
POC (umoll”! ) ¥ -12.0 (-28%) *
Tss(mg!"") ry 4p> $0.20 (459)
POC as % TSS 4 6.0 (-33%) *
k (m7) Y 4 +0.01 (+196)
SECCHI DEPTH (m) 4} +0.1 (+495)
DO CONC. (mg I!) 4} +0.3 (+596)
DO% SAT. {> 3396)
SALINITY (ppt) - “1.2 (-4%)
TEMPERATURE (°C) - 0.5 (-5%)

59

60 
For most of these variables, the decreases were largest during the seasons when
concentrations in the Harbor water-column tended to be seasonally greatest. This was
winter for the total and dissolved inorganic fractions of N and P, and for DIN:DIP, and
summer for the particulate fractions of N and P, and for chl- a , phaeophytin and POC. 
The second category included the 6 variables that showed no significant differences
between the two periods for the data averaged Harbor-wide, but did show significant
differences at individual stations. These variables included molar ratios of TN:TP,
concentrations of TSS, k and secchi depth, and the two DO variables (DO concentrations
and % saturation). 
For all 6 variables, the differences were significant at 4 or less stations, and the
differences were confined to localized areas of the Harbor. For many of these variables,
which compared to the first category of variables showed relatively small changes, the
subtraction yielded negative values (suggestive of decreases) in certain areas of the
Harbor, and positive values (suggestive of increases) in others. 
The third category of variables included only two variables, salinity and water
temperature. For both of these variables, the differences between the two periods were
not significant for either the data averaged Harbor-wide, or the data from each of the
individual stations (see Appendix A for data). The absence of detectable difference for
these variables, suggests that the differences we observed between periods for the other
variables, were not the result of large differences in background environmental conditions
between the periods. 
Seasonal data. Table 17 shows, for the same 20 variables, the differences in averages
between the 5-years and baseline, for each of the four seasons. For most of the N and P
variables, the decreases were significant for the data averaged Harbor-wide, for all four
seasons. For most of these variables, the decreases were largest during winters and fall,
the seasons when before transfer, average concentrations tended to be greatest.

Summary of seasonal differences in Harbor water-quality between the 5-years and baseline.

Table 17.

VARIABLE

SEASON

TEMPERATURE (°C)

SALINITY (ppt)
DO CONC. (mg L)
DO % SAT

SECCH| DEPTH (m)

k (m4)

POC as % TSS

POC (umol I!)

Tss (mg I"')
PHAEOPHYTIN (ug I"')
‘ACTIVE’ CHL-A (ug I")
TOTAL CHL-A (ug I")
DIN: DIP

TN :TP

DIP as % TP

DIP (umol 17)

TP (mol 1")

DIN as % TN

DIN (mol 1)

TN ( umol I")

ns Soins

ns

FORE a
**F#F#%osFF % 4

*%*FF#+s#tsog#F

VY

ee © FY O

7 4% 9 F 4

SPRING

ns

(*) (@) o>

ns

SUMMER

ns {> ns ns

ns

ns ns 4} 5 Sf ns
& 9

ns

FALL

ns

ns

ns

ns

*FFeeouvee © O OW

WINTER

61

For the three chl- a variables, the averages after transfer were significantly lower than
baseline for the data averaged Harbor-wide, during summers, and at individual stations
during spring. During fall, we were unable to detect significant differences between the
two periods, for the data averaged Harbor-wide or individual stations. During winters, 
For the three chl- a variables, the averages after transfer were significantly lower than
baseline for the data averaged Harbor-wide, during summers, and at individual stations
during spring. During fall, we were unable to detect significant differences between the
two periods, for the data averaged Harbor-wide or individual stations. During winters, 
certain stations showed significant increases in concentrations of the three fractions. 
For POC, and POC as % of TSS, the average values after transfer were significantly
different from baseline, for the data averaged Harbor-wide, during spring, summer and
fall. During winters, we were unable to detect significant differences between the two
periods for both variables, for the data averaged Harbor-wide or the individual stations. 
For TSS, k and secchi depth, at the level of the Harbor as a whole, the averages after
transfer were not significantly different from baseline during any of the four seasons. All
three variables, however showed significant differences at individual stations, during
specific seasons; during all four seasons for TSS, and two of the four seasons (summer
and winter) for k and secchi depth. 
For both bottom-water DO variables, the averages after transfer were significantly
different, and in this case greater than baseline, for the data averaged Harbor-wide, during
part of one of the four seasons, specifically ‘mid-summer’. During two of the other three
seasons, the differences were significant at individual stations, for one or both of the DO
variables depending on season. 
62

