&EPA
United States
Environmental Protection
Agency
Office of Marine
and Estuarine Protection
Region 10
Office of Puget Sound
Seattle WA 98101
Water
EPA 503/3-88-003
July 1988
Characterization of Spatial
and Temporal Trends in
Water Quality in Puget
Sound
Rnal Report
Sound Estuary Program
-------
FINAL REPORT
CHARACTERIZATION OF SPATIAL AND TEMPORAL TRENDS
IN WATER QUALITY IN PUGET SOUND
Contract No. 68-03-3319, Work Assignment 1-32
Contract No. 68-02-4341, Work Assignment 11
July 1988
Submitted to
U.S. ENVIRONMENTAL PROTECTION AGENCY
Region X
Seattle, Washington
Prepared by
Tetra Tech, Inc.
11820 Northup Way, Suite 100
Bellevue, Washington 98005
Under Contract to
BATTELLE
Ocean Sciences
397 Washington Street
Duxbury, Massachusetts 02332
-------
CONTENTS
Page
LIST OF FIGURES vi
LIST OF TABLES xvii
ACKNOWLEDGMENTS xix
EXECUTIVE SUMMARY xx
SPATIAL AND TEMPORAL TRENDS IN WATER QUALITY
IN PUGET SOUND xxi
Physical conditions xxi
Dissolved oxygen xxii
Nutrients xxiii
Indicators of Phytoplankton Growth xxiv
Pollutants xxv
SENSITIVITY TO NUTRIENT ENRICHMENT xxvi
RECOMMENDATIONS FOR ENVIRONMENTAL MONITORING IN PUGET SOUND xxvi
CHAPTER 1. WATER QUALITY CHARACTERIZATION STUDY FOR THE PUGEt
SOUND ESTUARY PROGRAM 1-1
INTRODUCTION 1-1
The Estuarine Environment 1-1
PUGET SOUND ESTUARY PROGRAM 1-2
THE PUGET SOUND WATER QUALITY CHARACTERIZATION PROJECT 1-3
Content and Scope of Work 1-3
Rationale 1-4
Characterization Work Group and Peer Review 1-5
CHAPTER 2. OVERVIEW OF PUGET SOUND 2-1
PHYSICAL ENVIRONMENT AND OCEANOGRAPHY OF PUGET SOUND 2-1
Location 2-1
Basin Configuration 2-1
Climatic Patterns 2-3
Water Sources 2-3
ii
-------
Patterns of Water Circulation 2-8
Patterns of Physical and Chemical Variation in Puget Sound 2-8
A HISTORY OF THE DEVELOPMENT OF THE PUGET SOUND AREA 2-13
FACTORS AFFECTING THE SENSITIVITY OF PUGET SOUND TO
NUTRIENT ENRICHMENT 2-19
CHAPTER 3. STUDY DESIGN 3-1
VARIABLES 3-1
Salinity 3-3
Water Temperature 3-3
Dissolved Oxygen Concentration 3-3
Dissolved Inorganic Nitrate 3-4
Dissolved Orthophosphate 3-4
Chlorophyll a 3-4
Percent Dissolved Oxygen Saturation 3-5
Secchi Disk Depth 3-5
Sulfite Waste Liquor 3-6
Fecal Coliform Bacteria 3-6
Climatic Variables 3-7
STUDY AREAS 3'7
DATA SOURCES 3'8
Study Design and the Amount of Usable Data 3-8
Analytical Techniques 3-11
DATA SETS USED 3-12
Database Quality Assurance Review 3-12
University of Washington 3-14
Washington Department of Ecology 3-14
Washington Department of Fisheries 3-15
Metro 3-15
Climatic Data 3-16
CHAPTER 4. DATA ANALYSIS PROCEDURES 4-1
DATABASE PREPARATION 4-1
Data Compatibility Among the Different Data Sources 4-2
Selection of Representative Stations in Each Study Area for
Pooling Data 4-6
Identification of the Annual Period for Algal Blooms in
Each Study Area 4-7
Standardization of Sampling Depths 4-7
-------
STANDARD ANALYTICAL PROTOCOL 4-8
Characterization of the Environment in the Study Areas 4-8
Analysis of Temporal Trends in Water Quality in Each
Study Area 4-10
CHAPTER 5. RESULTS AND DISCUSSION 5-1
ORGANIZATION OF THE CHAPTER 5-1
WEATHER DURING THE STUDY PERIOD 5-1
NORTHERN SOUND 5-2
Bellingham Bay 5-8
Summary of Results for the Northern Sound , 5-27
CENTRAL SOUND 5-31
Port Gardner 5-45
Point Jefferson 5-60
Sinclair Inlet 5-77
City Waterway 5-85
Summary of Results for the Central Sound 5-101
SOUTHERN SOUND 5-108
Carr Inlet 5-109
Nisqually Reach 5-132
Budd Inlet 5-142
Totten Inlet 5-159
Oakland Bay 5-169
Summary of Results for the Southern Sound 5-183
HOOD CANAL 5-189
Dabob Bay 5-201
Mid-Hood Canal 5-219
South Hood Canal 5-234
Summary of Results for Hood Canal 5-249
CHAPTER 6. SUMMARY AND RECOMMENDATIONS 6-1
SUMMARY OF WATER QUALITY TRENDS IN PUGET SOUND 6-1
Physical Conditions 6-4
Dissolved Oxygen 6-5
Nutrients 6-7
Indicators of Phytoplankton Growth 6-10
Pollutants 6-12
SENSITIVITY TO NUTRIENT ENRICHMENT 6-13
iv
-------
RECOMMENDATIONS FOR ENVIRONMENTAL MONITORING IN PUGET SOUND 6-14
Institutional Recommendations 6-14
Technical Recommendations 6-15
CHAPTER 7. REFERENCES 7-1
APPENDICES
APPENDIX A. HISTORY OF ANALYTICAL TECHNIQUES USED IN WATER
QUALITY STUDIES IN PUGET SOUND A-l
APPENDIX B. SUMMARY OF DATA SET QUALITY ASSURANCE REVIEWS B-l
APPENDIX C. SOURCES OF PUGET SOUND WATER QUALITY DATA C-l
APPENDIX D. COMPARABILITY OF DATA FROM DIFFERENT SOURCES AT
STATIONS WITH OVERLAPPING SAMPLING PERIODS D-l
APPENDIX E. DESCRIPTIVE STATISTICS FOR WATER QUALITY VARIABLES E-l
APPENDIX F. SUMMARY OF CORRELATION COEFFICIENTS BETWEEN WATER
QUALITY VARIABLES F-l
-------
FIGURES
Number Page
2.1 Map of Puget Sound 2-2
2.2 Monthly mean climatic conditions in the Puget Sound area
a. Air temperature (data from 1945 to 1985) 2-4
b. Percent of possible sunshine (data from 1965 to 1985) 2-4
c. Rainfall (data from 1945 to 1985) 2-5
d. Wind velocity (data from 1965 to 1985) 2-5
e. Wind direction (data from 1965 to 1985) 2-6
f. Total freshwater runoff to Puget Sound (data from 1930
to 1978) 2-6
2.3 Generalized vertical cross section of Puget Sound, showing
depth profiles and net circulation pattern 2-9
2.4 Monthly mean surface water temperatures at Pillar Point,
Point Jefferson, and Oakland Bay 2-11
2.5 Dynamics of phytoplankton in central Puget Sound 2-14
2.6 Dynamics of nutrient concentrations in central Puget Sound 2-15
2.7 Total population in the counties of the Puget Sound basin
from 1890 through- 1980 2-18
3.1 Map of Puget Sound showing locations of the study areas
in the water quality characterization project 3-9
5.1 Annual means of air temperature and the percent of possible
sunshine at Seattle-Tacoma International Airport 5-3
5.2 Annual totals of rainfall at Seattle-Tacoma International
Airport and runoff to Puget Sound 5-4
5.3 Annual mean wind velocity at Seattle-Tacoma International
Airport 5-5
5.4 Locations of the study area and sampling stations in the
northern sound 5-6
5.5 Mean salinity and water temperature values in the northern
sound study area during the algal bloom season 5-11
VI
-------
5.6 Mean concentrations of dissolved oxygen and dissolved
inorganic nitrate in the northern sound study area during
the algal bloom season 5-12
5.7 Mean concentrations of dissolved orthophosphate and the mean
percent saturation of dissolved oxygen at the surface in the
northern sound study area during the algal bloom season 5-13
5.8 Mean Secchi disk depth and log of geometric mean concentrations
of sulfite waste liquor and fecal coliform bacteria in the
northern sound study area during the algal bloom season 5-15
5.9 Salinity values at the surface and at 10-m depth in the
Bellingham Bay study area during the algal bloom season 5-18
5.10 Salinity values at 30-m depth and water temperature at the
surface in the Bellingham Bay study area during the algal
bloom season 5-19
5.11 Water temperatures at 10- and 30-m depths in the Bellingham
Bay study area during the algal bloom season 5-20
5.12 Concentrations of dissolved oxygen at the surface and at 10-m
depth in the Bellingham Bay study area during the algal bloom
season 5-22
5.13 Concentrations of dissolved oxygen at 30-m depth and dissolved
inorganic nitrate at the surface in the Bellingham Bay study
area during the algal bloom season 5-23
5.14 Concentrations of dissolved inorganic nitrate at 10- and 30-m
depths in the Bellingham Bay study area during the algal
bloom season 5-24
5.15 Concentrations of dissolved orthophosphate at the surface and
at 10-m depth in the Bellingham Bay study area during the
algal bloom season 5-25
5.16 Concentrations of dissolved orthophosphate at 30-m depth and
percent dissolved oxygen saturation at the surface in the
Bellingham Bay study area during, the algal bloom season 5-26
5.17 Secchi disk depth and log of concentrations of sulfite waste
liquor at the surface in the Bellingham Bay study area during
the algal bloom season 5-28
5.18 Log of concentrations of sulfite waste liquor at 10- and 30-m
depths in the Bellingham Bay study area during the algal
bloom season 5-29
-------
5.19 Log of concentrations of fecal coliform bacteria at the
surface in the Bellingham Bay study area during the algal
bloom season 5-30
5.20 Locations of study areas and sampling stations in the central
sound 5-33
5.21 Mean salinity and water temperature values in the central
sound study areas during the algal bloom season 5-36
5.22 Mean concentrations of dissolved oxygen and dissolved
inorganic nitrate in the central sound study areas during
the algal bloom season 5-37
5.23 Mean concentrations of dissolved orthophosphate and chloro-
phyll a in the central sound study areas during the algal
bloom season 5-38
5.24 Mean percent dissolved oxygen saturation at the surface and
Secchi disk depth in the central sound study areas during the
algal bloom season 5-39
5.25 Log of geometric mean concentrations of sulfite waste liquor
and fecal coliform bacteria in the central sound study areas
during the algal bloom season 5-40
5.26 Depth profiles of mean salinity and water temperature values
in the Point Jefferson study area during the algal bloom
season 5-41
5.27 Depth profiles of mean concentrations of dissolved oxygen in
the Point Jefferson study area during the algal bloom season 5-42
5.28 Salinity values at the surface and at 10-m depth in the
Port Gardner study area during the algal bloom season 5-49
5.29 Water temperatures at the surface and at 10-m depth in the
Port Gardner study area during the algal bloom season 5-50
5.30 Concentrations of dissolved oxygen at the surface and at 10-m
depth in the Port Gardner study area during the algal bloom
season 5-52
5.31 Concentrations of dissolved inorganic nitrate at the surface
and at 10-m depth in the Port Gardner study area during the
algal bloom season 5-53
5.32 Concentrations of dissolved orthophosphate at the surface and
at 10-m depth in the Port Gardner study area during the algal
bloom season 5.54
vm
-------
5.33 Percent dissolved oxygen saturation at the surface and Secchi
disk depth in the Port Gardner study area during the algal
bloom season 5-56
5.34 Log of concentrations of sulfite waste liquor at the surface
and at 10-m depth in the Port Gardner study area during the
algal bloom season 5-57
5.35 Log of concentrations of fecal coliform bacteria at the
surface in the Port Gardner study area during the algal bloom
season 5-59
5.36 Salinity values at the surface and at 10-m depth in the Point
Jefferson study area during the algal bloom season 5-64
5.37 Salinity values at 30- and 100-m depths in the Point
Jefferson study area during the algal bloom season 5-65
5.38 Salinity values at 150- and 200-m depths in the Point
Jefferson study area during the algal bloom season 5-66
5.39 Water temperatures at the surface and 10-m depth in the Point
Jefferson study area during the algal bloom season 5-67
5.40 Water temperatures at 30- and 100-m depths in the Point
Jefferson study area during the algal bloom season 5-68
5.41 Water temperatures at 150- and 200-m depths in the Point
Jefferson study area during the algal bloom season 5-69
5.42 Concentrations of dissolved oxygen at the surface and at
10-m depth in the Point Jefferson study area during the algal
bloom season 5-71
5.43 Concentrations of dissolved oxygen at 30- and 100-m depths in
the Point Jefferson study area during the algal bloom season 5-72
5.44 Concentrations of dissolved oxygen at 150- and 200-m depths
in the Point Jefferson study area during the algal bloom
season 5.73
5.45 Concentrations of chlorophyll a at the surface and at 10-m
depth in the Point Jefferson study area during the algal
bloom season 5.75
5.46 Percent dissolved oxygen saturation at the surface and Secchi
disk depth in the Point Jefferson study area during the algal
bloom season 5.75
5.47 Salinity values at the surface and at 10-m depth in the
Sinclair Inlet study area during the algal bloom season 5-80
IX
-------
5.48 Water temperatures at the surface and at 10-m depth in the
Sinclair Inlet study area during the algal bloom season 5-81
5.49 Concentrations of dissolved oxygen at the surface and at
10-m depth in the Sinclair Inlet study area during the algal
bloom season 5-82
5.50 Concentrations of dissolved inorganic nitrate at the surface
and at 10-m depth in the Sinclair Inlet study area during the
algal bloom season 5-83
5.51 Concentrations of dissolved orthophosphate at the surface and
at 10-m depth in the Sinclair Inlet study area during the
algal bloom season 5-84
5.52 Percent dissolved oxygen saturation at the surface and Secchi
disk- depth in the Sinclair Inlet study area during the algal
bloom season 5-86
5.53 Log of concentrations of sulfite waste liquor and fecal
coliform bacteria at the surface in the Sinclair Inlet study
area during the algal bloom season 5-87
5.54 Salinity values at the surface and at 10-m depth in the
City Waterway study area during the algal bloom season 5-92
5.55 Water temperatures at the surface and at 10-m depth in the
City Waterway study area during the algal bloom season 5-93
5.56 Concentrations of dissolved oxygen at the surface and at 10-m
depth in the City Waterway study area during the algal bloom
season 5-94
5.57 Concentrations of dissolved inorganic nitrate at the suKace
and at 10-m depth in the City Waterway study area during the
algal bloom season / 5-95
5.58 Concentrations of dissolved orthophosphate at the surface and
at 10-m depth in the City Waterway study area during the
algal bloom season 5-96
5.59 Concentrations of chlorophyll a at the surface and at 10-m
depth in the City Waterway study area during the algal bloom
season 5-98
5.60 Percent dissolved oxygen saturation at the surface and Secchi
disk depth in the City Waterway study area during the algal
bloom season 5-99
5.61 Log of concentrations of sulfite waste liquor at the surface
and at 10-m depth in the City Waterway study area during the
algal bloom season 5-100
-------
5.62 Log of concentrations of fecal coliform bacteria at the
surface in the City Waterway study area during the algal
bloom season 5-102
5.63 Locations of study areas and sampling stations in the south
sound 5-110
5.64 Mean salinity and water temperature values in the southern
sound study area during the algal bloom season 5-113
5.65 Mean concentrations of dissolved oxygen and dissolved
inorganic nitrate in the southern sound study areas during
the algal bloom season 5-114
5.66 Mean concentrations of dissolved orthophosphate and chloro-
phyll a in the southern sound study areas during the algal
bloom season 5-115
5.67 Mean percent dissolved oxygen saturation at the surface and
Secchi disk depth in the southern sound study areas during
the algal bloom season 5-116
5.68 Log of geometric mean concentrations of sulfite waste liquor
and fecal coliform bacteria in the southern sound study
areas during the algal bloom season 5-117
5.69 Salinity values at the surface and at 10-m depth in the
Carr Inlet study area during the algal bloom season 5-123
5.70 Salinity values at 30-m depth and water temperatures at the
surface in the Carr Inlet study area during the algal bloom
season 5-124
5.71 Water temperatures at 10- and 30-m depths in the Carr Inlet
study area during the algal bloom season 5-125
5.72 Concentrations of dissolved oxygen at the surface and at
10-m depth in the Carr Inlet study area during the algal
bloom season 5-126
5.73 Concentrations of dissolved oxygen at 30-m depth and
dissolved inorganic nitrate at the surface in the Carr
Inlet study area during the algal bloom season 5-127
5.74 Concentrations of dissolved inorganic nitrate at 10- and
30-m depths in the Carr Inlet study area during the algal
bloom season 5-129
5.75 Concentrations of dissolved orthophosphate at the surface
at the surface and at 10-m depth in the Carr Inlet study
area during the algal bloom season 5-130
xi
-------
5.76 Concentrations of dissolved orthophosphate at 30-m depth
and percent dissolved oxygen saturation at the surface in
the Carr Inlet study area during the algal bloom season 5-131
5.77 Secchi disk depth and log of concentrations of fecal
coliform bacteria at the surface in the Carr Inlet study
area during the algal bloom season 5-133
5.78 Salinity values at the surface and at 10-m depth in the
Nisqually Reach study area during the algal bloom season 5-136
5.79 Water temperatures at the surface and at 10-m depth in the
Nisqually Reach study area during the algal bloom season 5-138
5.80 Concentrations of dissolved oxygen at the surface and at
10-m depth in the Nisqually Reach study area during the
algal bloom season 5-139
5.81 Concentrations of dissolved inorganic nitrate at the surface
and at 10-m depth in the Nisqually Reach study area during
the algal bloom season 5-140
5.82 Concentrations of dissolved orthophosphate at the surface
and at 10-m depth in the Nisqually Reach study area during
the algal bloom season 5-141
5.83 Percent dissolved oxygen saturation at the surface and
Secchi disk depth in the Nisqually Reach study area during
the algal bloom season 5-143
5.84 Log of concentrations of fecal coliform bacteria at the
surface in the Nisqually Reach study area during the algal
bloom season 5-144
5.85 Salinity values at the surface and at 10-m depth in the
Budd Inlet study area during the algal bloom season 5-149
5.86 Water temperatures at the surface and at 10-m depth in the
Budd Inlet study area during the algal bloom season 5-150
5.87 Concentrations of dissolved oxygen at the surface and at
10-m depth in the Budd Inlet study area during the algal
bloom season 5-152
5.88 Concentrations of dissolved inorganic nitrate at the surface
and at 10-m depth in the Budd Inlet study area during the
algal bloom season 5-154
5.89 Concentrations of dissolved orthophosphate at the surface
and 10-m depth in the Budd Inlet study area during the algal
bloom season 5-155
xii
-------
5.90 Percent dissolved oxygen saturation at the surface awdJigefiM-
disk depth in the Budd Inlet study area during the algin
bloom season 5-156
5.91 Log of concentrations of sulfite waste liquor and fecal
coliform bacteria at the surface in the Budd Inlet study
area during the algal bloom season 5-158
5.92 Salinity values at the surface and at 10-m depth in the
Totten Inlet study area during the algal bloom season 5-162)
5.93 Water temperatures at the surface and at 10-m depth in the
Totten Inlet study area during the algal bloom season 5-163
5.94 Concentrations of dissolved oxygen at the surface and at
10-m depth in the totten Inlet study area during the algal
bloom season 5-165
5.95 Concentrations of dissolved inorganic nitrate at the surface
and at 10-m depth in the Totten Inlet study area during the
algal bloom season 5-166
5.96 Concentrations of dissolved orthophosphate at the surface
and at 10-m depth in the Totten Inlet study area during the
algal bloom season 5-167
5.97 Percent dissolved oxygen saturation at the surface and
Secchi disk depth in the Totten Inlet study area during the
algal bloom season 5-168
5.98 Log of concentrations of sulfite waste liquor and fecal
coliform bacteria at the surface in the Totten Inlet study
area during the algal bloom season 5-170
5.99 Salinity values at the surface and at 10-m depth in the
Oakland Bay study area during the algal bloom season 5-174
5.100 Water temperatures at the surface and at 10-m depth in the
Oakland Bay study area during the algal bloom season 5-175
5.101 Concentrations of dissolved oxygen at the surface and at
10-m depth in the Oakland Bay study area during the algal
bloom season 5-176
5.102 Concentrations of dissolved inorganic nitrate at the surface
and at 10-m depth in the Oakland Bay study area during the
algal bloom season 5-178
5.103 Concentrations of dissolved orthophosphate at the surface
and at 10-m depth in the Oakland Bay study area during the
algal bloom season 5-179
xiii
-------
5.104 Concentrations of chlorophyll a and percent dissolved oxygen
saturation at the surface in the Oakland Bay study area
during the algal bloom season 5-181
5.105 Secchi disk depth and log of concentrations of sulfite waste
liquor at the surface in the Oakland Bay study area during
the algal bloom season 5-182
5.106 Log of concentrations of fecal coliform bacteria at the
surface in the Oakland Bay study area during the algal bloom
season 5-184
5.107 Locations of study areas and sampling stations in Hood Canal 5-191
5.108 Mean salinity and water temperature values in the Hood Canal
study areas during the algal bloom season 5-194
5.109 Mean concentrations of dissolved oxygen and dissolved
inorganic nitrate in the Hood Canal study areas during the
algal bloom season 5-195
5.110 Mean concentrations of dissolved orthophosphate and chloro-
phyll a in the Hood Canal study areas during the algal bloom
season 5-196
5.111 Mean percent dissolved oxygen saturation at the surface and
Secchi disk depth in the Hood Canal study areas during the
algal bloom season 5-197
5.112 Log of geometric mean concentrations of sulfite waste liquor
and fecal coliform bacteria in the Hood Canal study areas
during the algal bloom season 5-198
5.113 Salinity values at the surface and at 10-m depth in the
Dabob Bay study area during the algal bloom season 5-205
5.114 Salinity values at 30-m depth and water temperature at the
surface in the Dabob Bay study area during the algal bloom
season 5-206
5.115 Water temperatures at 10- and 30-m depths in the Dabob Bay
study area during the algal bloom season 5-207
5.116 Concentrations of dissolved oxygen at the surface and at
10-m depth in the Dabob Bay study area during the algal
bloom season 5-209
5.117 Concentrations of dissolved oxygen at 30-m depth and
dissolved inorganic nitrate at the surface in the Dabob Bay
study area during the algal bloom season 5-210
-------
5.118 Concentrations of dissolved inorganic nitrate at 10- and
30-m depths in the Dabob Bay study area during the algal
bloom season 5-212
5.119 Concentrations of dissolved orthophosphate at the surface
and at 10-m depth in the Dabob Bay study area during the
algal bloom season 5-213
5.120 Concentrations of dissolved orthophosphate at 30-m depth
and chlorophyll a at the surface in the Dabob Bay study
area during the algal bloom season 5-214
5.121 Concentrations of chlorophyll a at 10- and 30-m depths in
the Dabob Bay study area during the algal bloom season 5-216
5.122 Percent dissolved oxygen saturation at the surface and
Secchi disk depth in the Dabob Bay study area during the
algal bloom season 5-217
5.123 Log of concentrations of fecal coliform bacteria at the sur-
face in Dabob Bay study area during the algal bloom season 5-218
5.124 Salinity values at the surface and at 10-m depth in the
Mid-Hood Canal study area during the algal bloom season 5-223
5.125 Salinity values at 30-m depth and water temperatures at the
surface in the Mid-Hood Canal study area during the algal
bloom season 5-224
5.126 Water temperatures at 10- and 30-m depths in the Mid-Hood
Canal study area during the algal bloom season 5-225
5.127 Concentrations of dissolved oxygen at the surface and at
10-m depth in the Mid-Hood Canal study area during the
algal bloom season 5-227
5.128 Concentrations of dissolved oxygen at 30-m depth and
dissolved inorganic nitrate at the surface in the Mid-Hood
Canal stufy area during the algal bloom season 5-228
5.129 Concentrations of dissolved inorganic nitrate at 10- and
30-m depths in the Mid-Hood Canal study area during-the
algal bloom season 5-229
5.130 Concentrations of dissolved orthophosphate at the surface
and at 10-m depth in the Mid-Hood Canal study area during
the algal bloom season 5-231
5.131 Concentrations of dissolved orthophosphate at 30-m depth
and percent dissolved oxygen saturation at the surface in
the Mid-Hood Canal study area during the algal bloom season 5-232
xv
-------
5.132 Secchi disk depth and log of concentrations of fecal coliform
bacteria at the surface in the Mid-Hood Canal study area
during the algal bloom season 5-233
5.133 Salinity values at the surface and at 10-m depth in the South
Hood Canal study area during the algal bloom season 5-238
5.134 Salinity values at 30-m depth and water temperatures at the
surface in the South Hood Canal study area during the algal
bloom season 5-239
5.135 Water temperatures at 10- and 30-m depths in the South Hood
Canal study area during the algal bloom season 5-240
5.136 Concentrations of dissolved oxygen at the surface and at
10-m depth in the South Hood Canal study area during the
algal bloom season 5-242
5.137 Concentrations of dissolved oxygen at 30-m depth and
dissolved inorganic nitrate at the surface in the South
Hood Canal study area during the algal bloom season 5-243
5.138 Concentrations of dissolved inorganic nitrate at 10- and
30-m depths in the South Hood Canal study area during the
algal bloom season 5-244
5.139 Concentrations of dissolved orthophosphate at the surface
and at 10-m depth in the South Hood Canal study area during
the algal bloom season 5-246
5.140 Concentrations of dissolved orthophosphate at 30-m depth and
percent dissolved oxygen saturation at the surface in the
South Hood Canal study area during the algal bloom season 5-247
5.141 Secchi disk depth and log of concentrations of fecal coliform
bacteria at the surface in the South Hood Canal study area
during the algal bloom season 5-248
6.1 Rates of change of phosphate concentrations in urban and
rural study areas since 1973 6-9
xvi
-------
TABLES
Number Page
1.1 Members of the work group for the Puget Sound water
characterization project 1-6
2.1 Location and flow of rivers discharging into Puget Sound 2-7
3.1 Water quality variables analyzed for the characterization
study of Puget Sound 3-2
3.2 Study areas in the water quality characterization project 3-10
4.1 Laboratory detection limits used in the characterization
database 4-3
4.2 Water quality standards applicable to the characterization
study areas 4-11
5.1 Sampling station numbers, data sources, and sampling periods
for the study area in the northern sound 5-7
5.2 Algal bloom seasons in the northern sound study area, as
defined by monthly mean and standard error of percent
dissolved oxygen saturation in surface water 5-9
5.3 Net change and percent change in the mean values of water
quality variables in the northern sound, based on ANOVA
comparisons of tiata taken before 1973 with data taken from
1973 to 1986 5-16
5.4 Slopes of statistically significant long-term and recent
regressions of water quality variables as a function of year
for the northern sound 5-17
5.5 Sampling station numbers, data sources, and sampling periods
for the study areas in the central sound 5-34
5.6 Algal bloom seasons for the central sound study areas, as
defined by monthly mean and standard error of percent
dissolved oxygen saturation in surface water 5.35
5.7 Net change and percent change in the mean values of water
quality variables in the central sound, based on ANOVA
comparisons of data taken before1 1973 with data taken from
1973 to 1986 5.43
xvn
-------
5.8 Slopes of statistically significant long-term and recent
regressions of water quality variables as a function of year
for the central sound 5-44
5.9 Sampling station numbers, data sources, and time periods
for the study areas in the southern sound 5-111
5.10 Algal bloom seasons for the southern sound study areas, as
defined by monthly mean and standard error of percent
dissolved oxygen saturation in surface water 5-112
5.11 Net change and percent change in the mean values of water
quality variables in the southern sound, based on ANOVA
comparisons of data taken before 1973 with data taken from
1973 to 1986 5-118
5.12 Slopes of statistically significant long-term and recent
regressions of water quality variables as a function of year
for the southern sound 5-119
5.13 Sampling station numbers, data sources, and sampling periods
for the study areas in Hood Canal 5-192
5.14 Algal bloom seasons for Hood Canal study areas, as defined
by monthly mean and standard error of percent dissolved
oxygen saturation in surface water 5-193
5.15 Net change and percent change in the mean values of water
quality variables in Hood Canal, based on ANOVA comparisons
of data taken before 1973 with data taken from 1973 to 1986 5-199
5.16 Slopes of statistically significant long-term and recent
regressions of water quality variables as a function of year
for Hood Canal 5-200
6.1 Summary of water quality trends in Puget Sound 6-2
xvm
-------
ACKNOWLEDGEMENTS
This document was prepared by Tetra Tech, Inc., under the direction of
Dr. Stephen K. Brown, for Battelle Ocean Sciences and the U.S. Environmental
Protection Agency (EPA), Region X, in partial fulfillment of EPA Contracts
Nos. 68-03-3319 and 68-02-4341. This project was funded through the National
Estuary Program under the authority of the Clean Water Act, as amended, and
by the Puget Sound Estuary Program. Funding was approved by the EPA Office
of Marine and Estuarine Protection.
Ms. Michelle Miller of the EPA Office of Marine and Estuarine Protection
and Dr. John Armstrong of EPA Region X were the EPA Work Assignment Managers.
Mr. Richard McGrath and Dr. Michael Connor of were the Technical Monitors for
Battelle.
The primary author of this report was Dr. Stephen K. Brown. Ms. Becky
A. Maguire served as the data manager and computer programmer. Dr. Alyn C.
Duxbury of Washington SeaGrant provided technical guidance and wrote
Appendices A and C. Peer review was provided by Dr. Gordon R. Bilyard of
Tetra Tech and Drs. Michael Connor and Eric Crecelius of Battelle. Ms.
Theresa Wood, Ms. Marcy Brooks-McAuliffe, and Dr. James Erckmann assisted in
technical editing and report production. Ms. Betty Dowd, Ms. Pamela
Charlesworth, Ms. Kim Reading, and Mr. Michael Rylko provided graphics
support. Word processing support was provided by Ms. Andi Manzo and Ms.
Nellie Johnson of EPA. Tetra Tech word processing support was provided by
Ms. Lisa Fosse, Ms. Debra Shlosser, Ms. Gail Singer, Ms. Patricia Canterbury,
Ms. Anna Bolstead, Ms. Gestin Suttle, Ms. Jo Graden, and Ms. Vivia Boe.
xix
-------
EXECUTIVE SUMMARY
The primary purpose of this characterization study is to assess whether
water quality in Puget Sound has changed over time. The major focus of this
study is nutrient enrichment and the enhancement of algal blooms. The
physical variables investigated are salinity and water temperature. The
chemical variables investigated are concentrations of dissolved oxygen,
inorganic nitrate, and orthophosphate. The intensity of algal blooms is
measured by the concentration of chlorophyll a, percent dissolved oxygen
saturation in surface water, and Secchi disk depth. The concentration of
sulfite waste liquor is evaluated as an index of pulp mill pollution. The
concentration of fecal coliform bacteria is evaluated as an index of sewage
contamination. Although toxic contaminants are an important environmenta1
concern in Puget Sound, they were not investigated during this study. The
project was sponsored by the U.S. Environmental Protection Agency (EPA),
Office of Marine and Estuarine Protection (OMEP); and the U.S. EPA Region X,
Office of Puget Sound as part of the estuarine characterization initiative.
Data for this study were compiled from existing data sources and n jw
exist in a unique database. Oceanographic data were obtained fr>. T, the
University of Washington Department of Oceanography, the Washington Depart-
ment of Ecology, the Washington Department of Fisheries, and the Municipality
of Metropolitan Seattle. Climatic data were obtained from the National
Oceanic and Atmospheric Administration. Sources of oceanographic data were
screened to determine where data were collected (i.e., location) and the
timeframe in which data were collected. Quality assurance reviews were also
performed to assess the validity of the field and laboratory techniques used
to generate the data. Before the above data sets were used, they were
examined for completeness and corrected for errors and inconsistencies in
data coding and units of measurement.
xx
-------
Numerous changes in the water quality of Puget Sound were observed
despite limited availability of high quality, long-term monitoring data.
Changes in station locations and data sources also limited data interpre-
tation in some areas (e.g., Port Gardner, Budd Inlet, Oakland Bay). Hence,
results should not be overinterpreted. The absence of detectable changes in
a study area may indicate that changes have not occurred or that the
available data did not provide sufficient resolution to detect changes that
have occurred. Also, substantial distances separating most of the sampling
stations from onshore pollutant sources may have limited the ability of the
analyses to reveal changes that may have occurred near such sources when
inputs from those sources charged. Moreover, because data were available at
most sites only for the upper 10-30 m of the water column, changes in condi-
tions below these depths could only be detected if they affected the water
column relatively close to the surface. Finally, changes detected at a
given sampling station do not demonstrate that the same types of changes
have occurred throughout the body of water that contains the station.
SPATIAL AND TEMPORAL TRENDS IN WATER QUALITY IN PUGET SOUND
Temporal trends in water quality were analyzed at 13 study areas around
Puget Sound. Study areas were located in northern Puget Sound (one),
central Puget Sound (four), southern Puget Sound (five), and Hood Canal
(three). Trend analyses were conducted for each study area during that
area's algal bloom season. Although oceanographic data are available dating
back to 1932, data sets for most of the study areas began in the 1950s.
Results are summarized below.
Physical Conditions
Salinity values decreased and water temperature values increased in
most study areas. Salinity values declined in 8 of the 13 study areas. No
explanation is available for the decreasing salinity values, but the
available rainfall data are inconsistent with the salinity declines. The
observed decline in rainfall in the Seattle area would have been expected to
increase, rather than to decrease, salinity in the sound. One possible
xx i
-------
explanation for the salinity declines is that decreased volumes of oceanic
water entering the sound from the Strait of Juan de Fuca may have contributed
to the salinity decreases.
Water temperature values increased in 7 of the 13 study areas and
decreased in only 2 of the 13 study areas. Water temperatures appear to be
influenced by climate. At the study sites where increases in water tempera-
tures were detected, data collection began during the cool periods of the
early 1930s and the early 1950s.
Dissolved Oxygen
Dissolved oxygen concentrations increased in 7 of the 13 study areas,
all of which are located in the southern sound or Hood Canal. The sites in
the southern sound were influenced by unusually high dissolved oxygen
concentrations in 1986, the last year included in the study. Although the
cause of these high dissolved oxygen concentrations could not be determined,
they may have occurred during intense algal blooms. Neither temporal
increases in dissolved oxygen concentrations, nor unusually high concentra-
tions of dissolved oxygen in 1986 were observed in the northern or central
sound study areas.
Very low dissolved oxygen concentrations were rarely observed during
the study. At all study areas except Point Jefferson, the deepest samples
were only collected from 10 or 30-m depth. Because of active circulation,
low dissolved oxygen concentrations are unlikely to occur in the Point
Jefferson study area. Also, minimum dissolved oxygen concentrations
typically do not occur during the algal bloom season. The lowest mean
dissolved oxygen concentration (4.3 mg/L) was observed at 30-m depth in the
South Hood Canal study area.
A single exception to the apparent absence of low dissolved oxygen
concentrations occurred at the Oakland Bay study area during the mid-1950s.
Discharges of sulfite waste liquor from the ITT-Rayonier pulp mill in the
City of Shelton lowered dissolved oxygen concentrations, occasionally down
xxii
-------
to zero. Very low dissolved oxygen concentrations were not found in the
Oakland Bay study area after this mill closed in 1957.
Nutrients
Nitrate data are available since the mid-1970s. Well-developed trends
appear to have occurred only in the study areas in Port Gardner, City
Waterway, Carr Inlet, and Hood Canal. Except for the City Waterway study
area, substantial decreases in nitrate concentrations were detected in all
these sites. In Hood Canal, progressively less-developed decreases were
detected in the Mid-Hood Canal and Dabob Bay study areas. It appears that
the factor affecting nitrate concentrations in Hood Canal [apparently algal
blooms (see below)] probably was most influential in southern Hood Canal.
The decrease in nitrate concentrations in the Carr Inlet study area also may
be attributed to increased algal blooms. No explanations were apparent for
the nitrate decrease in the Port Gardner study area or the nitrate increase
in the City Waterway study area.
Temporal changes in phosphate concentrations were detected at 11 study
areas. Long-term decreases (since the 1950s) were detected in seven of the
nine study areas from which long-term phosphate data are available. Recent
increases (since the mid-1970s) were detected in six study areas. Five of
these six increases were statistically significant at PO.05; the signifi-
cance level of the sixth increase was P=0.08. No statistically significant
(PO.05) decreases have been detected since the mid-1970s.
The cause(s) of the widespread decreases in phosphate concentrations
observed since the 1950s are unknown. One explanation involves decreased
inputs of phosphate from the Strait of Juan de Fuca. However, this
hypothesis has not been tested. Because the declines were detected in both
rural and urban study areas, anthropogenic influences do not explain these
results. Although it was not possible to calibrate the analytical techniques
used in the 1950s with those used recently, the older techniques generally
were reasonably accurate.
xxm
-------
All recent increases in phosphate concentrations occurred in urban
study areas: Bellingham Bay (Bellingham), Port Gardner (Everett), Sinclair
Inlet (Bremerton), City Waterway (Tacoma), Budd Inlet (Olympia), and
Oakland Bay (Shelton). The cause(s) of these phosphate increases are not
known. However, the absence of such increases in rural study areas suggests
that these increases may be attributed to anthropogenic factors. Changes in
the quantities and characteristics of pulp mill discharges since the mid-
1970s may have influenced phosphate concentrations in Bellingham Bay, Port
Gardner, and City Waterway. Substantial decreases in the discharges of
sulfite waste liquor occurred near the Bellingham Bay and Port Gardner study
areas. Because sulfite waste liquor removes phosphate from seawater
solution, increased phosphate concentrations in these two areas could have
been in response to decreased sulfite waste liquor concentrations. Phos-
phoric acid has been added to the effluent of the kraft pulp mill located on
Commencement Bay since 1977. This additional phosphate may have contributed
to the recent increase in phosphate concentrations observed at the nearby
City Waterway study area. No specific anthropogenic factors were identified
to explain the phosphate increases that have occurred since the mid-1970s in
the other three urban areas (Sinclair Inlet, Budd Inlet, Oakland Bay).
Indicators of Phvtoplankton Growth
Few credible trends in the values of the standard phytoplankton indi-
cators were detected in most of the study areas. Phytoplankton concentra-
tions appear to have increased in the Carr Inlet study area. A statistical
decrease in phytoplankton concentrations was detected at the Point Jefferson
study area, while a statistical increase in phytoplankton concentrations was
detected at the Nisqually Reach study area. However, both of these changes
appear to have been caused by erratic fluctuations, rather than by systematic
trends.
Increases in percent dissolved oxygen saturation and decreases in
nutrient concentrations were detected at depths of 10- and 30-m depths in the
Hood Canal study areas. These changes suggest that rates of photosynthesis
may have increased at depth in the Hood Canal study areas.
xx iv
-------
The lack of statistically significant changes in the values of the
standard phytoplankton indicator variables in most of the study areas may be
attributed in part to inadequacies in the database. Sampling frequencies
were insufficient to assess algal bloom dynamics. Data on cell density,
photosynthesis rates, or chlorophyll a concentrations would have provided
direct measures of phytoplankton abundance, but few such data are available.
The surrogate variables used as pKytoplankton indicators in this study (i.e.,
percent dissolved oxygen saturation at the surface and Secchi disk depth)
only provide information about conditions near the surface. Phytoplankton
maxima can occur well below the surface, particularly in areas with clear
water, such as Hood Canal. Also, both surrogate variables are affected by
variations in environmental variables other than phytoplankton abundance.
Percent dissolved oxygen saturation at the surface is affected by the oxygen
content of the source water, while Secchi disk depth is affected by the
turbidity associated with suspended sediments.
Pollutants
Concentrations of sulfite waste liquor declined in all four study areas
near pulp mills: Bellingham Bay, Port Gardner, City Waterway., and Oakland
Bay. The decline of sulfite waste liquor concentrations in the Oakland Bay
study area coincided with the closure of the ITT-Rayonier pulp mill in the
City of Shelton. Declines at the other sites generally coincided with
improvements in the effluent treatment procedures used by the local mills.
Temporal changes in concentrations of fecal coliform bacteria may be
attributed to changes in point sources -near the Bellingham Bay and Port
Gardner study areas. Declines at the Bellingham Bay site coincided with
improvements in the sewage treatment facilities and closures of combined
sewer overflows. An apparent increase in fecal coliform bacteria concentra-
tions in the Port Gardner study area followed the initiation of secondary
effluent treatment by the nearby Scott sulfite pulp mill. This increase
probably was due to discharges of the bacterium, Klebsiella. an organism
xxv
-------
that is detected in standard fecal coliform tests and that grows rapidly in
the secondary treatment facilities of sulfite pulp mills.