DISCUSSION 
Comparison with the first 36-months after transfer 
Table 18 compares the differences in water quality reported here for the first 5-years after
transfer with the differences reported earlier, by Taylor (2004) for the first 36-months
after transfer. For 14 of the 20 variables, the directions and levels of significance of the
differences for the 5-years (or lack of differences in the case of water-temperature), were
the same as for the first 36-months. 
The 14 variables included the total and dissolved inorganic fractions of N and P, the %
contributions of DIN to TN, molar DIN:DIP ratios, concentrations of phaeophytin, POC,
and TSS, secchi depths, the two DO variables, and water-temperature. For these
variables the fact that the differences during the 5-years were as for the 36-months,
suggests that the bulk of the differences after transfer occurred during the first 36-months. 
For an additional four of the 20 variables, the directions of the differences for the 5-years
were as for the 36-months, but the levels of significance for the 5-years were greater than
for the 36-months. For two of the four variables, total chl- a and acid-corrected chl- a , the
size of the difference was increased between the 36-months and 5-years. For the other
two variables, DIP as % TP and k , the increased n value for the post-transfer period, was
likely responsible for the increased significance of the differences, between the 36-
months and 5-years. 
For two other variables, TN:TP and salinity, the levels of significance of the differences
during the 5-years were smaller than for the 36-months. For these variables, background
environmental processes, in this case reduced river inflows during 2002, were likely
responsible for the greater differences for the 36-months than 5-years. River inflows to
the Harbor are enriched with TN relative to TP (MWRA unpublished data). 
63

Table 18. Summary of differences in Harbor water-quality between the 5-years and baseline.
Also shown are the differences between the first 36-months after transfer and baseline (from Taylor 2004).

VARIABLE 5- YEARS 36-MONTHS
TN( umoll"') ¥ -10,7 (-35%) B  -10.0(-27%)
DIN (umol I") ¥ 6.5 (-55%) ge 70159)
DIN as %TN y -22 (55%) #140370)
TP( umolI-!) & -0.58 {-28%) B 0.58 (28%)
DIP( umol I") t 0.40 (-38%) BH  -04(-38%)
DIP as % TP ay -7 (159%) ty 7-159)
TNAP vy -0.7 (-5%) e 130%)
DIN:DIP et “3,4 (-30%) BH  -38(-33%)
TOTAL CHL-A (ug !7!) g “1,7 (-26%) t3 -1,3 (-20%)
‘ACTIVE CHL-A ( pig I") ¥ -1.3 (-28%) G -0.9 (-19%)
PHAEOPHYTIN (ug!"") g +05 (-28%) H  -05(-26%)
POC (umol I"! ) ¥ -12.0 (-28%) #121 628%)
Tss(mg!"!) Ot +0.20 (+5%) n 4 40.25 (+7%)
POC as % TSS y 6.0 (-33%) BY 6.033%)
r

k(m7!) Vv 4) +0.01 (+19) = 9,01 (-19)
SECCHI DEPTH ( m) 4} +0.1 (+496) {> 40.1 (44%)
DO CONC. (mgI"!) ty +0.3 (+59) 4s +0.3 (4+5%)
DO% SAT. {> 363%) 4 +2 (42%)
SALINITY (ppt) - “1.2 (-4%) 04(1H1%)
TEMPERATURE (°C) - 0.5 (-5%) — 0.3 (-3%)

64

65 
Observed versus predicted changes 
Bays Hydrodynamic (BHM ) and Eutrophication Models (BEM) . Table 19 compares
for three variables, concentrations of DIN, acid-corrected chl- a , and DO, the differences
observed for the 5-years, and the differences predicted by HydroQual and Normandeau
(1995) using the 3-D BHM/BEM model. Note, only broad comparisons can be
conducted of the two sets of data, because the modeled predictions were generated using
1991 boundary data. 
Also, the model predictions that were reported for specific times of the year, for specific
locations within the Harbor, rather than for the year-round or seasonal, Harbor-wide
averages emphasized in this report. In this report we have considered the observed
differences to be ‘similar’ to the differences predicted by the BEM model, if the observed
differences fell within + 20% of the predicted differences. 
For DIN, the decreases we observed for the Harbor as a whole, of -6.5- μ mol l -1 (or -
55%), were almost identical to the decreases of -6.8- μ mol l -1 (or -57%) predicted by the
BEM model for Station 2. For year-round acid-corrected chl- a , the difference we
observed, -1.3 μgl -1 (or -28%), was again almost identical to the difference of -1.5 μgl -1 
(or -39%) predicted by the BEM model. 
For year-round bottom-water DO concentrations, the small (and in this case, non-
significant) value of +0.3 mg l -1 (or +3%) for the 5-years, was identical to that 
of +0.3-mg l -1 (or +4%) predicted by the BEM model. For salinity, the observed, but
non-significant difference of 1.2 ppt was slightly greater than the 0.5 ppt difference
predicted by the BHM (Signell et al . 2000). Both sets of salinity values, however, were
small, and even small differences in background (river or oceanic) conditions between
the two different periods covered by the two studies, could account for the difference. 
Comparison with 2-D box model . Table 20 compares for six variables (TN, DIN, DIN
as %TN, DIP, molar DIN:DIP, and TSS), the differences we observed for the 5-years for

VARIABLE OBSERVED 
Baseline 5 years Difference (%) 
PREDICTED 
Baseline 5 years Difference (%) 
DIN (μmol l -1 ) 
Chl- a (μg l -1 ) d 
DO conc. (mg l -1 ) 
11.8 + 6.4 a 5.3 + 3.5 a -6.5 (-55%) a 
(75) (62) 
4.7 + 3.1 a 3.3 + 2.2 a -1.3 (-28%) a 
(61) (63) 
8.6 + 1.2 a 8.9 + 1.3 a +0.3 (+3%) a 
(40) (64) 
11.9 + 5.5 c 5.1 + 4.1 c -6.8 (-57%) c 
(24) (24) 
3.8 + 2.1 c 2.3 + 1.1 c -1.5 (-39%) c 
(24) (24) 
9.2 + 1.0 e 9.5 + 1.0 e +0.3 (+4%) e 
(24) (24) 
a values are the averages for all stations sampled, for all depths for DIN and chl- a , and near-bottom depth alone for DO. Averages are for year-round
data; b from HydroQual (1995); c predicted values are for HydroQual and Normandeau station 2, located between Stations 106 and 142; d data are
acid-corrected, extracted chl- a ; e predicted DO data for Harbor are averages for mid- and bottom depths for HydroQual station 2. 
66 
Table 19. Observed versus predicted changes . Comparisons of the differences in average concentrations of DIN, chl- a and DO
observed for the 5-years, and the changes predicted by the coupled hydrodynamic/water quality BHM/BEM model.

the North West Harbor, with the differences predicted for the same area by Kelly (1998). 
Note, the ‘North Harbor’ referred to by Kelly is equivalent to the region we refer to as the
‘North West Harbor’. Also, all data in the Table cover the March through September
period, to be compatible with the period sampled by Kelly in 1994. 
Table 20. Comparison of the differences observed during the 5-years with the 
differences predicted by Kelly (1998) . 
Variable This study Kelly (1998) 
Observed Observed Difference Observed Predicted Difference 
before 5-years before post-transfer 
TN 27.4 19.3 -8.1 (-30%) 25 19 -6 (-24%) 
(μmol l -1 ) 
DIN 7.4 3.0 -4.4 (-59%) 11 5.5 -5.5 (-50%) 
(μmol l -1 ) 
DIN as 28 12 -16 (-57%) 44 29 -15 (-34%) 
% TN 
DIP 0.8 0.5 -0.3 (-39%) 2.5 1.5 -1.0 (-40%) 
(μmol l -1 ) 
Molar 9.6:1 6.3:1 -3:1 (-34%) 4.4:1 3.7:1 -0.7:1 (-16%) 
DIN:DIP 
TSS 4.0 4.1 +0.1 (+1%) 2.5 1.0 -1.5 (-60%) 
(mg l -1 ) 
For TN, we observed a decrease of -8.1-μmol l -1 (or -30%) between the 5-years and
baseline; Kelly predicted a decrease of -6-μmol l -1 (or -24%). The observed decrease
was, in terms of absolute concentrations, 35% greater than the modeled estimate, and
based on our 20% criterion different from the modeled estimate. For DIN and DIN as
%TN, the observed and predicted decreases were similar. 
67