A decline in fecal coliform bacteria concentrations was detected at the
Nisqually Reach study area. One relatively high value was recorded in 1978,
near the beginning of data collection for fecal coliform bacteria at this
site. This high value, which was detected in a sample collected near the
end of a heavy rainstorm, has been followed by very low values since 1978.
These results suggest that the high value in 1978 reflected an unusual
influence of contaminated runoff.
SENSITIVITY TO NUTRIENT ENRICHMENT
The sensitivity of an estuary to the deleterious effects of algal
blooms caused by nutrient enrichment depends on several factors, including
the amounts of inputs, flushing rates, and density stratification. Because
of their physical conditions and/or their proximities to large population
centers, the following areas were judged to be most vulnerable to nutrient
enrichment:
t Sinclair Inlet
• Budd Inlet
• Oakland Bay
• South Hood Canal.
RECOMMENDATIONS FOR ENVIRONMENTAL MONITORING IN PUGET SOUND
The water quality characterization study provides a unique opportunity
to assess existing monitoring programs by using the data in a water quality
trends analysis. Key recommendations derived from the results of this study
and comments of work group members are outlined below.
xxvi
-------
• One organization should oversee all water quality monitoring
programs to facilitate compatibility of field and laboratory
techniques and database formats, and to coordinate geographic
coverage. Use of the protocols recommended by the Puget
Sound Estuary Program (U.S. EPA 1986a) would standardize the
field and laboratory techniques of monitoring programs in
Puget Sound.
t Changes in field and laboratory techniques should be docu-
mented; new techniques should be calibrated with old tech-
niques.
• The goals of the monitoring program should be developed
quantitatively before the study design is finalized (e.g.,
how much change in dissolved oxygen concentrations should be
detectable over a given time period?). Alternative study
designs should be evaluated with statistical power analysis,
using existing data.
t The influence of physical factors (e.g., climate, bulk flows
of oceanic and fresh water) on water quality should be
monitored to improve understanding of ecosystem function and
to enable comparison of the impacts of physical and anthro-
pogenic factors on water quality.
t Environmental sampling should reflect the seasonal, temporal,
and spatial scales of variation of the individual variables
of most interest. The following suggestions are provided to
improve upon the historical monitoring programs in Puget
Sound:
Samples collected to assess the impacts of major
contaminant sources should be collected close to those
sources
xxvn
-------
Sampling plans designed to detect low dissolved oxygen
concentrations should emphasize frequent sampling of the
bottom waters at the heads of poorly flushed inlets
(e.g., Budd Inlet) during late summer
Sampling of algal blooms should occur frequently during
the peak bloom season of each individual study area
Sampling of episodic events (e.g., discharges from
combined sewer overflows or pulp mills) should occur
when the discharges occur.
• Water quality in Budd Inlet should be monitored to determine
whether nitrogen removal by the Lacey, Olympia, Tumwater, and
Thurston County (LOTT) sewage treatment plant reduces bloom
intensity.
• Specific and quantitative measures of phytoplankton density
(e.g., concentration of chlorophyll a, species-level
identification) and suspended sediments (e.g., concentration
of suspended solids) should be used in place of Secchi disk
depth. Species-level identification of phytoplankton
combined with chlorophyll a data provides the most sensitive
measure of changes in the phytoplankton community.
• A microbiological test is needed to distinguish between
bacterial contamination from sewage and secondary effluents
from sulfite pulp mills. Because current fecal coliform
tests cannot make this distinction, shellfish beds may be
closed because of exposure to pulp mill effluent rather than
exposure to sewage. Although not well studied in Puget
Sound, the health risk associated with Klebsiella contamina-
tion from secondary pulp mill effluent appears to be less
than the health risk associated with sewage contamination.
xxvm
-------
CHAPTER 1. WATER QUALITY CHARACTERIZATION STUDY
FOR THE PUGET SOUND ESTUARY PROGRAM
INTRODUCTION
This report presents the results of the analysis of spatial and
temporal trends in the water quality of Puget Sound. The sound is an
estuary located in western Washington State, USA (see Chapter 2 for a
detailed discussion). This study was sponsored by the U.S. Environmental
Protection Agency (U.S. EPA), Office of Marine and Estuarine Protection
(OMEP); and U.S. EPA Region X, Office of Puget Sound.
The objective of this project is to characterize the water column of
Puget Sound by analyzing historical and current water quality data.
Physical and chemical variables were analyzed to assess nutrient enrichment
and algal blooms in the water column. In addition, contaminants from pulp
mill and sewage discharges were analyzed.
This report comprises six chapters: Chapter 1 introduces the water
quality characterization study and the Puget Sound Estuary Program (PSEP);
Chapter 2 provides an overview of the physical environment, oceanography,
and history of Puget Sound; Chapter 3 describes the study design; Chapter 4
details the procedures used for data analysis; Chapter 5 presents the
results of the analysis of 13 selected study areas; and Chapter 6 includes
the summary of results and recommendations for monitoring in the sound.
The Estuarine Environment
An estuary is a semi-enclosed coastal body of water with an open
connection to the sea. The seawater in an estuary is diluted by fresh water
from upland sources. This fresh water typically provides substantial inputs
of nutrients to an estuary. Compared with the open ocean, most estuaries are
1-1
-------
shallow and protected. Thus, the estuarine environment usually is conducive
to high rates of biological productivity. Because estuaries are often
natural connecting points between oceanic transportation systems and onshore
populations, most coastal cities and sea ports are situated on or near
estuaries. Consequently, estuaries are vulnerable to the environmental
degradation that is often associated with an urbanized, industrial society.
The impact of pollution on an estuary depends on the amount and type of
inputs, and the capacity of the system to assimilate or export excesses. In
the case of nutrient enrichment, the initial effect is to stimulate primary
production, principally by phytoplankton in the water column. Moderate
increases in primary production may be beneficial because increased energy
inputs into the food chain may enhance fish and shellfish production.
However, excessive nutrient enrichment in estuaries can overstimulate
primary production. The subsequent sinking and decay of dead plant material
may cause declines in dissolved oxygen concentrations, which become increas-
ingly severe with depth (Neilson and Cronin 1981). Large inputs of organic
material (e.g., untreated raw sewage or pulp mill wastes) may also lower
dissolved oxygen concentrations at depth as the material decays. Low
dissolved oxygen concentrations in the water column can have severe dele-
terious effects on an estuary, causing declines or mortality of fish and
bottom-dwelling invertebrates, and the occurrence of fouled and malodorous
waters.
PUGET SOUND ESTUARY PROGRAM
PSEP was initiated in 1985 under OMEP's National Estuary Program. The
purpose of the National Estuary Program is to protect and restore water
quality and living resources in the nation's estuaries. The major partici-
pants in PSEP are the U.S. EPA Region X, Office of Puget Sound, the Puget
Sound Water Quality Authority (PSWQA); and the Washington Department of
Ecology (Ecology). Agencies involved with PSEP seek to maintain water
quality standards for Puget Sound that protect public health and welfare;
assure protection and propagation of fish, shellfish, and wildlife popula-
tions; and allow recreational activities.
1-2
-------
One of the highest priorities of PSEP is the comprehensive characteri-
zation (i.e., description) of Puget Sound. The goals of characterization
are to 1) identify environmental changes and determine relationships between
environmental conditions and resource use, 2) identify adverse changes in the
biological system, 3) identify the causes and importance of such adverse
changes, and 4) identify key measures needed to track changes and improve-
ments in the environment.
THE PUGET SOUND WATER QUALITY CHARACTERIZATION PROJECT
In keeping with the National Estuary Program's characterization
initiative, one focus of PSEP's characterization work is the synthesis of
historical data that have not been analyzed completely. Trends in water
quality emerged as a priority for historical characterization studies in
1987. This topic was chosen by the U.S. EPA's Office of Puget Sound after
extensive consultation with their Technical Advisory Committee, PSWQA, and
interested scientists from universities and other governmental agencies in
the region.
Content and Scope of Work
This water quality characterization report examines temporal trends in
several nutrient-related variables (i.e., conventional pollutants), including
turbidity and concentrations of nitrate, phosphate, dissolved oxygen, and
chlorophyll a in the water column of Puget Sound. Fecal coliform bacteria
and sulfite waste liquor are included as .indices of domestic and pulp mill
discharges. Salinity and water temperature data are also analyzed to
facilitate interpretation of the variables of interest. The study is
geographically comprehensive, with detailed analyses of the available data
from numerous locations throughout the sound.
The investigation of conventional pollutants in this project comple-
ments PSEP's studies of toxic contamination in Puget Sound (e.g., U.S. EPA
1986b). Toxic contaminants (i.e., metals and organic chemicals) cause
1-3
-------
serious problems in some urbanized areas of the sound [e.g., metals in the
sediments of Commencement Bay (Tetra Tech 1985)]. However, investigating
trends in contamination by toxic substances was beyond the scope of the
present study.
Because of constraints on time and level-of-effort, most of the work
consisted of analyzing existing data sets that were reasonably complete and
readily accessible. However, substantial efforts were devoted to completing
the entry of water column data from the University of Washington and the
Washington Department of Fisheries into STORET, U.S. EPA's computerized
database of water quality information.
Rationale
The purpose of this report is to fill a large gap in the existing body
of information on water quality in Puget Sound with a synthesis of historical
data. The sound is a critical resource that supports an abundance of
commercially and recreationally important fish and shellfish. The sound also
serves as a major shipping corridor (U.S. EPA 1984). However, growth of
population and industry in the Puget Sound region have caused the intro-
duction of large quantities of nutrients and other wastes into the sound.
This analysis of water quality trends throughout the sound will attempt to
provide an early warning of potential problems and a long-term historical
basis for interpreting future monitoring data. This information may also be
used to facilitate the design of the sampling program for the Puget Sound
monitoring plan presently being developed by PSWQA (Tetra Tech 1986).
Previous analyses of temporal trends in conventional water quality
variables in Puget Sound have not detected long-term degradation of water
quality caused by nutrient enrichment. However, most of these studies have
been limited to the central basin of the sound [e.g., Duxbury 1975; National
Oceanic and Atmospheric Administration (NOAA) 1985]. Jones and Stokes
(1984) reported that "nutrient loading can become, or is, a problem in
relatively shallow, poorly flushed embayments of Puget Sound," and indicated
that future population growth could cause deteriorating conditions in other
1-4
-------
regions of the sound. Localized effects of high nutrient concentrations,
which may be anthropogenic in origin, have been observed in certain areas
[e.g., in Budd Inlet (URS 1986a)]. Thus, a comprehensive study encompassing
all the regions of the sound was conducted.
Characterization Work Group and Peer Review
A work group consisting of local scientists with experience in water
quality research in Puget Sound was assembled to provide advice and peer
review (Table 1.1). Dr. John Armstrong of the U.S. EPA monitored the
project. External review was provided by Dr. Gordon Bilyard of Tetra Tech,
Inc., and Dr. Michael Connor, Dr. Eric Crecelius, and Mr. Richard McGrath of
Battelle. Dr. Alyn Duxbury of Washington Sea Grant was retained by PSEP as
a consultant for this project. He provided detailed recommendations on the
design of the study.
1-5
-------
TABLE 1.1. MEMBERS OF THE WORK GROUP FOR THE
PUGET SOUND WATER CHARACTERIZATION PROJECT
Name Affiliation
John Armstrong U.S. EPA
Chuck Boatman URS, Inc.
Ned Cokelet NOAA
Eugene Collias University of Washington (retired)
Jeffery Cox Evans-Hamilton, Inc.
Ralph Domenowske Metro
Alyn Duxbury Washington Sea Grant
Alan Mearns NOAA
Marvin Tarr Washington Department of Fisheries
Don Weston U.S. EPA
John Yearsley U.S. EPA
1-6
-------
CHAPTER 2. OVERVIEW OF PUGET SOUND
This chapter contains background information about Puget Sound. The
information provides a context in which the water quality analyses can be
interpreted. The chapter consists of three sections:
t Physical environment and oceanography of Puget Sound
• History of the development of the Puget Sound area
• Factors affecting the sensitivity of Puget Sound to nutrient
enrichment.
PHYSICAL ENVIRONMENT AND OCEANOGRAPHY OF PUGET SOUND
Location
Puget Sound is a fjord-like estuary (i.e., narrow and deep) with a
maximum depth of approximately 280 m (see Figure 2.1). It is connected to
the Pacific Ocean by the Strait of Juan de Fuca to the west and the Strait
of Georgia to the north. A shallow sill at Admiralty Inlet separates the
sound from the two straits.
Basin Configuration
South of Admiralty Inlet, Puget Sound is subdivided into the Main
Basin, Whidbey Basin, South Basin, and the Hood Canal Basin. The Main Basin
lies between sills at Admiralty Inlet and Tacoma Narrows. It contains about
60 percent of the total volume of the sound south of Admiralty Inlet.
Whidbey Basin lies between Whidbey Island to the west and the mainland to
the east; it is not bordered by a sill. The Hood Canal Basin is long and
narrow. It is oriented primarily north and south, with a major embayment
(Dabob Bay) to the west and an eastern arm at its head. The Hood Canal
2-1
-------
Basin is separated from Admiralty Inlet by a sill 50 m deep. The South
Basin lies south of the sill at Tacoma Narrows. This basin has a complex
arrangement of deep (down to nearly 200 m) channels, shallow embayments,
islands, and sills.
Climatic Patterns
The climate in Puget Sound has a strong seasonal component. Meteoro-
logical data collected at the Seattle-Tacoma International Airport from 1945
to 1985 (NOAA 1945-1985) show that July and August generally are warm, sunny,
and dry. Late autumn, winter, and early spring generally are cool, cloudy,
and damp (see Figure 2.2a-c). About three-quarters of the annual precipi-
tation occurs from October to March. These patterns are long-term averages;
individual years may deviate substantially from the norms. For example,
from 1945 to 1985, air temperature averaged 12.1° C during the warmest year
(1958) and 8.9° C during the coolest year (1955). Prevailing winds have the
highest average velocity during the wet season (Figure 2.2d) and tend to come
from the south and southeast during this period (Figure 2.2e). The calmest
winds occur during the summer, when the prevailing direction is from the
west.
Water Sources
Coastal seawater flows into the Strait of Juan de Fuca at depth
(Collias et al. 1974). Compared with water already in the sound, the
coastal seawater is relatively dense (a^ >26), salty (salinity >33 ppt), and
cold (temperature <8° C). Although most of the coastal seawater exiting the
Strait of Juan de Fuca flows north into the Strait of Georgia, a substantial
amount flows south through Admiralty Inlet and into the Main Basin.
Rivers that discharge into Puget Sound are its primary source of fresh
water (Collias et al. 1974). The average flow of fresh water into the
sound is approximately 1,600 m3/sec (Table 2.1). The largest rivers
discharging into Puget Sound are in the northern and central regions.
Because fresh water is less dense than salt water, it tends to remain near
the surface until turbulence mixes it into deeper layers.
2-3
-------
(a)
ac.
D
Q.
2
10
§ 0
-16 i
6 7
MONTH
ie
12
(b)
NOTE: Data collected at the Seattle-Tacoma International Airport.
Figure 2.2. Monthly mean climatic conditions in the Puget Sound area.
a. Air temperature (data from 1945 to 1985).
b. Percent of possible sunshine (data from 1965 to 1985).
2-4
-------
(c)
•40
30
<
u.
O.
ie
—I—
5
i—• i
6 7
MONTH
ie
11 18
(d)
*
Q
—i—'—i—'—i—'—i—'—i—'—i—'—i—'—r
23456789
MONTH
10 11 12
NOTE: Data collected at the Seattle-Tacoma International Airport.
Figure 2.2. (Continued). Monthly mean climatic conditions in the Puget Sound area.
c. Rainfall (data from 1945 to 1985).
d. Wind velocity (data from 1965 to 1985).
2-5
-------
(e)
300
§200
o
Ul
a:
5
o
* 100
6
9
10 11
MONTH
(f)
50001
4000
E 3000
u
!o
2000
1000
6 7
MONTH
10
11 12
NOTE' Wind data collected at the Seattle-Tacoma International Airport;
runoff data collected from 7-22 USGS gauging stations.
Figure 2.2. (Continued). Monthly mean climatic conditions in the Puget Sound area.
e. Wind direction (data from 1965 to 1985).
f. Total freshwater runoff to Puget Sound (data from 1930 to 1978).
2-6
-------
TABLE 2.1. LOCATION AND FLOW OF RIVERS DISCHARGING INTO PUGET SOUND
River
Skagit
Snohomish
Still aguamish
Nooksack
Puyallup
Nisqually
Sammamish-C'edar
Green-Duwamish
Elwha
Skokomish 1
Skokomish 2a
Deschutes
Dosewall ips
Duckabush
Hamma Hamma
Dungeness
Samish
Big Quilcene
Tahuya
Whatcom
Little Quilcene
Other
Location of
Discharge
Skagit Bay
Everett Harbor
Port Susan
Bell ingham Bay
Commencement Bay
Nisqually Reach
Shilshole Bay
Elliott Bay
Dungeness Bay
Southern Hood Canal
Southern Hood Canal
Budd Inlet
Central Hood Canal
Central Hood Canal
Central Hood Canal
Dungeness Bay
Samish Bay
Quilcene Bay
Southern Hood Canal
Bell ingham Bay
Quilcene Bay
Average
Flow
(nr/sec)
470
290
130
110
95
70
50
50
40
35
30
25
20
15
15
13
5
5
4
3
2
60
Percent
Total
Flow
30
18
8
7
6
5
3
3
2
2
2
1
1
1
1
1
<1
<1
<1
<1
<1
3
Total
1,590
100
a Skokomish 2 is the outlet from Cushman powerhouse No. 2.
Reference: Evans-Hamilton, Inc. and D.R. Systems, Inc. (1987).
2-7
-------
The volume of fresh water runoff entering the sound varies with volume
of rainfall, except during spring and early summer when mountain snowmelt
augments riverine inputs (see Figure 2.2f) (NOAA 1984a). Runoff from
snowmelt occurs primarily in the Whidbey, Main, and Hood Canal Basins.
Because the area around the southern sound contains few mountains, the early
summer rise in runoff from snowmelt has little effect in the southern sound.
Patterns of Water Circulation
A two-layered pattern of net water circulation occurs in Puget Sound
(Collias et al. 1974; Ebbesmeyer and Barnes 1980). Lighter, less saline
water flows seaward near the surface, while denser, more saline oceanic water
flows landward near the bottom (Figure 2.3). The seaward surface flow is
driven by riverine inputs. Vertical mixing, particularly at the heads of
embayments, entrains deeper oceanic water up toward the surface, driving the
landward flow of the deeper layers and providing salt to the surface layers.
This two-layered circulation pattern is complicated by the presence of
islands and shallow sills.
Large, oscillating tidal currents are superimposed on the net circu-
lation pattern of surface outflow and deep inflow. Tidal exchange drives
much of the vertical antl horizontal mixing in Puget Sound, particularly at
the sills at Admiralty Inlet and Tacoma Narrows. This mixing causes a
substantial amount of low-salinity surface water to reflux into deeper
layers (Figure 2.3). Approximately one-third to one-half of the surface
water passing over the sill at Admiralty Inlet recirculates into the Main
Basin via the deep layer before exiting the sound (PSWQA 1986a).
Patterns of Physical and Chemical Variation in Puget Sound
Sal inity--
The outflow of surface waters removes water and salt from the sound.
Salt is replenished by the inflow of oceanic water along the bottom.
Surface salinity values are near zero at river mouths and reach approximately
2-8
-------
ro
I
Z 160
260 •:•:•:•:•:
•X-; BOO
STRAIT OF JUAN DE FUCA | ADMIRALTY INLET| PUGET SOUND, MAIN BASIN [SOUTHERN BASINI
Reference: PSWQA(1986b).
Figure 2.3. Generalized vertical cross section of Puget Sound, showing depth profile and net circulation pattern.
-------
32 ppt in the Strait of Juan de Fuca. Bottom salinity values range from 28
to 33 ppt in the Main Basin and in the deep (>50 m) inlets. Bottom water
with the highest salinity values is found in the western (Pacific) end of
the Strait of Juan de Fuca and in the northern sound. Bottom water with the
lowest salinity values is found at the heads of shallow (<20 m) inlets,
particularly in the southern sound. Typically, there is a replacement of
bottom water in the landward basins by dense, high-salinity seawater during
the late summer or early autumn (Collias et al. 1974).
Depth gradients of salinity are affected by storms and wind direction.
Winter storms and the prevailing winter southerly winds promote vertical
mixing and break down vertical salinity gradients in the water column. The
calmer summer weather and the prevailing summer westerly winds reduce rates
of vertical mixing. These factors allow the development of density strati-
fication in the water column. In areas where inputs of fresh water are
substantial, density stratification results in vertical salinity gradients
(particularly during calm periods), with the least dense and lowest salinity
water at the surface.
Water Temperature--
The annual pattern of variation in surface water temperature usually
lags behind the annual pattern of variation in air temperature (see Figures
2.2a and 2.4). Highest surface water temperatures typically occur in July or
August. Differences among annual mean surface water temperatures can reach
approximately 1.0° C. Temperature at depth is less variable than at the
surface, responding more to advective .processes than to heat exchange.
Thermal depth stratification is well developed during the summer, except in
areas such as Admiralty Inlet, where turbulence caused by currents passing
over sills mixes the water column. Thermal depth stratification breaks down
during the winter because of storms and the prevailing wind direction.
Ranges in surface water temperatures vary geographically. Generally,
the more enclosed, shallow, sluggishly circulating areas undergo more summer
warming and winter cooling than the more open, deeper, turbulently mixed
areas (Figure 2.4). The maximum summer monthly mean surface water tempera-
2-10
-------
19
18
M 16
.
«
fe.15
14
LJ
§13
CL
2 11
u
£ 10
u
I9
8
7
6
* PILLAR POINT
*. POINT JEFFERSON
4 OAKLANDBAY
•-*.
23456789
MONTH
10 11 12
Figure 2.4. Monthly mean surface water temperatures at Pillar Point, Point Jefferson,
and Oakland Bay.
2-11
-------
ture is approximately 18.3° C at Oakland Bay, a shallow southern embayment.
However, the maximum summer monthly mean surface water temperature is only
about 11.2° C at Pillar Point, a deep area in the Strait of Juan de Fuca.
Winter surface water temperatures vary over a smaller range. Minimum winter
monthly mean surface water temperature is approximately 6.5° C at Oakland
Bay and 7.7° C at Pillar Point.
Dissolved Oxygen--
Surface gas exchange and photosynthesis are the principal sources of
dissolved oxygen in Puget Sound. Decay of organic material consumes oxygen,
particularly in deeper waters. These factors produce vertical gradients of
dissolved oxygen concentrations with lower values at depth. This gradient
is enhanced by vertical density stratification during the warmer months,
which prevents mixing of the surface and bottom layers. These effects,
along with the entrainment of oxygen-poor bottom waters into the heads of
embayments by net circulation, can cause periods of low dissolved oxygen in
the heads of Dabob Bay, Lynch Cove, Port Susan, and northern Saratoga
Passage.
Results from recent studies of Budd Inlet (URS 1986a) suggest that
diurnal vertical migration of dinoflagellates may also contribute to
vertical gradients of dissolved oxygen concentrations during dinoflagellate
blooms in southern embayments. During such blooms, oxygen produced by
photosynthesis during the day may cause surface waters to become super-
saturated. At night the dinoflagellates consume oxygen in deeper waters.
This concentration gradient is then maintained by density stratification.
Phytoplankton and Nutrients--
In central Puget Sound, phytoplankton growth usually is controlled by
the amount of light available for photosynthesis and by vertical mixing
rates. Vertical mixing removes algal cells from the photic zone. Winter et
al. (1975) reported that over 50 percent of the chlorophyll a in the water
column of central Puget Sound was below the photic zone, and that at the end
of phytoplankton blooms, nitrate concentrations in the photic zone fell
2-12
-------
below levels needed for phytoplankton growth for periods of 2-3 days. They
also reported that phosphate and silicate concentrations dropped during
blooms, but remained above levels needed for phytoplankton growth. Lower
rates of vertical mixing in the sluggishly circulating embayments may reduce
the rates of removal of algal cells from the photic zone and the rates of
renewal of nutrients to the surface layers from deeper waters (Collias et
al. 1974; Duxbury 1975; Anderson et al. 1984; NOAA 1985). Therefore,
limitation of algal growth by low nutrient concentrations may be more
important in the southern embayments and in Hood Canal than in central Puget
Sound.
The annual phytoplankton cycle begins with a spring diatom bloom.
Diatom abundance tends to drop off in midsummer, although a secondary diatom
bloom may occur in late summer. Dinoflagellate abundances gradually
increase through the spring and reach a peak bloom in midsummer. Photosyn-
thetic rates and algal standing stocks tend to be highest near the summer
solstice and lowest near the winter solstice (Figures 2.5a,b) (Anderson et
al. 1984).
Nutrient concentrations generally reflect the annual cycle of algal
growth and dieoff. Lowest concentrations of nitrate and phosphate usually
occur near midsummer, and highest concentrations usually occur near the
winter solstice (Figure 2.6a,b) (Collias et al. 1974; Anderson et al. 1984).
Phosphate in Puget Sound is replenished via the late summer replacement of
resident water by upwelled oceanic water that is high in phosphate. Another
factor that augments phosphate concentrations is the decay of sinking
organic particles. Subsequently, phosphate-rich deep water is entrained to
the surface by net circulation. Nitrate regeneration results primarily from
riverine inputs during periods of high runoff in the winter (Robinson and
Brown 1983).
A HISTORY OF THE DEVELOPMENT OF THE PUGET SOUND AREA
The potential for water quality deterioration resulting from anthro-
pogenic inputs is influenced by a variety of sociological factors, including
population size, land use patterns, and economic activities. Environmental
2-13
-------
June "82
January '83
June '83
January '84
(b)
50
-------
(b)
June U2
January '83
June'83
January '84
NOTE: Data collected at Seahurst Bight.
Reference: Anderson etal. (1984).
Figure 2.6. Dynamics of nutrient concentrations in central Puget Sound.
a. Dissolved inorganic nitrate.
b. Dissolved orthophosphate.
2-15
-------
impacts can be affected not only by changes in these factors (e.g., popula-
tion growth may increase the amount of sewage discharged into a receiving
environment), but also by changes in governmental regulations and programs.
For example, construction of a regional sewage treatment system by the
Municipality of Metropolitan Seattle (Metro) reduced the number of raw
sewage discharges into Puget Sound from 46 to zero between the late 1950s
and 1970 (Metro 1969).
Human population activities may affect water quality directly by the
production of domestic wastes and by other activities (e.g., industrial
production) that generate wastes. Land use patterns affect the distribution
and types of impacts caused by the population. Dredging or filling of
wetlands to increase navigable waterways and usable land area can reduce the
capacity of an estuary to trap sediments and absorb nutrients. Urban
development concentrates the human population and local environmental
impacts, increases runoff by increasing the amount of impervious land
surface (e.g., roads, buildings), and increases nonpoint contamination from
sources such as automobiles, farms, and households. While concentration of
the urban population may facilitate the development of municipal sewage
treatment systems, decentralization of the urban population may cause
environmental problems to become more diffuse and less easily solved by
local institutions and centralized treatment facilities.
Historical and current economic activities may have produced or may be
producing characteristic types of environmental impacts in Puget Sound. For
example, manufacturing processes that have been used in the region can
introduce oxygen-demanding wastes and toxic wastes from point sources (e.g.,
pulp and lumber mills). Agricultural activities in the region may cause
releases of organic wastes, nutrients, and pesticides from nonpoint sources.
A forestry-based economy may lead to problems with upland erosion, sedimen-
tation, and contamination by wood debris and chemicals that are used in the
manufacture of products derived from wood (e.g., pulp).
The following historical information on the settlement and economic
development of the Puget Sound region has been summarized from three
2-16
-------
sources: Chasan (1981), NOAA (1985), and PSWQA (1986b). Detailed infor-
mation is available from these sources.
European settlement of the Puget Sound region began in 1845. Timber
products and shipping were the major industries during the nineteenth
century. Fisheries and agriculture were also important industries. By
1900, the population of the region had reached 200,000 (Figure 2.7). After
1910, resource-based industries (e.g., fishing, logging) peaked and began to
decline, while manufacturing, transportation, and service industries began
to grow. By 1920, substantial portions of the wetlands in Puget Sound had
been diked and filled for agriculture (Shapiro and Associates 1983). During
the 1920s, adverse impacts of pulp mills were becoming apparent. Communities
around Lake Washington began treating sewage in the 1930s. The first
routine monitoring of Puget Sound by the University of Washington began in
1932. Population growth continued in both urban and agricultural areas.
Total population of the region reached 1 million by the mid-1940s.
A^ter World War II, the economy became more diversified, with emphasis
on the shipping, aerospace, and manufacturing sectors. The Washington State
Pollution Control Commission was established in 1945. Resource-based and
agricultural sectors continued to decline, although forestry-based employment
and manufacturing remained important. In 1958, Metro was formed to develop
and operate a regional sewage treatment system in the Seattle area.
Industrial activity and residential land development became increasingly
decentralized during the post-war period. For example, the average
population density in pre-1960 residential areas was 20.5 persons/ha. In
residential developments after 1960, the population density was approximately
11 persons/ha. The West Point Sewage Treatment Plant, which treats the bulk
of Seattle's municipal sewage, became operational in 1966. During the 1950s
and 1960s, 80,000 ha of commercially managed forest land was converted to
urban or industrial use, roads, and farms. After 1970, the service sector
became the largest component of the economy. Aerospace and military
employment also remained high, while resource-based sectors, including
forestry products, continued to decline. Population growth continued to
expand into outlying counties.
2-17
-------
ro
I
t—'
CO
2.3T
2.2.
2.1-
2.0.
1.9.
1.8.
1.7.
1.6.
— 1.5.
w
§1 I-3-
lc12-
Q. « 1-1-
£ | ..o.
e 0.9.
9^0.8.
0.7.
0.6.
0.5.
0.4.
0.3.
0.2.
0.1
1890 1900 1910 1920 1930 1940 1950 1960
Year
1970
1980
Reference: NOAA(1985).
Figure 2.7. Total population in the counties of the Puget Sound basin from 1890 through 1980.
-------
By 1985, the population of the Puget Sound region reached approxi-
mately 2.9 million. Population growth rates have fallen slightly in the
past 15 yr, to below 50,000/yr. "Suburbanization" has continued, along with
the associated increases in the coverage of land by impervious surfaces.
Between 1965 and 1985, the amount of intensively used urban land in the
region nearly doubled, from 135,000 to 260,000 ha. Of the 91 km2 of coastal
wetlands that were present in Puget Sound before European settlement,
approximately 60 percent has been converted to other uses. In 1985, the
U.S. EPA made the decision to deny all applications for waivers of secondary
sewage treatment requirements in the Puget Sound region. The intent of this
decision was to forestall future degradation of water quality in the sound
that might be caused by future population expansion in the region.
Forecasts suggest that the recent trends in population and economic
growth that may affect water quality in Puget Sound will continue (PSWQA
1986b). The population of the region is expected to reach 3.7-3.9 million by
the year 2000. The continued growth of low-density residential areas
suggests that increases in nonpoint sources of pollution will continue.
With the continued shift to a service-based, rather than a manufacturing-
based economy, the rate of increase of industrial pollutants may begin to
decline. Other future impacts on the water quality of the sound may be
related to increased recreational and commercial marine traffic, and an
increase in Naval port facilities.
FACTORS AFFECTING THE SENSITIVITY OF PUGET SOUND TO NUTRIENT ENRICHMENT
The impact of pollution on an estuary depends on pollutant input rates
and on the physical conditions controlling the capacity of the system to
assimilate or export excesses. Because nutrient inputs are affected by the
size of the population that discharges wastes, urbanized areas of Puget
Sound may be more sensitive to the effects of nutrient enrichment than are
rural areas. Major urbanized areas of the sound include Elliott Bay (near
Seattle), Commencement Bay (near Tacoma), Bellingham Bay (near Bellingham),
Possession Sound/Port Gardner (near Everett), Sinclair Inlet (near Bremer-
ton), Budd Inlet (near Olympia), and Oakland Bay (near She!ton).
2-19
-------
The capacity of an estuary to assimilate a given input of nutrients
without excess algal blooms and subsequent low dissolved oxygen concentra-
tions depends, in part, on the rate of vertical mixing in the water column.
As discussed previously, rapid vertical mixing removes phytoplankton from
the photic zone, thereby limiting the rate of algal growth, and facilitating
the dispersal of excess nutrients. Slow vertical mixing, which is often
associated with a stable density stratification of the water column, allows
algal cells to remain in the photic zone where their growth rate is high.
Slow vertical mixing also reduces the dispersal of excess nutrients.
Stability of the water column is maximal in sheltered areas during warm,
calm weather. Because a lack of vertical mixing may allow an algal bloom to
exhaust the nutrient supply near the surface (Winter et al. 1975), anthro-
pogenic nutrient enrichment during a bloom in an area with a stratified
water column could enhance an algal bloom by artificially replenishing the
diminished nutrient supply (URS 1986a).
The capacity of an estuary to export excess inputs of nutrients before
algal blooms and dissolved oxygen problems occur is influenced by the
flushing rate, or the residence time of water. Rapid flushing removes
nutrient inputs before high concentrations can accumulate, and also disperses
the nutrients before they can be consumed by the algae. Slow flushing
allows nutrient inputs to accumulate, and allows the algae to remain in an
area with an enhanced nutrient concentration, which could stimulate the
growth of a bloom.
Within Puget Sound, the capacity of an area to assimilate excess
nutrients and the capacity of an area to export excess nutrients are closely
linked (Strickland 1983). Sheltered embayments frequently have stable water
columns and low flushing rates, while areas that are offshore or in the
major basins of the sound tend to be both well flushed and well mixed.
Thus, areas likely to be vulnerable to nutrient enrichment because of the
physical environment include the southern embayments (Carr Inlet, Case
Inlet, Henderson Inlet, Budd Inlet, Eld Inlet, Totten Inlet, and Oakland
Bay), Hood Canal, Port Susan, Liberty Bay, Dyes Inlet, and several other
small embayments, coves, and harbors with limited circulation. Areas not
likely to be vulnerable to nutrient enrichment because of the physical
environment include the Strait of Juan de Fuca, Admiralty Inlet, Nisqually
Reach, and central areas of the Main Basin.
2-20
-------
CHAPTER 3. STUDY DESIGN
This chapter describes the components of the study design: 1) the
variables investigated to characterize the study areas, 2) the criteria used
to select study areas and evaluate data sets, and 3) the quality assurance
review and final selection of data sets. The data available for use in this
study are from sites scattered throughout the sound. Thus, it was only
feasible to characterize environmental conditions and water quality trends
at particular sites. It was not possible to characterize the sound, or even
portions of the sound, on a large geographic scale (i.e., entire embayments).
For example, the results obtained from the Bellingham Bay study area are
representative only of the immediate study area, and may not be represen-
tative of Bellingham Bay as a whole.
VARIABLES
The purpose of the y/ater quality characterization study was to investi-
gate temporal and spatial variation in water quality variables that are
related to conventional pollutants in Puget Sound. These pollutants
typically are derived from point discharges of nutrients and oxygen-demanding
materials (i.e., municipal and industrial wastes) and from nonpoint sources
such as agricultural runoff. Toxic pollutants such as heavy metals,
polychlorinated biphenyls (PCBs), and pesticides are also an important
concern for Puget Sound, but were not part of this particular study. The
major variables chosen for analysis are directly related to nutrient
enrichment and algal blooms, and were considered by the work group to be key
measures of, or indicators of conventional pollution in Puget Sound.
Physical variables were included for descriptive purposes and because
physical conditions in an area affect the dynamics of pollutant impacts.
The water quality variables investigated during this study are listed
in Table 3.1. The importance of each is summarized below. In addition,
climatic data were obtained to facilitate interpretation of the water quality
3-1
-------
TABLE 3.1. WATER QUALITY VARIABLES ANALYZED FOR THE
CHARACTERIZATION STUDY OF PUGET SOUND
Variable Category
Variable Analyzed
Physical conditions
Dissolved oxygen
Nutrients
Indicators of
phytoplankton growth
Pollutants
Salinity
Water temperature
Dissolved oxygen concentration
Dissolved inorganic nitrate
Dissolved orthophosphate
Chlorophyll a concentration
Percent dissolved oxygen
saturation at the surface
Secchi disk depth
Sulfite waste liquor
Fecal coliform bacteria
3-2
-------
trends detected in the study areas. Climatic variables are summarized
following the discussion of the water quality variables.
Salinity
Salinity is the concentration of dissolved salts in a water sample. It
is used in this study to determine the extent of density stratification and
vertical mixing of the water column. Information on density stratification
is important because vertical mixing can remove phytoplankton from the
photic zone, reducing the likelihood that a phytoplankton bloom will occur.
Alternatively, a stratified water column occurs when vertical mixing rates
are low. Stratification reduces the rate of phytoplankton removal from the
photic zone and increases the likelihood that a phytoplankton bloom will
occur. In an estuary, salinity data also provide an index of seawater
dilution and are needed to calculate percent dissolved oxygen saturation.
Water Temperature
Water temperature is used to evaluate climatic influences on the water
column, including vertical mixing, density stratification, and the origins
of water masses. Warm water and sunshine enhance photosynthetic rates,
increasing the likelihood of algal blooms. A well-mixed water column does
not exhibit a substantial depth gradient in water temperature. However, a
density-stratified water column generally exhibits a depth gradient in water
temperature, with warmer water, heated by the sun, near the surface. Water
temperature also affects dissolved oxygen concentration because the solu-
bility of oxygen is lower in warm water.
Dissolved Oxygen Concentration
Dissolved oxygen concentrations directly affect the ability of organisms
(e.g., fish and invertebrates) to live in the water. Changes in dissolved
oxygen concentration are caused by the decay of organic material; by the
respiration of pelagic and benthic organisms, both of which consume oxygen;
and by photosynthesis in the water column, which produces oxygen. The
photosynthetic production of oxygen occurs primarily in near-surface waters.
3-3
-------
Dissolved oxygen concentrations may be impacted by anthropogenic inputs of
nutrients and oxygen-demanding wastes. Nutrient enrichment can cause an
increase in dissolved oxygen concentration in the surface waters if photo-
synthetic rates are increased by the additional nutrients. Nutrient
enrichment can also cause a decrease in dissolved oxygen at depth when an
algal bloom caused by enhanced nutrient concentrations dies and decays.
Oxygen-demanding wastes (e.g., raw sewage and untreated pulp mill discharges)
also reduce dissolved oxygen concentrations as they decay.
Dissolved Inorganic Nitrate
Dissolved inorganic nitrate is a major algal nutrient present in sewage
(including primary and secondary effluent) and in agricultural runoff.
Nitrogen is often the phytoplankton nutrient in lowest supply (i.e., the
limiting nutrient) in Puget Sound (Anderson et al 1984). Therefore,
anthropogenic inputs of nitrate into the sound may increase the intensity of
algal blooms.
Dissolved Orthophosphate
Like nitrate, dissolved orthophosphate is a major algal nutrient
present in sewage (including primary and secondary effluent) and in agricul-
tural runoff. Phosphate often limits phytoplankton growth in fresh water,
but it rarely limits phytoplankton growth in estuaries. However, because
phosphate generally is not consumed as rapidly as nitrate during phytoplank-
ton blooms in estuaries, concentrations during blooms generally remain above
analytical detection limits. Thus, although phosphate may be less important
ecologically than nitrate in Puget Sound, phosphate may be a useful index of
changes in nutrient concentrations.
Chlorophyll a
Chlorophyll a is a measure of the concentration of photosynthetic
pigments in water. It is a rough, but easily obtained, measure of the
standing stock of phytoplankton. Chlorophyll a concentration is not a
particularly good measure of photosynthetic rate because the photosynthetic
3-4
-------
rate of a cell containing a given amount of chlorophyll a can vary.
Chlorophyll a concentration also is not a good measure of the concentration
of living algal cells because dead cells may retain their pigments for a
substantial period of time (Winter et al. 1975). However, chlorophyll a has
been widely measured in other studies of Puget Sound [e.g., Winter et al.
(1975) and Anderson et al. (1984)], and is the only direct measurement of
algal concentration for which data are available for a sufficient length of
time to warrant trend analysis.
Percent Dissolved Oxvaen Saturation
The percent saturation of dissolved oxygen and the dissolved oxygen
concentration are closely related. Dissolved oxygen concentration is
affected by salinity and temperature, while the calculation of percent
saturation compensates for differences in salinity and temperature. Because
most water quality studies analyze oxygen concentration rather than oxygen
saturation, concentration was the major dissolved oxygen variable analyzed
in this study. However, data on percent dissolved oxygen saturation were
also analyzed because oxygen saturation above 100 percent in surface water
often indicates photosynthetic enhancement of dissolved oxygen.
Values of percent dissolved oxygen saturation were calculated using the
method of Weiss (1970). Values were obtained from the ratio of the actual
dissolved oxygen concentration in a water sample vs. the concentration that
would be found in a water sample of the same salinity and temperature at
100-percent saturation. Percent saturation is calculated by multiplying the
above ratio by 100.