For DIP, the decrease of -0.3 -μmol l -1 (or -39%) we observed was smaller, and in terms
of absolute concentrations about one-third the size of the predicted decrease. This
difference might simply be the result of the DIP concentrations observed in the Harbor
during baseline being one-third the size of the concentrations measured during 1994.
Expressed as a percent of baseline, the observed decrease of -39% for DIP was identical
the predicted decrease of -39%. 
For DIN:DIP, the observed decrease of -3.1:1 was similar to the predicted decrease of -
3.7:1; in percent terms, however, the observed decrease was larger than the predicted
decrease. As noted earlier, the DIN:DIP ratios in the Harbor vary widely year-to-year,
and the difference could simply reflect background differences between the 5-years and
the period modeled by Kelly. 
For TSS, the fact that we were unable to detect a significant difference was unlike the
decrease of -1.5 mg l -1 (or -60%) predicted by Kelly. This could be because our
comparison of TSS concentrations during the periods before and offshore transfer does
not capture the fact that the bulk of the decrease in wastewater loadings to the Harbor
occurred with upgrades to treatment, before offshore transfer. 
Both sets of models predicted the ‘effects’ of the transfer of the wastewater discharges by
numerically ‘eliminating’ the wastewater discharges to the Harbor. The output from
these models can therefore be used as a ‘reference’, to verify the changes we observed in
the Harbor. It can also be used to verify whether the transfer of the WWTF discharges
were responsible for the sizes and nature of any changes observed in the Harbor. 
Conclusions 
The primary purpose of this report was to compare Harbor water-quality during the first
5-years after offshore transfer with water quality during a baseline period, and then
compare the observed differences with the differences predicted by others using 
68

numerical models. The study was able to demonstrate significant differences in the
Harbor water quality between the two periods. 
For the few variables where comparisons could be made, the observed changes were
similar to the changes predicted by others using the BHM/BEM models. The model
predictions were made be ‘numerically eliminating’ the wastewater discharges to the
Harbor. Therefore the similarity of the observed and predicted changes would indicate
that the changes observed in the Harbor were, at least for these variables, the result of the
transfer of the WWTF discharges offshore. 
ACKNOWLEDGEMENTS 
Grateful thanks are extended to Drs Andrea Rex and Mike Delaney, and to Steve Rhode for
providing the administrative and logistic support for this work. Thanks are extended to Kelly
Coughlin for management of field monitoring, and for data management and QA/QC. Thanks are
also due to Keary Berger, Laura Ducott, Nicole O’Neill and others for field sampling, and to
Laura Ducott, Nicole O’Neill, N. McSweeney and Lisa Wong for laboratory QA/QC. Keary
Berger, Rob Rabideau and others operated the sampling vessels. Numerous others, including T.
Smirnova, C. Blodget, M. Gofsteyn and R. Warot, conducted laboratory analyses. Data
management was provided by W. Leo, D. Hersh, P. Ralston, S-J Liang, and J. Lobuglio. 
69