Secchi Disk Depth
Secchi disk depth is an easily obtained measurement of the transparency
or turbidity of the water column near the surface. The depth of the photic
zone (i.e., the portion of the water column where light levels are suffi-
ciently high for photosynthesis to occur) is roughly twice the Secchi depth
(Preisendorfer 1986). The Secchi disk depth is the depth at which a white
disk (usually 30 cm in diameter) disappears from view. Secchi disk depth
3-5
-------
decreases as the amount of suspended particulates in the water increases.
Changes in phytoplankton standing stock, which can be affected by nutrient
enrichment, can be detected by changes in Secchi disk depth.
Several limitations affect the interpretation of Secchi disk depth
data. The Secchi disk depth is influenced by concentrations of both
phytoplankton and suspended sediment. In addition, the amount of available
light and the visual capabilities of the observer affect the data. Moreover,
the Secchi disk does not provide information about conditions deeper than
the Secchi disk depth. Thus, although Secchi disk depth is a widely used
index of water clarity, changes in Secchi disk depth may be difficult to
interpret.
Sulfite Waste Liquor
The concentration of sulfite waste liquor is a measurement of the
amount of waste discharged from pulp mills that use sulfites. The sulfites
are used to separate cellulose fibers from wood for the production of paper.
Sulfite waste liquor is toxic in high concentrations and contains large
amounts of organic material that consume oxygen as it decays. Sulfites also
react directly with oxygen (Strickland and Parsons 1972) and dissolved
orthophosphate (Westley and Tarr 1978) in seawater. Because pulp mills have
been an important industry in the Puget Sound area since about 1920, the
measurement of sulfite waste liquor is an indicator of a major industrial
contaminant that affects concentrations of dissolved oxygen and nutrients
and that has been discharged for many years into the sound.
Fecal Coliform Bacteria
Fecal coliform bacteria are present in inadequately chlorinated sewage
and in runoff from pastures and other agricultural facilities that contain
large amounts of animal wastes. These organisms are not directly harmful,
but their concentration is used as an -index of contamination by pathogens
from sewage and runoff from agricultural facilities.
3-6
-------
Climatic Variables
The following variables are included in the climatic database: air
temperature, wind direction, wind velocity, rainfall, the percentage of
possible sunshine (i.e., the percentage of time that the sun shines,
unobscured by clouds, between sunrise and sunset), and total estimated runoff
to Puget Sound. All these variables can affect algal growth rates and
physical conditions in the water column. They may affect algal growth rates
directly by influencing the rate of photosynthesis, or indirectly by
influencing vertical mixing rates of the water column.
STUDY AREAS
Sites were selected for the characterization study to optimize geo-
graphical and temporal coverage. Because data availability varied greatly
among candidate study sites, it was not possible to use rigid criteria for
site selection. Based on the consensus of the work group, candidate study
areas were ranked as high, medium, or low priority within each region of the
sound (Strait of Juan de Fuca; northern, central, and southern Puget Sound;
and Hood Canal). Only high priority areas were included in the study. The
criteria used to evaluate sites are listed below in 'the approximate order of
importance:
• Inclusion of sites from all regions of the sound
• Availability of long-term monitoring data, with sampling
extending through 1986
• Inclusion of a wide range of environments (e.g., rural or
urban, well or poorly mixed)
• Potential for detecting long-term changes in water quality
without interference from excessive, short-term variation
(e.g., a station near the mouth of a tidal river could have
salinity fluctuations over a tidal cycle that greatly
exceeded the magnitude of any possible long-term change)
3-7
-------
• Potential for anthropogenic nutrient enrichment to affect
water quality
• Computer accessibility of the data.
Thirteen study areas were chosen for inclusion in the characterization
study (Figure 3.1). Brief descriptions of each area are given in Table 3.2.
Detailed discussions of these 13 areas are provided in Chapter 5. No study
area was selected from the Strait of Juan de Fuca because no site had both
long-term historical data and present day monitoring.
DATA SOURCES
Two factors were evaluated to determine which of the many Puget Sound
water quality data sets would be included in the water quality characteri-
zation study:
0 Study design and the amount of usable data available for each
potential study site
• Analytical techniques used in each study.
Study Design and Amount of Usable Data
Study designs were evaluated to determine the frequency and water depth
of sampling. Studies were used that provided at least several data points
per year at a site. However, sampling frequency tended to be lower in the
earlier studies. For example, the University of Washington monitoring
program often sampled once per season from the 1930s through the 1950s.
Although this sampling frequency is inadequate by current standards, the
University of Washington data set is the only substantial data source for
this early period. Therefore, the University of Washington data were used.
The depths sampled varied greatly among studies, but studies that included
routine sampling to at least a 30-m depth were preferred.
3-8
-------
O
BELUNGHAU
V < -
? - ' -
o «£•'
-A-
-Si
1 Bellingham Bay
2 Port Gardner
3 Point Jefferson
Sinclair Inlet
City Waterway
Carr Inlet
Nisqually Reach
8 Budd Inlet
9 Totten Inlet
1 0 Oakland Bay
11 Dabob Bay
1 2 Mid-Hood Canal
1 3 South Hood Canal
4
5
6
7
SHELTON TT
,f j-
V-
"V
rOfc.--
V . Js,.
.-m
•? _•' i •••
'-OLVMPIA
0 5 10
Figure 3.1. Map of Puget Sound showing locations of the study areas in the water
quality characterization project.
3-9
-------
TABLE 3.2. STUDY AREAS IN THE WATER QUALITY CHARACTERIZATION PROJECT
Study Area
Environmental Characteristics
Bellingham Bay
Port Gardner
Point Jefferson
Sinclair Inlet
City Waterway
Carr Inlet
Nisqually Reach
Budd Inlet
Totten Inlet
Oakland Bay
Dabob Bay
Mid-Hood Canal
South Hood Canal
NORTHERN SOUND
Urban, moderately deep embayment
CENTRAL SOUND
Urban, deep embayment
In greater Seattle area, but not highly
urbanized; deep, open basin
Urban, moderately shallow embayment
Urban, at mouth of commercial waterway on
deep, open embayment (Commencement Bay)
SOUTHERN SOUND
Rural, deep embayment, lacks major
freshwater source
Rural, in mid-southern basin near a sill
Urban, shallow sluggishly circulating
embayment
Rural, shallow sluggishly circulating
embayment
Urban, very shallow, sluggishly
circulating embayment
HOOD CANAL
Rural, deep, sluggishly circulating
embayment
Rural, deep, sluggishly circulating,
in narrow basin
Rural, deep, very sluggishly circulating,
near head of narrow basin
3-10
-------
Data sets were chosen to provide the longest possible period of
coverage at study sites throughout the sound. Reports and data summaries
from governmental agencies, academic institutions, and the private sector
were examined to determine the locations and time periods surveyed and the
variables measured. Because the University of Washington Department of
Oceanography has the oldest large source of data, other data sources usually
were selected to increase the amount of data available for locations
initially surveyed by the University of Washington.
Analytical Techniqups
Analytical techniques were identified to the extent possible from
reports and interviews with the scientists who worked on the original
projects. Generally, the techniques applied in the major studies of water
quality were widely accepted at the time they were used. However, early
measurements for some of the variables suffered from relatively poor accuracy
and precision. The techniques used in the studies included in the water
quality characterization project are summarized and evaluated in Appendix A.
More detailed discussions of these methods can be found in Barnes (1959),
Strickland and Parsons (1972), Riley (1975), and Preisendorfer (1986).
Generally, the historical data for salinity, water temperature, and
Secchi depth determinations are highly reliable. The difference between
Secchi depths determined with a standard 30-cm diameter Secchi disk and the
20-cm diameter Secchi disk used by Ecology (Singleton, L., 22 September
1987, personal communication) is probably only about 1 percent (Preisendorfer
1986). That difference is probably not large enough to be detected in this
characterization study. Dissolved oxygen measurements using the Winkler
titration method are reliable. Dissolved oxygen measurements made by
electronic measuring devices (e.g., oxygen probe) also are reliable when
proper calibration and equilibration procedures have been followed. Early
measurements of concentrations of nutrients, chlorophyll a, and fecal
coliform bacteria may have been less reliable because earlier techniques
were less accurate and precise. A modest amount of phosphate data from the
1930s and 1950s was retained because the quality of those data was judged to
3-11
-------
be acceptable. The oldest data judged to be acceptable for concentrations
of nitrate, chlorophyll a, and fecal coliform bacteria were collected in the
mid-1960s.
Studies of water quality in Puget Sound have been independently con-
ducted by many groups during a 55-yr period. During the assessment of the
analytical techniques used in the historical studies, it was assumed (unless
other information was available) that sampling and analysis were performed
correctly by trained, professional personnel. Where serious errors in the
performance of accepted techniques were detected, data were excluded from
analysis. Unfortunately, changes in techniques used in the historical water
quality studies were often not well documented. In some cases, changes in
analytical techniques were not adopted in the interest of obtaining higher
quality data, but to increase sampling efficiency (e.g., use of electronic
oxygen probes). Generally, minor changes in techniques resulting from
turnover in laboratory and field personnel and changes in equipment could
not be detected reliably in the historical data sets. Therefore, the
effects of these factors could not be assessed or corrected.
DATA SETS USED
Database Quality Assurance-Review
All data files used in this study were subjected to a quality assurance
review. Sources of information about the data included existing documenta-
tion (as provided by the relevant agencies) and interviews with investigators
and database managers.
Initially, the contents of the computer files from the data sources
were simply examined. The contents of the computer files were then checked
to verify that the time period in the computer files agreed with the time
period described in the documentation. If major discrepancies were dis-
covered, corrective actions were taken (e.g., rereading data cards into
computer files, adding new data into existing computer files).
3-12
-------
Units of measurement for the variables of each data set were checked.
Data values were plotted against date for each variable of interest. The
ranges of the data values in the plots were compared to the ranges that
would have been expected, based on the units cited in the available documen-
tation. Discrepancies were noted and corrected in the computer files. For
example, Ecology sampled at depths of 0, 10, and 30 m, but their computer
file (STORE!) reported depths of 0, 32, and 98 (no units given). Evidently,
the depth data had been converted from meters to feet. The depths were
reconverted to meters.
Some raw data were entered into computer files for this project. These
data were double-punched (i.e., entered twice and compared). Printouts of
the newly created files were manually compared to tables of raw data for
selected high-priority stations.
The final step of the quality assurance review was to check for
outliers. Because this project includes a diverse collection of study areas
sampled in all four seasons at various depths, simple range checks were not
appropriate. (For example, a water temperature of 20° C at the surface of
Budd Inlet recorded in July would not be unusual, but a 20° C temperature at
200-m depth at Point Jefferson in February would be highly unusual.) The
data from each source were separated by season and depth, and each variable
of interest was plotted by date. Points that appeared to have extreme
values (based on visual scans of these stratified plots) were reviewed by
contacting the investigators who were involved with the particular monitoring
program. Using this approach, a few points were deemed unreasonable or were
found to be based on erroneous laboratory procedures. These points were
dropped from the characterization database.
Data sets collected by the University of Washington Department of
Oceanography, Ecology, Washington Department of Fisheries, and Metro were
used for the characterization study. These data sets are described below.
Results of the quality assurance reviews for these data sets are summarized
in Appendix B. All other data sets were either too restricted in temporal
coverage, or did not include suitable study sites. Other potentially useful
ancillary water quality data sets are described in Appendix C. These data
3-13
-------
sets were evaluated, but not used in this study. The sources of climatic
data are described in this chapter following the descriptions of the water
quality data sets.
University of Washington
The University of Washington Department of Oceanography monitored water
quality throughout Puget Sound from 1932 to 1971. Sampling was interrupted
by World War II. Data coverage is summarized in Collias (1970). [Some of
the data referred to in Collias (1970) actually were collected by the
Washington Department of Fisheries or by various Canadian agencies. These
data are not in the original STORET database.] A portion of the physical
and chemical data collected by the University of Washington is presented
graphically in Collias et al. (1974). Overall, approximately 300 stations
were occupied at least once. Several stations were sampled throughout most
of the monitoring program (e.g., Pillar Point, Point Jefferson, Devil's
Head), but most stations were sampled sporadically or for only a few years.
Water temperature, salinity, and dissolved oxygen were measured in most
surveys; phosphate, nitrate, and Secchi disk depth were measured occasion-
ally. In most cases, a wide range of depths (i.e., surface to bottom) was
sampled.
Washington Department of Ecology
Ecology has routinely monitored water quality in Puget Sound since
1967. Winter months were not monitoring during most years, and there was a
gap in monitoring during 1971 and 1972. Results of these studies have not
been published. Historically, 167 stations were sampled; long-term data are
available through 1986 for 52 stations located throughout the sound. Water
temperature, salinity, and dissolved oxygen were measured throughout the
study and sulfite waste liquor was measured from 1967 to 1984. Monitoring
of several nutrients and measurements of bacterial contamination was added
in 1973. Measurements of Secchi disk depth and chlorophyll a were added in
1977 and 1979, respectively. From 1967 to 1970, samples were taken at the
surface and at a depth of 6 m. After 1973, samples were taken at depths of
0, 10, and 30 m. All data collected through 1986 are available from STORET.
3-14
-------
Washington Department of Fisheries
Two water quality data sets were obtained for this project from the
Washington Department of Fisheries. Both surveys were conducted under the
direction of the University of Washington Department of Oceanography
(Collias 1970). One data set spans the late 1950s and contains data from 20
stations in southern Puget Sound and 27 stations in northern Puget Sound.
Information from this data set for Bellingham Bay, Totten Inlet, and Oakland
Bay were incorporated into the characterization database. Water temperature,
salinity, dissolved oxygen, and sulfite waste liquor typically were measured
at the surface and at a depth of 6 m. Some deeper samples and some phosphate
data also were collected in the northern sites. Prior to the characteri-
zation project, these data existed only in unpublished reports by the
Washington Department of Fisheries (Westley 1957a,b, 1958; Westley and Tarr
1959, 1960). (Copies of these reports were provided by M.A. Tarr.) The
data were entered into computer database for this project.
The second set of water quality data from the Washington Department of
Fisheries includes several southern embayments that were sampled from 1964
to 1971 (Case, Eld, and Totten Inlets; Oakland and Quilcene Bays; and Burley
Lagoon). In most-cases, sampling was performed at three or four stations
near the heads of the embayments. Variables included water temperature,
salinity, dissolved oxygen, phosphate, nitrate, chlorophyll a, Secc.hi disk
depth, and several other nutrients and physical variables. Samples were
collected at depths of 1 and 3 m. The data were not summarized prior to the
characterization project. M.A. Tarr organized the data into tables [adjust-
ing chlorophyll a calculations to conform to Strickland and Parsons (1972)].
The data were entered into the computer database for this project.
Metro
The Water Quality Division of Metro has been monitoring water quality
in the central basin of Puget Sound since 1965. Long-term, routine moni-
toring data are available for approximately 70 stations. Periodically,
Metro has published summaries of their data (e.g., Metro 1986). Variables
3-15
-------
surveyed include water temperature, salinity, dissolved oxygen, Secchi
depth, and fecal coliform bacteria. Depth profiles covered in Metro's
monitoring program include the entire water column at many stations. Some
data on nitrogen, phosphorus, and phytoplankton standing stock were also
recorded. Methods for measuring nitrogen and phosphorus differed from the
methods used in the other major water quality studies of Puget Sound.
Phosphorus was measured as hydrolyzable phosphorus, and nitrogen was
measured as nitrate plus nitrite. These two variables were only measured
from 1967 to 1972 (Dalseg, R., 17 September 1987, personal communication).
Because this limited temporal coverage was deemed to be insufficient to
warrant analysis in the characterization study, these data were dropped.
Metro provided their data to this project on a magnetic tape. Much of their
data are also in STORET.
Climatic Data
A climatic database containing data on both weather conditions and
runoff was developed for the characterization project. The data were
obtained from U.S. government reports. Weather data were from Local
Climatological Data Reports produced and distributed by NOAA's National
Climatic Data Center in Asheville, NC. Runoff data were obtained from NOAA
(1984a).
Weather data are recorded at Seattle-Tacoma International Airport. A
continuous record was available from 1945 to 1985. Because climatic data in
the Local Climatological Data Reports indicated substantial variability in
the weather among locations around the sound, the data from the airport
serve only as a general index of the climate in the central Puget Sound
area. The following weather variables were included in the climatic
database as monthly means: air temperature, wind direction, wind velocity,
and the percentage of possible sunshine. Monthly totals for rainfall and
runoff were also included in the climatic database.
The runoff data were estimates of monthly total runoff to Puget Sound
from 1930 to 1978. These estimates were based on data from seven United
States Geological Survey (USGS) gaging stations located in large rivers
3-16
-------
draining into the sound. Station locations (all in Washington State) were
Newhalen and Concrete (Skagit River with 60 percent of the volume assumed to
exit through Deception Pass), Arlington (Stillaquamish River), Gold Bar
(Skykomish River), Carnation (Snoqualmie River), Puyallup (Puyallup River),
and Union (Skokomish River). These data serve only as a general index of
variation in runoff. The patterns at particular locations may have differed
from the patterns analyzed in this report.
3-17
-------
CHAPTER 4. DATA ANALYSIS PROCEDURES
A series of procedures was used to prepare the database for analysis.
Standard analytical protocols were then implemented to characterize the
environment and analyze temporal trends in the water quality of the 13 study
areas.
DATABASE PREPARATION
Five major procedures were used to prepare the database for analysis:
t Identification and correction of data values below analytical
detection limits
• Evaluation of the comparability of the data from the different
data sources
t Selection of representative sampling stations at each study
area for pooling data
• Identification of the annual period during which algal blooms
occurred in each study area
• Standardization of sampling depths.
Some data records in the water quality data sets contained undetected
values for certain variables. This situation occurred when the concentration
of the substance being measured was too low to be detected by laboratory
procedures. When an undetected value was reported, the data field in the
particular record usually contained the actual detection limit, accompanied
by a STORET code in another data field'that indicated that the value was
below the given detection limit. However, because detection limits were not
handled in the same manner by the different agencies from which data were
4-1
-------
obtained, it was necessary to standardize the way in which undetected values
were listed in the characterization database (see below). Moreover, some
errors in data entry occurred for undetected values because these values
were sometimes entered as zeroes or "missing value." These situations were
identified by examining plots of the data by date, and by examining printouts
of the data in the forms of frequency tables and hard copies of the computer
files. Investigators that worked on the monitoring programs used as data
sources were also contacted to help resolve the problems with the detection
.limits.
Three variables (i.e., nitrate, phosphate, and fecal coliform bacteria)
had data values below detection limits. For the two nutrients, the problem
was most prevalent for nitrate in surface samples, where concentrations were
often very low during algal blooms. For each of the above three variables,
data values below the detection limits were standardized by converting to the
highest detection limit used in any of the data sources (Table 4.1). Using
the highest detection limit avoided the possibility of introducing artificial
spatial or temporal trends into the data that were actually caused by
changes in the detection limits.
Data Comparability Among the Different Data Sources
A comparability analysis was conducted to determine whether the data
produced by the different monitoring programs could be used together in the
same analysis without correction or calibration. Unfortunately, the
monitoring programs used as data sources did not conduct side-by-side
sampling and laboratory analyses. Sources of variation other than differ-
ences in laboratory and field techniques (among the studies) may have
affected the data being compared (e.g., date, time of day, and stage of
tide for sampling).
To the extent possible, the four data sources used in the characteri-
zation study were compared in a pair-wise fashion. Each of the data sets
was scanned to identify stations where sampling overlapped (i.e., where
sampling was done in the same season and at or near the same location). It
was not possible to control for differences in sampling date or time of day.
4-2
-------
TABLE 4.1. LABORATORY DETECTION LIMITS USED IN
THE CHARACTERIZATION DATABASE
Detection Limit
0.714 ug-at/L
Phosphate 0.323 ug-at/L
Fecal coliform bacteria 1 organism/100 mL
4-3
-------
Moreover, only data from surface samples were included, because of the lack
of data from deeper water.
For the pairs of studies where sampling overlapped at a location,
values of the variables included in both studies were compared. Because
variances were highly variable (i.e., variances differed by more than a
factor of 10), nonparametric statistics were used for these comparisons.
The data in the two data sets being compared were stratified by season. The
data for each variable were then ranked within the seasons, and the ranks
were compared using a one-way analysis of variance (ANOVA). This procedure
is equivalent to a Kruskal-Wallis test (SAS 1985, p. 608). Statistical
significance of differences between pairs of data sources was determined
using Tukey's multiple comparison test.
With the exception of the University of Washington and Washington
Department of Fisheries, one pair of stations with overlapping sampling was
found for all the pairs of agencies from which data were obtained. However,
because much of the monitoring performed by Washington Department of
Fisheries was actually conducted for the University of Washington, the
absence of a comparison between the University of Washington data and
Washington Department of Fisheries data may not be critical. The only
location where sampling by University of Washington and Ecology overlapped
was near Alki Point (Stations PSB318 and PSB002, respectively), which is not
a location included in this trend analysis. Those stations were sampled
from 1967 to 1970. University of Washington and Metro both sampled at Point
Jefferson from 1966 to 1972 (Stations PSB305 and KSPB01, respectively), and
Washington Departments of Fisheries and Ecology both sampled in Oakland Bay
from 1967 to 1970 (Stations 23 and OAK004, respectively).
The only variables for which data were collected at pairs of stations
with overlapping sampling were dissolved oxygen, salinity, and water
temperature. Variables for which data were not collected at both stations,
and for which comparisons were not possible, included Secchi disk depths and
concentrations of nutrients, chlorophyll a, fecal coliform bacteria, and
sulfite waste 1iquor.
4-4
-------
Results of the data source comparisons are summarized in Appendix D. No
significant differences (P<0.05) were found between measurements of dissolved
oxygen, salinity, and water temperature made by Washington Department of
Fisheries and Ecology during the overlapping sampling period in Oakland Bay.
However, Ecology sample sizes were small, with only 10 observations for the
period of overlap.
For the overlapping University of Washington and Metro data from Point
Jefferson, salinity observations made by the University of Washington were
significantly (P<0.05) higher (approximately 1 ppt) than those made by Metro
during spring, summer, and autumn. However, they were not significantly
different from those made by Metro during the winter. Water temperature
observations made by the University of Washington were significantly higher
(P<0.05) during the spring, but did not differ significantly (P>0.05) from
those made by Metro during other seasons. No significant differences
(P>0.05) were detected among dissolved oxygen concentrations.
Although sample sizes for the University of Washington and Ecology were
small for the overlapping sampling period at Alki Point, some differences
were detected. Dissolved oxygen measurements made by Ecology during the
spring were significantly higher (P<0.05) than those made by the University
of Washington. However, dissolved oxygen measurements made by the University
of Washington during the autumn were significantly higher (P<0.05) than those
made by Ecology. Salinity measurements made by the University of Washington
were significantly higher (P<0.05) than those made by Ecology during the
spring and summer, with an average difference of approximately 2 ppt.
Salinity measurements made by the two agencies during the autumn and winter
did not differ significantly (P>0.05). No significant differences (P>0.05)
were detected for water temperature.
In summary, most of the paired comparisons did not detect consistent
differences among the data sources, although limitations of the data
prevented a thorough analysis of data comparability. Dissolved oxygen,
salinity, and water temperature data from the different data sources were,
therefore, used together in the characterization study with a minimum of
caveats. The most consistent difference detected between agencies was the
4-5
-------
salinity determinations made by Metro and the University of Washington.
Values from Metro usually were slightly lower. No conclusions can be drawn
about the concentrations of nutrients, chlorophyll a, sulfite waste liquor,
or fecal coliform bacteria, or about Secchi disk depths due to the lack of
paired data for statistical comparisons.
Selection of Representative Stations in Each Study Area for Pooling Data
Data were pooled (when possible) from several adjacent stations to
characterize conditions in a study area. Pooling data increases the number
of observations available for analysis and provides better coverage of short-
term variations in the measured variables (e.g., salinity changes caused by
the tides) because sampling is spread over a longer period of time. In
addition, the data are more representative of the general area because they
come from more than one location. Combining stations also extends the
period of time over which sampling occurred because different stations often
were sampled during different years.
Combining water quality data from adjacent stations within a study area
requires that conditions at the stations be as similar as possible. For
each study area, data were compared statistically for periods of overlap in
sampling. All data available for salinity; water temperature; concentrations
of dissolved oxygen, nitrate, phosphate, chlorophyll a, sulfite waste
liquor, and fecal coliform bacteria; and Secchi disk depth were analyzed.
Data from the candidate stations within a given study area were ranked
during overlapping years for each variable. Ranking was conducted for the
calendar summer data only because most of the rest of the year would not be
expected to have algal blooms. Station ranks were then compared using a
nonparametric one-way ANOVA. Statistical differences among stations were
evaluated using Tukey's multiple comparison test (SAS 1985). Stations that
differed significantly (P<0.05) from the other stations within a study area
for any of the water quality variables were dropped. These analyses were
repeated with the reduced station list until no significant differences
(P<0.05) remained among stations within each study area.
4-6
-------
Identification of the Annual Period for Algal Blooms in Each Study Area
Because water quality problems caused by nutrient enrichment are often
associated with algal blooms, data collected during the annual period of
maximal growth of phytoplankton were analyzed to identify temporal changes.
An obvious method to identify the period of maximal growth would be to
examine the annual profile of chlorophyll a concentrations. Unfortunately,
there were insufficient chlorophyll a data from most of the stations in the
characterization study to use this approach. Therefore, the annual period
of high percent dissolved oxygen saturation at the surface was chosen as a
surrogate for algal standing stock because photosynthesis increases the
percent dissolved oxygen saturation at the surface during algal blooms
(Winter et al. 1975; Collias and Lincoln 1977).
Because the algal bloom season may occur at different times and for
varying lengths of time in different areas, the annual period of occurrence
of algal blooms was identified separately for each study area. Monthly
means for percent dissolved oxygen saturation at the surface were calculated
for each study area. The three consecutive months with the highest- means
were chosen as the algal bloom season for many of the study areas. For
other areas, the bloom (as indicated by elevated surface dissolved oxygen
saturation values) appeared to continue over 4 mo. When a fourth month just
before or just after the highest three consecutive months had a higher mean
surface percent dissolved oxygen saturation than the one of the highest
three consecutive months, it was considered part of the algal bloom season.
Standardization of Sampling Depths
Because nearly all data collected since the mid-1960s and used in the
characterization study came from Ecology's routine monitoring database,
sampling depths from the other data sources were adapted to conform to the
design of Ecology's program. The Ecology sampling protocol consisted of
collecting samples from the surface and from depths of 10 and/or 30 m.
(From 1967 to 1970, Ecology collected samples from the surface and from
6.1-m depth.) Therefore, the standard depths analyzed in this study were
limited to the surface and depths of 10 and 30 m. In this study, surface
4-7
-------
samples were defined as samples taken at depths between 0 and 2 m. Simi-
larly, samples taken from between 9- and 11-m depths were defined as 10-m
samples, and samples taken from between 29- and 31-m depths were defined as
30-m samples. Sampling programs other than Ecology's often used different
sampling depths. In these situations, data were adjusted to the standard
depths by linear interpolation when the available sampling depths bracketed
the standard depths. When the sampling depths did not bracket the standard
depths, the data were not used.
Because more comprehensive data are available for the Point Jefferson
study area, several additional depths were investigated at this site. The
standard depths were analyzed as described above, and additional analyses
were conducted for selected depths down to 200 m.
STANDARD ANALYTICAL PROTOCOL
A series of analytical procedures was conducted for each study area to
characterize the environment and to detect temporal trends in water quality.
The time period investigated was the algal bloom period for each individual
study area. The analyses were conducted for depths of 0, 10, and 30 m. In
addition, possible exceedances of water quality standards were assessed for
surface waters. Details are given below.
Characterization of the Environment in the Study Areas
Graphical Analysis--
The environment in each study area was characterized by examining
histograms depicting the mean values for each water quality variable at 0-,
10-, and 30-m depths. Back-up tables containing standard errors and coef-
ficients of variation for these variables were also examined. To facilitate
comparisons among study areas, four sets of histograms were produced, each
of which contained the data from all the study areas within one region of
the sound as defined in the characterization study (i.e., the northern,
central, and southern sound, and Hood Canal).
4-8
-------
Raw data values were used in the histograms depicting the means of all
variables except fecal coliform bacteria and sulfite waste liquor. Because
the frequency distributions of the data for these two variables were
positively skewed, the data for these two variables were transformed. Values
for fecal coliform bacteria were log transformed (Greenberg et al. 1985).
Values for sulfite waste liquor were transformed to log (X+l) because the
database contained values of zero (Steel and Torrie 1960). Thus, the data
shown for fecal coliform bacteria and sulfite waste liquor are logs of
geometric means.
Statistical Analysis--
Possible cause-and-effect relationships were investigated using cor-
relation analysis to support interpretations of the histograms depicting the
environmental data at each site (described above). Product-moment cor-
relation coefficients (Zar 1974) were calculated for each study depth for
all possible pairs of the following water quality variables: salinity,
water temperature, dissolved oxygen concentration, percent dissolved oxygen
saturation (at the surface), Secchi disk depth, nitrate concentration, and
phosphate concentration. The minimum data requirement for conducting the
correlation analysis on a pair of variables was 10 data points.
Because the data for each variable in each study area were used in
several correlation analyses, a conservative approach was adopted for the
assessment of statistical significance. The chosen significance level for
each correlation coefficient was scaled using the Bonferroni inequality, a
simple but highly conservative method that preserves the experiment-wise
error rate when the same data are used to calculate several correlation
coefficients (Snedecor and Cochran 1980). Using the Bonferroni inequality,
each correlation coefficient was interpreted at a significance level of 0.05
and 0.01 divided by the number of correlations investigated using the same
data. For example, if dissolved oxygen data were correlated with five other
variables in the correlation matrix for a particular station and depth,
statistical significance for each of the five individual correlation
coefficients involving the oxygen data would be determined at P<0.05/5=0.0T
and P<0.01/5=0.002.
4-9
-------
Analysis of Temporal Trends in Water Quality in Each Study Area
Graphical Analysis--
Temporal trends were investigated within each study area by plotting
water quality data for the algal bloom period by year. These plots were
examined visually to identify possible periods of changes in water quality
and to assess exceedance of water quality standards. When two or more
observations during the algal bloom period were available for a year, the
individual observations were plotted along with the mean and the standard
error. When only one observation was available, that value was simply
plotted as a point. Data for fecal coliform bacteria and sulfite waste
liquor were plotted as the log of the geometric means, with the standard
errors calculated on log-transformed data (Greenberg et al. 1985). Regres-
sion lines were included on these plots when significant temporal trends
were detected by regression of a variable against year (see below).
Historical Causes of Changes in Water Quality—When a temporal change
in the values of a variable plotted against year was evident from visual
inspection, an attempt was made to determine whether an historical event
could explain the apparent change. For example, if the amount of sulfite
waste liquor dropped in an area, inquiries were made to determine when
changes in pulp mill discharges might have occurred.
Exceedance of Water Quality Standards — Plots of water quality variables
through time were examined visually to detect possible exceedances of water
quality criteria for surface waters. Dissolved oxygen concentration and the
concentration of fecal coliform bacteria are the two variables in the
characterization study for which water quality standards have been estab-
lished for Puget Sound by Washington State. The water quality standards
applicable to the study areas are given in Table 4.2. These standards are
included in the descriptions of each study area in Chapter 5.
4-10
-------
TABLE 4.2. WATER QUALITY STANDARDS APPLICABLE
TO THE CHARACTERIZATION STUDY AREAS
Classification
of Water3
Dissolved
Oxygen
Standard
(mg/L)
Fecal Coliform
Bacteria Standard
(organisms/100 mL)
Characterization
Study Area
AA
7.0
6.0
5.0
100b'd
Bellingham Bay
Point Jefferson
Nisqually Reach
Carr Inlet
Dabob Bay
Mid-Hood Canal
South Hood Canal
Sinclair Inlet
Budd Inlet (northern
portion of study
area)
Totten Inlet
Port Gardner
(University of
Washington stations)
Budd Inlet (southern
portion of study
area)
Oakland Bay
City Waterway
Port Gardner
(Ecology stations)
a AA = extraordinary; A = excellent; B = good.
b Geometric mean.
c No more than 10 percent of samples can exceed 43 organisms/lOOmL.
d No more than 10 percent of samples can exceed 200 organisms/lOOmL.
Reference: WAC 173-201-045(2) and WAC 173-201-085(2).
4-11
-------
Statistical Analysis--
Nonparametric ANOVA and linear regression were used to detect temporal
trends in water quality and climate.
Nonparametric ANOVA--A nonparametric ANOVA was conducted on the water
quality data from each of the study areas. The purpose of this analysis was
to determine whether significant differences in the values of each variable
existed through time. Data collected prior to 1973 were compared with data
collected from 1973 through 1986. The choice of 1973 as the year dividing
the earlier data from the more recent data was made primarily because Ecology
updated their monitoring program in 1973.
The procedure used to calculate the ANOVA comparing early and recent
data was to compute an analogue of the Kruskal-Wallis test (SAS 1985,
p. 608). The entire data set for each variable was ranked within each study
area. The set of ranked values was then divided into pre-1973 and 1973-1986
subsets. A one-way ANOVA was conducted comparing ranks in the two periods
of time.
Linear Regression—Temporal trends were also analyzed by linear
regressions of the values of each variable against year of collection. The
minimum data requirement for conducting the regression analysis was 5 yr of
data, including data through 1986. The reason for conducting the regressions
was to provide a measurement of the rate of change of each variable. This
rate of change is estimated by the slope of the line. Multiplication of the
rate of change of a variable by the number of years that data have been
collected gave an estimate of the amount the variable has changed over the
time period that it has been measured. Statistically significant regressions
(P<0.05) were plotted on the graphs of data values plotted against year (as
described above).
Two regressions were conducted for each variable within each study
area. One regression was used to detect changes in values over the whole
time period sampled between 1932 and 1986. In that analysis, the actual
period analyzed was dependent on the number of years sampled in a given
4-12
-------
study area. The second regression was used to detect significant changes in
values over the more recent time period of 1973-1986. Again, the actual
period analyzed was dependent on the number of years sampled in a given
study area. A temporal trend was considered significant if the slope of the
regression was statistically significant (P<0.05). A significant positive
slope indicates increasing values of the variable, while a significant
negative slope indicates decreasing values of the variable. A nonsig-
nificant slope indicated that no overall trend was detected by regression.
Raw data values were used in the regressions of all variables except
fecal coliform bacteria and sulfite waste liquor. The frequency distribu-
tions of the data for these two variables were positively skewed. Values
for fecal coliform bacteria were log transformed (Greenberg et al. 1985).
Values for sulfite waste liquor were transformed to log(X+l) because the
database contained values of zero (Steel and Torrie 1960).
Because recent data are a subset of the long-term data, a recent change
in the data could introduce an apparent long-term change that would be
detected statistically by the long-term regression, even though the change
actually occurred recently. For example, if dissolved oxygen concentration
in an area averaged 8 mg/L from 1932 to 1980, and then averaged 4 mg/L from
1981 to 1986, a declining trend for the entire period of 1932 to 1986 might
be detected statistically, even though the actual change would have occurred
only in the 1981 to 1986 portion of the long-term data.
4-13
-------
CHAPTER 5. RESULTS AND DISCUSSION
ORGANIZATION OF THE CHAPTER
A brief discussion of long-term weather patterns in Puget Sound precedes
the discussion of results of the graphical, regression, and correlation
analyses performed for this characterization study. The results are
presented according to geographic regions of the sound: northern, central,
and southern sound, and Hood Canal. Except for the northern sound, these
regions correspond to the basins of Puget Sound as described in Chapter 2.
Because neither long-term nor recent data were available for the Strait of
Juan de Fuca, no results are presented for this area. Following the
discussion of results within each geographic region, a summary of the major
findings is provided.
As discussed in Chapter 3, the results for a particular study area are
representative only of the immediate area in which the sampling stations were
located. For example, trends in water quality detected in the Bellingham
Bay study area may or may not be indicative of trends in water quality in
all of Bellingham Bay.
Statistical statements in the text are based on the following conven-
tions. Regressions of data against year were considered significant if the
slope of the regression line was statistically significant at P<0.05. The
nonparametric ANOVA also was considered statistically significant at
p<0.05. Correlation coefficients were considered significant if PO.05
after sealing with the Bonferroni inequality.
WEATHER DURING STUDY PERIOD
Plots of air temperature, percentage of possible sunshine, rainfall,
runoff, and wind velocity data by year for the Puget Sound area are given in
5-1
-------
Figures 5.1-5.3. These data provide general information on basic weather
patterns in the central Puget Sound area during the study period.
Several long-term trends are evident. Between 1945 and 1985, a
significant increase in mean air temperature was detected (slope=+0.3° C
per year). The temperature increase may be attributable to a cool period
that occurred between 1948 and 1955. Total annual rainfall declined
significantly between 1945 and 1985 (slope=-0.45 cm/yr). The decrease in
rainfall may be attributable to a wet period that occurred during the 1950s
and a dry period that occurred from 1976 through the 1980s. Some years had
unusual weather. For example, 1955 was cool, while 1958 was warm.
Similarly, 1950 was wet, while 1952, 1976, and 1985 were dry, and 1978 was
cloudy, while 1982 was sunny.
NORTHERN SOUND
The northern sound is defined in this study as the region encompassing
the eastern end of the Strait of Juan de Fuca, the southern end of the
Strait of Georgia, and the area around the San Juan Islands (see Figure 2.1).
The northern sound is the only study region north of Admiralty Inlet, and is
the region most subject to oceanic influences. The northern sound is
typically over 100 m deep in the Straits of Juan de Fuca and Georgia.
Extensive tidelands and sheltered embayments are located along the mainland
shore. Water movements are complicated by an abundance of islands.
Approximately 60 percent of the flow of the Skagit River, the largest river
in the Puget Sound basin (see Table 2.1), discharges into the eastern end of
the Strait of Juan de Fuca. The remaining volume flows into the Main Basin
of Puget Sound through Skagit Bay and Possession Sound (NOAA 1984a). Major
population centers are the Cities of Bellingham and Anacortes. The major
historical sources of pollutants in the northern sound have been saw mills,
pulp mills, and canneries near Anacortes and Bellingham, and a large oil
refinery near Anacortes (Chasan 1981).
Bellingham Bay is the only study area in the characterization project
that is located in the northern sound. Station locations are shown in
Figure 5.4. Data sources are given in Table 5.1. The algal bloom season
5-2
-------
13
o
12
2
UJ
8-1
1945 1950 1955 I960
1965 1970
YEAR
1975 1980 1985 1990
53
52'
51
50
49
48
47
46
45
44
43
421
41
40
39
381
37
1945 1950 1955 1960 1965 1970 1975 1980 1985 1990
YEAR
Figure 5.1. Annual means of air temperature and the percent of possible sunshine
at Seattle-Tacoma International Airport.
5-3
-------
150
130
^ 120
j-110
I
: 100
90
80
70
60
T 1 1 T—
1945 1950 1955 1960
—i 1 1—
1965 1970 1975
YEAR
1980 1985 1990
18
7 16
E 14
u
liB
u
o iw
1 8
4
1 1 1 1 1 1—
1945 195* 1955 196* 1965 197* 1975
YEAR
1 1 r
1989 1985 19M
Figure 5.2. Annual totals of rainfall at Seattle-Tacoma International Airport and runoff
to Puget Sound.
5-4
-------
O . c
Q 1 .5
(0
E
*H^>
1.4
O
O 1.3
UJ
>
^ 1.3
1.1
1.0
1945 1959 19S5 1966
1965 1979
YEAR
1975 1989 1985 1990
Figure 5.3. Annual mean wind velocity at Seattle-Tacoma International Airport.
5-5
-------
Bellingham Bay
Figure 5.4. Locations of the study area and sampling stations in the northern sound.
5-6
-------
TABLE 5.1. SAMPLING STATION NUMBERS, DATA SOURCES, AND SAMPLING
PERIODS FOR THE STUDY AREA IN THE NORTHERN SOUND
Study Area
Bellingham Bay
Station
Number
BLL755
BLL759
BLL008
Data
Source
uwa
UW
Ecology
Sampling Period
1956-63 (includes data
1956-63 (includes data
1967-70, 1973-86
from WDFb)
from WDF)
a UW = University of Washington.
WDF = Washington Department of Fisheries.
5-7
-------
is given in Table 5.2. Based on the percent dissolved oxygen saturation at
the surface, algal blooms were most prevalent in Bellingham Bay from May
through July.
Bellinqham Bay
The study area is located off Post Point (Figure 5.4). The northern
portion of Bellingham Bay is bordered by the City of Bellingham. Class A
water quality standards apply at the site. Water depth at the study area is
about 32 m. The Nooksack River discharges into the head of the bay, about
10 km northwest of the study area. The Nooksack is the fourth largest river
flowing into Puget Sound and discharges approximately 7 percent of the total
freshwater flow reaching the sound. Several creeks also flow into Bell ing-
ham Bay, including Padden Creek, which enters the bay approximately 1 km
north of the study area. On a flooding tide, low salinity water from the
Nooksack River is recirculated to the northeast, toward Post Point (City of
Bellingham 1984).
A series of improvements have been made to the waste treatment facili-
ties in the Bellingham Bay area. A primary sewage treatment plant began
operating near the study site at Post Point in 1974. It replaced the old
City of Bellingham plant that discharged into the Whatcom Waterway in the
inner harbor north of Post Point (City of Bellingham 1984). The Post Point
plant treats municipal wastes, including discharges from vegetable and
seafood processors from July through December. When the Post Point plant
became operational, at least two outfalls near the study area that discharged
raw sewage from a service population of over 6,000 in South Bellingham were
closed (Thomas, K., 27 October 1987, personal communication). From 1979
through the 1980s, several combined sewage overflows that drained into
Bellingham Bay were also closed. One remains open within the City of
Bellingham, well north of the study area. The Georgia-Pacific pulp mills in
Bellingham reduced the biological oxygen demand (BOD) in their effluents by
more than 1 order of magnitude during the 1970s. These mills upgraded to
secondary effluent treatment in 1979 '(NOAA 1985).
5-8
-------
TABLE 5.2. ALGAL BLOOM SEASONS IN THE NORTHERN SOUND STUDY AREA,
AS DEFINED BY MONTHLY MEAN AND STANDARD ERROR OF PERCENT
DISSOLVED OXYGEN SATURATION IN SURFACE WATER
Percent Dissolved Oxygen Saturation
Bellingham
Month Bay
April 98 +/- 2
May 114 +/- 2a
June 115 +/- 4a
July 121 +/- 3a
August 100 +/- 4
September 100 +/- 5
a Months included in the algal bloom season.
5-9
-------
Environmental Conditions in the Study Area--
Mean salinity and water temperature values during the algal bloom period
are depicted in Figure 5.5. Summary statistics are given in Appendix E.
Data are available from 1958 through 1986, with limited coverage from the
mid-1960s through the mid-1970s. Substantial vertical gradients in both
salinity and water temperature values indicated that density stratification
was well developed in the study area. The mean salinity value at the
surface was 4.2 ppt lower than the mean at 10-m depth, while the mean water
temperature at the surface was 3.4° C higher than the mean at 10-m depth.
The negative correlation between salinity values and water temperature
values at the surface (Appendix F) indicates that warm water tended to be
low in salinity and that cold water tended to be high in salinity. This
correlation probably reflects fluctuating freshwater inputs from the
Nooksack River. During periods of density stratification and relative
stability of the water column, solar heating of the surface would be most
effective in heating up the low salinity surface water.
The vertical distributions of dissolved oxygen and nutrients appear to
have been strongly influenced by density stratification of the water column
(Figures 5.6 and 5.7). The concentration of dissolved oxygen was approxi-
mately 10 percent higher at the surface than at 10-m depth. The concen-
tration of nitrate was only about 40 percent as high at the surface as at
10-m depth, while the concentration of phosphate was 60 percent as high the
surface as at 10-m depth. Although considerable, these depth gradients were
less developed than those in more sheltered embayments that lacked substan-
tial inputs of fresh water, such as Sinclair and Carr Inlets.
Correlations for surface waters (see Appendix F) suggest that when
dissolved oxygen concentrations were low, salinity values were also low and
water temperature values were high. Freshwater sources in the area probably
have lower concentrations of dissolved oxygen than does the seawater in the
area.
The moderate elevation of the percent dissolved oxygen saturation at
the surface (i.e., 115 percent) suggests that only moderate algal blooms
5-10
-------
BELLINGHAM
BAY
DEPTH (m)
STUDY AREA
e ie 30
BELLINGHAM
BAY
DEPTH (m)
STUDY AREA
Figure 5.5. Mean salinity and water temperature values in the northern sound study
area during the algal bloom season.
5-11
-------
BELLINGHAM
BAY
DEPTH (m)
STUDY AREA
e ie 3e DEPTH (m)
STUDY AREA
BELLINGHAM
BAY
Figure 5.6. Mean concentrations of dissolved oxygen and dissolved inorganic nitrate
in the northern sound study area during the algal bloom season.
5-12
-------
UJ
16 36 DEPTH (m)
BELLINGHAM
BAY
STUDY AREA
200
O ::
x 3
O
si
>H
-J<
ooc
COD
WH
5<
CO
100
0
BELLINGHAM
BAY
DEPTH (m)
STUDY AREA
Figure 5.7. Mean concentrations of dissolved orthophosphate and the mean percent
saturation of dissolved oxygen at the surface in the northern sound study
area during the algal bloom season.
5-13
-------
occurred in the study area (Figure 5.7). Secchi disk depth readings
(Figure 5.8) were relatively low (i.e., 3 m) compared with areas such as
Point Jefferson, presumably because of suspended particulate material
carried into Bellingham Bay by the Nooksack River. The limited photic zone
and relatively rapid flushing of upper Bellingham Bay, which averages about
4 days (City of Bellingham 1984), may limit the growth rates of
phytoplankton.
The geometric mean concentration of sulfite waste liquor was high
(22.1 Pearl Benson Index) in the study area, particularly at the surface
(Figure 5.8). Historically., two Georgia-Pacific pulp mills discharged
sulfite waste liquor into Whatcom Waterway in inner Bellingham Harbor.
Generally, the sulfite waste liquor remained in the top 6 m of the water
column (Federal Water Pollution Control Administration and Washington State
Pollution Control Commission 1967).
Concentrations of fecal coliform bacteria have remained low (geometric
mean <2.5 organisms/100 mL) in the study area (Figure 5.8). The Federal
Water Pollution Control Administration and the Washington State Pollution
Control Commission (1967) reported that fecal coliform concentrations were
markedly elevated in the vicinity of Whatcom Waterway before the City of
Bellingham sewage treatment plant was replaced by the Post Point Pollution
Control Plant.
Water Quality Trends in the Study Area--
A summary of comparisons between water quality data collected before and
after 1973 is given in Table 5.3. Slopes of statistically significant
long-term and recent regressions of the values of water quality variables
by year are given in Table 5.4.
Physical Conditions—Plots of salinity and water temperature data by
year are shown in Figures 5.9-5.11. Significant long-term declines (P<0.05)
in salinity values were detected at the surface and at 10-m depth. These
declines in salinity values were detected by both the ANOVA comparisons of
the data reported before and after 1973 (Table 5.3) and the long-term
5-14
-------
x? a
OQ.
OLU
UJQ
w e
BELLINGHAM
BAY
STUDY AREA
0 16 30 DEPTH (m)
STUDY AREA
BELLINGHAM
BAY
e 10 ae
BELLINGHAM
BAY
DEPTH (m)
STUDY AREA
Figure 5.8. Mean Secchi disk depth and log of geometric mean concentrations of sulfite
waste liquor and fecal coliform bacteria in the northern sound study area
during the algal bloom season.
5-15
-------
TABLE 5.3. NET CHANGE AND PERCENT CHANGE IN THE MEAN VALUES OF WATER QUALITY
VARIABLES IN THE NORTHERN SOUND, BASED ON ANOVA COMPARISONS OF DATA
TAKEN BEFORE 1973 WITH DATA TAKEN FROM 1973 TO 1986
Bel 1i ngham Bay
Depth Change
(m) Net Percent
Salinity (ppt)
0 -3.35 13.2
10 -1.11 3.8
30 naa
Water Temperature (° C)
0 -1.57 10.4
10 NSb
30 na
Dissolved Oxygen (mg/L)
0 NS
10 NS
30 na
Nitrate (ug-at/L)
0 na
10 na
30 na
Phosphate (ug-at/L)
0 NS
10 NS
30 na
Chlorophyll a (ug/L)
0 na
10 na
30 na
Surface Dissolved Oxygen Saturation (Percent)
0 NS
Secchi Disk Depth (m)
NS
Sulfite Waste Liquor (Pearl Benson Index)
0 -39.94
10 +7.21
30 na
Fecal Col i form
0
10
30
68.7
236.4
Bacteria (No.
na
na
na
7100 ml)
a na Results of the statistical test were not available because of a lack of
data.
b NS - The pre-1973 and 1973-1986 values were not significantly different at
P<0.05, based on a nonparametric one-way ANOVA.
5-16
-------
TABLE 5.4. SLOPES OF STATISTICALLY SIGNIFICANT LONG-TERM
AND RECENT REGRESSIONS OF WATER QUALITY VARIABLES AS
A FUNCTION OF YEAR FOR THE NORTHERN SOUND
Depth
(m)
0
10
30
S I opes
Bellingham Bay
Long-term Recent
Salinity (ppt)
-0.130
-0.060
na5
Water Temperature (°
0
10
30
-0.086
NS
na
NSa
NS
NS
C)
0.194
NS
0.148
Dissolved Oxygen (mg/L)
0
10
30
0
10
30
0
10
30
NS
NS
na
Nitrate (ug-at/L)
na
na
na
Phosphate (ug-at/L)
NS
NS
na
NS
NS
NS
NS
NS
NS
NS
0.064
NS
Surface Dissolved Oxygen Saturation (Percent)
0 NS NS
Secchi Disk Depth (m)
NS NS
Sulfite Waste Liquorc (Pearl Benson Index)
0
10
30
Fecal Col i form
0
10
30
-0.037
0.025
NS
Bacteria
na
na
na
-0.068
NS
NS
(No./100mL)
-0.061
na
na
a NS - Not significant at P<0.05.
** na = Results of the statistical test were not available because of a
lack of data.
c Data were subjected to a log(X+1) transformation for the regressions.
Data were subjected to a log transformation for the regressions.
5-17
-------
40
a
a
10
<5
r
0 0
T1 I I I I I 1 I
1950 19S5 I960 1965 1970 1975 1989 1985 1996
YEAR
40-
30-
a
a
t 20
3
1/1
10
0
ANNUAL MEAN
I STANDARD ERROR
0 INDIVIDUAL OBSERVATION
—- SIGNIFICANT REGRESSION LINE
(P < 0.05)
1950 1955 1966 1965 1976
YEAR
1975 1986 1985 1996
Figure 5.9. Salinity values at the surface and at 10-m depth in the Bellingham Bay study
area during the algal bloom season.
5-18
-------
30
.
a
z
_J
10
0
ANNUAL MEAN
I STANDARD ERROR
O INDIVIDUAL OBSERVATION
—- SIGNIFICANT REGRESSION LINE
(P < 0.05)
1958 1955 I960 1965
1970
YEAR
1975 1980 1985
1990
24
23
22
21
^19
§15
S'«
£ 13
g 181
11
10
'1
;
1950
1955 I960 1965
1970
YEAR
1 1- 1—
1975 1980 1985
1990
Figure 5.10. Salinity values at 30-m depth and water temperatures at the surface in the
Bellingham Bay study area during the algal bloom season.
5-19
-------
24
23-
22"
21
8 18
T)
2"
£ 13
S 12
10
9
8
ANNUAL MEAN
I STANDARD ERROR
0 INDIVIDUAL OBSERVATION
—- SIGNIFICANT REGRESSION LINE
(P < 0.05)
1950 1955 1960 1965 1970 1975 1980 1985
YEAR
1990
24-
23
22
21
| 15
< 14
£ 13^
2 12
*~ 11 ^
10-
9-
8
7-
1950 1955 1960 1965 1970 1975 1980 1985 1990
YEAR
Figure 5.11. Water temperatures at 10- and 30-m depths in the Bellingham Bay study area
during the algal bloom season.
5-20
-------
regressions of salinity values by year (Table 5.4). No explanation is
available for the apparent declines in salinity values. Based on the
declines in total annual rainfall values at the Seattle-Tacoma International
Airport (Figure 5.2). increases in salinity values, rather than decreases,
would be expected.
A significant long-term decline in surface water temperatures, detected
by both ANOVA (Table 5.3) and regression (Table 5.4), appears to have been
driven by the high values recorded in 1958. The highest annual mean and
temperature of Seattle-Tacoma International Airport for the period of 1945-
1985 was recorded in 1958 (Figure 5.1). Recent increasing trends (Table 5.4)
appear to have been associated with cool periods in 1974 through 1976 and
warm periods in 1985 and 1986 (Figure 5.1).
Dissolved Oxvaen--Plots of dissolved oxygen concentration by year are
shown in Figures 5.12 and 5.13. Violations of the Class A water quality
standard (see Table 4.2) were recorded only in 1960 at 10-m depth. No
significant changes in dissolved oxygen concentrations were detected.
Nutrients—Plots of nitrate concentrations by year are given in
Figures 5.13 and 5.14. Because data are only available since 1974,
comparisons between data from before 1973 and data from 1973 through 1986,
and long-term regressions of nitrate concentration by year, were not
possible. Recent regressions against year were not significant (Table 5.4).
Concentrations at the surface were low, often near the analytical detection
limit (0.7 ug-at/L) and highly variable.
Plots of phosphate concentration by year are given in Figures 5.15
and 5.16. One point is available from 1960, along with data from 1974
through 1986. A positive slope in the regression for data from 10-m depth
since 1974 was the only significant (P<0.05) trend detected (Table 5.4).
Although the underlying cause of the rise in phosphate concentrations is
unclear, the decline in sulfite waste liquor discharges from the
Georgia-Pacific pulp mills that occurred when the mills adopted secondary
treatment in 1979 (NOAA 1985) may have contributed to this trend. As noted
in Chapter 3, sulfite waste liquor causes the inorganic phosphate contained
5-21
-------
30-
19
18
16
i 15
J-13
on
g!0
Q 9
£ 8
7
6
0 5
4
3
a
i
1950 1955 1968 1965
1978
YEAR
1975 1988 1985
1998
30-
19
18
171
16
"oil 4
£13
oil
g!0
Q 9
2 8
3-71
i/> c
i/2 b
0 5
41
3
2
1
8
ANNUAL MEAN
] STANDARD ERROR
O INDIVIDUAL OBSERVATION
—- SIGNIFICANT REGRESSION LINE
(P < 0.05)
1958 1955 1968 1965 1978 1975 1988 1985 1998
YEAR
Figure 5.12. Concentrations of dissolved oxygen at the surface and at 10-m depth in the
Bellingham Bay study area during the algal bloom season.
5-22
-------
DA -
C.V
19
18
17
16
<15
oi!4
£13
2 12'
UJ
on
X" * ft '
o 10
o 9
2 8
o 7
v> e .
(fi b
o 5-
4'
3
a
i
0
ANNUAL MEA
I STANDARD E
0 INDIVIDUAL (
SIGNIFICANT
(P < 0.05)
*i «tAt
Am?
I V 4
t ¥ *
o
• i i i 1 1 i —
1950 1955 1960 1965 1970 1975 1980
YEAR
40
C?
\
'o
i
?30-
LJ
t-
u.
i—
z
y 20-
z cc
4
O
Ct
O
z
o
> 10-
O
V)
w
o
A '
0
" | 1 1 1 1 1 T
1950 1955 1960 1965 1970 1975 1980
YEAR
M
RROR
»SERVATION
REGRESSION LINE
$ ° t
AiA
•V0y0v
1
o
1 r
1985 1990
o o
ji
rioo
1985 1990
Figure 5.13. Concentrations of dissolved oxygen at 30-m depth and dissolved inorganic
nitrate at the surface in the Bellingham Bay study area during the algal
bloom season.
5-23
-------
40-
1
'o
1
?30
N_S
LJ
<
|20
o
o
z
o
u
o
V)
5
0
ANNUAL MEAN
J STANDARD ERROR
0 INDIVIDUAL OBSERVATION
— - SIGNIFICANT REGRESSION LINE
(P < 0.05)
o
0
t]
o '
<
\
O (
/
i
o »
S
<
ki/
Ur
IV,
> H
O
O
< t
A
/'
°0(.
1950 1955 1960 1965 1970 1975 1980 1985 1990
YEAR
40"
z
o
cc
O
Z
o
CO
to
o o
o o
1950
1955 1960
1965 1970
YEAR
1975 1980 1985 1990
Figure 5.14. Concentrations of dissolved inorganic nitrate at 10- and 30-m depths in
the Bellingham Bay study area during the algal bloom season.
5-24
-------
o>
3
•v**
LJ
I-
f 3
(ft
o
a.
o
i ,
a
LJ
Vt
5
ANNUAL MEAN
I STANDARD ERROR
O INDIVIDUAL OBSERVATION
—- SIGNIFICANT REGRESSION LINE
(P
•3
o
o
UI
O 1
tn 1
i/>
5
1958 1955 I960 1965
1979
YEAR
1975 19M 1985 199«
Figure 5.15. Concentrations of dissolved orthophosphate at the surface and at 10-m
depth in the Bellingham Bay study area during the algal bloom season.
5-25
-------
01
tfl
o
Q.
O
o
a
LJ
O .
1
VI
5
ANNUAL MEAN
J STANDARD ERROR
O INDIVIDUAL OBSERVATION
SIGNIFICANT REGRESSION LINE
(P < 0.05)
O O O
o o
1958 1955 196e 1965 1976 1975 1980 1985
YEAR
i99e
300
•
a
o:
3
(/I
O
O 100
Q
U
O
I/)
e
1950 1955 1960 1965
1970
YEAR
1975 1980 1985 1999
Figure 5.16. Concentrations of dissolved orthophosphate at 30-m depth and percent
dissolved oxygen saturation at the surface in the Bellingham Bay study
area during the algal bloom season.
5-26
-------
in seawater to precipitate (Westley and Tarr 1978). The decline in sulfite
waste liquor emissions might have allowed dissolved inorganic phosphate
concentrations to recover over time. Alternatively, the increase in
phosphate concentrations may have been influenced by changes in other
anthropogenic sources or oceanic sources.
Indicators of Phytoolankton Growth—Data on chlorophyll a concentra-
tions are not available. Percent dissolved oxygen saturation at the
surface and Secchi disk depth are plotted against year in Figures 5.16
and 5.17. No significant changes were detected for either variable.
Pollutants — Plots of sulfite waste liquor concentrations by year are
shown in Figures 5.17 and 5.18. The most important change was a substantial
decline in sulfite waste liquor concentration at the surface (Tables 5.3
and 5.4). This decline appears to have coincided with the onset of secondary
treatment by the Georgia-Pacific pulp mill in 1979. A statistically
significant long-term increase in sulfite waste liquor was detected at
10-rm depth. No explanation is available for this increase.
Concentrations of fecal coliform bacteria in surface water are plotted
by year in Figure 5.19. Class A water quality standards were not violated
after 1978. A significant decline in the concentrations of fecal coliform
bacteria since 1974 was detected by regression (Table 5.4). High values
were reported in 1974 and 1978, but values reported in 1985 and 1986
represented "undetected" concentrations. The data from 1974 were obtained
before the Post Point plant became operational, and probably reflect
conditions that existed when raw sewage was still discharged near the study
area. Subsequent declines in coliform concentrations may have reflected
closures of combined sewer overflows that occurred in the early 1980s
(Thomas, K., 27 October 1987, personal communication).
Summary of Results for the Northern Sound
Because only one area was investigated, summaries of environmental
conditions and trends in water quality would simply repeat the foregoing
5-27
-------
16-
14
I12
X
LU
o
o
o
III
(A
ANNUAL MEAN
I STANDARD ERROR
O INDIVIDUAL OBSERVATION
---- SIGNIFICANT REGRESSION LINE
(P < 0.05)
1950 1955 I960 1965 1970 1975 1980 1985 1990
YEAR
CC
O
3
g_ 3
LU
t •
LU CO
si
u-t
o
o
1950 1955 1960 1965
1970
YEAR
1975 1989 1985 1990
Figure 5.17. Secchi disk depth and log of concentrations of sulfite waste liquor at the
surface in the Bellingham Bay study area during the algal bloom season.
5-28
-------
DC
O
3
g
-••x
in 0)
t «
LL CD
51
o
O
2-
1-
ANNUAL MEAN
STANDARD ERROR
INDIVIDUAL OBSERVATION
SIGNIFICANT REGRESSION LINE
(P < 0.05)
1950 1955 I960 1965
19?e
YEAR
1975 1989 1985
1990
cc
O
g
-"*
Ei
U_ CQ
_J _
O
O
1-
\
1959
1955 1969 1965
1970
YEAR
1975 1980 1985
1990
Figure 5.18. Log of concentrations of sulfite waste liquor at 10- and 30-m depths in
the Bellingham Bay study area during the algal bloom season.
5-29
-------
DC
LU
O
<
m
E j
oE
§1
O ,.
O »
< o
O2
O
O
1-
ANNUAL MEAN
I STANDARD ERROR
0 INDIVIDUAL OBSERVATION
SIGNIFICANT REGRESSION LINE
(P < 0.05)
o o*
1959 1955 I960 1965 1970 1975 1980 1985 1990
YEAR
Figure 5.19. Log of concentrations of fecal coliform bacteria at the surface in the
Bellingham Bay study area during the algal bloom season.
5-30
-------
information. Hence, summaries of these topics are not presented for
northern Puget Sound.
Sensitivity to Nutrient Enrichment--
The capacity of Bellingham Bay to export or assimilate pollutants
without deleterious effects is probably higher than the capacities of the
more sheltered embayments of Puget Sound that lack substantial freshwater
inputs (e.g., Sinclair and Carr Inlets). The range of estimated flushing
times for the bay, 1-10 days (City of Bellingham 1984), and the moderate
depths of the bay (considerable portions of the bay are deeper than 100 m),
suggest that nutrients would be removed or diluted more effectively than in
the more sheltered areas mentioned above. However, the flushing rate and
dilution capability of Bellingham Bay are lower than those found in open,
mid-channel areas, such as Point Jefferson and Nisqually Reach.
CENTRAL SOUND
The central sound is defined herein as the area encompassing the Main
Basin of Puget Sound from Admiralty Inlet to Tacoma Narrows, including the
embayments west of Bainbridge Island (see Figure 2.1). Virtually all oceanic
waters enter the central sound over the sill at Admiralty Inlet. Much of
the central sound is relatively deep and well flushed, except for some
urban waterways and the embayments west of Bainbridge Island. Substantial
inputs of fresh water enter the central sound from the Skagit River (via
Possession Sound) and from the Snohomish, Duwamish, and Puyallup Rivers
(see Table 2.1). The central sound contains approximately 60 percent of the
volume, 46 percent of the surface area, 33 percent of the shoreline, and
22 percent of the tidelands that occur within Puget Sound south of Admiralty
Inlet (Burns 1985). Most of the original tidelands in the urban and
agricultural areas of the central sound have been diked or filled (Shapiro
and Associates 1983).
Most of the population of the Puget Sound basin lives in the vicinity
of the central sound. Consequently, most .of the pollutant loadings
(nutrients, toxic substances) that reach Puget Sound are discharged into the
5-31
-------
central sound (PSWQA 1986b). The major cities and industrial centers in this
region are Everett, Seattle, Bremerton, and Tacoma.
Four study areas included in the characterization project are located in
the central sound: Port Gardner, Point Jefferson, Sinclair Inlet, and
City Waterway in Commencement Bay. Station locations are shown in
Figure 5.20. Data sources are given in Table 5.5. Algal bloom seasons for
the study areas are given in Table 5.6. Histograms summarizing the water
quality variables are given in Figures 5.21-5.27. Back-up tables of summary
data are provided in Appendix E. The ANOVAs comparing the water quality
variables before and after 1973 are summarized in Table 5.7. Long-term and
recent regressions are summarized in Table 5.8.
The study areas in the central sound are all in urbanized areas. They
represent a wide range of environments. The Port Gardner study area is
located in a fairly large, deep embayment that is affected by significant
inputs of fresh water. However, tidal volumes are low and tidal currents
are weak in the area (Federal Water Pollution Control Administration and
Washington State Pollution Control Commission 1967). The Point Jefferson
study area is in an open, deep part of the Main Basin. It is characterized
by a large volume and substantial flux of water. The Sinclair Inlet study
area is in a sheltered, shallow embayment, with little freshwater input and
a low flushing rate. The City Waterway study area is at the mouth of a
manmade waterway in the southeastern corner of Commencement Bay, a deep and
relatively open embayment. Although the Puyallup River influences the
circulation of Commencement Bay, the study area is approximately 1.2 km
south of the mouth of the river. Previously, Dames and Moore (1981)
reported that the study area was not greatly influenced by the freshwater
plume of the Puyallup River.
Based on the percent dissolved oxygen saturation at the surface, algal
blooms occurred in all the central sound study areas during May and June
(Table 5.6). The bloom period began and ended early in Port Gardner, and
ended late (August) in Sinclair Inlet. The blooms appeared to be most
intense in Sinclair Inlet and least intense in Port Gardner and City
Waterway.
5-32
-------
SUZ606
PSS602
PSS006
PSS006
PSS007
PSS008
JSPH
JSNK 01XV
JSPR 01 "
JSUQ 01
JSVS 01
JSWT01
JSTS 01
KSBP 01
KSHK 01
PSB305
PSB306
City Waterway
CMB006
Figure 5.20. Locations of study areas and sampling stations in the central sound.
5-33
-------
TABLE 5.5. SAMPLING STATION NUMBERS, DATA SOURCES, AND SAMPLING
PERIODS FOR THE STUDY AREAS IN THE CENTRAL SOUND
Study Area
Port Gardner
Point Jefferson
Sinclair Inlet
City Waterway
Station
PSS602
SUZ605
PSS005
PSS006
PSS007
PSS008
PSB305
PSB306
JSVS01
JSWT01
JSUQ01
JSYS01
KSHK01
JSPH01
JSNK01
JSPR01
KSBP01
SIN001
CMB006
Data
Source
uwa
UW
Ecology
Ecology
Ecology
Ecology
UW
UW
Metro
Metro
Metro
Metro
Metro
Metro
Metro
Metro
Metro
Ecology
Ecology
Sampling Period
1952-62
1952-53, 1956-57, 1960-62, 1969-71
1967-70, 1973-76
1967-70
1967-70
1967-70, 1980-86
1933-71
1965-67
1965-86
1965-86
1965-86
1965-86
1965-86
1966-67
1966-86
1965-67
1966-75, 1985-86
1967-70, 1973-74, 1976, 1978-86
1967-70, 1973-86
a UW = University
/ of Washingt
on.
5-34
-------
TABLE 5.6. ALGAL BLOOM SEASONS FOR THE CENTRAL SOUND STUDY AREAS,
AS DEFINED BY MONTHLY MEAN AND STANDARD ERROR OF PERCENT
DISSOLVED OXYGEN SATURATION IN SURFACE WATER
Percent Dissolved Oxyqen Saturation
Month
April
May
June
July
August
September
Port
Gardner
102 +/- 3a
115 +/- 3a
102 +/- 3a
98 +/- 5
93 +/- 6
72 +/- 6
Point
Jefferson
103 +/- 2
116 +/- 2a
118 +/- 2a
122 +/- 2a
102 +/- 2
86 +/- 1
Sinclair
Inlet
116 +/- 4
140 +/- 3a
123 +/- 7a
129 +/- 7a
143 +/- 13a
119 +/- 4
City
Waterway
95 +/- 2
103 +/- 8a
99 +/- 4a
110 +/- 6a
88 +/- 6
88 +/- 4
Months included in the algal bloom season.
5-35
-------
301
e ie 39 e
le 39 0 ie 30
DEPTH (m)
PORT
GARDNER
POINT
JEFFERSON
SINCLAIR
INLET
STUDY AREA
e ie ae
PORT
GARDNER
e ie 30 e 10 3e
DEPTH (m)
POINT
JEFFERSON
SINCLAIR
INLET
STUDY AREA
6 16 36
CITY
WATERWAY
0 10 30
CITY
WATERWAY
Figure 5.21. Mean salinity and water temperature values in the central sound study areas
during the algal bloom season.
5-36
-------
e te ae
ae e l
DEPTH (m)
ae e ie ae
PORT
GARDNER
POINT
JEFFERSON
SINCLAIR
INLET
CfTY
WATERWAY
STUDY AREA
10 38
ie ae e le ae
DEPTH (m)
PORT
GARDNER
POINT
JEFFERSON
SINCLAIR
INLET
e ie ae
crrv
WATERWAY
STUDY AREA
Figure 5.22. Mean concentrations of dissolved oxygen and dissolved inorganic nitrate
in the central sound study areas during the algal bloom season.
5-37
-------
LLJ
e 10 3e
PORT
GARDNER
(01 10
X J
S|
o:3
o
0 10 38
PORT
GARDNER
0 10 30 0 10 30
DEPTH (m)
POINT
JEFFERSON
SINCLAIR
INLET
e ie 30
CITY
WATERWAY
STUDY AREA
0 10 30 0 10 30
DEPTH (m)
POINT
JEFFERSON
SINCLAIR
INLET
0 10 30
CITY
WATERWAY
STUDY AREA
Figure 5.23. Mean concentrations of dissolved orthophosphate and chlorophyll a.
in the central sound study areas during the algal bloom season.
5-38
-------
tu o>
Se
> 0)
x 3
O
si
>h-
5^<
Ooc
WD
WH
o<
-------
uj^SS
i_ DC 73
£° =
*|
oo
HI < O
u. m o
"~5 *•
QC ii
CJQ •
Ou. o
o"
o
PORT
GARDNER
18 30
PORT
GARDNER
POINT
JEFFERSON
SINCLAIR
INLET
STUDY AREA
POINT
JEFFERSON
SINCLAIR
INLET
STUDY AREA
0 10 30 e ie 30 0 10 30 e ie 30
DEPTH (m)
CITY
WATERWAY
0 10 30 e 10 39 9 19 39
DEPTH (m)
CITY
WATERWAY
Figure 5.25. Log of geometric mean concentrations of sulfite waste liquor and fecal coliform
bacteria in the central sound study areas during the algal bloom season.
5-40
-------
2*
e
e ie
ae tee
DEPTH (m)
ise see
131
io'8J
^« 11-
o ie-
5*if
•o
0, g-
8
7
ie ae lee
DEPTH (m)
ise aee
Figure 5.26. Depth profiles of mean salinity and water temperature values in the Point
Jefferson study area during the algal bloom season.
5-41
-------
11
0)
LU
O
O
LU
>
9
8
7
6
5
3:
2-
e
0 10 30 100
DEPTH (m)
150 800
Figure 5.27. Depth profiles of mean concentrations of dissolved oxygen in the Point
Jefferson study area during the algal bloom season.
5-42
-------
TABLE 5.7. NET CHANGE AND PERCENT CHANGE IN THE MEAN VALUES OF WATER QUALITY
VARIABLES IN THE CENTRAL SOUND, BASED ON ANOVA COMPARISONS OF DATA
TAKEN BEFORE 1973 WITH DATA TAKEN FROM 1973 TO 1986
Depth
(m)
0
10
30
100
150
200
0
10
30
100
150
200
0
10
30
100
150
200
0
10
30
0
10
30
0
10
30
Port Gardner
Change
Net Percent
-3.11 15.0
-0.81 2.8
na
na
na
na
NS
NS
na
na
na
na
NS
NS
na
na
na
na
na
na
na
-0.16 20.9
-0.55 28.3
na
na
na
na
Point Jefferson
Change
Net Percent
Salinity (ppt)
+0.53 2.0
NS
NS
NS
NS
na
Water Temperature (° C)
NS
+0.46 4.1
+0.73 7.2
+0.72 7.4
NS
na
Dissolved Oxygen (mg/L)
NS
NS
NS
NS
NS
na
Nitrate (ug-at/L)
na
na
na
Phosphate (ug-at/L)
na
na
na
Chlorophyll a (ug/L)
NS
NS
na
Sinclair Inlet
Change
Net Percent
NSa
nab
na
na
na
na
NS
na
na
na
na
na
NS
na
na
na
na
na
na ,
na (
na
na
na
na ,
na
na
na
Ci ty
Net
NS
na
na
na
na
na
NS
na
na
na
na
na
NS
na
na
na
na
na
na
na
na
na
na
na
na
na
na
Waterway
Change
Percent
Surface Dissolved Oxygen Saturation (Percent)
0
NS
NS
NS
Seech i Disk Depth (m)
NS
NS
na
NS
na
Sulfite Waste Liquor (Pearl Benson Index)
0
10
30
-35.38 76.4
NS
na
na
na
na
NS
na
na
-7.32
na
na
51.8
Fecal Coliform Bacteria (No./100 mL)
0
10
30
na
na
na
na
na
na
na
na
na
na
na
na
a NS The pre-1973 and 1973-1986 values were not significantly different at P<0.05, based on a
nonparametric one-way ANOVA.
b na - Results of the statistical test were not available because of a lack of data.
5-43
-------
TABLE 5.8. SLOPES OF STATISTICALLY SIGNIFICANT LONG-TERM AND RECENT REGRESSIONS
OF WATER QUALITY VARIABLES AS A FUNCTION OF YEAR FOR THE CENTRAL SOUND
Depth
(m)
0
10
30
100
150
200
0
10
30
100
150
200
0
10
30
100
150
200
0
10
30
0
10
30
0
10
30
Port Gardner
Long-term
NSa
-0.027
na
na
na
na
NS
0.035
na
na
na
na
NS
NS
na
na
na
na
na
na
na
NS
-0.017
na
na
na
na
Recent
0.450
0.085
na
na
na
na
NS
0.150
na
na
na
na
NS
NS
na
na
na
na
NS
-0.709
na
NS
0.063
na
na
na
na
SI ooes
Point Jefferson
Sinclair
Long-term Recent Long-term
Salinity (ppt)
NS NS
-0.017 NS
NS -0.077
NS -0.115
-0.009 -0.165
NS na
Water Temperature (° C)
NS -0.068
0.028 NS
0.027 NS
NS NS
NS NS
NS na
Dissolved Oxygen (mg/L)
NS -0.078
NS NS
NS NS
NS NS
NS NS
na na
Nitrate (ug-at/L)
na na
na na
na na
Phosphate (ug-at/L)
na na
na na
na na
Chlorophyll a (ug/L)
NS na
NS na
na na
NS.
nab
na
na
na
na
NS
na
na
na
na
na
NS
na
na
na
na
na
na
na
na
na
na
na
na
na
na
Inlet
Recent
NS
NS
na
na
na
na
NS
NS
na
na
na
na
NS
NS
na
na
na
na
NS
NS
na
NS
0.056
na
na
na
na
City Waterway
Long -term
NS
NS
na
na
na
na
NS
na
na
na
na
na
NS
na
na
na
na
na
na
na
na
na
na
na
na
na
na
Recent
NS
NS
na
na
na
na
NS
na
na
na
na
na
NS
NS
na
na
na
na
NS
0.452
na
0.090
0.086
na
NS
NS
na
Surface Dissolved Oxygen Saturation (Percent)
NS NS -0.285 -2.091 NS NS
Seechi Disk Depth (m)
-0.086 NS NS 0.102 na NS
Sulfite Waste Liquorc (Pearl Benson Index)
NS
a NS Not significant at P<0.05.
b na - Results of the statistical test were not available because of a lack of data.
c Data were subjected to a log(X+1) transformation for the regressions.
" Data were subjected to a log transformation for the regressions.
NS
NS
0
10
30
0
10
30
-0.044
-0.052
na
na
na
na
NS
-0.099
na
Fecal
0.122
na
na
na
na
na
Col i form
na
na
na
na
na
na
Bacteriad
na
na
na
NS
na
na
(No. /100 mL)
na
na
na
na
na
na
NS
na
na
-0.059
-0.047
na
na
na
na
-0.079
-0.047
na
NS
na
na
5-44
-------
Port Gardner
The Port Gardner study area is located in the Whidbey Basin
(Figure 5.20). It is near the industrialized City of Everett and is
relatively close to shore. Historically, several pulp mills have discharged
wastes into the area (NOAA 1985). The earlier University of Washington
sampling stations were farther from shore than the more recently sampled
Ecology stations. Depths range from 100 to 150 m for the University of
Washington stations and average about 90 m for the Ecology stations.
Class A water quality standards apply in the area of the University of
Washington stations, while Class B water quality standards apply in the area
of the Ecology stations.
Tidal currents are weak in the Port Gardner area. The Snohomish River
flows into Possession Sound about 4.5 km north of the study area. This river
is the second largest river discharging into Puget Sound, contributing
approximately 18 percent of the total volume of fresh water that enters the
sound (see Table 2.1). A net southward surface flow .is caused by the input
from the Snohomish River. Net motion is generally northward at mid-depths
and generally southward near the bottom (Federal Water Pollution Control
Administration and the Washington State Pollution Control Commission 1967).
The quantity of wastes discharged from pulp and paper mills in the
Everett area decreased progressively through the 1960s and 1970s (Ecology
1976; NOAA 1985; Loehr, L., 21 July 1987, personal communication). These
changes reduced discharges of both chemical wastes and BOD. Major discharge
reductions were achieved by both the Scott and Weyerhaeuser sulfite mills in
1975. In 1978, the Weyerhaeuser mill closed. The Scott sulfite mill
adopted secondary effluent treatment beginning in 1980, replacing a system
that had discharged approximately half of the plant's effluent without
treatment and half of the plant's effluent after primary clarification
(Bechtel, T., 22 March 1988, personal communication).
5-45
-------
Environmental Conditions in the Study Area--
Mean salinity and water temperature values during the algal bloom period
are depicted in Figure 5.21. Data are available for surface water from the
early 1950s through 1986, but less coverage is available for deeper water.
The depth gradients for both salinity and water temperature were substantial
(Appendix E). At the surface, the mean salinity value was approximately
8 ppt lower than at the 10-m depth, while the mean water temperature value
was approximately 2.3° C higher than at the 10-m depth. The reduced
salinities observed at the surface presumably were caused by the inputs of
fresh, low density water from the Snohomish River. The negative correlation
between salinity and water temperature at the surface (Appendix F) suggests
that the fresh water from the Snohomish River tended to be warmer than the
salt water from the sound.
A vertical gradient in dissolved oxygen concentrations was also
detected. Average dissolved oxygen concentration was approximately 10.1 mg/L
at the surface and 8.9 mg/L at the 10-m depth (Figure 5.22). Data from
deeper than 30 m are not available for the characterization study, but
anoxic sediments containing material from log yards and pulp mills have been
reported in the past (Federal Water Pollution Control Administration and
Washington State Pollution Control Commission 1967).
Average surface concentrations of nitrate and phosphate were approxi-
mately one-third as high at the surface (6.0 ug-at/L and 0.7 ug-at/L,
respectively) as they were at the 10-m depth (18.2 ug-at/L and 1.6 ug-at/L,
respectively) (Figures 5.22 and 5.23). Negative correlations between water
temperature and both nitrate and phosphate concentrations at the surface
(Appendix F) may have been due to the seasonal rise in temperature and the
seasonal decline in nutrient concentrations that occur during the spring and
early summer. From April through June, the monthly mean water temperature
rose by 4.8° C, while the mean nitrate concentration fell by a factor of
four and the mean phosphate concentration fell by a factor of nearly two.
These seasonal drops in nutrient concentrations also probably caused the
positive correlation between the nitrate and phosphate concentrations at the
surface (Appendix F).
5-46
-------
The relatively low mean percent dissolved oxygen saturation at the
surface (106 percent) suggests that algal blooms in the Port Gardner study
area were less intense than those in most of the other study areas (e.g.,
Point Jefferson and Sinclair Inlet). Although density stratification of the
water column was well developed, the net southward drift of surface water
probably does not allow intense blooms to develop around Port Gardner. The
low value for mean Secchi disk depth (2.7 m) was probably due to suspended
particulate materials from the Snohomish River, rather than to high
concentrations of phytoplankton. Low transparency of the water column would
restrict the depth of the photic zone, limiting the growth of algal blooms.
Suspended material in pulp mill effluents did not have a major influence on
the Secchi depth values because secondary treatment was instituted in the
local mills by the time most of the Secchi depth data were collected.
The geometric mean of the concentrations of sulfite waste liquor
(measured by the Pearl Benson Index) at the surface was higher in the Port
Gardner study area (17.3) than in any other central sound study area
(Figure 5.25). However, the geometric mean of the surface concentrations of
sulfite waste liquor was higher in the Bellingham Bay study area (22.1) in
the northern sound. At Port Gardner, the average sulfite waste liquor
concentration at 10-m depth was less than half that of the surface,
reflecting the tendency of sulfite waste liquor released in surface water in
a density stratified system to remain in the surface layer. The high
concentration detected at 30-m depth (geometric mean=22.0) was based on
only three data points from the mid-1950s and early 1960s. During that
period, the Scott and Weyerhaeuser mills both discharged highly concentrated
sulfite wastes from a deep water diffuser off Port Gardner (Federal Water
Pollution Control Administration and Washington State Pollution Control
Commission 1967). However, substantial reductions of sulfite waste liquor
discharges from this diffuser have occurred since those data were obtained
(see above). Thus, this limited set of data for 30-m depth may have been
correct for the time period during which sampling occurred. The concentra-
tion of sulfite waste liquor probably has been much lower at this depth
during the 1980s.
5-47
-------
The mean concentration of fecal coliform bacteria at the surface
appeared to be relatively high in the Port Gardner study area (geometric
mean=22.9 organisms/100 ml) (Figure 5.25). As discussed in the next
section on water quality trends, this result probably was not caused by
detection of inadequately treated sewage. Instead, this apparent elevation
in the concentration of fecal coliform bacteria probably was the result of
detecting the bacterium, Klebsiella. which is often released in large
quantities in secondary pulp mill effluent (Johnson, B., 21 July 1987,
personal communication).
Water Quality Trends in the Study Area--
A summary of comparisons between water quality data collected prior to
and after 1973 is given in Table 5.7. Slopes from statistically significant
regressions of long-term and recent water quality data by year are given in
Table 5.8.