LITERATURE CITED 
Alber, M. and Chan, A. B. 1994. Sources of contaminants to Boston Harbor: revised
loading estimates. Massachusetts Water Resources Authority Environmental Quality
Dept. Technical Report No 94-1. 
Bendschneider, K. and Robinson, R. J. 1952. A new spectrophotometric determination
of nitrate in seawater. Journal of Marine Research 11: 87-96. 
Bricker, S. B., C. G. Clement, D. E. Pirhalla, S.P. Orlando and D.R.G. Farrow. 1999.
National Estuarine Eutrophication Assessment: Effects of Nutrient Enrichment in the
Nations Estuaries. NOAA, National Ocean Service, Special Projects Office and the
National Centers for Coastal Ocean Science. Silver Spring, MD: 71pp. 
Clesceri, L. S., A. E. Greenberg, and A. D. Eaton. 1998. Standard Methods for the
Examination of Water and Wastewater. 20th Edition. American Public Health
Association, American Water Works Association, Water Environment Federation. 
Fiore, J. and O'Brien, J. E. 1962. Ammonia determination by automatic analysis.
Wastes Engineering. 33: 352. 
Holm-Hanson. O, Lorenzen, C. J, Holmes, R. W, and Strickland, J. D. H. 1965.
Fluorometric determination of chlorophyll. J. Cons. Int. Explor. Mer. 30: 3-15. 
Hydroqual and Normandeau 1995. A water quality for Massachusetts and Cape Cod
bays: calibration of the Bays Eutrophication Model (BEM). MWRA Environmental
Quality Department Technical Report No. 95-8. Massachusetts Water Resources
Authority, Boston MA 402 
70

Kelly, J. R. 1997. Nitrogen flow and the interaction of Boston Harbor with
Massachusetts Bay. Estuaries, 20, 365-380. 
Kelly, J. R. 1998. Quantification and potential role of ocean nutrient loading to Boston
Harbor, Massachusetts, USA. Marine Ecology Progress Series: 17353-65. 
Libby P. S, Geyer W. R, Keller A. A, Turner J. T, Mansfield A. D, Borkman D. G. and
Oviatt C. A. 2006. 2004 Annual Water Column Monitoring Report. Boston:
Massachusetts Water Resources Authority. Report 2006-15. 177 p. 
Maciolek N. J, Diaz R. J, Dahlen D. T. and Williams I. P. 2006. 2004 Boston Harbor
benthic monitoring report. Boston: Massachusetts Water Resources Authority. Report
2006-01. 244 p. 
Murphy, J. and Riley, J. 1962. A modified single solution for the determination of
phosphate in natural waters. Anal. Chim. Acta. 27:31. 
Nixon, S. W. 1995. Coastal marine eutrophication: A definition, social causes, future
concerns. Ophelia 41: 199-219. 
Rex, A. C. and Taylor, D. I. 2000. Combined Work/Quality Assurance Project Plan for
Water Quality Monitoring and Combined Sewer Overflow Receiving Water Monitoring
in Boston Harbor and its Tributary Rivers 2000. Boston: Massachusetts Water Resources
Authority. Technical Report MS-067 
Rex, A.C, Wu, D, Coughlin, K, Hall, M, Keay, K and D. I. Taylor. 2002. The State of
Boston Harbor; Mapping the Harbor’s Recovery. Boston: Massachusetts Water
Resources Authority, Technical Report ENQUAD 2002-09. 42 p. 
71

Signell, R. P., H. L. Jenter and A. F. Blumberg. 2000. Predicting the physical effects of
relocating Boston outfall. J. Estuarine, Coastal and Shelf Sci., 50: 59-72. 
Solarzano, L, and Sharp, J. H. 1980 a . Determination of total dissolved phosphorus and
particulate phosphorus in natural waters. Limnology and Oceanography, 25, 754-758. 
Solarzano, L, and Sharp, J. H. 1980b. Determination of total dissolved nitrogen in
natural waters. Limnology and Oceanography, 25, 750-754. 
SPSS 2002. SPSS Advanced Statistics 10.0. Chicago. 
Stolzenbach, K. D, and Adams, E. E. 1998. Contaminated Sediments in Boston Harbor.
(pp. 170). Massachusetts Institute for Technology Sea Grant Publications, Cambridge,
USA. 
Sung, W. 1991. Observations on the temporal variations of dissolved copper and zinc in
Boston Harbor. Civil Engineering Practice, Spring 1991, 99-110. 
Taylor, D.I. 2001. Trends in water quality in Boston Harbor during the 8-years before
offshore transfer of Deer Island flows. Boston: Massachusetts Water Resources
Authority. Report ENQUAD 2001-05. 54 p. 
Taylor, D.I. 2004. Boston Harbor: a comparison of eutrophication-related water-quality
before, and 36-months after ‘offshore transfer’. Boston: Massachusetts Water Resources
Authority. Report ENQUAD 2004-04. 63 p. 
Taylor, D.I. 2005. Patterns of wastewater, river and non-point source loadings to Boston
Harbor, 1995 – 2003. Boston: Massachusetts Water Resources Authority. Report
ENQUAD 2005-08. 52 p. 
72