Physical Conditions—Plots of salinity and water temperature data by
year are shown in Figures 5.28 and 5.29. A long-term decline in salinity
values is evident, although salinity values increased after 1974 (see
Tables 5.7 and 5.8). The long-term decline in salinity values was probably
caused by a change in station locations from the offshore stations that were
sampled by the University of Washington in the 1950s and early 1960s to the
inshore stations (closer to the mouth of the Snohomish River) that were
sampled by Ecology since the late 1960s. The apparent increase in salinity
values since 1974 also may have been driven by changes in station location.
The values from 1974 through 1976 were recorded at Station PSS005, while the
values since 1981 were recorded at Station PSS008 (see Figure 5.20). The
later samples came from a station at the mouth of a manmade waterway that
was sheltered from the Snohomish River by an earthen breakwater.
The long-term increase in water temperature values at 10-m depth
appears to have been driven by the recent increase in water temperature
values at this depth. This recent increase also may have been caused by
changes in the locations of sampling stations. The most recent data were
taken from the mouth of the waterway mentioned above. The sheltering effect
5-48
-------
30
a
a
t a0-
V}
ie
e
1950 1955 1966 1965 1970 1975 1980 1985 1990
YEAR
30
I20
10
ANNUAL MEAN
I STANDARD ERROR
0 INDIVIDUAL OBSERVATION
— SIGNIFICANT REGRESSION LINE
(P < 0.05)
1 1 1 1 1 1 1 r
1950 1955 1960 1965 1970 1975 1980 1985 1990
YEAR
Figure 5.28. Salinity values at the surface and at 10-m depth in the Port Gardner study
area during the algal bloom season.
5-49
-------
24
23'
22"
21
201
19
181
17
16
15
H
13
12
11
ie
9
8
7
T 1 1
1950 1955 1966
—i 1 1 1 1—
1965 1979 1975 1988 1985
YEAR
1990
24-
23
22
21
18
17
16
15
14
13
12
11
ie
9
8
7
ANNUAL MEAN
I STANDARD ERROR
0 INDIVIDUAL OBSERVATION
SIGNIFICANT REGRESSION LINE
(P < 0.05)
1950 1955 1960 1965
1970
YEAR
1975 1980 1985
1990
Figure 5.29. Water temperatures at the surface and at 10-m depth in the Port Gardner
study area during the algal bloom season.
5-50
-------
of the breakwater may reduce the rate of vertical mixing in the waterway,
which could allow more effective solar heating of the near-surface water.
Dissolved Qxvaen--P1ots of dissolved oxygen concentration by year are
shown in Figure 5.30. There is no evidence for violations of the Class A
water quality standard (see Table 4.2). No significant temporal trends were
detected in dissolved oxygen concentrations (see Tables 5.7 and 5.8) at the
surface or at 10-m depth. The possibility that changes in station location
influenced the dissolved oxygen data could not be assessed. As discussed
above, anoxic sediments have been reported in the past, but long-term data
for water column depths below 10 m are not available.
Nutrients—Plots of nitrate concentrations against year are shown in
Figure 5.31. Data are available since 1974. Nitrate concentrations have
declined significantly (PO.05) at the 10-m depth. Changes in nitrate
concentration do not appear to have coincided with changes in station
location that occurred between 1976 and 1981. No explanation is available
for this decrease.
Statistically significant (PO.05) declines in phosphate concentra-
tions have occurred since the 1950s (Figure 5.32, Tables 5.7 and 5.8),
although the apparent change was greater at 10-m depth than at the surface.
The decline at 10-m depth could indicate that a long-term change has
occurred in phosphate concentrations. This long-term decline is consistent
with the declines detected in most other study areas. However, this
apparent decline may have been influenced by changes in station locations or
analytical techniques. Values averaged approximately 2 ug-at/L for the
University of Washington samples, which were collected from offshore
stations during the mid-1950s through the early 1960s. Values averaged
approximately 1.5 ug-at/L for the Ecology samples, which were collected
since 1968 from stations located closer to shore. Because it was not
possible to calibrate the methods used for phosphate analyses by University
of Washington and Ecology, the possibility that analytical differences
between University of Washington and Ecology introduced changes in phosphate
concentrations into the data cannot be assessed. However, the University of
5-51
-------
o>
E
19;
18
17"
16
15
14
13
la
n
10
9
8
7
6
5
4
3
Z
1
e
1959 1955 I960 1965 1979 1975 1980 1985 1999
YEAR
20
19
18
17
ie
13
IB
11
9
8
7
6
5
4
3
Z
1
ANNUAL MEAN
I STANDARD ERROR
O INDIVIDUAL OBSERVATION
—- SIGNIFICANT REGRESSION LINE
(P < 0.05)
o o
1950 1955 1960 1965 1970 1975 1980 1985
YEAR
1990
Figure 5.30. Concentrations of dissolved oxygen at the surface and at 10-m depth in
the Port Gardner study area during the algal bloom season.
5-52
-------
I20
o
o:
O
10
o
in
0-1,
1950
ANNUAL MEAN
J STANDARD ERROR
0 INDIVIDUAL OBSERVATION
---- SIGNIFICANT REGRESSION LINE
(P < 0.05)
0 °
1955 1960 1965 1970 1975 1980 1985 1990
YEAR
X.
'o
?30
LJ
t-
o
z
o
I"
10
o o
1950 1955 1969
1965
1970
YEAR
1975 1980 1985 1990
Figure 5.31. Concentrations of dissolved inorganic nitrate at the surface and at 10-m
depth in the Port Gardner study area during the algal bloom season.
5-53
-------
4
o>
3
I3
W
o
0.
o
o
o
u
e
ANNUAL MEAN
I STANDARD ERROR
O INDIVIDUAL OBSERVATION
---- SIGNIFICANT REGRESSION LINE
(P < 0.05)
1959 1955 I960
1965
1970
YEAR
1975 1980 1985
1990
en
3
O
a.
o
I
-«/>
XI
V)
o
0
IS
0 \_
<
0 0
i i t i i i i i i
150 1955 1960 1965 1970 1975 1980 1985 1990
YEAR
Figure 5.32. Concentrations of dissolved orthophosphate at the surface and at 10-m
depth in the Port Gardner study area during the algal bloom season.
5-54
-------
Washington data were produced with a spectrophotoroeter (Appendix A), which
probably yielded reasonably accurate results.
Phosphate concentrations at 10-m depth appear to have increased since
1974. This increase may have been affected by changes in station location
in Ecology's monitoring program. However, changes in location that occurred
since 1974 were less drastic than those that occurred when the data sources
changed from University of Washington to Ecology (see Table 5.5 and
Figure 5.20). As discussed in Chapter 3, sulfite waste liquor removes
dissolved orthophosphate from seawater, and secondary pulp mill waste
treatment facilities came on line in the Port Gardner area in 1980. The
negative correlation between the concentrations of phosphate and sulfite
waste liquor at 10 m (r=-0.41), while not statistically significant (P<0.05)
when scaled with the Bonferroni inequality, suggests that reductions in the
discharge of sulfite waste liquor may have contributed to the increase in
phosphate concentrations detected since 1974. Alternatively, changes in
other anthropogenic factors or oceanic inputs may have influenced the
phosphate data.
Indicators of Phytoplankton Growth—No substantial changes were
detected in the indicators of phytoplankton growth.. Data on chlorophyll a
concentrations are not available. No trends were detected in the percent
saturation of dissolved oxygen in the surface water (Figure 5.33). A
statistically significant long-term decline in Secchi disk depth since 1961
was detected (Figure 5.33, Table 5.8), but this decline was driven by one
very high value recorded in 1961. This observation was obtained at an
offshore station located relatively far from the influence of the Snohomish
River. Because observations obtained since 1968 were from inshore stations
relatively close to the mouth of the Snohomish River and the influences of
the Port of Everett, the apparent decline in Secchi disk depth probably was
an artifact of changes in the location of sampling stations.
Pollutants—Sionificant temporal declines (P<0.05) in the concentra-
tion of sulfite waste liquor were detected (Figure 5.34, Tables 5.7 and
5.8). The sharp declines in concentrations that occurred in the mid-1970s
and early 1980s coincided with the discharge changes mentioned above.
5-55
-------
300-
z
g 300
o:
i-
10
Z
UJ
o
o 100
o
UJ
>
_i
O
16-
H
IS
'10
•
o
o
Ul
ANNUAL MEAN
J STANDARD ERROR
0 INDIVIDUAL OBSERVATION
—- SIGNIFICANT REGRESSION LINE
(P < 0.05)
1950 1955 1960 1965 1970 1975 1980 1985 1990
YEAR
1950 1955 1960 1965 1970 1975 1980 1985 1990
YEAR
Figure 5.33. Percent dissolved oxygen saturation at the surface and Secchi disk depth
in the Port Gardner study area during the algal bloom season.
5-56
-------
oc
o
x
0)
s
is 8
£ 00
(0
O
O
CC
o
*
4)
55 g
o
o
1 '
ANNUAL MEAN
J STANDARD ERROR
0 INDIVIDUAL OBSERVATION
—- SIGNIFICANT REGRESSION LINE
(P < 0.05)
0 O
o o
o oo
1958 1955 1960 1965 1970 1975 1980 1985 1990
YEAR
o o <
1950 1955 1960 1965 1970 1975 1980 1985 1990
YEAR
Figure 5.34. Log of concentrations of sulfite waste liquor at the surface and at 10-m
depth in the Port Gardner study area during the algal bloom season.
5-57
-------
The actual magnitude of the decrease in sulfite waste liquor concen-
trations in the Port Gardner area may have been greater than the decrease
detected in the characterization database because the samples taken before
1968 were collected farther offshore (and farther from the discharge point)
than the samples taken after 1968. Changes in station location probably
biased the database to show increasing concentrations of sulfite waste
liquor. Therefore, the improvements in sulfite waste liquor discharges
were very substantial in the Port Gardner area.
A significant increase (P<0.05) was detected in the concentration of
fecal coliform bacteria since 1974 (Figure 5.35, Table 5.8). Two distinct
periods were evident in the data, as concentrations observed from 1974
through 1976 were much lower than concentrations observed since 1981. This
change coincides with the conversion of the Scott sulfite pulp mill to
secondary waste treatment in 1980. The fecal coliform bacteria detected
after this conversion probably were of the genus Klebsiella (Bechtel, T.,
22 March 1988, personal communication). This organism can be detected in
fecal coliform tests (Johnson, B., 21 July 1987, personal communication) -
Klebsiella grows rapidly in secondary treatment facilities of sulfite pulp
mills, which contain high concentrations of complex polysaccharides that
Klebsiella can metabolize rapidly. Concentrations of Klebsiella as high as
2.1 x 10^/100 mL have been reported in discharges from pulp mill treatment
ponds (Knittel 1975). Thus, the fecal coliform bacteria detected recently
in the Port Gardner study area probably were not indicative of contamination
by sewage effluent. Increases in seal and sea lion populations may also
have influenced fecal coliform bacteria concentrations, but no data are
available to investigate this possibility in the study area.
Although Klebsiella is a known human pathogen that can exist in the
guts of warm blooded animals, the presence of Klebsiella is probably not a
substantial environmental concern. Storm (1981) conducted a literature
review to determine whether dredging of sediments containing Klebsiella in
Gray's Harbor, Washington represented a serious threat to human health.
Storm (1981) concluded that Klebsiella was not a high risk human pathogen in
that situation. Moreover, because the reproductive capacity of Klebsiella
5-58
-------
4
o
ffi
§
_i t-
o ..
< 0
uz
u.
o
ANNUAL MEAN
I STANDARD ERROR
O INDIVIDUAL OBSERVATION
—- SIGNIFICANT REGRESSION LINE
(P < 0.05)
oo
195* 1955 1969 1965 197* 1975 19M 1985 199«
YEAR
Figure 5.35. Log of concentrations of fecal coliform bacteria at the surface in the Port
Gardner study area during the algal bloom season.
5-59
-------
is dependent on the availability of high concentrations of organic wastes,
this organism would be unlikely to persist in high concentrations in the
marine environment. However, because current Ecology regulations do not
distinguish Klebsiella from other fecal coliform bacteria, fecal coliform
guidelines for marine waters in Classes A and B (see Table 4.2) frequently
were violated in the Port Gardner area during the 1980s.
Point Jefferson
The study area for Point Jefferson is near the middle of the Main Basin
of Puget Sound, approximately even with the northern border of the City of
Seattle (see Figure 5.20). Class AA water quality standards apply in the
area. The sound is quite deep in this region, ranging from 37 to 285 m.
Although a substantial volume of water moves through the study area, current
velocities are only moderate because flow is not restricted by geographic
features. Thus, the currents do not cause substantial mixing of the water
column (Lincoln and Collias 1975). Wind stress can increase mixing and
retard the development of algal blooms in the area. Alternatively, an
extended period of calm winds and sunshine allows density stratification to
occur and enhance the development of algal blooms (Winter et al. 1975).
The Point Jefferson study area is not strongly affected by freshwater
inputs. The nearest large source of fresh water is the Duwamish River,
which contributes approximately 2 percent of the total flow of fresh water
to the sound. The Duwamish River empties into Elliott Bay approximately
22 km southeast of the study area. The West Point sewage treatment plant,
which has provided primary treatment for most of the sewage from the City of
Seattle since the mid-1960s, discharges approximately 10 km south of the
study area. Neither of these major sources of fresh water greatly influence
water quality in the Point Jefferson study area.
Because of the absence of major pollutant sources, Point Jefferson is a
reference area in this study. All the water that transits the sound south
of Point Jefferson must pass by the study area. The discharges from the
major urban areas on the sound south of the City of Everett also pass
through the study area. In addition, the study area is in the middle of the
5-60
-------
Main Basin of Puget Sound, well removed from local, small-scale influences.
Because circulation is not restricted at Point Jefferson, the area does not
appear to be particularly sensitive to excess nutrient enrichment. A
substantial data set exists for this site, providing coverage from the
surface to 200-m depth as far back as 1932.
Environmental Conditions in the Study Area--
Mean salinity and water temperature values for the surface and for
depths of 10 and 30 m are plotted in Figure 5.21. This information is
combined with data from depths of 100, 150, and 200 m in Figure 5.26. Data
are available since 1932, although coverage decreases with depth (Appen-
dix E).
Moderate changes of salinity and water temperature values were evident
with depth. At the surface, the mean salinity value was approximately
1.0 ppt lower than at 10-m depth, while the mean water temperature value was
approximately 0.9° C higher than at 10-m depth. Salinity varied less with
depth at Point Jefferson than at several other areas with nearby sources of
fresh water. For example, at Port Gardner the mean salinity value at the
surface was over 8 ppt lower than the mean salinity value at 10-m depth.
Water temperature varied less with depth at Point Jefferson than it did at
sites with nearby sources of fresh water. For example, at Port Gardner, the
mean water temperature value at the surface was approximately 2.3° C higher
than the mean value at 10-m depth. Water temperatures also varied less with
depth at Point Jefferson than at sites with more limited rates of
circulation. For example, at Sinclair Inlet, the mean water temperature
value at the surface was approximately 1.7° C higher than the mean value at
10-m depth. The rates of change of the salinity and temperature values were
lower at depths below 100 m than at depths closer to the surface. For
example, the mean salinity value at 150-m depth was only 0.2 ppt lower than
the mean value at 200-m depth, and the mean water temperature value at 150-m
depth was only 0.26° C higher than the mean value at 200-m depth.
The vertical distribution of dissolved oxygen concentrations
(Figures 5.22 and 5.27) indicated that photosynthetic enhancement of
5-61
-------
dissolved oxygen was restricted to near-surface waters. The mean
concentrations of dissolved oxygen were approximately 11.0 mg/L at the
surface, 9.7 mg/L at 10-m depth, and 8.3 mg/L at 30-m depth. Mean dissolved
oxygen concentrations dropped slowly below 100-m depth, remaining slightly
above 7 mg/L at 200-m depth.
As discussed in Chapter 3, nutrient data for Point Jefferson, which
would have been obtained from the University of Washington and Metro, were
not analyzed because of inconsistencies in the variables measured. Algal
blooms appear to have been moderately well developed at Point Jefferson (see
Figures 5.23 and 5.24). Chlorophyll a concentrations were significantly
higher (t-test, PO.001) in the Point Jefferson study area than they were in
the City Waterway study area, with highest concentrations occurring near the
surface. Mean percent dissolved oxygen saturation at the surface
(119.4 percent) was moderately elevated, although this value was less than
in the Sinclair Inlet study area (134.0 percent). Mean Secchi disk depth
was high (4,7 m) in the Point Jefferson study area. Secchi disk depth was
negatively correlated with surface chlorophyll a concentration and with
percent dissolved oxygen saturation at the surface (Appendix F). These
correlations suggest that transparency at Point Jefferson is influenced
primarily by phytoplankton growth. Because there is no nearby large source
of fresh water, suspended particulates probably had a relatively small
effect on Secchi disk depth.
It was not possible in this study to investigate pollutants in the
Point Jefferson study area. Because no large pulp mill has existed in the
area, it is unlikely that pollution by sulfite waste liquor has been a
problem. However, data on sulfite waste liquor concentrations are not
available for this study area. Data on the concentrations of fecal coliform
bacteria have been recorded by Metro since 1966. However, the analytical
techniques used to measure this variable changed from the Most Probable
Number (MPN) Method to the Membrane Filtration Method in 1977, and the
detection limit subsequently dropped from 10 organisms/100 mL to
1 organism/100 mL in 1980 (Hayward, A., 24 July 1987, personal
communication). Therefore, incompatibility of analytical methods prevented
5-62
-------
the analysis of temporal trends in the concentrations of fecal coliform
bacteria.
Water Quality Trends in the Study Area--
Temporal trends in water quality near Point Jefferson were evaluated
by Duxbury (1975) and by Collias and Lincoln (1977). Duxbury (1975)
concluded that changes in dissolved oxygen saturation and phosphate
concentration that occurred at 10-m depth between 1933 and 1973 were related
to oceanographic factors, rather than to increases in the amounts of waste
discharged into the sound. Collias and Lincoln (1977) reached a similar
conclusion in a more comprehensive study that used data collected through
1975.
A summary of comparisons between water quality data collected before and
after 1973 is given in Table 5.7. Slopes from statistically significant
long-term and recent regressions of values of the water quality variables by
year are given in Table 5.8.
Physical Conditions—Plots of salinity and water temperature data by
year are shown in Figures 5.36-5.41. The mean surface salinity from 1932
through 1972 was approximately 2 percent lower than the mean surface
salinity from 1973 through 1986 (Table 5.7). However, a significant slope
was not detected for the plot of this variable by year (Table 5.8).
Increasing salinity is consistent with the rainfall data from the Seattle-
Tacoma International Airport, which showed that total annual rainfall has
declined since the late 1940s (see Figure 5.2). Although mean salinity
values before and after 1973 were not significantly different (P>0.05) at
depths below the surface (Table 5.7), most of the regressions of salinity
against year had negative slopes. At 100- and 150-m depth, salinity values
appear to have been higher in the 1930s and to have dropped in 1986.
The cause of the apparent declines in salinity values at depth is not
known. Salinity changes may have been influenced by variations in oceanic
inputs. Data on rainfall and runoff to Puget Sound do not explain these
declines. The rainfall data show an overall decrease since 1945, which is
5-63
-------
40
a
a
20
e
T
T 1 1 1 1 1 1 1 1 1 1 r
1930 1935 1940 1945 1950 1955 I960 1965 1970 1975 1989 1985 1999
YEAR
30
a.
a
t 20
_
10
10
0
ANNUAL MEAN
STANDARD ERROR
INDIVIDUAL OBSERVATION
SIGNIFICANT REGRESSION LINE
(P < 0.05)
1930 1935 1940 1945 1959 1955 1969 1965 1979 1975 1989 1985 1999
YEAR
Figure 5.36. Salinity values at the surface and at 10-m depth in the Point Jefferson study
area during the algal bloom season.
5-64
-------
30
a
a.
I I |» i r r i r r i I T
1930 1935 1949 1945 1950 1955 I960 1965 1970 1975 1980 1985 1990
YEAR
30
a
a
_
VI
10
ANNUAL MEAN
J STANDARD ERROR
0 INDIVIDUAL OBSERVATION
—- SIGNIFICANT REGRESSION LINE
(P < 0.05)
1930 1935 1940 1945 1950 1955 1960 1965 1970 1975 1980 1985 1990
YEAR
Figure 5.37. Salinity values at 30- and 100-m depths in the Point Jefferson study area
during the algal bloom season.
5-65
-------
40
Se-
_
l/l
le
e
T 1 1 1 1 1 1 1 1 1 1 1 r
1938 1935 194e 1945 1959 1955 1966 1965 1970 1975 1980 1985 1990
YEAR
30
a
a
t ae
z
10
I*****
ANNUAL MEAN
J STANDARD ERROR
O INDIVIDUAL OBSERVATION
—- SIGNIFICANT REGRESSION LINE
(P < 0.05)
T T
1930 1935 1940 1945 1950 1955 1960 1965 1970 1975 1980 1985 1990
YEAR
Figure 5.38. Salinity values at 150- and 200-m depths in the Point Jefferson study area
during the algal bloom season.
5-66
-------
24
23-
22"
21 '
20
19
18
17"
16
15
14
13
121
11
ie
91
8
71
ANNUAL MEAN
J STANDARD ERROR
0 INDIVIDUAL OBSERVATION
—- SIGNIFICANT REGRESSION LINE
(P<0.05)
oo o
I I
1938 193S 1946 194S 1956 1955 1969 1965 1979 1975 1989 1985 1999
YEAR
24
23
22
21
~20
S19
• IP
o>
13
12
11
10
9
8
7
1939 1935 1949 1945 1959 1955 1969 1965 1979 1975 1989 1985 1999
YEAR
Figure 5.39. Water temperatures at the surface and at 10-m depth in the Point Jefferson
study area during the algal bloom season.
5-67
-------
23
22
21 1
~20
U.19
•18
I"
«16
15
2"
£ 13
2 12
" 11
9
8
7
ANNUAL MEAN
J STANDARD ERROR
O INDIVIDUAL OBSERVATION
-— SIGNIFICANT REGRESSION LINE
(P < 0.05)
—i 1 1 1 1 1 1 1 1 1 i r
1935 1940 1945 1954 1955 196* 1965 1979 1975 1980 1985 1990
YEAR
24-
23
221
21
20
19
18
17
16
15
H
131
12
11
10
9
8
7i
I I T I I I 1 1 I I I T
1930 1935 1940 1945 1950 1955 1960 1965 1970 1975 1980 1985 1990
YEAR
Figure 5.40. Water temperatures at 30- and 100-m depths in the Point Jefferson study
area during the algal bloom season.
5-68
-------
34 '
S3'
S3
31 '
,20"
. 19
5l8j
*17'
16"
15"
14
13
(E
U
0.
w 15
11
10
9
8
ANNUAL MEAN
J STANDARD ERROR
0 INDIVIDUAL OBSERVATION
—- SIGNIFICANT REGRESSION LINE
(P < 0.05)
1 1 1 1 1 1 1 1 1 1 1 r
1930 1935 1940 1945 1950 1955 1960 1965 1970 1975 1980 1985 1990
YEAR
-------
contrary to the apparent trend. Runoff data for 1930 through 1978 do not
contain any statistically significant (P>0.05) temporal trend. Changes in
station location probably did not influence these trends markedly because
all data collected through the mid-1960s came from the same station
(Station PSB305). However, for the station near Alki Point that was sampled
during overlapping years and seasons by the University of Washington and
Metro, the salinity values reported by Metro tended to be lower than the
salinity values reported by the University of Washington (see Chapter 4).
Thus, it is possible that differences in analytical technique between the
University of Washington, which provided the older data, and Metro, which
provided the recent data, could have introduced an apparent decline in
salinity values into the data.
Some moderate changes were detected in water temperature. A negative
slope of -0.07° C/yr was found for the regression of surface water
temperature by year since 1973 (Table 5.8). However, mean surface water
temperatures from before and after 1973 were not significantly different
(P>0.05) (Table 5.7). The pattern of decreasing surface water temperatures
in recent years was likely influenced by some high water temperatures
reported in the early 1970s (Figure 5.39). Increases in water temperature
were found at depths of 10, 30, and 100 m (Tables 5.7 and 5.8). These
increases were likely influenced by low water temperatures at depth in the
1950s.
Dissolved Oxygen—Plots of dissolved oxygen concentration against year
are shown in Figures 5.42-5.44. There is no evidence that the Class AA water
quality standard (see Table 4.2) was violated. The only statistically
significant change (PO.05) in dissolved oxygen concentration was a decline
in surface water since 1973 (Tables 5.7 and 5.8). However, this apparent
decline appears to have been caused by some high values recorded from the
mid-1970s and by erratic variations that have occurred since the mid-1970s,
including some low values in 1986.
Nutrients—As discussed previously, temporal trends in nutrient
concentrations were not analyzed due to the limited amount of available
5-70
-------
20
19 '
18-
17
16
on
X10
Q 9
£ 8
8 7
$. 6
<=> 5
4
3
2
1 1
Oft O
O 0
1935 1946 1945 19Se 19S5 1966 1965 1970 1975 1980 1985 1990
YEAR
20
19
18
17
16
^14
gl0
Q 9
8
o 7
vt c
1/1 b
3"
2
1
e
ANNUAL MEAN
J STANDARD ERROR
O INDIVIDUAL OBSERVATION
—- SIGNIFICANT REGRESSION LINE
(P < 0.05)
1930 1935 1940 1945 1950 1955 1960 1965 1970 1975 1980 1985 1990
YEAR
Figure 5.42. Concentrations of dissolved oxygen at the surface and at 10-m depth in the
Point Jefferson study area during the algal bloom season.
5-71
-------
w
19
181
17
16
15
14
13
512
on
gie
o »
U o
* 7
o 7
ft 6
5 5
3
1
e
ANNUAL MEAN
I STANDARD ERROR
0 INDIVIDUAL OBSERVATION
SIGNIFICANT REGRESSION LINE
(P < 0.05)
T
T 1 1 1 1 1 1 T i i i r
1939 1935 1949 1945 1959 1955 1969 1965 1979 1975 1989 1985 1999
YEAR
20
19
18
17
16
13
oil
glj
o »
2 8
3 7
g 6
o 5
4
3
a
i
9
1939 1935 1949 1945 1959 1955 1969 1965 1979 1975 1989 1985 1999
YEAR
Figure 5.43. Concentrations of dissolved oxygen at 30- and 100-m depths in the Point
Jefferson study area during the algal bloom season.
5-72
-------
19
18
171
16
<15
raH
I 13
512
oil
X10
81
7
6
5
4
3
a
i
ANNUAL MEAN
STANDARD ERROR
INDIVIDUAL OBSERVATION
SIGNIFICANT REGRESSION LINE
(P < 0.05)
i i i i f i IT i ~r \ i T
1930 1935 1940 1945 1950 1955 1960 1965 1970 1975 1980 1985 1990
YEAR
ae
19
18
17
16
£13
£12
oil
gi0
Q 9
LJ o
5 5
p 7
6
Q 5
4
3
a
i
0
193e 1935 1940 1945 1950 1955 1960 1965 1970 1975 1980 1985 1990
YEAR
Figure 5.44. Concentrations of dissolved oxygen at 150- and 200-m depths in the Point
Jefferson study area during the algal bloom season.
5-73
-------
data and the incompatibility of the data collected by the University of
Washington and Metro.
Indicators of Phvtoplankton Growth—The evidence discussed below
suggests that algal production has declined recently in the Point Jefferson
study area. However, the evidence is somewhat weak and is not unequivocal.
It appears that this decline, if it was a real phenomenon, was merely
short-term variation within the normal range of production, and not a well-
established, long-term trend.
Chlorophyll a concentrations are plotted against year in Figure 5.45.
Unfortunately, data are available only for the years 1966 through 1975, and
cannot be used to corroborate the tentative results discussed above. No
temporal trends were detected in chlorophyll a concentrations from 1966
through 1975.
The percent dissolved oxygen saturation at the surface exhibited both
long-term and recent declines (Figure 5.46, Table 5.8). However, the
overall averages from before and after 1973 were not significantly different
(Table 5.7). From examination of Figure 5.46, it appears that the values
were highest in the 1930s (e.g., up to 220 percent saturation), and that the
values recorded in 1985 and 1986 were low (averaging approximately 70 percent
saturation). The recent statistically significant (PO.05) increase in
Secchi disk depth (Figure 5.46, Table 5.8) also suggests that the drop in
percent dissolved oxygen saturation at the surface represented a decline in
algal production. However, as with the dissolved oxygen saturation data,
unusual Secchi disk depth data were reported in 1985 and 1986. The mean
Secchi disk depth reported during the 1986 algal bloom season was
approximately 9 m, which is the highest seasonal mean observed in the Point
Jefferson study area. Thus, the changes in the percent dissolved oxygen
saturation at the surface and Secchi disk depth appear to have been caused
by unusual conditions in 1985 and 1986, rather than by systematic changes
through time.
Pollutants—As discussed previously, analyses of concentrations of
sulfite waste liquor and fecal coliform bacteria could not be conducted.
5-74
-------
30
201
i
a.
o
cc
O
ei
1930 1935 1940 1945 1950 1955 1966 1965 1978 1975 1980 1985 1990
YEAR
301
20
a.
o
o
ANNUAL MEAN
I STANDARD ERROR
« INDIVIDUAL OBSERVATION
—- SIGNIFICANT REGRESSION LINE
(P < 0.05)
1930 1935 1940 1945 1950 1955 1960 1965 1970 1975 1980 1985 1990
YEAR
Figure 5.45. Concentrations of chlorophyll 3 at the surface and at 10-m depth in the
Point Jefferson study area during the algal bloom season.
5-75
-------
300
200
10
z
u
o
O 100
o
LJ
>
o
ANNUAL MEAN
J STANDARD ERROR
« INDIVIDUAL OBSERVATION
—- SIGNIFICANT REGRESSION LINE
(P < 0.05)
oo
1 1 1 1 1 1 1 1 1 1 1 r
1939 1935 1946 1945 1959 1955 I960 1965 1970 1975 1986 1985 1996
YEAR
16
14
12
I10
Q.
Ld
<=> 8
5
z 6
o
LJ
0
1939 1935 1940 1945 1950 1955 I960 1965 1970 1975 1980 1985 1990
YEAR
Figure 5.46. Percent dissolved oxygen saturation at the surface and Secchi disk depth in
the Point Jefferson study area during the algal bloom season.
5-76
-------
Sinclair Inlet
The Sinclair Inlet study area is on the western side of the central
sound, separated from the Main -Basin of Puget Sound by Bainbridge Island
(see Figure 5.20). It is located in midchannel about two-thirds of the way
from the mouth to the head of Sinclair Inlet, off the City of Bremerton and
the Puget Sound Naval Shipyard. The maximum depth of Sinclair Inlet is
approximately 65 m at the mouth; depths generally become shallower from the
mouth to the head. Salt marshes and mudflats extend out approximately
0.6 km into the inlet from the head. Average depth in the study area is
approximately 12 m. Class A water quality standards apply in the area (see
Table 4.2). Contamination of the sediments by heavy metals has been
detected near the naval shipyard (U.S. EPA 1986b).
The principal forces that produce currents in Sinclair Inlet are tidal
(Lincoln and Collias 1975). Generally, weak tidal currents oscillate in
direction, moving water in and out of the inlet. Two small creeks provide
most of the freshwater input to Sinclair Inlet, so the flushing rate is
low, especially during neap tides. In addition, wind stress substantially
affects water transport. Southwesterly winds often force surface water out
of the inlet, which draws replacement water into the inlet at depth.
Improvements in wastewater treatment facilities in and around the City
of Bremerton were completed in 1985 (Baker, D., 29 October 1987, personal
communication; Poppe, J., 9 November 1987, personal communication).
Effluent previously discharged from two primary sewage treatment plants is
now consolidated and given secondary treatment prior to discharge into
Sinclair Inlet near the City of Bremerton. Most combined sewer overflows
that discharged to Port Washington Narrows (on the eastern side of
Bremerton), or to Sinclair Inlet (on the western end of the naval shipyard),
were closed in 1985. However, several combined sewer overflows still exist
in the City of Bremerton and the naval shipyard. It is anticipated that
these remaining combined sewer overflows will be closed in the next few
years (Baker, D., 13 November 1987, personal communication). One of the
small creeks mentioned above, Gorst Creek, is a known source of contamination
5-77
-------
by fecal coliform bacteria (Struck, P., 9 November 1987, personal
communication).
Environmental Conditions in the Study Area--
Mean salinity and water temperature values during the algal bloom
period are shown in Figure 5.21. Data are available from 1967 through
1986. More data are available for the surface water than for 10-m depth.
The salinity gradient over depth was small (i.e., the mean salinity value at
the surface was only 0.5 ppt lower than the mean salinity value at 10-m
depth). This gradient reflects the lack of large inputs of fresh water into
Sinclair Inlet. The temperature gradient over depth was large, with the
surface temperature averaging approximately 1.7° C higher than the
temperature at 10-m depth. Thus, density stratification over the water
column was caused principally by the temperature gradient. The magnitude of
the surface warming suggests that vertical mixing rates are low, comparable
to the vertical mixing rates in City Waterway (see below) and less than
those at Point Jefferson.
Depth gradients in the concentrations of dissolved oxygen and nutrients
were well developed (Figures 5.22 and 5.23). Mean dissolved oxygen concen-
tration at the surface in .Sinclair Inlet was the highest of any central
sound study area (11.3 mg/L), while mean dissolved oxygen concentration at
10-m depth was the lowest of any central sound study area (8.9 mg/L). Mean
nitrate concentrations were quite low at the surface (<2.7 ug-at/L) and at
10-m depth (8.1 ug-at/L). The mean phosphate concentrations at Sinclair
Inlet were not markedly different from those at the other central sound
study areas. The significant negative correlations (P<0.05) between
dissolved oxygen and nitrate concentrations (Appendix F) suggest that
nitrate concentrations were strongly influenced by photosynthetic rates.
Intense algal blooms appear to have occurred in the study area. The
mean percent dissolved oxygen saturation at the surface (134 percent) was
the highest of any area studied in this characterization study. The next
highest mean value for this variable (128.5 percent) was detected in Carr
Inlet (Appendix E). Also, the mean Secchi disk depth (3.5 m) was relatively
5-78
-------
low (Figure 5.24). Although Secchi disk depths may have been influenced by
disturbances from the City of Bremerton and the naval shipyard, the absence
of substantial freshwater inputs suggests that the major factor limiting
Secchi disk depths in Sinclair Inlet was phytoplankton abundance.
Geometric mean concentrations of sulfite waste liquor (3.6 Pearl Benson
Index) and fecal coliform bacteria (1.9 organisms/100 ml) were low in the
study area (Figure 5.25). There was no source of sulfite waste liquor near
the study area, and raw sewage was discharged only through combined sewer
overflows during the study period.
Water Quality Trends in the Study Area—
A summary of comparisons between water quality data collected prior to
and after 1973 is given in Table 5.7. Slopes from statistically significant
long-term and recent regressions of the water quality data against year are
given in Table 5.8.
Physical Conditions—Plots of salinity and water temperature by year
are shown in Figures 5".47 and 5.48. No temporal trends were detected for
either variable.
Dissolved Oxygen—Plots of dissolved oxygen concentrations by year are
shown in Figure 5.49. Violations of the Class A water quality standard (see
Table 4.2) were recorded at 10-m depth in 1974 and 1980. No temporal trends
were detected.
Nutrients—Plots of concentrations of nitrate and phosphate by year are
shown in Figures 5.50 and 5.51, respectively. Nutrient data are available
since 1973. No temporal trends were detected in nitrate concentrations, but
an increase in phosphate concentrations at 10-m depth was detected
(Table 5.8). No explanation was readily apparent for the increased phosphate
concentrations, but it did not appear to be influenced by improvements in
the sewage treatment system implemented in 1985. It is possible that
changes in other anthropogenic factors or in oceanic inputs influenced the
phosphate data.
5-79
-------
30
a
a
-^
t
z
_i
e
1950 1955 I960
1965
1979
YEAR
1975
1986 1985 1996
30
a
o.
E 20
z
10
1!
I
0
....
)50 1955 1966 1965 1976 1975
YEAR
ANNUAL MEAN
STANDARD ERROR
INDIVIDUAL OBSERVATION
SIGNIFICANT REGRESSION LINE
(P
-------
24
23
221
21
•181
J»
§15
£ 13 J
ui 12-
"~ 11 '
10-
9
8
7-1
1959
1 1 1
1955 1960 1965 1970 1975 1980 1985 1990
YEAR
23
22
21
^20
^19
• 18
I"
§15
12 ^
11
10;
9-
8
71
ANNUAL MEAN
I STANDARD ERROR
0 INDIVIDUAL OBSERVATION
SIGNIFICANT REGRESSION LINE
(P < 0.05)
O O
1950 1955 1960 1965 1970 1975 1980 1985 1990
YEAR
Figure 5.48. Water temperatures at the surface and at 10-m depth in the Sinclair Inlet
study area during the algal bloom season.
5-81
-------
20-
19"
18
17"
16'
SIS'
^ . A •
O> 14
1 13
z 13-
3 n
gi0-
o 9
2 8
3 7
£ 6
5 5
3
a
i
0
o
0
Ov ^ 1
\I/ \ ° °jnc?
V' ^ — *^~4A i °ri
o Vo 1
0 I *
*
T 1 i 1 1 1 i i r
1950 1955 1960 1965 1970 1975 1980 1985 1990
30 '
19-
18-
17-
16
x15
512
on
>•
x 10
o 9
^ 8
* 7
irt e
in b
0 5
3
a
i
a
YEAR
ANNUAL MEAN
} STANDARD ERROR
0 INDIVIDUAL OBSERVATION
-- -- SIGNIFICANT REGRESSION LINE
(P < 0.05)
0
6 JK o 0
A ^ ^ -* ^ * X
i/\s\r ^J^
*> 1 I °
I * 0
0 0
i I i I i i i I I
1950 1955 1960 1965 1970 1975 1980 1985 1990
YEAR
Figure 5.49. Concentrations of dissolved oxygen at the surface and at 10-m depth in
the Sinclair Inlet study area during the algal bloom season.
5-82
-------
40
o
be
I so
o
a.
O
z
o
LJ'
> 19
o
in
e
ANNUAL MEAN
I STANDARD ERROR
0 INDIVIDUAL OBSERVATION
—- SIGNIFICANT REGRESSION LINE
(P < 0.05)
1950 1955 I960 1965
—i—
1970
YEAR
—i 1 1 r
1975 1980 1985 1990
LJ
>—
<
20
o
ct
O
Q
LJ
1950
—i 1 1—
1955 1960 1965
T
-T
1 I
1970 1975 1980 1985 1990
YEAR
Figure 5.50. Concentrations of dissolved inorganic nitrate at the surface and at 10-m depth
in the Sinclair Inlet study area during the algal bloom season.
5-83
-------
Ld
t-
£3
I/I
o
Q.
o
o
o
1959
ANNUAL MEAN
I STANDARD ERROR
O INDIVIDUAL OBSERVATION
SIGNIFICANT REGRESSION LINE
(P < 0.05)
O O
o o
1955 I960 1965
—i—
i9?e
YEAR
1975
198* 1985 1996
I3
t/1
O
Q.
O
o
o
o .
in 1
e
'o o
o o
o o
1959 1955 I960 1965
1976
YEAR
1975 1980 1985 1990
Figure 5.51. Concentrations of dissolved orthophosphate at the surface and at 10-m depth
in the Sinclair Inlet study area during the algal bloom season.
5-84
-------
Indicators of Phvtoplankton Growth—Chlorophyll a data are not
available. Percent dissolved oxygen saturation at the surface and Secchi
disk depth are plotted by year in Figure 5.52. No temporal trends were
detected.
Pollutants—Data for sulfite waste liquor and fecal coliform bacteria
are plotted by year in Figure 5.53. Sulfite waste liquor data are only
available from 1968 through 1976; no changes were detected. Trends in the
concentration of fecal coliform bacteria were not statistically signifi-
cant, but a few high values, in violation of Class A water quality standards
(Table 4.2), were detected from 1978 through 1983. No explanation was
available to explain this phenomenon, although the combined sewer overflows
that were closed in 1985 may have contributed to the earlier elevations in
fecal coliform bacteria.
City Waterway
The study area is located in the mouth of City Waterway in the
southeastern corner of Commencement Bay (see Figure 5.20). Commencement Bay
is a deep (over 150" m), open embayment. City Waterway is a manmade
commercial waterway bordered by the industrial City of Tacoma. The depth
near the study area has been maintained by dredging at approximately 10 m.
The Puyallup River empties into Commencement Bay approximately 1.2 km north
of City Waterway. The Puyallup River discharges 6 percent of the total
volume of fresh water entering into the sound (see Table 2.1). It carries a
heavy load of sediment, creating a delta at its mouth and a highly turbid
surface layer in the bay (City of Tacoma 1983a,b; NOAA 1987).
Water movements in Commencement Bay are highly variable, and are
influenced by tides, the flow of the Puyallup River, and winds (Dames and
Moore 1981; City of Tacoma 1983a,b; NOAA 1986b, 1987). On ebbing tides, the
plume from the river exits out along the central axis of the bay as a turbid
surface flow. On flooding tides, the flow of the river is deflected and
backs up, causing low salinities to occur along the northern shoreline and
in the southeastern corner of the bay. Winds principally affect surface
5-85
-------
300
2 see
X
O 100
o
O
in
vt
o
e
ANNUAL MEAN
J STANDARD ERROR
0 INDIVIDUAL OBSERVATION
—- SIGNIFICANT REGRESSION LINE
(P c 0.05)
1959 1955 I960 1965
1976
YEAR
1975
1988 1985 1998
a.
16"
14
12'
10
8
52
o
x 6
o
o
UJ
1950
1955 I960 1965
1970
YEAR
1975 1988 1985 1998
Figure 5.52. Percent dissolved oxygen saturation at the surface and Secchi disk depth
in the Sinclair Inlet study area during the algal bloom season.
5-86
-------
oc
o
3
O
-3*1
Uj 0
LLjg
o
o
oc
UJ
1?
§1
O i.
.
< o
oz
UJ —
LL
U.
O
o
o
3
e
ANNUAL MEAN
I STANDARD ERROR
0 INDIVIDUAL OBSERVATION
—- SIGNIFICANT REGRESSION LINE
(P < 0.05)
l T
T I I I
1950 1955 I960 1965 1970 1975 19M 1985 1990
YEAR
31
1950 1955 1960 1965 1970 1975 1980 1985 1990
YEAR
Figure 5.53. Log of concentrations of sulfite waste liquor and fecal coliform bacteria at
the surface in the Sinclair Inlet study area during the algal bloom season.
5-87
-------
waters (e.g., northerly and westerly winds may force surface waters back
into the bay and into the waterways). At depth in Commencement Bay,
nontidal flows are weak and erratic, but the net movement of sub-surface
water typically is onshore. The depth of no net horizontal movement in the
bay has been estimated to be between 10 and 20 m (Dames and Moore 1981).
Because City Waterway is sheltered, water movements are weak and
erratic. The major forces influencing water movements in the waterway are
tides and wind stress (Dames and Moore 1981). The influence of the plume
from the Puyallup River has been debated, but recent evidence suggests that
water from the Puyallup River can enter the mouths of the waterways (see
below). The frequency with which the plume influences City Waterway and the
distance over which the plume water may penetrate into City Waterway are
unknown. Dames and Moore (1981) concluded that City Waterway is largely
isolated from the influences of the Puyallup River by the effects of a back
eddy in the southeastern corner of the bay. Tetra Tech (1985) noted that
the water at the mouth of City Waterway contained lower levels of total
suspended solids than the water in the other waterways of Commencement Bay.
This observation supports the interpretation that City Waterway is not
affected substantially by the Puyallup River. However, NOAA (1986b) showed
that fresh water and suspended particulate matter from the Puyallup River
plume can enter the mouths of the waterways, including City Waterway, along
the surface. In this characterization study, surface salinity values at the
mouth of City Waterway appear to be quite low (the mean surface salinity was
23.3 ppt) relative to the salinities found in the bay. Average surface
salinity values near the center of Commencement Bay exceed 29 ppt (NOAA
1987). Because there is no other substantial source of fresh water for the
mouth of City Waterway, water from the Puyallup River appears to influence
salinity values at the study site.
Water quality in City Waterway has been affected by numerous historical
and present day waste discharges. The waterway currently receives input
from over 50 storm drains and at least seven industrial discharges permitted
by the National Pollutant Discharge Elimination System (NPDES) (Tetra Tech
1985). Pulp and wood product industries have been present in Tacoma since
the late nineteenth century. The Simpson Tacoma Kraft pulp mill, which
5-88
-------
discharges about 750 m northwest of the mouth of City Waterway, began
primary effluent treatment in 1970 and secondary effluent treatment in 1977
(Tetra Tech 1985). Historically, at least six combined sewer overflows
drained into City Waterway. Between 1969 and 1979, the amount of sanitary
wastes discharged through these combined sewer overflows was progressively
reduced and then eliminated. However, effluent from the Central Waste Water
Treatment Plant is discharged to the Puyallup River, about 2 km above the
river's mouth (City of Tacoma 1983a). Also, effluent from the North End
Wastewater Treatment Plant is discharged at Ruston, along the southern
shoreline of Commencement Bay. Sub-surface flow from this area might reach
the mouth of City Waterway (City of Tacoma 1983b). Although organic
enrichment has caused sediments to become anoxic near the head of City
Waterway, anoxic sediments are less of a problem near the study area (Tetra
Tech 1985).
Environmental Conditions in the Study Area--
Mean salinity and water temperature values during the algal bloom
period are shown in Figure 5.21. Data are available from 1968 through
1986. All the data came from a single station, Ecology's Station CMB006.
There was a large gradient of salinity over depth, with a difference of
approximately 5.1 ppt between the surface and 10-m depth (Appendix E). The
magnitude of the salinity depth gradient probably reflects freshwater inputs
from the Puyallup River plume to the mouth of City Waterway (see above).
The depth gradient in water temperature was also well developed. The
average temperature at the surface was approximately 1.6° C higher than the
average temperature at 10-m depth. The relatively large depth gradients of
salinity and water temperature suggest that vertical mixing rates were low
in the study area. The low rate of vertical mixing presumably results
because the study area is sheltered from turbulence and because the rate of
circulation in the waterway is low (Dames and Moore 1981).
Depth gradients in the concentrations of dissolved oxygen and nutrients
in the City Waterway study area were less well developed than those in any
other study area in the northern or central sound (i.e., Bellingham Bay,
Port Gardner, Point Jefferson, and Sinclair Inlet). The mean concentrations
5-89
-------
of dissolved oxygen at the surface and at 10-m depth only differed by
0.3 mg/L, while the concentrations of nitrate and phosphate at the surface
and at 10-m depth only differed by 1.8 ug-at/L and 0.03 ug-at/L,
respectively. The poorly developed gradients appear to have resulted from
conditions at the surface in City Waterway. Surface concentrations of
dissolved oxygen were relatively low at this site, while surface nutrient
concentrations were relatively high. Evidently, intense algal blooms that
would increase surface dissolved oxygen concentrations and decrease surface
nutrient concentrations rarely developed in the City Waterway study area.
The interpretation that algal blooms were of low intensity in the City
Waterway study area is supported by the relatively low average percent
dissolved oxygen saturation at the surface (104.5 percent) (see Figure 5.24).
Also, the mean concentration of chlorophyll a (4.8 ug/L) was significantly
lower (PO.001) than the mean concentration reported for the Point Jefferson
study area (5.6 ug/L) (see Figure 5.23). High turbidity in City Waterway,
as indicated by the low mean Secchi disk depth (2.9 m) (Figure 5.24), may
have limited the depth of the photic zone such that intense algal blooms
could not develop.
The apparent geometric mean concentration of sulfite waste liquor at
the City Waterway site was low (4.7 Pearl Benson Index) (Figure 5.25).
Sulfite waste liquor was measured by Ecology using the Pearl Benson Index,
but kraft mills, such as the Simpson-Tacoma mill, do not release sulfite
waste liquor. However, the effluent from such mills contains substances
that are detected by the Pearl Benson Index (Felicetta and McCarthy 1963;
Henry, C., 17 November 1987, personal communication). Thus, the sulfite
waste liquor detected in City Waterway probably reflected the presence of
effluent from the Simpson-Tacoma mill.
The geometric mean concentration of fecal coliform bacteria in City
Waterway (13.8 organisms/100 ml) was the second highest of any study area in
the characterization study (see Figure 5.25 and Appendix E). Port Gardner
had a higher geometric mean. However, the fecal coliform values at Port
Gardner were probably inflated by high concentrations of Klebsiella from the
secondary treatment system of the Scott sulfite pulp mill, and were probably
5-90
-------
not indicative of sewage contamination. A similar elevation of Klebsiella
would not be expected from a kraft mill, such as the Simpson Tacoma mill.
It is possible that the fecal coliform bacteria in City Waterway came from
combined sewer overflows (see below).
Water Quality Trends in the Study Area—
A summary of comparisons between water quality data collected before
and after 1973 is given in Table 5.7. Slopes from statistically significant
long-term and recent regressions of the water quality data against year are
given in Table 5.8.
Physical Conditions—Plots of salinity and water temperature values by
year are shown in Figures 5.54 and 5.55. No temporal trends were detected
for either variable (Tables 5.7 and 5.8).
Dissolved Oxygen—Plots of dissolved oxygen concentration by year are
shown in Figure 5.56. There is no evidence that the Class B water quality
standard (see Table 4.2) was violated in the study area, although a few
values below the Class AA standard (7 mg/L) were detected prior to 1981. No
statistically significant temporal trends were detected in the concentrations
of dissolved oxygen.
Nutrients—Plots of nitrate concentrations by year are shown in
Figure 5.57. Increasing concentrations of nitrate were detected
statistically at 10-m depth (Table 5.8). However, it cannot be determined
whether this apparent increase was caused by an actual change in
environmental conditions. It is possible that the statistical increase in
nitrate concentrations at 10-m depth was driven by erratic variation in the
data, which included some low values near the beginning of the data set and
some high values near the end of the data set. Plots of phosphate
concentration by year are shown in Figure 5.58. A significant positive
slope (P<0.05) was detected both at the surface and at 10 m depth
(Table 5.8). Despite considerable scatter in the data, generally increasing
trends seem to be evident in the data.
5-91
-------
40-
30
a
a
t 20
z
10
o o
. 1 1 1 1 I— 1 1—
1950 1955 1960 1965 1970 1975 1980 1985
YEAR
1990
40-
30
a
a
Gee
z
_l
1/1
ANNUAL MEAN
J STANDARD ERROH
0 INDIVIDUAL OBSERVATION
-— SIGNIFICANT REGRESSION LINE
(P e 0.05)
1950 1955 1960 1965 1970 1975 1980 1985 1990
YEAR
Figure 5.54. Salinity values at the surface and at 10-m depth in the City Waterway study
area during the algal bloom season.
5-92
-------
23-
22
21
^191
•.IB-
S'17
< 14
£ 13
2 12
k-
11
10
9
8
7'
24
23
22
21 1
?19
•181
S1!?'
iis
a:
SI 13"
u 12
K 11
101
9
8
7-\
ANNUAL MEAN
J STANDARD ERROR
0 INDIVIDUAL OBSERVATION
SIGNIFICANT REGRESSION LINE
(P < 0.05)
°
1950 1955 1960 1965 1970 1975 1980 1985 1990
YEAR
o o
1950 1955 1960 1965 1970 1975 1980 1985 1990
YEAR
Figure 5.55. Water temperatures at the surface and at 10-m depth in the City Waterway
study area during the algal bloom season.
5-93
-------
20-
19
18
17'
cr is-
oiM-
Il3-
512
oll-
gl0-
Q 9
£ 8
o 7-
(/i c .
vi 6
Q 5
4 '
3
2
1
0 '
o
/ ° 1 °°
O' y/ \ 0 J\t/IsLl
f V^A/ ^
A « 1° j
* * *
1959 1955 1969 1965 1979 1975 1989 1985 1999
20 -
19'
18
17
18
~ is
Il3
z 12
0 11
x 10
O
Q 9
2 8
o 1
(n ,.
tfl O
0 5
4
3
2
1
a
YEAR
ANNUAL MEAN
I STANDARD ERROR
0 INDIVIDUAL OBSERVATION
— - SIGNIFICANT REGRESSION LINE
(P < 0.05)
O
O T
» I °
tfi j^ AtU4
X rj \ l\^!* T
rr\ 1 \ I **!
iN4 \/
«
1950 1955 1969 1965 1979 1975 1989 1985 1999
YEAR
Figure 5.56. Concentrations of dissolved oxygen at the surface and at 10-m depth in
the City Waterway study area during the algal bloom season.
5-94
-------
!?30
o:
O
CC
O
z
o
10
vt
a
0
ANNUAL MEAN
J STANDARD ERROR
0 INDIVIDUAL OBSERVATION.
---- SIGNIFICANT REGRESSION LINE
(P < 0.05)
1950 1955 1960 1965 1970 1975 1980 1985 1990
YEAR
x.
'o
o:
i^
z
z 2e
o
o:
O
Q
LJ
> 10
O
VI
VI
Q
1950 1955 I960 1965 1970 1975
YEAR
1985 1990
Figure 5.57. Concentrations of dissolved inorganic nitrate at the surface and at 10-m
depth in the City Waterway study area during the algal bloom season.
5-95
-------
5 '
x.
O)
3
(A
O
0.
O
o
o
1
in 1
(/)
o
ANNUAL MEAN
I STANDARD ERROR
0 INDIVIDUAL OBSERVATION
—- SIGNIFICANT REGRESSION LINE
(P < 0.05)
0 O
004
T 1
1950 1955
—i 1—
1965 1970
YEAR
—i 1 1 r
1975 1989 1985 1990
o>
3
o
I
CL
O
o
o
UJ
>
1950 1955 1960 1965 1970 1975 1980 1985 1990
YEAR
Figure 5.58. Concentrations of dissolved orthophosphate at the surface and at 10-m
depth in the City Waterway study area during the algal bloom season.
5-96
-------
The cause(s) of the apparent increases in nutrient concentrations in
the City Waterway study area are unknown. Possible contributing factors
include increased nutrient loadings in runoff and in effluent from the
Tacoma Central and North End Wastewater Treatment Plants. Also, the
Simpson-Tacoma pulp mill has added phosphoric acid to their effluent since
secondary effluent treatment was instituted in 1977 (Henry, C., 17 November
1987, personal communication). The increases in phosphate concentrations
detected in City Waterway might have been influenced by this practice.
Information to test these hypotheses is not available for this characteriza-
tion study.
Indicators of Phvtoplankton Growth—Chlorophyll a concentrations at the
surface and 10-m depth are plotted by year in Figure 5.59. Plots of
percent dissolved oxygen saturation at the surface and Secchi disk depth are
plotted by year in Figure 5.60. No significant temporal trends were detected
for any of these variables. Data on the concentration of chlorophyll a. are
only available since 1982. Although the changes in chlorophyll a
concentrations were not statistically significant (P>0.05), high concentra-
tions (up to 19 ug/L) were recorded at the surface in 1986. However,
percent dissolved oxygerr saturation at the surface and Secchi disk depth did
not appear to be affected by the high concentrations of chlorophyll a in
1986. It appears that the elevation of chlorophyll a did not affect
transparency or photosynthetic production of oxygen, at least at the time of
sampling.
Pollutants.--Plots of sulfite waste liquor concentration by year are
shown in Figure 5.61. Statistically significant declines (P<0.05) were
detected at both the surface and at 10-m depth (see Tables 5.7 and 5.8).
As discussed above, kraft pulp mills do not discharge sulfite waste liquor.
However, the Pearl Benson Index, which is used to detect sulfite waste
liquor, also detects kraft wastes. Therefore, the declines in the Pearl
Benson Index probably reflect the declines in waste discharges that occurred
when the Simpson Tacoma Kraft plant adopted secondary waste treatment in
1977.
5-97
-------
ANNUAL MEAN
J STANDARD ERROR
0 INDIVIDUAL OBSERVATION
—- SIGNIFICANT REGRESSION LINE
(P < 0.05)
o.
o
K
o
1959 1955 196* 1965
197*
YEAR
1975 19M 1985 1999
01
d
>•
0.
1959 1955 1969 1965
1979
YEAR
1975 19M 1985 199*
Figure 5.59. Concentrations of chlorophyll a at the surface and at 10-m depth in the
City Waterway study area during the algal bloom season.
5-98
-------
300-
2 300
a:
3
(/I
o
O 100
o
ui
_J
o
V)
o
o »
ANNUAL MEAN
J STANDARD ERROR
O INDIVIDUAL OBSERVATION
SIGNIFICANT REGRESSION LINE
(P < 0.05)
1950 1955 I960 1965 1970 1975 1980 1985 1990
YEAR
16
14"
12"
*•—\
J101
Q.
UJ
Q 8
o
x 6
o
u
0
o o o
0 O O
1950 1955 I960
1965 1970 1975 1980 1985 1990
YEAR
Figure 5.60. Percent dissolved oxygen saturation at the surface and Secchi disk depth
in the City Waterway study area during the algal bloom season.
5-99
-------
cc
o
g
li
Ujg
t- 0)
m_ CD
O
O
3-
T
0 OO
T
~r
1 1 1 1—
1956 1955 1966 1965 1976 1975 1986 1985 1996
YEAR
CC
O
=)
Uj 4)
fel
Q. CD
Ss
o
o
3
1950 1955 1966 1965 1976 1975 1986 1985 1996
YEAR
Figure 5.61. Log of concentrations of sulfite waste liquor at the surface and at 10-m depth
in the City Waterway study area during the algal bloom season.
5-100
-------
Concentrations of fecal coliform bacteria are plotted by year in
Figure. 5.62. No statistically significant (P>0.05) changes were detected.
However, a few high concentrations (up to 1,000 organisms/100 ml) were
recorded before 1981, some of which violated Class B water quality standards
(Table 4.2). The absence of high' concentrations of fecal coliform bacteria
since 1981 may reflect the cessation of discharges of raw sewage through
combined sewer overflows into City Waterway in 1979. In addition,
improvements in the chlorination facilities at Tacoma's North End and
Central Wastewater Treatment Plants were completed in 1982. The North End
plant discharges along the southern shoreline of Commencement Bay at Ruston,
while the Central plant discharges 2-km upstream in the Puyallup River
(City of Tacoma 1983a,b).
Summary of Results for the Central Sound
This section summarizes the major findings of this report for the
central sound. Environmental conditions in the study areas are summarized
and compared. A brief assessment of the sensitivity of the central sound
study areas to pollution is provided. Temporal trends in water quality are
also summarized.
Environmental Conditions--
Salinity depth gradients were well developed in the study areas that
have substantial sources of fresh water: Port Gardner and City Waterway.
Salinity values at 10-m depth were similar in all four study areas
(approximately 28.3 ppt). Substantial depth gradients of water temperature
were present in all the central sound study sites. The thermal gradient was
least developed at Point Jefferson, where vertical mixing rates were
probably highest. Mean temperatures at both the surface (14.5° C) and at
10-m depth (12.8° C) were highest in the Sinclair Inlet site. The large
thermal depth gradients at Sinclair Inlet and City Waterway suggest that
mixing rates were lowest in those two areas.
Depth gradients of dissolved oxygen concentrations reflected differences
in photosynthetic enhancement of dissolved oxygen in near-surface waters.
5-101
-------
EC
HI
CO
II
§•
O .-
o«
_J .
< O
oz
~~
u.
O
ANNUAL MEAN
J STANDARD ERROR
0 INDIVIDUAL OBSERVATION
SIGNIFICANT REGRESSION LINE
(P c 0.05)
OO O
1 f 1 1 1—
195« 1955 I960 1965 1979 1975 1980 1985 199«
YEAR
Figure 5.62. Log of concentrations of fecal coliform bacteria at the surface in the City
Waterway study area during the algal bloom season.
5-102
-------
The largest and smallest depth gradients of dissolved oxygen concentrations
were detected at the Sinclair Inlet and City Waterway study areas,
respectively. The differences between mean dissolved oxygen concentrations
at the surface and at 10-m depth were 2.5 mg/L at the Sinclair Inlet site
and 0.4 mg/L at the City Waterway site. The mean concentration of dissolved
oxygen at 10-m depth was elevated at Point Jefferson (9.7 mg/L), possibly
because the clarity of the water column was sufficient to allow substantial
photosynthesis to occur at this depth.
Extremely low dissolved oxygen concentrations at depth were rarely
observed. Dissolved oxygen concentrations averaged over 7 mg/L down to
200-m depth in the Point Jefferson area. Unfortunately, data from water
deeper than 30 m were not available from the other central sound study
areas. Problems with low dissolved oxygen concentrations at depth could have
occurred in any of those areas. Sinclair Inlet had intense blooms, with low
flushing and circulation rates. Die-off and decay of algal blooms could
cause problems with low dissolved oxygen concentrations at depth in this
area. City Waterway has low flushing rates and anoxic sediments near its
head. Oxygen-demanding wastes could accumulate on the bottom of City
Waterway, causing problems with low dissolved oxygen concentrations at
depth. Port Gardner has somewhat better flushing than do Sinclair Inlet and
City Waterway, but circulation along the bottom is slow, and large accumula-
tions of organic matter from log yards and discharges from pulp mills have
been found in the area in the past. Therefore, problems with low dissolved
oxygen concentrations at depth also could occur in this area.
Nutrient data are available from the Port Gardner, Sinclair Inlet, and
City Waterway study areas. Mean concentrations of nitrate were much lower
in Sinclair Inlet than at the other sites, especially at the surface (e.g.,
less than one-fifth of the concentration of the City Waterway site).
Geographic variation in phosphate concentrations was less conspicuous than
the geographic variation in nitrate concentrations. The lower concen-
trations of nitrate in the Sinclair Inlet study area may have been due to
the higher intensity of the algal blooms in this area (see below). The
paucity of freshwater sources that drain into Sinclair Inlet may also have
influenced the nitrate concentrations because rivers are a major source of
5-103
-------
nitrogen for Puget Sound (Robinson and Brown 1983). The largest depth
gradients in phosphate concentrations were found at the Sinclair Inlet and
Port Gardner sites. Phytoplankton blooms probably lowered phosphate
concentrations near the surface, especially in Sinclair Inlet. The surface
nutrient concentrations at the Port Gardner site were probably influenced by
flows from the Snohomish River.
Based on the percent dissolved oxygen saturation in surface water, the
intensity of algal blooms was greatest at the Sinclair Inlet site.
Phytoplankton blooms at Point Jefferson were also well developed, while the
blooms at Port Gardner and City Waterway were less intense. The limited
chlorophyll a data indicated that these concentrations were higher in the
Point Jefferson study area than in the City Waterway study area.
Because the transparency of surface water can be affected by phytoplank-
ton density and by the concentration of suspended particulate material,
geographic variation in Secchi disk depths was not consistent with the above
interpretation of geographic variation in the intensity of algal blooms.
Mean Secchi disk depths were lowest at the Port Gardner and City Waterway
sites (<3 m), where blooms appeared to be least developed. Presumably, the
Secchi disk depths in these two areas were influenced by suspended
particulate material from the Snohomish and Puyallup Rivers. The high mean
Secchi disk depth at Point Jefferson (4.7 m) probably reflected the absence
of nearby large sources of fresh water and suspended particulates, rather
than the absence of algal blooms, because Secchi disk depths and chlorophyll
a concentrations were negatively correlated (P<0.05) in this area. Compared
with the other central sound study areas, Secchi disk depths were
intermediate in the Sinclair Inlet study area. This result also probably
reflects the lack of sources of suspended particulate material from rivers
and streams, rather than the intensity of algal blooms in the area.
Sensitivity to Nutrient Enrichment--
Based on inherent limitations ,in its capacity to export or assimilate
pollutants without deleterious ecological effects, Sinclair Inlet is
probably the most sensitive of the central sound study areas to impacts
5-104
-------
from excess nutrients. The intense blooms and low nitrate concentrations
that occur at the Sinclair Inlet site suggest that further nutrient enrich-
ment could increase the intensity of the algal blooms in the area. Because
the flushing rate of Sinclair Inlet is quite low, the rate of export of
pollutants from Sinclair Inlet also would be low.
City Waterway also may appear to be sensitive to the enhancement of
algal blooms by nutrient enrichment, given its small volume and low flushing
rate. However, intense blooms did not appear to develop in City Waterway.
Bloom intensity in the study area may be limited by some factor other than
nutrient concentrations, such as turbidity.
Point Jefferson probably has the best capacity of any central sound
study area to export or assimilate nutrient inputs. The volume of water in
this area and the lack of restrictions on water movements both facilitate
assimilation (by dilution) and export of nutrients. However, because
phytoplankton blooms were moderately well developed in the Point Jefferson
study area, nutrient enrichment in this area might enhance bloom intensity.
Unfortunately, sufficient nutrient data are not available from Point
Jefferson site to determine whether low nutrient concentrations occurred
during phytoplankton blooms in this area.
Based on both the volume and the exchange of water, the Port Gardner
area appears to be less sensitive to the effects of nutrient enrichment
than are the Sinclair Inlet and City Waterway areas. However, the Port
Gardner area appears to be more sensitive to nutrient enrichment than the
Point Jefferson area. Intense phytoplankton blooms did not appear to be
prevalent in the Port Gardner area.
Trends in Water Quality--
The interpretation of the statistical data is summarized in Tables 5.7
and 5.8. A few interpretable patterns of environmental change were evident
in the individual study areas within the central sound. Problems in
interpretation caused by changes in station location and data sources only
occurred in the Port Gardner study area. The most readily detected trends
5-105
-------
were changes in the concentrations of sulfite waste liquor in study areas
near sulfite pulp mills.
Physical Conditions—The temporal trends in salinity values that were
detected in the Port Gardner study area may have been artifacts of changed
station locations. Average surface salinity values in the Point Jefferson
study area appear to have been slightly higher since 1973 than from 1932
through 1972. However, subsurface salinity values appear to have been
decreasing in the Point Jefferson area. The salinity data collected at the
Point Jefferson area could have been affected by actual environmental
changes and by artifacts in the data caused by differences in the analytical
procedures used by the University of Washington and Metro. Salinity changes
were not detected at the Sinclair Inlet or City Waterway sites, where data
are available only since the late 1960s.
Temporal changes in water temperature at Port Gardner also appear to
have been caused by changes in station locations. At Point Jefferson,
surface temperature appears to have decreased slightly since 1973. However,
increased temperatures have been detected at depths from 10 to 100 m since
1932. These temperature increases at depth at Point Jefferson may have been
influenced by the cool period that occurred in the late 1940s and early
1950s.
Dissolved Qxvaen--There was no substantial evidence that dissolved
oxygen concentrations have changed in the central sound study areas. The
only statistically significant (PO.05) trend was a decline in the dissolved
oxygen concentration at the surface in the Point Jefferson study area since
1973. However, this decline appears to have been caused by erratic
variations in dissolved oxygen concentrations that by chance included some
high values in the mid-1970s and some low values in 1986.
Nutrients—Substantial changes in nitrate concentrations were apparent
in the Port Gardner and City Waterway study areas. (Unfortunately, nutrient
data are not available from the Point Jefferson study area.) The apparent
increase in nitrate concentrations at 10-m depth in City Waterway may be
attributable to increased nutrient inputs. Alternatively, this increase may
5-106
-------
be attributable to erratic fluctuations in the data that introduced an
apparent change into the data that did not reflect an underlying change in
the environment. No explanation for the decline in nitrate concentrations at
10-m depth in the Port Gardner study area is known. This decline probably
was not an artifact of changes in station location.
Decreases in phosphate concentrations since the 1950s were detected in
the Port Gardner study area, the only central sound study area for which
long-term data on phosphate concentrations are available. Increases in
phosphate concentrations since the mid-1970s were detected in the Port
Gardner, Sinclair Inlet, and City Waterway study areas. These three sites
are in urban areas, suggesting that anthropogenic factors may have
influenced the data. However, oceanic influences cannot be ruled out. The
above study sites may have been affected by changes in nutrient inputs from
point sources or runoff. Increased phosphate concentrations at Port Gardner
may have been influenced by reductions in emissions of sulfite waste liquor
in the area. At the City Waterway site, the apparent increase in phosphate
concentrations might have reflected additions of phosphoric acid to the
secondary effluent discharged by the Simpson-Tacoma Kraft mill. This mill
adopted secondary effluent treatment in 1977. However, testing the above
hypotheses was beyond the scope of this study. No explanation involving a
point source is apparent for increased phosphate concentrations in the
Sinclair Inlet study area.
Indicators of Phvtoplankton Growth—Data on chlorophyll a concentra-
tions are available for Point Jefferson from 1966 through 1975 and for City
Waterway from 1982 through 1986. No temporal trends in chlorophyll a
concentrations were detected at either site. At Point Jefferson, percent
dissolved oxygen saturation at the surface has declined recently, while
Secchi disk depths have increased recently. These results suggest that
algal production has declined in the Point Jefferson area. However, these
temporal changes in oxygen saturation and Secchi disk depth were not well
developed in the data. Thus, the determination of whether the apparent
decline in primary production at Point Jefferson reflected short-term
variation within the normal range, or the beginning of a long-term trend
must await future studies.
5-107
-------
Pollutants—Declines in the concentrations of sulfite waste liquor were
detected at Port Gardner and at the mouth of City Waterway. Data on this
variable are not available from Point Jefferson. The decline in the Port
Gardner study area appeared to coincide with improvements in effluent
treatment by the local pulp mills. The sulfite waste liquor declines
detected in City Waterway could have been related to improvements in the
effluent treatment by the Simpson-Tacoma Kraft mill. This mill has never
discharged sulfite waste liquor. However, kraft effluent from this mill may
contain material that is detected by the Pearl Benson Index test.
The only statistically significant (PO.05) change that was detected in
concentrations of fecal coliform bacteria was an apparent increase that
occurred in the Port Gardner study area since 1981. This increase coincided
with the initiation of secondary effluent treatment by the Scott sulfite
pulp mill at Port Gardner. The organism detected in the fecal coliform
tests in this area probably was the bacterium, Klebsiella. This organism
grows rapidly in the secondary effluent treatment facilities of sulfite pulp
mills. Thus, the apparent increases in the concentrations of fecal coliform
bacteria in the Port Gardner area probably reflected secondary treatment by
the Scott mill, rather than increased contamination from sewage effluent.
SOUTHERN SOUND
The South Sound is defined herein as all of Puget Sound upstream of
Tacoma Narrows (see Figure 2.1). This region of Puget Sound (exclusive of
Hood Canal) is the most removed from direct oceanic influences. Most of the
region is relatively shallow and poorly flushed. Numerous shallow embayments
and large islands are present. The southern sound contains 16 percent of
the surface area, 29 percent of the shoreline, and 21 percent of the
tidelands of Puget Sound, but only 9 percent of the volume of Puget Sound
south of Admiralty Inlet (Burns 1985). Population centers are the Cities of
Olympia and Shelton, located on Budd Inlet and Oakland Bay, respectively.
Most of the remaining southern sound region is sparsely populated.
5-108
-------
Five study areas were located in the South Sound: Carr Inlet, Nisqually
Reach, Budd Inlet, Totten Inlet, and Oakland Bay. Station locations are
shown in Figure 5.63; data sources are given in Table 5.9. Algal bloom
seasons for the study sites are given in Table 5.10. Histograms summarizing
the water quality variables are given in Figures 5.64-5.68. Back-up tables
of the summary data are given in Appendix E. The ANOVAs comparing the water
quality variables before and after 1973 are summarized in Table 5.11.
Long-term and recent regressions are summarized in Table 5.12.
With the exception of Nisqually Reach, the study areas in the southern
sound are located in sheltered embayments. Because of a limited capacity to
assimilate or export contaminants, these areas may be vulnerable to
deleterious effects of pollution. Carr Inlet is relatively deep, averaging
about 92 m deep. Budd Inlet, Totten Inlet, and Oakland Bay are shallower
than Carr Inlet. The depth of Oakland Bay is less than 5 m over much of its
area. Although Nisqually Reach is in the main channel of the South Basin,
it is near a sill that is about 36 m deep. Circulation is sluggish in the
four embayments, but it is more rapid and turbulent at Nisqually Reach.
Based on the percent dissolved oxygen saturation at the surface, algal
blooms were most prevalent in the southern sound study areas from May through
August. However, in Oakland Bay the algal blooms were best developed from
April through June (see Table 5.10). Algal blooms appear to have been more
intense in Carr, Budd, and Totten Inlets, and less intense in Oakland Bay
and Nisqually Reach.
Carr Inlet
The study area is located approximately half way up the axis of Carr
Inlet, off Green Point (Figure 5.63). The region is rural, and sometimes
serves as a reference area for studies of contaminated urban bays (e.g.,
Tetra Tech 1985). Class AA water quality standards apply in the area
(Table 4.2). Because Carr Inlet is a deep (approximately 92 m) embayment,
tidal flushing is slower than in the shallower southern embayments, such as
Budd Inlet (URS 1986b). The study area has no nearby source of fresh water.
Net current velocities are low in the study area (e.g., 0.6 cm/sec at 5-m
5-109
-------
OAK 484
OAK 023
OAK 485
OAK 001
OAK 002
OAK 003
OAK 004
CRR421
CRR 419
CRR001
CRR 415
CRR 416
CRR 417
CRR 418
Oakland
Bay
TOT 472
TOT 001
Totten
Inlet
Gok/sbomugh Creek
NSQ406
NSQ001
Nisqually
Reach
BUD 006
BUD 004
BUD 003
BUD 463
W£ Olympia 122 50
Figure 5.63. Locations of study areas and sampling stations in the southern sound.
5-110
-------
TABLE 5.9. SAMPLING STATION NUMBERS, DATA SOURCES, AND TIME PERIODS
FOR THE STUDY AREAS IN THE SOUTHERN SOUND
Study Area
Carr Inlet
Nisqually Reach
Budd Inlet
Totten Inlet
Oakland Bay
Station
Number
CRR415
CRR416
CRR417
CRR418
CRR419
CRR421
CRR001
NSQ406
NSQ001
BUD463
BUD003
BUD004
BUD005
TOT472
TOT001
OAK484
OAK485
OAK001
OAK002
OAK003
OAK004
23
Data
Source
uwa
UW
UW
UW
UW
UW
Ecology
UW
Ecology
UW
Ecology
Ecology
Ecology
UW
Ecology
UW
UW
Ecology
Ecology
Ecology
Ecology
WDF
Sampl
1954-55
1954-62
1954-55
1954-55
1935-41,
1953-62
1967-70,
1932-41,
1967-70,
1957-58
1967-70,
1967-70,
1967-70,
1956-60
1967-70,
1956-57
1956-58
1967-70
1967-70
1967-70
1967-70,
1964-71
ing Period
1950-67
1977-86
1949-62
1977-86
1973-77
1976-77
1973-86
(includes data
1977-86
(includes data
(includes data
1975, 1978-86
from WDFb)
from WDF)
from WDF)
a UW = University of Washington.
b WDF = Washington Department of Fisheries.
5-111
-------
TABLE 5.10. ALGAL BLOOM SEASONS FOR THE SOUTHERN SOUND STUDY AREAS, AS
DEFINED BY MONTHLY MEAN AND STANDARD ERROR OF PERCENT
DISSOLVED OXYGEN SATURATION IN SURFACE WATER
Percent Dissolved
Month
April
May
June
July
August
September
Carr
Inlet
105 +/-
141 +/-
121 +/-
126 +/-
123 +/-
116 +/-
3
5a
4a
4a
7a
5
Nisqually
Reach
96 +/-
107 +/-
103 +/-
103 +/-
117 +/-
91 +/-
3
4a
2a
3a
8a
4
Oxyqen Saturation
Budd
Inlet
107
131
115
117
121
111
+/-
V-
+/-
+/-
+/-
+/-
2
8a
5a
6a
6a
10
Totten
Inlet
102 +/-
121 +/-
114 +/-
113 +/-
117 +/-
106 +/-
Oakland
Bay
2
4a
4a
3a
4a
3
102 +/-
105 +/-
104 +/-
96 +/-
92 +/-
83 +/-
la
3a
6a
4
3
3
a Months included in the algal bloom season.
5-112
-------
301
7/1
CARR
INLET
18 36 6 16 36 6 16 36
DEPTH (m)
NISQUALLY
REACH
BUDO
INLET
STUDY AREA
TOTTEN
INLET
6 16 36
OAKLAND
BAY
o
I
I
6 16 36 6 16 36 8 16 36 6 16 36
DEPTH (m)
CARR
INLET
NISQUALLY
REACH
BUDO
INLET
STUDY AREA
TOTTEN
INLET
6 16 36
OAKLAND
BAY
Figure 5.64. Mean salinity and water temperature values in the southern sound study
areas during the algal bloom season.
5-113
-------
CARR
INLET
e le 39
DEPTH (m)
e le ae e 10 30
NISQUALLY
REACH
BUDD
INLET
STUDY AREA
TOTTEN
INLET
OAKLAND
BAY
0 10 30 e le 30 e le 30 e le 30 e 10 30
DEPTH (m)
CARR
INLET
NISQUALLY
REACH
BUDD
INLET
STUDY AREA
TOTTEN
INLET
OAKLAND
BAY
Figure 5.65. Mean concentrations of dissolved oxygen and dissolved inorganic nitrate
in the southern sound study areas during the algal bloom season.
5-114
-------
HI
0 10 30 0 10 30 0 10 30 0 10 30
DEPTH (m)
CARR
INLET
NISQUALLY
REACH
BUDD
INLET
TOTTEN
INLET
OAKLAND
BAY
STUDY AREA
(01
10
I S
Sf
cc 3>
o
_i
x
o
0
0 10 30
0 10 30 0 10 30 0 10 30
DEPTH (m)
CARR
INLET
NISQUALLY BUDO TOTTEN
REACH INLET INLET
STUDY AREA
0 10 30
OAKLAND
BAY
Figure 5.66. Mean concentrations of dissolved orthophosphate and chlorophyll a. in
the southern sound study areas during the algal bloom season.
5-115
-------
UJ fl>
Ss
>-«
x Q-
o
si
>l-
-J<
ooc
COD
tOH
5<
CO
0
1
I
I
I
I
CARR
INLET
BUDO
INLET
NISQUALLY
REACH
OAKLAND
BAY
TOTTEN
INLET
STUDY AREA
~ 8
E
UJ
Q
to
a
o
UJ
CO
0
1
CARR
INLET
BUDO
INLET
NISQUALLY
REACH
OAKLAND
BAY
TOTTEN
INLET
STUDY AREA
Figure 5.67. Mean percent dissolved oxygen saturation at the surface and Secchi disk
depth in the southern sound study areas during the algal bloom season.
5-116
-------
H«I
Egl
= 2g
«j»
ssi
e2c
LOG OF FECAL
COLIFORM BACTERI
(N 100 mL)
per
I
0 10 30 0 10 30 0 10 30 0 10 30 0 10 30
DEPTH (m)
CARR
INLET
NISQUALLY
REACH
BUDO
INLET
TOTTEN
INLET
OAKLAND
BAY
STUDY AREA
e 10 3e e le 30 0 10 30 e ie 30 e 10 30
DEPTH (m)
CARR
INLET
NISQUALLY
REACH
BUDO
INLET
STUDY AREA
TOTTEN
INLET
OAKLAND
BAY
Figure 5.68. Log of geometric mean concentrations of sulfite waste liquor and fecal coliform
bacteria in the southern sound study areas during the algal bloom season.
5-117
-------
TABLE 5.11. NET CHANGE AND PERCENT CHANGE IN THE MEAN VALUES OF WATER QUALITY
VARIABLES IN THE SOUTHERN SOUND, BASED ON ANOVA COMPARISONS OF DATA
TAKEN BEFORE 1973 WITH DATA TAKEN FROM 1973 TO 1986
Depth
(m)
0
10
30
0
10
30
0
10
30
0
10
30
0
10
30
0
10
30
Cam Inlet
Change
Net Percent
-0.39 1.4
-0.38 1.3
-0.42 1.4
+1.51 11.4
+1.12 10.1
+1.16 11.3
NS
NS
NS
na
na
na
NS
-0.23 15.5
-0.61 31.9
na
na
na
Nisquatty Reach Budd Inlet
Change Change
Net Percent Net Percent
-1.49 5.5
-0.50 1.7
na
+1.14 9.4
+0.90 8.0
na
NS
+0.67 7.8
na
na
na
na
NS
NS
na
na
na
na
Salinity (ppt)
NSf*
nab
na
Water Temperature (° C)
-1.97 11.6
na
na
Dissolved Oxygen (mg/L)
+1.39 16.2
na
na
Nitrate (ug-at/L)
na
na
na
Phosphate (ug-at/L)
-0.68 32.5
na
na
Chlorophyll a (ug/L)
na
na
na
Totten Inlet
Change
Net Percent
NS
NS
na
NS
NS
na
NS
NS
na
na
na
na
-0.54 32.5
-0.48 27.6
na
na
na
na
Oakland Bay
Change
Net Percent
NS
na
na
NS
na
na
+1.80 23.1
na
na
NS
na
na
-0.60 36.
na
na
na
na
na
2
NS
Surface Dissolved Oxygen Saturation (Percent)
+6.87 6.7 NS NS
Seechi Disk Depth (m)
na NS NS
Sulfite Waste Liquor (Pearl Benson Index)
+23.24 27.3
NS
0
10
30
0
10
30
na
na
na
na
na
na
na
na
na
na
na
na
NS
na
na
Fecal Coliform Bacteria (No./100 mL)
na
na
na
na
r\&
na
na
na
na
NS
na
na
na
na
na
a NS = The pre-1973 and 1973-1986 values were not significantly different at P<0.05, based on a nonparametric
one-way ANOVA.
b na Results of the statistical test were not available because of a lack of data.