Tucker J, Kelsey S, Giblin A. E. and C. S. Hopkinson J. 2005. 2004 annual benthic
nutrient flux monitoring report. Boston: Massachusetts Water Resources Authority.
Report 2005-11. 68 p. 
Werme, C. and C. Hunt. 2003. 2002 Outfall monitoring overview. Boston:
Massachusetts Water Resources Authority. Report ENQUAD 2003-12. 80 p. 
73

APPENDIX A 
Salinity and water temperature 
Salinity. During the 5-years, Harbor-wide average salinity was not significantly different
from baseline (Table A-1, Figure A-1). The same applied for each of the individual
stations (Fig. A-2). Harbor-wide average salinity was also not significantly different
from baseline for each of the four seasons (Table A-2). 
Table A-1. Salinity and temperature . Comparison of volume-weighted Harbor-
wide average values between the 5-years after transfer, and baseline. 
Variable Average values during Difference p 
between baseline 
Baseline 5-years and 5-years 
Salinity 30.3 + 1.1 29.1 + 1.2 -1.2 (-4%) 0.17 
(ppt) (75) (64) 
Water 9.9 + 5.8 9.3 + 6.1 -0.5 (-5%) 0.21 
temperature (75) (64) 
( o C) 
Table A-2. Differences in salinity and temperature between periods, by season . 
Variable Difference (ppt and o C, and then % baseline) 
Spring Summer Fall Winter 
Salinity -0.3 (-1%) ns -0.1 (<-1%) * -0.4 (-1%) ns +0.45 (+1%) ns 
Temperature +0.3 (+3%) ns -0.2 (-1%) * -0.3 (-3%) ns -1.2 (-36%) * 
74

94 95 96 97 98 99 00 01 02 03 04 05
26
27
28
29
30
31
32
33
34 
Fig. A-1. Salinity . Time-series plot of volume-weighted, Harbor-wide average
salinity, 1994 through 2005. Note, data here are average monthly values.
Offshore
transfer
AVERAGE MONTHLY SALINITY
(PPT.) 
1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005
27
28
29
30
31
32
33 Salinity 
Offshore
transfer
Fig. A-2. Annual Harbor-wide average salinity , 1994 -2005.
ANNUAL AVERAGE SALINITY
( PPT ) 
75

Water temperature. Harbor-wide average water temperatures during the 5-years were
also not significantly different from baseline (Table A-1, Fig. A-4, Fig. A-5). During
2002, annual average temperatures were greater than during the other four years after
transfer. 2002 was also the year when average concentrations of chl- a , and to a lesser
extent POC, were elevated among the years after transfer. 
At none of the 10 stations, as for salinity, were the average temperatures for the 5-years
significantly different from baseline. At all stations, subtraction yielded small negative
values, but at none of the stations were the differences significant (MWRA unpublished
data, plots not shown). 
76

1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005
-5
0
5
10
15
20
25
30 
Offshore
transfer 
Fig. A-4. Water temperature . Time-series plot of volume-weighted, Harbor-wide
average values. Vertical arrow shows completion date of offshore flow transfer.
TEMPERATURE ( o C ) 
1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005
6
8
10
12 Temp. 
Offshore
transfer
Fig. A-5. Annual Harbor-wide average water temperatures , 1994 -2005.
ANNUAL AVERAGE TEMPERATURE
( o C ) 
77

Massachusetts Water Resources Authority
Charlestown Navy Yard
100 First Avenue
Boston, MA 02129
(617) 242-6000
http://www.mwra.state.ma.us