5-118
-------
TABLE 5.12. SLOPES OF STATISTICALLY SIGNIFICANT LONG-TERM AND RECENT REGRESSIONS
Of WATER QUALITY VARIABLES AS A FUNCTION OF YEAR FOR THE SOUTHERN SOUND
Depth
(m)
0
10
30
0
10
30
0
10
30
0
10
30
0
10
30
0
10
30
Cam Inlet
Long-term
-0.017
-0.014
-0.016
0.050
0.031
0.03.2
NS
NS
0.024
na
na
na
NS
NS
-0.020
na
na
na
Recent
-0.067
NS
NS
NS
NS
NS
0.291
0.306
0.303
-0.687
NS
NS
NS
NS
NS
na
na
na
Ni squat ly
Long-term
-0.046
-0.035
nab
NS
NS
na
NS
0.022
na
na
na
na
NS
NS
na
na
na
na
S I opes
Reach Budd Inlet
Recent Long-term Recent
Salinity (ppt)
NSa NS 0.242
NS NS NS
na na na
Water Temperature (° C)
NS NS NS
NS NS NS
na na na
Dissolved Oxygen (mg/L)
0.267 0.105 0.245
NS NS NS
na na na
Nitrate (ug-at/L)
NS na NS
NS na NS
na na na
Phosphate (ug-at/L)
NS NS NS
NS na NS
na na na
Chlorophyll a (ug/L)
na na na
na na na
na na na
Totten
Long-term
-0.013
NS
na
NS
NS
na
NS
NS
na
na
na
na
-0.021
-0.018
na
na
na
na
Inlet
Recent
NS
NS
na
NS
NS
na
NS
0.285
na
NS
NS
na
NS
NS
na
na
na
na
Oakland
Long-term
0.131
na
na
NS
na
na
0.135
na
na
NS
na
na
-0.023
na
na
NS
na
na
Bay
Recent
NS
NS
na
NS
NS
na
NS
NS
na
NS
NS
na
0.051
0.065
na
na
na
na
Surface Dissolved Oxygen Saturation (Percent)
NS 3.457 NS 2.921 1.190 3.154 NS NS 1.565 NS
Seechi Disk Depth (m)
NS -0.481 na NS NS 0.177 NS NS NS 0.117
Sulfite Waste Liquor0 (Pearl Benson Index)
0
10
30
0
10
30
na
na
na
na
na
na
na
na
na
-0.019
na
na
na
na
na
na
na
na
na
na
na
Fecal Col i form
-0.038
na
na
NS
NS
na
Bacteria
na
na
na
NS
NS
na
(NO./100 mL)
-0.095
na
na
NS
na
na
na
na
na
na
na
na
NS
na
na
-0.041
na
na
na
na
na
na
na
na
NS
na
na
a NS = Not significant at P<0.05.
na = Results of the statistical test were not available because of a lack of data.
c Data were subjected to a log(X+1) transformation for the regressions.
"Data were subjected to a log transformation for the regressions.
5-119
-------
depth) (NOAA 1984b). Current velocities in other regions of Carr Inlet,
where tidal flows are restricted (e.g., Hale Passage), are 1 order of
magnitude higher than those in the study area. The Nisqually River
contributes approximately 5 percent of the annual river flow into Puget
Sound and is the largest river entering the southern sound. The Nisqually
River discharges about 14 km southwest of the mouth of Carr Inlet. When
large volumes are discharged due to snowmelt during the late spring (USGS
1985), southerly winds may occasionally force Nisqually River water northward
into Carr Inlet (Duxbury, A.C., 15 October 1987, personal communication).
Environmental Conditions in the Study Area--
Mean salinity and water temperature values during the algal bloom
period are depicted in Figure 5.64. Data are available from 1950 through
1986, with the best coverage during the mid-1950s and from 1977 through
1986. Substantial vertical stratification of water temperature was evident,
as indicated by the temperature gradient (mean water temperature value was
approximately 2.3° C higher at the surface than at 10-m depth). The vertical
salinity gradient was small (mean salinity value was approximately 0.2 ppt
lower at the surface than at 10-m depth), presumably because no substantial
source of fresh water is near the study area. The steepness of the vertical
gradient of water temperature.suggests that rates of vertical mixing are low
in the study area. The stability of the water column during the bloom
season suggests that algal blooms could become well developed in Carr Inlet.
The vertical distribution of concentrations of dissolved oxygen and
nutrients in Carr Inlet appears to have been strongly influenced by stability
of the water column during the algal bloom season. Mean values of dissolved
oxygen concentrations were 11.1 mg/L at the surface and 9.6 mg/L at 10-m
depth (Figure 5.65). Dissolved oxygen concentrations were above 100 percent
saturation at both the surface and 10-m depth (Figure 5.67 and Appendix E).
Nutrient concentrations were much lower at the surface than at 10- or 30-m
depths (e.g., mean values of nitrate concentrations were 3.6 ug-at/L at the
surface and 10.7 ug-at/L at 10-m depth) (Figures 5.65 and 5.66). The
negative correlations between nutrient concentrations and percent dissolved
oxygen saturation presumably were due to the enhanced uptake of nutrients in
5-120
-------
the surface waters during algal blooms, and the reduced uptake of nutrients
at depth where algal blooms were not well developed. Positive correlations
between phosphate concentrations and salinity values at the surface and at
30-m depth may have been caused by the following two factors: 1) high
phosphate concentrations can occur during occasional periods of high
salinity when the water column is unstable and deeper, high-phosphate water
is advected toward the surface, and 2) low phosphate concentrations can
occur during algal blooms, when density stratification causes vertical
exchange rates to be low, allowing near-surface nutrient concentrations to
fall. Occasional upwelling may also explain the negative correlations
between dissolved oxygen concentrations and both salinity and water
temperature values at 10- and 30-m depths (i.e., upwelled water would be
high in salinity and phosphate, but low in temperature and dissolved
oxygen).
High values for percent saturation of dissolved oxygen in the surface
waters support the interpretation that intense algal blooms occurred in the
Carr Inlet study area (Figure 5.67). Based on a very limited data set,
chlorophyll a concentrations appear to have been highest at 10-m depth
(Figure 5.66). Mean Secchi disk depth was over 6 m (Figure 5.67), which also
suggests that high phytoplankton concentrations occurred below the surface.
The data for mean Secchi disk depth also suggest that the photic zone
averaged over 12 m deep (Preisendorfer 1986). Although the negative
correlation (r=-0.52) between Secchi disk depth and percent dissolved
oxygen saturation at the surface was not significant (P=0.07) when scaled
with the Bonferroni inequality, the magnitude of this correlation coefficient
suggests that Secchi disk depths were influenced by the intensities of
algal blooms.
As would be expected for a rural site, the concentrations of the
pollutants analyzed in this study were low in the study area (Figure 5.68).
The geometric mean value for the concentration of sulfite waste liquor at
the surface was only 3.6 (Pearl Benson Index). The geometric mean value for
the concentration of fecal coliform bacteria at the surface was
1.1 organism/100 ml. This mean value is only 10 percent above the analytical
detection limit.
5-121
-------
Water Quality Trends in the Study Area—
A summary of comparisons between water quality data collected before and
after 1973 is given in Table 5.11. Slopes from statistically significant
long-term and recent regressions of the water quality data against year are
given in Table 5.12.
Physical Conditions — Plots of salinity and water temperature values by
year are shown in Figures 5.69-5.71. Significant declines (p<0.05) in
salinity values were detected at the surface, and at 10- and 30-m depths
(Tables 5.11 and 5.12). The long-term trend in surface salinity values, a
decline of about 0.61 ppt over the period of 1950-1986, appears to have been
driven by the decline of about 0.60 ppt that occurred over the period of
1977 to 1986. This recent decline was detected using data from only one
sampling station, and could not have been an artifact of changing station
locations.
Significant increases (P<0.05) in water temperature values were
detected at all depths (Tables 5.11 and 5.12). These increases appear to
have coincided with the pattern of climatic change evident in the climate
data collected at the Seattle-Tacoma Airport. The data set for Carr Inlet
begins in 1950, during a cool period (see Figure 5.1).
Dissolved Oxygen—Plots of dissolved oxygen concentration by year are
shown in Figures 5.72 and 5.73. The Class AA water quality standard (see
Table 4.2) was always met in the surface waters. However, violations at
10-m depth occurred during two years in the 1950s, one year in the 1960s, and
one year in the 1980s. Violations at 30-m depth occurred during four years
in the 1950s, one year each in the 1960s and 1970s, and two years in the
1980s. Dissolved oxygen concentrations do not appear to have changed
substantially in Carr Inlet from 1950 through 1986 because no significant
differences were detected between dissolved oxygen concentrations recorded
before and after 1973 (see Table 5.11). The positive slopes of the long-term
regression at 30-m depth and the recent regressions at 0-, 10-, and 30-m
depths may have been influenced by variations in dissolved oxygen
5-122
-------
40-
30
Q.
a
s™^
b
z
_i
ie
0
ANNUAL MEAN
I STANDARD ERROR
O INDIVIDUAL OBSERVATION
SIGNIFICANT REGRESSION LINE
(P < 0.05)
1959 1955 I960 1965
197*
YEAR
1975
1980 1985 1990
40-
30 >
a
a
^*
fc
z
_i
10
1950 1955 1960 1965
1970
YEAR
1975 1980 1985 1990
Figure 5.69. Salinity values at the surface and at 10-m depth in the Carr Inlet study area
during the algal bloom season.
5-123
-------
a
a
I 29
z
_i
t/i
ANNUAL MEAN
I STANDARD ERROR
0 INDIVIDUAL OBSERVATION
-— SIGNIFICANT REGRESSION LINE
(P < 0.05)
1959 1955 1966 1965
—i—
1970
YEAR
1975
1986 1985 1996
1956 1955
1985 1996
Figure 5.70. Salinity values at 30-m depth and water temperatures at the surface in the
Carr Inlet study area during the algal bloom season.
5-124
-------
ANNUAL MEAN
I STANDARD ERROR
O INDIVIDUAL OBSERVATION
-— SIGNIFICANT REGRESSION LINE
(P < 0.05)
1959 1955
1968 1965 197C 1975 1986 1985 1996
24-
23;
22"
21
o
19"
3 18
13-
ia-
11
ie-
9;
s-
7-
1956 1955 1960 1965
1970 1975 1989 1985
YEAR
1996
Figure 5.71. Water temperatures at 10- and 30-m depths in the Carr Inlet study area during
the algal bloom season.
5-125
-------
ae
19
18 1
17
16
is
113
oil
x 10
a 9
UJ o
- ?
o 7
i/i c
-------
ANNUAL MEAN
J STANDARD ERROR
O INDIVIDUAL OBSERVATION
---- SIGNIFICANT REGRESSION LINE
(P < 0.05)
1950 1955
1985 1990
X
"5
i
u
t-
<
20
o
ac
O
z
(/>
o
1950 1955 1960 1965 1970 1975 1980 1985 1990
YEAR
Figure 5.73. Concentrations of dissolved oxgyen at 30-m depth and dissolved inorganic
nitrate at the surface in the Carr Inlet study area during the algal bloom season.
season.
5-127
-------
concentrations that occurred near the beginning and the end of the data set.
High values (up to 15.0 mg/L at the surface) were detected in 1986, while
low values were detected near the beginning of both the long-term data set
at 30-m depth (e.g., less than 6 mg/L) and the recent data sets at all three
depths (e.g., as low as 8 mg/L at the surface). The high dissolved oxygen
concentrations of 1986 (observed on 19 May) may have been due to some
unusual condition (e.g., an intense algal bloom) that occurred that year.
Alternatively, their earlier absence could have been a consequence of the
infrequency of sampling, which might have missed previous similar high
concentrations.
Nutrients—Plots of nitrate values by year are shown in Figures 5.73
and 5.74. Data are available only since 1977, so comparisons between data
collected before and after 1973, and long-term regressions by year could not
be performed. The recent regression of surface nitrate concentration by
year (Table 5.12) detected a significant negative slope, with concentrations
at or near detection limits during 1985 and 1986 (Figure 5.73). This recent
decline in surface nitrate concentrations may have been caused by a recent
increase in algal concentrations (see below). This result suggests that
surface nitrate concentrations did not limit algal growth in this study
area, at least during the late 1970s. Similar declines in nitrate were not
detected at 10 and 30 m.
No significant changes in surface phosphate concentrations were
detected (Tables 5.11 and 5.12, Figures 5.75 and 5.76). A moderate decline
in phosphate concentrations was detected by ANOVA at 10-m depth, and a more
substantial decline was detected by both ANOVA and regression at 30-m depth.
These declines may be attributable to changes in natural or anthropogenic
inputs of phosphate. These declines in phosphate concentrations also may
have been influenced by changing sampling station locations and data sources
(see Figure 5.63 and Table 5.9). Because the early University of Washington
data were probably generated with a spectrophotometer (Appendix A), the
actual values probably were reasonably accurate. The actual changes in
location of the sampling stations over time would not seem to be likely to
have caused an overall decline in phosphate concentrations (Figure 5.63,
Table 5.9). In summary, the decreases in phosphate concentration may have
5-128
-------
o
I
O) •
ac.
o
o:
O
z
O
UJ
v\
o
0
ANNUAL MEAN
I STANDARD ERROR
O INDIVIDUAL OBSERVATION
—- SIGNIFICANT REGRESSION LINE
(P
-------
5 :
T
Q.
I/)
O
Q.
O
I
O
Q
O 4
I/I 1
ANNUAL MEAN
I STANDARD ERROR
0 INDIVIDUAL 06SERVATION
SIGNIFICANT REGRESSION LINE
(P < 0.05)
V
T
1956
1
1955
1
1966
1
1965
1
1976
YEAR
1
1975
i
1986
i
1985
r
1996
I
o>
3
fc3
O
O.
O
I a
O
Q
dl
3
o ,
* O O O O
1956 1955 1966 1965 1976 1975 1986 1985 1996
YEAR
Figure 5.75. Concentrations of dissolved orthophosphate at the surface and at 10-m
depth in the Carr Inlet study area during the algal bloom season.
5-130
-------
Ol
3
£3
I/I
o
Q.
o
a
UJ
1
(/) 1
v;
o
0-1
ANNUAL MEAN
J STANDARD ERROR
O INDIVIDUAL OBSERVATION
SIGNIFICANT REGRESSION LINE
(P < 0.05)
O O
1950 1955 I960 1965
1970
YEAR
—i 1 1—
1975 1980 1985
1990
300 '
2 200
cc
I-
ISI
z
Ud
o
O 100
I/)
a
1950
1955 I960 1965
1970
YEAR
1975 1980 1985
1990
Figure 5.76. Concentrations ot dissolved orthophosphate at 30-m depth and percent
dissolved oxygen saturation at the surface in the Carr Inlet study area
during the algal bloom season.
5-131
-------
been real phenomena, but it is also possible that changes in station
location or data sources could have affected the data.
Indicators of Phvtoplankton Growth—Because chlorophyll a data are
available only for 1979-1981, they are not plotted by year. Long-term
changes in percent dissolved oxygen saturation at the surface were not
detected. However, a significant (P<0.05) increasing trend in percent
dissolved oxygen saturation at the surface was detected since 1977
(Table 5.12, Figure 5.76). This increase was influenced by high values
recorded in 1986. A concomitant decrease in Secchi disk depth since 1977
was also detected. Thus, an increase in algal density since 1977 may have
returned this variable to levels that occurred in the 1950s.
A problem in interpreting these results is that the Secchi disk only
measures transparency from the surface down to the depth at which the disk
disappears from view. In a habitat such as Carr Inlet, where water clarity
is high and vertical mixing rates are low, changes in algal density could
occur principally at depths below the Secchi disk depth.
Pollutants—Very little data on sulfite waste liquor are available,
although a Boise-Cascade pulp mill is located across Puget Sound from the
mouth of Carr Inlet. Data on fecal coliform bacteria are available for
surface water since 1977 (Figure 5.77). Values were generally at the
detection limit (1 organism/100 mL). However, a significant decrease
(PO.05) was detected (Table 5.12) and may be attributed to a single high
value recorded in 1977. There is no known large source of bacterial
contamination in the Carr Inlet area.
Nisquallv Reach
The Nisqually Reach study area is in a rural region of the southern
sound (see Figure 5.63). Class AA water quality standards apply in the
area. Nisqually Reach is located near a sill, and is shallow (36 m) for a
main channel site. Turbulent mixing of the water column in the study area
is caused by the rapid currents south of Tacoma Narrows, and by the
proximity of the sills at Nisqually Reach and Tacoma Narrows. The Nisqually
5-132
-------
z
HI
O
CO
5
I
CO
16'
H '
IS
10
8
1 1 1 1 1 1 1 r
1950 1955 I960 1965 1970 1975 1980 1985 1990
YEAR
tr
LLJ
o
m
l
o ,.
.
< o
oz
ANNUAL MEAN
I STANDARD ERROR
0 INDIVIDUAL OBSERVATION
SIGNIFICANT REGRESSION LINE
(P < 0.05),
V»^y»»> *.«
1950 1955 I960 1965 1979 1975 1980 1985 1990
YEAR
Figure 5.77. Secchi disk depth and log of concentrations of fecal coliform bacteria at
the surface in the Carr Inlet study area during the algal bloom season.
5-133
-------
River enters the sound approximately 1.5 km south of the study area. It
contributes only 5 percent of the total freshwater input to Puget Sound,
but is the largest river in the southern sound (Table 2.1). Flow rates are
high in the winter through the late spring, when snowmelt occurs (USGS
1985). Nisqually Reach was included in the study because changes at various
locations in the southern sound might be expected to be integrated in this
region.
Environmental Conditions in the Study Area--
Mean salinity and water temperature values during the algal bloom period
are depicted in Figure 5.64. Data are available from 1932 to 1986, with the
best coverage during the mid-1950s and since 1977. Only surface and 10-m
depth data were suitable for analysis. Moderate density stratification was
evident. The mean surface salinity value was 26.4 ppt, while the mean value
at 10-m depth was 2.2 ppt higher. The mean surface water temperature value
was 12.6° C, while the mean value at 10-m depth was 0.8° C lower. The
salinity gradient was greater than that observed at Carr Inlet (0.2 ppt
difference between the mean salinity values at the surface and at 10-m
depth), presumably because of the proximity of the Nisqually Reach study
area to the outlet of the Nisqually River. The thermal depth gradient at
Nisqually Reach was smaller than the thermal gradients observed at either
Carr or Budd Inlets (over 2.3° C difference between the mean water
temperature values at the surface and at -10-m depth), probably because of
the higher turbulence and rates of vertical mixing of the water column at
Nisqually Reach.
The concentrations of dissolved oxygen and nutrients, and the potential
for the development of intense algal blooms, were affected by the moderate
density stratification and the propensity for vertical mixing (Figures 5.65
and 5.66). Although no chlorophyll a data are available, algal blooms did
not seem to become highly developed in the Nisqually Reach study area. The
percent saturation of dissolved oxygen at the surface was only about
105 percent (Figure 5.67 and Appendix E). The percent saturation of
dissolved oxygen was still near 100 percent at 10-m depth, which supports
the hypothesis of well-developed vertical mixing in the study area.
5-134
-------
Although the water column was well mixed, mean Secchi disk depth was high,
over 6.5 m (Figure 5.67). This deep Secchi disk depth supports the
assessment that algal densities were not high at Nisqually Reach. Moreover,
nutrient concentrations generally were higher than in the other southern
sound sites (e.g., mean nitrate concentration at the surface was 10.3 ug/L).
These high nutrient concentrations suggest that rates of nutrient uptake by
the phytoplankton in the area, and, thus, rates of algal growth in the area,
generally were lower than in the other southern sound sites.
Results of the correlations between pairs of water quality variables at
the surface (Appendix F) suggest that the moderate blooms of the Nisqually
Reach study area occurred when salinity was low and the water column was
stratified. The positive correlation between surface salinity values and
Secchi disk depths indicates that water clarity was lower when surface
salinity was low. Moreover, the negative correlation between nitrate
concentrations and water temperature values at the surface suggests that
algal blooms occurred when thermal gradients were present.
Pollution in the Nisqually Reach area does not appear to have been a
severe problem (Figure 5.68). Limited data on sulfite waste liquor are
available, but concentrations were lower than in any other southern sound
study area. The geometric mean concentration of sulfite waste liquor for
surface waters was only 2.0 (Pearl Benson Index). The geometric mean
concentration of fecal coliform bacteria also was low (1.5 organisms/100 ml
for surface waters).
Water Quality Trends in the Study Area--
A summary of comparisons between water quality data collected before and
after 1973 is given in Table 5.11. Slopes of statistically significant
long-term and recent regressions of the water quality data against year are
given in Table 5.12.
Physical Conditions—Plots of salinity values by year are shown in
Figure 5.78. There was a long-term decrease in salinity values at both the
surface and at 10-m depth. These changes appear to have been a steady
5-135
-------
30
I20
ie
e
1 1 1 r 1 1 1 1 1 i ir
1939 1935 1940 1945 1959 1955 I960 1965 1976 1975 1980 1985 1990
YEAR
40
30
a
a
z
_j
<
in
e
ANNUAL MEAN
I STANDARD ERROR
0 INDIVIDUAL OBSERVATION
SIGNIFICANT REGRESSION LINE
(P < 0.05)
1930 1935 1940 1945 1950 1955 I960 1965 1970 1975 1980 1985 1990
YEAR
Figure 5.78. Salinity values at the surface and at 10-m depth in the Nisqually Reach
study area during the algal bloom season.
5-136
-------
decline over the entire period of available data. The locations of the
sampling stations at Nisqually Reach used by the University of Washington
and Ecology are nearly identical, so changing station locations probably
did not contribute to the pattern of changing salinity values. At the
surface, the mean salinity value for the period 1973-1986 was approximately
1.5 ppt lower than the mean salinity value for the period 1932-1972. The
decreases in salinity values at Nisqually Reach do not appear to have been
caused by increases in the flow of the Nisqually River (USGS 1985).
Plots of water temperature values by year are shown in Figure 5.79.
Mean temperatures at both the surface and at 10-m depth were higher for the
period 1973-1986 than for the period 1932-1972 (Table 5.11). The mean
values were 9.4 percent (surface) and 8.0 percent (10-m depth) higher during
the recent period. However, neither the long-term nor the recent
regressions of water temperature by year had statistically significant
(P>0.05) slopes (Table 5.12). Apparently temperatures during the cool
period of the 1950s lowered the overall mean temperature for the period
1932-1972.
Dissolved Oxygen — Plots of dissolved oxygen concentrations by year are
shown in Figure 5.80. Concentrations did not fall below the Class AA
standard (see Table 4.2) at either the surface or 10-m depth. There was a
general pattern of increasing dissolved oxygen concentrations at both the
surface and 10-m depth. Both of these trends were influenced by high
dissolved oxygen concentrations reported in 1986 (e.g., 15.0 mg/L at the
surface on 19 May 1986). These high dissolved oxygen concentrations may
have been caused by an intense algal bloom. No changes in the discharges of
anthropogenic oxygen-demanding pollutants to the Nisqually River or to the
Nisqually Reach area were identified.
Nutrients—Plots of nitrate and phosphate concentrations by year are
shown in Figures 5.81 and 5.82. No significant temporal trends were
detected. Although the amount of data collected before 1977 was limited,
phosphate concentrations do not appear to have changed substantially since
the 1930s. Analytical techniques used in the 1950s were reliable, but the
5-137
-------
24 -
23
22
21 1
,20
•18
I"
515
£ 13
5 12
*~ 11
10
9
8
7
ANNUAL MEAN
J STANDARD ERROR
0 INDIVIDUAL OBSERVATION
SIGNIFICANT REGRESSION LINE
(P e 0.05)
1 1 1 1 1 1 1 1 1 ' I r
1930 1935 1940 1945 1950 1955 I960 1965 1970 1975 1980 1985 1999
YEAR
24'
23
221
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
I I
1930 1935 1949 1945 1959 1955 1969 1965 1979 1975 1989 1985 1999
YEAR
Figure 5.79. Water temperatures at the surface and at 10-m depth in the Nisqually Reach
study area during the algal bloom season.
5-138
-------
X.
Ol
E
ae •
19
18
17 '
16-
15
14
13
ia
11
10
9
81
7
6
5"
4 '
3
2-
1
e
i i 1 1 1 1 1 1 1 1 1 r
1938 1935 1949 1945 195e 1955 i960 1965 1970 1975 1980 1985 1990
YEAR
ae-
19
is 1
17
16
oil
gl0-
9
8
7
6
5
4
31
2
1
0-
ANNUAL MEAN
I STANDARD ERROR
O INDIVIDUAL OBSERVATION
SIGNIFICANT REGRESSION LINE
(P<0.05)
1930 193S 1940 1945 1950 1955 1960 1965 1970 1975 1980 1985 1990
YEAR
Figure 5.80. Concentrations of dissolved oxygen at the surface and at 10-m depth in
the Nisqually Reach study area during the algal bloom season.
5-139
-------
"5
e
K
o
z
o
o
10
ANNUAL MEAN
STANDARD ERROR
INDIVIDUAL OBSERVATION
SIGNIFICANT REGRESSION LINE
(P<0.05)
T
T
T
"i 1 1 1 1 1 1 r
1936 1935 194* 1945 195* 1955 1969 1965 1979 1975 19M 1985 1999
YEAR
40
x.
o
I
o:
o
o:
O
Z
-------
5 '
O>
3
UJ
K
£3
(A
o
a.
O
o
o
u
ii
yt
o
ANNUAL MEAN
I STANDARD ERROR
0 INDIVIDUAL OBSERVATION
SIGNIFICANT REGRESSION LINE
(P < 0.05)
O 0
T i 1 1 1 1 1 1 1 1 1 1 r
1930 1935 1940 1945 1950 1955 1960 1965 1970 1975 1980 1985 1990
YEAR
5
X
"5 .
CD
3
£3
t/1
o
o.
o
a.
O
1
e
1930 1935 1940 1945 1950 1955 1960 1965 1970 1975 1980 1985 1990
YEAR
Figure 5.82. Concentrations of dissolved orthophosphate at the surface and at 10-m
depth in the Nisqually Reach study area during the algal bloom season.
5-141
-------
techniques used in the 1930s relied upon visual color comparisons, so the
early data may be somewhat less accurate (Appendix A).
Indicators of Phvtoplankton Growth—No chlorophyll a data are avail-
able. An increase in surface percent dissolved oxygen saturation is evident
since 1977, although this increase was influenced by the generally high
dissolved oxygen concentrations observed in 1986 in the southern sound.
Secchi disk depth has not changed significantly since 1977 (Figure 5.83 and
Tables 5.11 and 5.12). There is no substantial evidence to suggest that the
intensity of algal blooms has changed at Nisqually Reach.
Pollutants—The quantity of sulfite waste liquor data was insufficient
for trends analysis. Concentrations of sulfite waste liquor were very low,
although the Boise-Cascade pulp mill is located approximately 11 km northeast
of Nisqually Reach. Fecal coliform bacteria have been monitored at the
surface since 1977 (Figure 5.84). As evidenced by a significant regression
against year, concentrations of fecal coliform bacteria have declined
(p<0.05) over the study period (Table 5.12). Concentrations did not violate
Class AA standards, except on 23 August 1978. This single high
contamination event probably drove the statistical significance of the
declining trend. That day had the heaviest rainfall of that particular
month, and the surface salinity recorded on that date was the lowest in the
entire data set for Nisqually Reach (15.4 ppt). These two factors suggest
that the source of the bacterial contamination was storm runoff from the
agricultural areas in the Nisqually River basin.
Budd Inlet
The study area is located in the southern portion af Budd Inlet, a
shallow (average depth under 10 m), sluggishly circulating, southern
embayment (see Figure 5.63). Budd Inlet is classified as a stratified,
partially mixed estuary (URS 1986a). Flushing rates are rather low,
particularly near the head of the inlet. Stations are located near Priest
Point, from 1.5 to 3 km north of the Port of Olympia. Class A water quality
standards apply to the northern portion of the study area. Class B standards
apply to the southern portion of the study area, closer to the City of
5-142
-------
300
g see
Z
U)
o
o
LJ
>
o
vt
e
ANNUAL MEAN
J STANDARD ERROR
O INDIVIDUAL OBSERVATION
-— SIGNIFICANT REGRESSION LINE
(P < 0.05)
i 1 1 1 1 \ 1 1 1 1 1 1 r
1939 1935 1940 1945 1950 1955 I960 1965 1979 1975 1980 1985 1990
YEAR
16
14
12
- 10
Q.
Ul
o 8
VI
Q
U
O
Id
Z
e
0 OO
1930 1935 1940 1945 1950 1955 1960 1965 1970 1975 1980 1985 1990
YEAR
Figure 5.83. Percent dissolved oxygen saturation at the surface and Secchi disk depth in
the Nisqually Reach study area during the algal bloom season.
5-143
-------
IT
HI
O
m
_
o w
< d
uz
O
O
ANNUAL MEAN
J STANDARD ERROR
0 INDIVIDUAL OBSERVATION
SIGNIFICANT REGRESSION LINE
(P c 0.05)
1930 1935 1940 1945 1959 1955 1969 1965 1979 1975 1989 1985 1999
YEAR
Figure 5.84. Log of concentrations of fecal coliform bacteria at the surface in the Nisqually
Reach study area during the algal bloom season.
5-144
-------
Olympia. The Deschutes River is an important source of fresh water for Budd
Inlet, contributing approximately 1 percent of the total freshwater flow to
Puget Sound (Table 2.1). The Deschutes River discharges into Capitol Lake.
The depth of Capitol Lake is regulated by a dam that discharges into the
head of Budd Inlet. Other sources of fresh water include several small
streams.
Environmental Conditions in the Study Area--
The inner portion of Budd Inlet is prone to periods of low dissolved
oxygen in near-bottom waters, particularly during the late summer (URS
1986a). In the past, low dissolved oxygen has been attributed to the decay
of diatom blooms. Recently, URS (1986a) determined that the spring diatom
bloom causes supersaturation of dissolved oxygen throughout the water column
in Budd Inlet. Based on the results of limited field and modeling studies,
URS (1986a) suggested that low dissolved oxygen concentrations in late
summer are caused by a combination of factors, including high temperatures,
high sediment oxygen demand, and low flushing rates. However, the influence
of the diurnal vertical migration patterns of dinoflagellates had to be
included in the URS (1986a) model to account for the depth gradient of
oxygen concentrations in the late summer. In this model, the dinoflagellates
functioned as an oxygen "pump," producing oxygen near the surface during the
day and consuming oxygen near the bottom during the night.
The contribution of anthropogenic sources to the dissolved oxygen
problems in Budd Inlet has been investigated recently. Modeling studies
have suggested that anthropogenic nitrogen inputs could be increasing the
magnitude of spring diatom blooms and summer dinoflagellate blooms by
30-50 percent (URS 1986a). The LOTT plant is the major point source of
nitrogen to Budd Inlet (URS 1986a). URS (1986a) has recommended that
nitrogen removal be implemented by the LOTT plant during the summer to reduce
anthropogenic enhancement of algal blooms in the area.
Point sources of biological oxygen demand do not appear to have a
substantial impact on dissolved oxygen concentrations. In 1979, point
sources of biological oxygen demand contributed less than 10 percent to the
5-145
-------
total sediment oxygen demand (Kruger 1979). This figure has declined since
1979 because of improvements to the sewage treatment facilities in the area.
The biological oxygen demand discharged from the Olympia primary sewage
treatment plant in the summer of 1979 averaged approximately 500 mg/L. The
LOTT secondary sewage treatment plant became operational at the same
location in 1981. Biological oxygen demand discharged from the new plant
in the summer of 1986 averaged approximately 10 mg/L (Singleton, L.(
7 August 1987, personal communication). Because the new plant also adds
ozone to the effluent, the dissolved oxygen concentration in the effluent
typically is 8 mg/L or higher.
Mean salinity and water temperature values during the algal bloom season
are depicted in Figure 5.64. Scattered data are available from the 1950s
and the 1960s, with nearly continuous coverage from the early 1970s through
1986. Density stratification of the water column was well developed, as
salinity values were substantially lower and temperature values were
substantially higher at the surface. The difference between the mean
temperature values at the surface and at 10-m depth was particularly large,
approximately 2.6° C. Mean salinity values were approximately 2.0 ppt lower
at the surface than at 10-m depth. The stability of the water column
indicates that algal blooms could develop readily, and that excess nutrients
might not be readily flushed from the head of the inlet, particularly when
the flow rate of the Deschutes River is low during the summer.
The distributions of dissolved oxygen and nutrients over depth were
strongly affected by water column stability (Figures 5.65 and 5.66). The
mean percent dissolved oxygen saturation was over 114 percent at the
surface, but was only approximately 90-percent at 10-m depth (Appendix E).
Dissolved oxygen concentrations below 3 mg/L, which can cause mortality in
sensitive biota (NOAA 1986a), were rarely seen during the algal bloom
season. The consistent presence of dissolved oxygen concentrations above
3 mg/L may be due to several factors related to sampling and station
location. Sampling was relatively infrequent (once per month by Ecology)
and did not extend to the bottom.' Thus, the sampling could have missed
short-term low dissolved oxygen events (i.e., the water most likely to be
low in dissolved oxygen was not sampled). Also, the sampling stations were
5-146
-------
not located at the head of the inlet, where the most severe problem exists
(URS 1986a).
Water clarity and the vertical distributions of nutrient concentrations
suggest that algal blooms were quite intense in the Budd Inlet study area.
Water clarity was only about half that of Carr Inlet and Nisqually Reach, as
mean Secchi disk depth was only about 3.1 m (Figure 5.67). Nitrate concen-
trations in the Budd Inlet study area were much lower than in the Nisqually
Reach study area. The mean surface nitrate concentration was 1.95 ug-at/L
at the Budd Inlet site and 10.8 ug-at/L at the Nisqually Reach site. These
results for Budd Inlet resemble the concentrations found in the shallow
Totten Inlet and Oakland Bay study areas. Phosphate concentrations were
similar to those observed in Carr Inlet (e.g., mean surface concentrations
were 1.4 ug-at/L at the Carr Inlet site and 1.5 ug-at/L at the Budd Inlet
site).
Nitrate concentrations in Budd Inlet were only 40 percent as high at
the surface as at 10-m depth. However, phosphate concentrations were
97 percent as high at the surface as at 10-m depth. The low nitrate
concentrations recorded in Budd Inlet, especially at the surface, suggest
that anthropogenic enrichment of nitrogen could enhance algal blooms in Budd
Inlet by supplementing the supply of available nutrients. This
interpretation is consistent with the conclusions of URS (1986a).
The relationships among the water quality variables provide further
insight into the role of algal blooms in the Budd Inlet ecosystem
(Appendix F). Negative correlations between nitrate concentrations and
water temperature values at the surface and at 10-m depth, and between
dissolved oxygen concentrations and water temperature values at 10-m depth
were probably caused by blooms that occurred during warm, calm, and sunny
weather. The positive correlation between percent dissolved oxygen
saturation at the surface and water temperature also probably was due to
enhanced photosynthetic rates that occurred during warm weather. The
positive correlations between surface nitrate concentration and Secchi disk
depth, and between nitrate concentrations and dissolved oxygen concentrations
at 10-m depth also may be attributable to the waxing and waning of algal
5-147
-------
blooms. Transparency and nutrients would both be high when blooms were not
well developed. Nutrient concentrations and dissolved oxygen concentrations
at depth would both be low when a bloom declined.
Pollution in the Budd Inlet study area by sulfite waste liquor and
fecal coliform bacteria does not appear to be a severe problem (Figure 5.68).
The geometric mean of the concentration of sulfite waste liquor was only 3.9
(Pearl Benson Index). The geometric mean concentration of fecal coliform
bacteria was not as high as that found in the City Waterway study area,
although it was the highest of any southern sound study area
(4.2 organisms/100 ml). URS (1986a) found that a small creek discharging
near the head of the inlet (Moxlie Creek) was the major point source of fecal
coliform bacteria to Budd Inlet.
Water Quality Trends in the Study Area--
A summary of comparisons between water quality data collected before and
after 1973 is given in Table 5.11. Slopes from statistically significant
long-term and recent regressions of the water quality data against year are
given in Table 5.12.
Physical Conditions—Plots of salinity and water temperature values by
year are shown in Figures 5.85 and 5.86. Significant differences (PO.05)
between salinity data collected before and after 1973 were not detected.
However, a significant (PO.05) positive slope in the regression of salinity
values against year was detected for surface water since 1973. The most
plausible explanation for the apparent recent salinity increase in the Budd
Inlet study area is that station locations changed over time, and that an
unusual low salinity event occurred in 1974, near the beginning of the time
period analyzed in the recent regression. During the period when all three
Ecology stations (BUD003, BUD004, BUD005) were sampled (1967-1970, 1976-
1977), salinity values did not differ significantly (PO.05) among the three
stations. However Station BUD003 was the station closest to the mouth of
the Deschutes River and to Capitol Lake (Figure 5.63). One very low surface
salinity value (12.7 ppt) was detected at this station in August 1974.
Because Stations BUD003 and BUD004 were dropped by Ecology after 1977, the
5-148
-------
40"
.30-
a
a
t 20-
10
o
0 0
1950 1955 1960 1965 1970 1975 1980 1985 1990
YEAR
40'
30
a.
a
t/i
10
ANNUAL MEAN
I STANDARD ERROR
0 INDIVIDUAL OBSERVATION
—- SIGNIFICANT REGRESSION LINE
(P < 0.05)
f
1950 1955 1960 1965 1970 1975 1980 1985 1990
YEAR
Figure 5.85. Salinity values at the surface and at 10-m depth in the Budd Inlet study area
during the algal bloom season.
5-149
-------
24
23
aa
31
80
19
18
17
16
15
H
13
12
11
10
9
8
1950 1955 196e 1965 1970 1975 1980 1985
YEAR
1990
33
aa-
ai
ae
191
18
17
16'
15-
14
13
121
11
1«
9
8
7"
I
0
ANNUAL MEAN
STANDARD ERROR
INDIVIDUAL OBSERVATION
SIGNIFICANT REGRESSION LINE
(P < 0.05)
60 o o
o o
1950 1955 1960 1965 1970 1975 1980 1985
YEAR
1990
Figure 5.86. Water temperatures at the surface and at 10-m depth in the Budd Inlet
study area during the algal bloom season.
5-150
-------
only station from which data are available after 1977 is BUD005, which is the
station most distant from the principal source of fresh water. The effects
of freshwater inputs from the Deschutes River and from the flushing of
Capitol Lake on the salinity data were diminished in the more recent data,
which could have introduced an apparent increasing trend into the salinity
data.
The only change detected for water temperature was from the comparison
of data collected before 1973 with data collected from 1973 through 1986
(Table 5.11). That comparison indicated that a decline in surface
temperature had occurred. However, the regressions of water temperature by
year were not significant (P>0.05) (Table 5.12). This contradiction may be
resolved by noting that the data collected before 1973 contained data from
only a few years, and that data collected in 1968 and 1969 had the highest
mean water temperatures recorded for the entire Budd Inlet data set
(Figure 5.86). Thus, no substantial temporal change in water temperature
was noted for Budd Inlet.
Dissolved Oxygen—Plots of dissolved oxygen concentrations by year are
shown in Figure 5.87. "Violations of the Class B water quality standard (see
Table 4.2) were recorded at the surface and 10-m-depth during the 1970s.
The mean dissolved oxygen concentration for the period 1973-1986 was
approximately 16 percent higher than the mean for the period 1957-1972
(Table 5.11). Statistically significant (P<0.05) increasing dissolved
oxygen concentrations were detected at the surface in both the long-term and
recent regressions (Table 5.12). This increase seems to have been due in
part to the absence of very low values after 1977, the last year that
Ecology sampled the stations nearest the head of Budd Inlet. Dissolved
oxygen concentrations typically are lowest near the head of the inlet (URS
1986a). Because the sampling stations near the head of Budd Inlet were
dropped at the same time the low dissolved oxygen values disappeared from
the data, the apparent rise in dissolved oxygen concentrations probably was
influenced by changing sampling station locations over time. Aside from any
apparent effects of changes in sampling station locations, dissolved oxygen
concentrations may have continued to increase in the Budd Inlet study area
during the 1980s (Figure 5.87). When the LOTT sewage treatment plant became
5-151
-------
20-
19
13
16
o
1956 1955 1966 1965
1976
YEAR
—i 1 1 r
1975 1986 1985 1996
20
19
18
17
16
^14 '
oil
o »
LU n
> 8
7
6
<=> 5
4
3
1
8
ANNUAL MEAN
J STANDARD ERROR
0 INDIVIDUAL OBSERVATION
SIGNIFICANT REGRESSION LINE
(P<0.05)
1950 1955 1966 1965
1976
YEAR
1975 1986 1985
1996
Figure 5.87. Concentrations of dissolved oxygen at the surface and at 10-m depth in
the Budd Inlet study area during the algal bloom season.
5-152
-------
operational, removal of biological oxygen demand and addition of ozone to
the effluent may have contributed to the apparent increase in dissolved
oxygen concentrations. Also, as was observed in the other study areas in
the southern sound, high dissolved oxygen concentrations, possibly caused by
an intense algal bloom, were reported in 1986.
Nutrients — Plots of nitrate concentrations by year are shown in
Figure 5.88. Because data are available only since 1977, comparisons
between data collected before and after 1973 could not be performed. The
regressions of nitrate by year were not significant (P>0.05) at either the
surface or at 10-m depth (Table 5.12). Plots of phosphate concentrations by
year are shown in Figure 5.89. The mean phosphate concentration at the
surface was approximately 33 percent lower for the period 1973-1986 than for
the period 1957-1972 (Table 5.11). However, the data collected from 1957
to 1972 consist of only five observations taken in only 2 yr, and the long-
term regressions of phosphate concentrations against year were not
significant (P=0.35). Thus, the evidence for decreasing phosphate
concentrations since the 1950s in Budd Inlet is weak. The recent regressions
(since 1973) of phosphate concentrations against year had a positive slope
with a statistical significance probability of P=0.08. This result suggests
that phosphate concentrations may have increased since 1973.
Indicators of Phvtoplankton Growth—No chlorophyll a data are
available. Percent dissolved oxygen saturation in surface water is plotted
by year in Figure 5.90. Statistically significant (P<0.05) positive slopes
were found in the long-term and recent regressions of surface dissolved
oxygen saturation by year (Table 5.12). The greater net increase, approxi-
mately 41 percent, was detected by the recent regression. The increase in
surface percent dissolved oxygen saturation was probably influenced by
several factors. Some of this increase appears to result from the absence of
very low values after 1977 (Figure 5.90), which was the last year that
Ecology sampled the stations nearest the head of Budd Inlet. Other factors
that could have affected the percent dissolved oxygen saturation include
high dissolved oxygen concentrations observed in 1986 (13.7 mg/L on 23 June),
and improvements in the sewage treatment facilities used by the City of
Olympia and the surrounding region. Thus, the available evidence concerning
5-153
-------
401
z
o
z
cc
o
o
LJ
>
_J
o
V)
WJ
Q
10
ANNUAL MEAN
STANDARD ERROR
INDIVIDUAL OBSERVATION
SIGNIFICANT REGRESSION LINE
(P < 0.05)
o
'8
1950
1955 1966 1965
1970
YEAR
1975 1989 1985 1990
40
20
a:
O
z
O
I/I
e
00
1950 1955 I960
1965
1970
YEAR
1975 1980 1985
1990
Figure 5.88. Concentrations of dissolved inorganic nitrate at the surface and at 10-m depth
in the Budd Inlet study area during the algal bloom season.
5-154
-------
5
^
\
'o 4
1 ~
I
o»
3
U
a 3
O
i
a.
0
I
O
a
Ul
0 , .
tn 1
o
S
0-
ANNUAL MEAN
J STANDARD ERROR
0 INDIVIDUAL OBSERVATION
SIGNIFICANT REGRESSION LINE
(P < 0.05)
0
A
\
0
\ O O O
\
B 00
w I
1 Y o o ^
O \ f 0 i » ii . JO
Wo o Y\ f °
i 1
| i A O
o o o o
If
0
0
1950 1955 I960 1965
1970
YEAR
1975 1980 1985 1990
5 '
£3
O
a.
o
CL
o
o
LJ
0
1950 1955 I960 1965
1970
YEAR
1975 1980 1985 1990
Figure 5.89. Concentrations of dissolved orthophosphate at the surface and at 10-m depth
in the Budd Inlet study area during the algal bloom season.
5-155
-------
300
e
a
OL
D
l/l
Z
UJ
O
O 100
o
u
ANNUAL MEAN
I STANDARD ERROR
0 INDIVIDUAL OBSERVATION
SIGNIFICANT REGRESSION LINE
(P < 0.05)
0
1950 1955 1960 1965 1970 1975 1980 1985 1999
YEAR
16
14 -
12"
o
- 6
u
ui
i/i 4
0
1950 1955 1960 1965 1970 1975 1980 1985 1990
YEAR
Figure 5.90. Percent dissolved oxygen saturation at the surface and Secchi disk depth
in the Budd Inlet study area during the algal bloom season.
5-156
-------
the intensity of algal blooms does not suggest that changes have occurred in
the photosynthetic rates of Budd Inlet. This interpretation is further
supported by the Secchi disk data (see below), which did not show declining
transparency, as would be expected if phytoplankton density in the water
column had increased.
Secchi depth data from the late 1950s and 1977-1986 are plotted by year
in Figure 5.90. Mean Secchi disk depths measured before and after 1973 were
not significantly different (Table 5.11). However, a positive slope was
found in the regression of Secchi disk depth against year since 1973. This
positive slope seems to have been caused by both low Secchi disk depth values
near the beginning of the recent time period and by occasional high Secchi
disk depth values since 1980. In addition, low values are absent from the
database after 1977. It appears that changing station locations after 1977
also may have affected the Secchi disk depth data. When the stations near
the head of Budd Inlet were dropped from Ecology's ambient monitoring
program in 1977, the low Secchi disk depths disappeared from the data.
These stations probably would have exhibited lower transparency due to
proximity to the head of the inlet.
Pollutants—Data on sulfite waste liquor in ,the surface waters are
available from the late 1950s and from 1969 to 1977 (Figure 5.91). Data for
10-m depth, available only from 1973 to 1977, are not plotted. Values were
low [geometric mean surface concentration was 3.9 (Pearl Benson Index)] and
no changes were detected at either depth.
Data on concentrations of fecal coliform bacteria for surface waters,
available from 1973 to 1986, are plotted by year in Figure 5.91. A
significant (PO.05) decrease was detected at the surface. High values,
frequently well in excess of Class B water quality standards, were reported
from 1973 through 1977. Lower values, generally well below Class A water
quality standards, were reported from 1978 through 1986.
The apparent decline in the concentrations of fecal coliform bacteria
may have had more than one cause. The trend may be an artifact of changing
sampling station locations because the sampling stations nearest the head of
5-157
-------
tr
O
3 >-
CO g
U.t
O
o
o
cr
LJJ
m
E J
ccE
§1
O *.
o «>
Q.
_l .
< O
02
O
o
ANNUAL MEAN
STANDARD EFWOR
INDIVIDUAL OBSERVATION
SIGNIFICANT REGRESSION LINE
(P < 0.05)
8 o
o o
oo o
19Se 1955 I960 1965 1970 1975 1986 1985 1990
YEAR
e
1956 1955 1966 1965 1979 1975 1986 1985 1999
YEAR
Figure 5.91. Log of concentrations of sulfite waste liquor and fecal coliform bacteria at the
surface in the Budd Inlet study area during the algal bloom season.
5-158
-------
the inlet and the known point sources of fecal coliform bacteria (URS 1986a)
were dropped after 1977. However, the upgrades in the sewage treatment
facilities discharging near the head of Budd Inlet may also have contributed
to the decreased concentrations of fecal coliform bacteria. Prior to the
improvements completed in 1981, a raw sewage lift station on the Deschutes
River was known to fail frequently (Singleton, L., 7 August 1987, personal
communication).
Totten Inlet
The study area is near Windy Point, in the middle of a shallow
(approximately 15 m), sluggishly circulating, southern embayment (see
Figure 5.63). Class A water quality standards apply in this rural area
(Table 4.2). There is no large source of fresh water for Totten Inlet,
although small creeks flow into the heads of each branch of the inlet.
Extensive mudflats are found in Oyster Bay, at the southern head of the
inlet. Totten Inlet is highly productive for shellfish.
Environmental Conditions in the Study Area--
Mean salinity and water temperature values are depicted in Figure 5.64.
Data are available from the late 1950s, sporadically from the 1960s and
1970s, and regularly from 1978 through 1986. Mean salinity values at the
surface and 10-m depth were very similar, approximately 28.0 ppt. Surface
water was moderately warmer (approximately 1.0° C) than water at 10-m
depth. Mean surface temperatures were similar at the Totten Inlet and Budd
Inlet study areas (15.5° C), but the difference in the mean temperature
values at the surface and 10-m depth was considerably smaller at the Totten
Inlet site (1.0° C for Totten Inlet; 2.7° C for Budd Inlet). This difference
suggests that more vertical mixing occurred at the Totten Inlet site.
Vertical mixing may occur more readily in Totten Inlet than in Budd Inlet
because Totten Inlet does not have a large freshwater source that contributes
fresh water to the surface layers (i.e., there is no density gradient to
inhibit mixing). Solar heating may also be more effective in Totten Inlet
than in Budd Inlet because water clarity is greater in Totten Inlet
5-159
-------
(Figure 5.67). Mean Secchi disk depth was 4.3 m in the Totten Inlet site
and 3.1 m in the Budd Inlet site.
Both dissolved oxygen and nitrate concentrations exhibited only slight
concentration gradients between the surface and 10-m depth (Figure 5.65).
Mean dissolved oxygen concentrations were 10.0 mg/L at the surface and
10.2 mg/L at 10-m depth. Mean nitrate concentrations were 1.8 ug-at/L at the
surface and 2.5 ug-at/L at 10-m depth. The nitrate concentration at the
surface was 72 percent of the mean nitrate concentration at 10-m depth.
These percentages were only 40 percent at the Budd Inlet site and 34 percent
at the Carr Inlet site. These results also suggest that vertical mixing was
substantial in the Totten Inlet study area. The higher concentration of
dissolved oxygen at 10-m depth than at the surface suggests that there may
be a source of dissolved oxygen at depth. This source is unknown, but it
could have been photosynthesis by benthic diatoms.
The low nutrient concentrations (especially nitrate; Figures 5.65
and 5.66) and the high mean percent dissolved oxygen saturation (120 percent)
at the surface (Figure 5.67) suggest that the water column in the Totten
Inlet study area had high rates of nutrient uptake and primary production.
Nitrate inputs also may have been low because of the lack of a large fresh-
water source that could serve as a nitrate source (see Chapter 2). The
existing high rate of primary production and the low nitrate concentrations
in Totten Inlet suggest that additional inputs of nutrients would be rapidly
utilized by algae, causing further increases in the already substantial
algal blooms.
Phosphate concentrations were positively correlated with salinity and
water temperature values at the surface and at 10-m depth, although the
surface correlation was not statistically significant when scaled with the
Bonferroni inequality (Appendix F). Nitrate concentrations were not
correlated with either salinity or water temperature values. Results of
these correlation analyses can be explained by seasonal changes in salinity
and water temperature values, and by the contrasting sources of nitrate and
phosphate. Phosphate concentrations probably were positively correlated
with both salinity and water temperature values because the main source of
5-160
-------
phosphate replenishment is oceanic water that replaces the existing water in
the southern sound in late summer (Collias et al. 1974). (In the Totten
Inlet data, phosphate concentrations reached the lowest monthly mean in
June, and began to increase in July.) Thus, high phosphate concentrations
could occur when both salinity and water temperature values were high.
However, nitrate probably is replenished later, after the algal bloom season
(see Chapter 2). Nitrate concentrations remained low throughout the entire
algal bloom season.
Pollutant concentrations in the Totten Inlet study area were low
(Figure 5.68). The geometric mean sulfite waste liquor concentration at the
surface was 2.2 (Pearl Benson Index). The geometric mean concentration of
fecal coliform bacteria at the surface was 1.04 organisms/100 ml. Both of
these values are near analytical detection limits.
Water Quality Trends in the Study Area--
A summary of comparisons between water quality data collected before and
after 1973 is given in Table 5.11. Slopes from statistically significant
long-term and recent regressions of the water quality data against year are
given in Table 5.12.
Physical Conditions—Plots of salinity and water temperature values by
year are shown in Figures 5.92 and 5.93. A decline in salinity values at
the surface was detected, but no changes were detected in water temperature
values (Tables 5.11 and 5.12). The decline in salinity values could have
been a real phenomenon, or it could have been an artifact of changes in
station location over time. The early higher salinity samples were collected
at the University of Washington's Station TOT472. That station was located
somewhat downstream from Ecology's Station TOT001, which is where the recent
lower salinity samples were collected. However, the horizontal salinity
gradient in Totten Inlet is not steep or consistent (01 cay 1959) because the
freshwater inputs into the head of Totten Inlet are small (USGS 1985).
Hence, the effect of the changes in station locations on the salinity data
cannot be assessed unequivocally.
5-161
-------
30
a
a.
^x
±20
z
10
T
T
1950
1955 I960 1965
1970
YEAR
1975
1980 1985 1999
40-
30
a
a
£20
z
10
ANNUAL MEAN
I STANDARD ERROR
0 INDIVIDUAL OBSERVATION
—- SIGNIFICANT REGRESSION LINE
(P < 0.05)
1950 1955 1960 1965
1970
YEAR
1975 1980 1985 1990
Figure 5.92. Salinity values at the surface and at 10-m depth in the Totten Inlet
study area during the algal bloom season.
5-162
-------
24
23'
22-
si
5 181
§15
13
12
11
ie
9
8
7
1959 1955 1969 1965 1979 1975 1989 1985 1999
YEAR
24
23-
22"
21
18
13-
12
11
18
9
8
7-
ANNUAL MEAN
I STANDARD ERROR
0 INDIVIDUAL OBSERVATION
SIGNIFICANT REGRESSION LINE
(P < 0.05)
1959 1955 1969 1965 1979 1975 1989 1985
YEAR
1999
Figure 5.93. Water temperatures at the surface and at 10-m depth in the Totten Inlet
study area during the algal bloom season.
5-163
-------
Dissolved Oxygen—Plots of dissolved oxygen concentration by year are
shown in Figure 5.94. Dissolved oxygen concentrations below the Class A
water quality standard (see Table 4.2) were observed only once, in 1970.
The only significant temporal trend in dissolved oxygen concentrations in
the Totten Inlet study area was an increase at 10-m depth since 1978. This
increase appears to have been caused largely by one very high concentration
observed in 1986. This high value may be attributed to a particularly
intense algal bloom.
Nutrients—Nitrate data are available since 1978. Although no
significant changes in nitrate concentrations were detected (Figure 5.95),
long-term declines in phosphate concentrations were detected at both the
surface and at 10-m depth (Figure 5.96, Tables 5.11 and 5.12). No changes
in phosphate concentrations were detected since 1978. No site-specific
explanation was available for the long-term declines in phosphate
concentrations.
The apparent long-term declines in phosphate concentrations were
probably real phenomena. The effect of changing station locations probably
would be to increase the apparent phosphate concentrations over time,
contrary to the decline that was observed. The more recent samples were
taken closer to the head of the inlet (Table 5.9, Figure 5.63), and phosphate
concentrations during the bloom season typically are higher closer to the
head of the inlet (Olcay 1959). The decline in phosphate concentrations
detected over time occurred despite the interfering influence of changing
station locations. However, it was not possible to assess the effects of
changes in analytical techniques used by University of Washington and
Ecology to measure phosphate. Thus, the possibility that changes in the
data sources over time that may have influenced the data could not be
evaluated.
Indicators of Phvtoplankton Growth—No chlorophyll a data are
available. Percent dissolved oxygen saturation at the surface and Secchi
disk depth are plotted by year in Figure 5.97. Because no temporal changes
were detected for either variable (Tables 5.11 and 5.12), overall changes in
algal abundance do not appear to have occurred.
5-164
-------
80
19
18
17"
16
512
on
git
Q 9
> 8
o 7
in 6
5 5
1950 1955 I960 1965 1970 1975 1980 1985 1990
YEAR
ciW
19
18
17
16-
—
I
0
—
ANNUAL MEAN
STANDARD ERROR
INDIVIDUAL OBSERVATION
SIGNIFICANT REGRESSION LINE
(P < 0.05)
*14
£10
° 9
l*J O
* 5
o 7
m f
in b
5 5
3'
8
1
0
1950 1955 1960 1965
1970
YEAR
1975 1980 1985 1990
Figure 5.94. Concentrations of dissolved oxygen at the surface and at 10-m depth in
the Totten Inlet study area during the algal bloom season.
5-165
-------
40-
<
t—
z
o
o:
O
o
Ifl
ANNUAL MEAN
J STANDARD ERROR
0 INDIVIDUAL OBSERVATION
—- SIGNIFICANT REGRESSION LINE
(P < 0.05)
1950
1955 1966 1965
1978
YEAR
1975 1980 1985 1996
40-
z
^ 20
O
z
o
l/l
19S0
1955 I960 1965
1970
YEAR
1975 1980 1985 1990
Figure 5.95. Concentrations of dissolved inorganic nitrate at the surface and at 10-m
depth in the Totten Inlet study area during the algal bloom season.
5-166
-------
5
01
U
t-
o
Q.
O
o
Q
(/>
in
ANNUAL MEAN
I STANDARD ERROR
O INDIVIDUAL OBSERVATION
SIGNIFICANT REGRESSION LINE
(P < 0.05)
O O
* o
1950 1955 I960 1965 1970
YEAR
1975 1980 1985 1990
o
Q.
O
o
LJ
81
•V-
1950
o o
1955 1960 1965 1970 1975 1980 1985 1990
YEAR
Figure 5.96. Concentrations of dissolved orthophosphate at the surface and at 10-m
depth in the Totten Inlet study area during the algal bloom season.
5-167
-------
300-
z
2 300
o
O 100
o
u
o
t/>
o
1950
ANNUAL MEAN
STANDARD ERROR
INDIVIDUAL OBSERVATION
SIGNIFICANT REGRESSION LINE
(P < 0.05)
1955 I960 1965
—i—
1970
YEAR
1975 1980 1985 1990
16
14 1
12
10
a.
LJ
o
o
UJ
0
O 0
O 0
1950 1955 I960 1965 1970 1975 1980 1985 1990
YEAR
Figure 5.97. Percent dissolved oxygen saturation at the surface and Secchi disk depth
in the Totten Inlet study area during the algal bloom season.
5-168
-------
Pollutants.--Values of sulfite waste liquor and fecal coliform bacteria
for surface water are plotted by year in Figure 5.98. Temporal trends were
not detected for either variable (Tables 5.11 and 5.12).
Oakland Bay
The study area is located near the intersection of Oakland Bay and
Hammersley Inlet (see Figure 5.63). Sampling stations are located from
Eagle Point to southwestern Oakland Bay, near Goldsborough Creek and the
City of Shelton. Class B water quality standards apply in the area.
Historically the area was affected by the ITT-Rayonier sulfite pulp mill,
which operated from 1928 to 1957 (NOAA 1985). A primary sewage treatment
plant that discharged into the inner portion of Shelton Harbor was replaced
in 1979 by a secondary sewage treatment plant that discharges near Eagle
Point (Singleton, L., 20 October 1987, personal communication).
Circulation in the study area is sluggish and erratic because Oakland
Bay is connected to Puget Sound only through the shallow and narrow
Hammersley Inlet. Oakland Bay is shallow, averaging approximately 3 m deep
over much of its area. Extensive mudflats border most of the bay. The
study area ranges from 3 to 15 m deep. Two stations were too shallow to
have 10-m data (University of Washington's Station OAK484 and Ecology's
Station OAK003).
Environmental Conditions in the Study Area--
Mean salinity and water temperature values during the algal bloom
period are depicted in Figure 5.64. Data are available from 1956 through
1986. Temporal coverage was variable and data for 10-m depth are available
only since 1975. Salinity and water temperature values were affected by the
timing of the algal bloom period. This period was shorter in duration and
occurred earlier in the year in Oakland Bay than in the other southern sound
study areas (Table 5.10). Salinity values in the Oakland Bay study area
were the lowest of all the southern sound study areas at both the surface
and at 10-m depth. The Oakland Bay site also exhibited the steepest
5-169
-------
cc
o
U. CD
3 "C
., a
C3
O
ANNUAL MEAN
STANDARD ERROR
INDIVIDUAL OBSERVATION
SIGNIFICANT REGRESSION LINE
(P < 0.05)
1950
1955 196e 1965
1970
YEAR
1975 1980 1985 1990
cc
LU
ffi
Is
O ..
O o
a.
_i .
< o
uz
u_
o
o
o
o o
» • • » yy»vi
19S0 1955 1960 1965 1970 1975 1980 1985 1990
YEAR
Figure 5.98. Log of concentrations of sulfite waste liquor and fecal coliform bacteria at
the surface in the Totten Inlet study area during the algal bloom season.
5-170
-------
vertical salinity gradient, a difference of 3.3 ppt between the mean
salinity values at the surface and at 10-m depth. Density stratification
appears to have been well developed in the study area, possibly because the
fresh water from Goldsborough Creek supplied fresh water to the surface in
the study area.
The thermal depth gradient in the Oakland Bay study area was not as
large as the gradients observed in Carr and Budd Inlets (Figure 5.64). In
the Oakland Bay site, the mean water temperature was 0.7° C higher at the
surface than at 10-m depth. The equivalent differences in Garr and Budd
Inlets were 2.3° C and 2.7° C, respectively. The relatively small thermal
depth gradient may be due, in part, to the shallowness and small volume of
Oakland Bay. These characteristics would allow solar warming to be effective
throughout the water column. The mean water temperatures during the algal
bloom season in the Oakland Bay study area were lower than those in the
other southern sound study areas (e.g., 13.6° C at the surface at the
Oakland Bay site and 15.5° C at the surface at the Totten Inlet site). This
temperature difference was probably an effect of the timing of the algal
bloom season in the Oakland Bay study area. Algal blooms occurred from
April through June in the Oakland Bay site and from May through August in
the other southern sound sites. Mean water temperature values increased at
the Oakland Bay site during midsummer, after the bloom season, averaging
approximately 1.5° C higher in the Oakland Bay site than in the Totten Inlet
site.
Figure 5.65 shows a concentration gradient of dissolved oxygen over
depth that is reversed from the typical estuarine bloom condition. The mean
dissolved oxygen concentration at the surface was approximately 1.4 mg/L
lower than the mean dissolved oxygen concentration at 10-m depth. Most of
this apparent difference was caused by the presence of sulfite waste liquor
in many of the surface water samples collected in 1956 and 1957, when the
ITT-Rayonier pulp mill was still in operation. As discussed in Chapter 3,
sulfite waste liquor lowers the dissolved oxygen concentration in seawater.
After the ITT-Rayonier pulp mill closed, the depth gradient in
dissolved oxygen concentrations observed in the Oakland Bay study area was
5-171
-------
similar to the gradient observed in the Totten Inlet study area. In both
sites, the mean dissolved oxygen concentration was approximately 0.2 mg/L
lower at the surface than at 10-m depth. Moreover, the mean dissolved
oxygen concentration at 10-m depth was higher in Oakland Bay (9.9 mg/L) than
in Budd Inlet (7.9 mg/L), Nisqually Reach (8.1 mg/L), or Carr Inlet
(8.4 mg/L). As in Totten Inlet, the cause of the higher dissolved oxygen
concentration at 10-m depth is unknown. Possible explanations include high
photosynthetic oxygen production by benthic diatoms and advection of high
dissolved oxygen water along the bottom through Hammers ley Inlet.
Depth gradients of nutrient concentrations were fairly typical of par-
tially stratified estuaries. Mean concentrations at the surface were lower
than at 10-m depth (Figures 5.65 and 5.66). At 10-m depth, the mean nitrate
concentration in the Oakland Bay study area during the algal bloom period
was lower (5.8 ug-at/L) than in the Carr Inlet (10.7 ug-at/L) or the
Nisqually Reach (12.9 ug-at/L) study areas. However, it was slightly higher
than in the Budd Inlet (4.8 ug-at/L) or the Totten Inlet (2.5 ug-at/L) study
areas. Phosphate concentrations in the Oakland Bay study area were slightly
lower than in the other southern sound study areas (e.g., mean surface
phosphate concentration was 1.2 ug-at/L at the Oakland Bay site and 1.5 ug-
at/L at the Budd Inlet site).
The positive correlation between surface phosphate concentrations and
Secchi disk depths (Appendix F) probably reflects variation in both variables
caused by the waxing and waning of algal blooms. Low nutrient concen-
trations would tend to occur during blooms, and blooms also would reduce
transparency of the water column. When blooms were absent, both nutrient
concentrations and transparency would be high.
The algal blooms that occurred in Oakland Bay did not seem to be as
intense as those in the other southern sound study areas (Table 5.10). [The
limited amount of chlorophyll a data available indicates that similar
standing stocks of phytoplankton were found in the Carr Inlet and Oakland
Bay sites (Figure 5.66). However, because data were collected during
different time periods, a direct comparison is not very meaningful.] The
algal bloom in the Oakland Bay study area does not appear to have been
5-172
-------
limited by nutrient availability. Although nutrient concentrations were
always higher in the Oakland Bay site than in the Totten Inlet site, algal
blooms appeared to have been more intense in the Totten Inlet site. Higher
turbidity of the water column in Oakland Bay is a possible explanation for
the more limited intensity of the algal bloom, when compared with Totten
Inlet. Both areas probably support extensive populations of benthic
diatoms, but the lower transparency of the water column in Oakland Bay may
reduce light penetration to the bottom, thereby lowering photosynthetic
rates and dissolved oxygen concentrations at the surface.
Water Quality Trends in the Study Area--
A summary of comparisons between water quality data collected before and
after 1973 is given in Table 5.11. Slopes of statistically significant
long-term and recent regressions of the values of water quality variables by
year are given in Table 5.12.
Physical Conditions—Plots of salinity and water temperature values by
year are shown in Figures 5.99 and 5.100. The nonparametric ANOVA detected
no significant differences (P>0.05) between data collected before 1973 and
data collected from 1973 to 1986 for either surface salinity or surface
water temperature. The long-term regression of surface salinity values by
year had a significant (P<0.05) positive slope. This pattern was probably
caused by some very low salinity values observed in 1956 and 1957. The low
values were observed at Station OAK484, the station closest to Goldsborough
Creek and the point of discharge for the ITT-Rayonier pulp mill. Data were
not collected at this station after 1957. Salinity values at Station OAK484
did not differ significantly from salinity values at Station OAK485 during
periods of overlapping samples, but were lower on average. Therefore, the
apparent increase in salinity in the Oakland Bay study area may have been
caused at least in part by changing station locations over time.
Dissolved Oxygen—Plots of dissolved oxygen concentrations by year are
shown in Figure 5.101. The Class B dissolved oxygen water quality standard
(see Table 4.2) has been met since the ITT-Rayonier pulp mill closed in
1957. A long-term increase in dissolved oxygen concentrations was observed
5-173
-------
30'
a
a
_
10
1950 1955 I960 1965
1976
YEAR
1 - 1 - 1 - r
1975 1980 1985 1999
30
a
a.
50
10
ANNUAL MEAN
I STANDARD ERROR
O INDIVIDUAL OBSERVATION
—- SIGNIFICANT REGRESSION LINE
(P < 0.05)
1959 1955 1960 1965 1970 1975 1980 1985 1990
YEAR
Figure 5.99. Salinity values at the surface and at 10-m depth in the Oakland Bay
study area during the algal bloom season.
5-174
-------
24
23-
22'
21
^2g-
• 18-
Ot
• 17-
TJ
LJ 16
§15-
< 14 "
£ 13"
i! 12-
*~ 11
10-
9-
8"
7-
0
f
6 ° /
3 o ^ /\ O O /
iii oij0 IIT o ,t
ffj, M o ,^4{ f JJ
OA |\ f\ ° "° '
S 0 *V
0*° 0°°° ° ° °°
O * ..
0 °
i i ~~i i 1 1 i 1 r
1950 1955 1960 1965 1970 1975 1980 1985 1990
D A '
24
23
22'
21
.-^ 20
° 10 J
n 19
V
• 18'
• 1?
T>
^ 16
§15
I1"
£ 13
u 12-
^
11
ie
9
8
7-
YEAR
ANNUAL MEAN
J STANDARD ERROR
0 INDIVIDUAL OBSERVATION
SIGNIFICANT REGRESSION LINE
(P < 0.05)
O
A
0 O 0 O /
o o o/
_ f /
""I «
/V4 1 7
.,/ 1\ 4 /ff
' t M
\/, 1
o [»
0 O
o
o
1950 1955 1960 1965 1970 1975 1980 1985 1990
YEAR
Figure 5.1 00. Water temperatures at the surface and at 10-m depth in the Oakland Bay
study area during the algal bloom season.
5-175
-------
20
19
18
17
16
15
14
13
12
11
ie
9
8
7
6
5
4
3
2
1
0
1950 1955 I960 1965 1970
YEAR
1975
1980 1985 1990
28-
19-
18
17"
16"
on
S 8
7"
6"
0 5
4
3"
2"
1 '
0
ANNUAL MEAN
I STANDARD ERROR
0 INDIVIDUAL OBSERVATION
SIGNIFICANT REGRESSION LINE
(P < 0.05)
1950 1955 1960
1965 1970 1975 1980 1985 1990
YEAR
Figure 5.101. Concentrations of dissolved oxygen at the surface and at 10-m depth in
the Oakland Bay study area during the algal bloom season.
5-176
-------
at the surface (Tables 5.11 and 5.12). This increase appears to have been
caused by very low values observed during 1956 and 1957, and by very high
values observed in 1986. The high values reported in 1986 may have been
caused by an intense algal bloom. The early data actually include some zero
values. The same samples that contained no dissolved oxygen had Pearl
Benson Index values of sulfite waste liquor in excess of 200. As discussed
in Chapter 3, pulp mill wastes react with dissolved oxygen, as well as with
the reagents used to measure dissolved oxygen. Therefore, the low early
values probably were caused by contamination with pulp mill wastes.
Nutrients — Plots of nitrate and phosphate concentrations by year are
shown in Figures 5.102 and 5.103, respectively. No statistically significant
changes were detected in nitrate concentrations (Tables 5.11 and 5.12).
Results of the nonparametric ANOVA indicated that average surface phosphate
concentrations were lower from 1973 through 1986 than from 1958 through 1972
(Table 5.11). The slope of the long-term regression for surface phosphate
concentration against year was negative (Table 5.12). Although statis-
tically significant (PO.05), these trends were based on a sparse data set
for the period of 1958 through 1975. In contrast to the long-term decline,
phosphate concentrations have increased significantly (PO.05) since 1975.
The high phosphate concentrations detected in 1958 probably can be
attributed to natural variation in phosphate concentrations. Alternatively,
these high concentrations could have been influenced by residual effects of
the ITT-Rayonier pulp mill, which closed in 1957- Changing station locations
probably did not influence the data substantially. The early phosphate
data were obtained from the University of Washington's Station OAK485,
which was located near the stations sampled recently by Ecology. Because the
analytical techniques used by Washington Department of Fisheries to generate
the 1958 data could not be calibrated with Ecology's techniques, analytical
differences could have influenced the data. However, the techniques used
for phosphate analyses by Washington Department of Fisheries in the late
1950s are considered to provide fairly accurate data (Appendix A).
The recent (since 1975) increase in phosphate concentrations probably
was a real phenomenon. No specific factors were identified that could have
5-177
-------
40 -
<
ct
o
0:
O
o
1/1
ANNUAL MEAN
I STANDARD ERROR
O INDIVIDUAL OBSERVATION
SIGNIFICANT REGRESSION LINE
(P c 0.05)
I I
1950 1955
I960 1965 1970 1975 1980 1985 1990
YEAR
_J
X.
"5
i
LJ
I—
<
Q:
t—
z
1 20
o:
O
z
Q
LJ
_J
O
l/l
l/l
5
0
0
o
\ °
^s °
< ' O
* ' \ Q I / ^ ^ ^
\® A \ u \\ 1 T
L L' ^^ «| 1 1
O \Jr 1 fl I
I 11^*
• O O
A "1
v L! 1 | ! ! 1 , i p
1950 1955 1960 1965 1970 1975 1980 1985 1990
YEAR
Figure 5.102. Concentrations of dissolved inorganic nitrate at the surface and at 10-m
depth in the Oakland Bay study area during the algal bloom season.
5-178
-------
5 '
01
3
£3
(A
o
i
a.
O
o
o
i/i
o
ANNUAL MEAN
I STANDARD ERROR
0 INDIVIDUAL OBSERVATION
---- SIGNIFICANT REGRESSION LINE
(P < 0.05)
1959 1955 I960 1965 1970 1975 1980 1985 1990
YEAR
£3
I/I
o
i
a.
O
o
o
i/i
o
0
1950 1955 1960 1965
1970
YEAR
1975 1980 1985 1990
Figure 5.103. Concentrations of dissolved orthophosphate at the surface and at 10-m
depth in the Oakland Bay study area during the algal bloom season.
5-179
-------
contributed to this increase. All the samples collected since 1975 were
collected at Ecology's Station OAK004, (i.e., changes in station location
did not occur). Also, Ecology's analytical techniques have not changed
substantially since 1975.
Indicators of Pnvtoolankton Growth—Chlorophyll a data are available
for the Oakland Bay study area from 1964 through 1971 (Figure 5.104). No
temporal trend was detected. The percent dissolved oxygen saturation at the
surface has increased since 1958 (Figure 5.104). This increase was
influenced by low values recorded in 1956 and 1957 (presumably dissolved
oxygen saturation percentages near zero percent were due to high
concentrations of sulfite waste liquor), and by high values (over
180 percent) recorded in 1986. The highest value for surface percent
dissolved oxygen saturation was observed on 23 June 1986. A very high
surface temperature and a substantial thermal depth gradient was also
observed. These conditions suggest that an intense algal bloom was occurring
on that date. The single high data point recorded in 1986 may not have been
indicative of a temporal trend. However, it had a substantial influence on
the positive slope of the regression because it occurred in the most recent
year of the data set.
Secchi disk depth data -are plotted by year in Figure 5.105. No
long-term change in Secchi disk depths was detected (Tables 5.11 and 5.12).
However, the increases in the values observed since 1978 were statistically
significant (P<0.05). The highest mean Secchi disk depth readings were
recorded in 1959 and 1986. These values suggest that transparency may have
decreased after 1957 and increased back to earlier levels during the 1980s.
Alternatively, the high readings at the beginning and end of the data set
could be due to inherent high variability or to changes in station
locations.
Pollutants—Sulfite waste liquor data from the surface are plotted by
year in Figure 5.105. A very large decline in sulfite waste liquor
concentrations coincided with the closirlg of the ITT-Rayonier pulp mill in
1957. Levels remained low in sporadically collected samples until sampling
ceased in 1975. Sulfite waste liquor data (Figure 5.105) indicate that the
5-180
-------
01
d
!
o
ANNUAL MEAN
STANDARD ERROR
INDIVIDUAL OBSERVATION
SIGNIFICANT REGRESSION LINE
(P e 0.05)
19S« 1955 196f 1965
197«
YEAR
1975 19M 1985 199*
300
c
u
c
Q.
g 200
a:
t-
z
iAl
U
o 100
Q
UI
>
O
1950
1955 I960 1965
1970
YEAR
1975 1980 1985 1990
Figure 5.104. Concentrations of chlorophyll a'and percent dissolved oxygen saturation
at the surface in the Oakland Bay study area during the algal bloom season.
5-181
-------
16
H
~ 12
E
Q.
LU
Q
10
CO
Q
I 6
o
o
(t
o
S«
t «
[^ CD
3 ~
u-S:.
O
O
1
ANNUAL MEAN
I STANDARD EHROfl
0 INDIVIDUAL OBSERVATION
—- SIGNIFICANT REGRESSION LINE
(P < 0.05)
> O O
1950 195S I960 1965 19?e 1975 1980 1985 1996
YEAR
19Se 1955 1960 1965 1979 1975 1980 1985 1990
YEAR t
Figure 5.105. Secchi disk depth and log of concentrations of sulfite waste liquor at the
surface in the Oakland Bay study area during the algal bloom season.
5-182
-------
release of sulfite waste liquor by the ITT-Rayonier pulp mill was episodic
(e.g., values of the Pearl Benson Index ranged from 3 to 800 in 1956).
Fecal col iform data from the surface waters are plotted by year in
Figure 5.106. The Class B fecal coliform standard was never exceeded during
the algal bloom periods of the years for which data are available. Although
the negative slope of the regression by year was not statistically
significant (P=0.07), concentrations generally appeared to be lower in the
mid-1980s than they were in the late 1970s and early 1980s. Too few data
from 10-m depth were available for analysis.
Summary of Results for the Southern Sound
Major findings for the southern sound are compiled in this section.
Environmental conditions in the study areas are summarized and compared. A
brief assessment of the sensitivity of the southern sound study areas to
pollution is provided. Apparent trends in water quality are also summarized.
Environmental Conditions--
Differences between mean salinity values at the surface and at 10-m
depth were >2 ppt in the study areas that have substantial sources of fresh
water (i.e., Nisqually Reach, Budd Inlet, and Oakland Bay) and were < 0.2 ppt
in the other areas (i.e., Carr and Totten Inlets). Vertical gradients of
water temperature were present in all five southern sound study areas, and
were best developed, in Budd and Carr Inlets. The differences between the
mean water temperatures at the surface and at 10-m depth exceeded 2.3° C in
these two areas. Due to the substantial density stratification in Budd and
Carr Inlets, vertical mixing appeared to be limited in both areas. Mean
water temperature values at 10-m depth were highest (above 12.8° C) in the
Budd Inlet, Totten Inlet, and Oakland Bay sites, which are all quite
shallow. Vertical mixing appears to have been well developed at Nisqually
Reach. Although the salinity gradient was well developed at the Nisqually
Reach site, the thermal gradient was not as large as those in Carr and Budd
Inlets. The difference between the mean water temperature values at the
surface and 10-m depth was only 0.8° C at the Nisqually Reach site.
5-183
-------
oc
LU
o
m
3
o
o»
_l .
< o
or
o
o
ANNUAL MEAN
J STANDARD ERROR
O INDIVIDUAL OBSERVATION
---- SIGNIFICANT REGRESSION LINE
(P < 0.05)
<
H
> <
o (
N
>
<
/
><
>
o
f O ^
Vi
y°
> * <
V.
i r
1956 1955 1969 1965 1976 1975 1989 1985 1996
YEAR
Figure 5.106. Log of concentrations of fecal coliform bacteria at the surface in the
Oakland Bay study area during the algal bloom season.
5-184
-------
Extremely low dissolved oxygen concentrations at depth were rare, even
in Budd Inlet where low dissolved oxygen values have been recorded in other
studies (e.g., URS 1986a). Gradients in dissolved oxygen concentrations
with depth were steepest in Budd -and Carr Inlets, reflecting photosynthetic
enhancement of dissolved oxygen near the surface. The mean concentrations of
dissolved oxygen were <1.6 mg/L higher at the surface than at 10-m depth in
these two sites. Turbulent mixing apparently reduced the magnitude of the
vertical oxygen gradient at Nisqually Reach. Little variation in dissolved
oxygen concentrations with depth was detected in the Totten Inlet and
Oakland Bay study areas, except when the surface dissolved oxygen
concentration' was lowered in the Oakland Bay study area by sulfite waste
liquor discharge from the ITT-Rayonier pulp mill. The mean dissolved
oxygen concentration at 10-m depth was approximately 0.1 mg/L higher than at
the surface in the Totten Inlet site and, after the ITT-Rayonier pulp mill
closed, 0.2 mg/L higher than at the surface in the Oakland Bay site. The
elevated dissolved oxygen concentrations at 10-m depth in these two sites
might have been due to the shallowness of these areas, which would have
allowed the photic zone to extend to the bottom and to support a photo-
synthetical ly active benthic diatom community.
Concentrations of nitrate and phosphate differed noticeably among study
areas. Nitrate concentrations were distinctly lower in the Totten Inlet,
Budd Inlet, and Oakland Bay study areas than in the Carr Inlet and Nisqually
Reach study areas. For example, mean nitrate concentrations at 10-m depth
were <6 ug-at/L in the Totten Inlet, Budd Inlet, and Oakland Bay study areas
and were >10.6 ug-at/L in the Carr Inlet and Nisqually Reach study areas. A
possible explanation is that Budd and Totten Inlets are highly productive,
which would account for the lower nutrient concentrations. Also, these two
areas are shallower than Carr Inlet and Nisqually Reach, and presumably have
less nitrate potentially available from deeper water. Phosphate
concentrations did not vary greatly among the areas. Lowest mean
concentrations were observed in Totten Inlet and Oakland Bay.
The propensity for algal blooms (based principally on the percent
dissolved oxygen saturation at the surface) appears to have been highest in
5-185
-------
Carr, Totten, and Budd Inlets. Algal density probably was lower at Nisqually
Reach because turbulence tends to remove algal cells from the photic zone,
thereby reducing growth rates. The reason for poorly developed algal blooms
at the Oakland Bay site is unknown, although low nutrient concentrations and
low water transparency may have been contributing factors.
Sensitivity to Nutrient Enrichment--
Based on their limited capacities to export or assimilate pollutants
without deleterious ecological effects, Budd Inlet, Totten Inlet, and
Oakland Bay appear to be sensitive to inputs of excess nutrients. Nitrate
concentrations were low in these areas, and, at least in Budd and Totten
Inlets, algal blooms were quite intense. Additional amounts of nutrients
probably would increase the magnitude of the algal blooms in these areas.
Furthermore, because the volumes of these three areas are rather small,
additional nutrients would not be diluted effectively. Tidal flushing is
comparatively rapid in these areas, on the order of a few days, even when
freshwater inputs are low (URS 1986b). However, considerable refluxing of
water occurs at Dana Passage (up to 60 percent), so the rate of net transport
out of these embayments is low.
Nisqually Reach is probably the least sensitive of the southern sound
study areas to ecological problems caused by nutrient enrichment because it
has a greater capacity to export and assimilate excess nutrients. Mixing
prevents intense blooms from developing, even though refluxing of southern
sound water occurs at the sills of Nisqually Reach and Tacoma Narrows.
Assimilative capacity may be substantial at Carr Inlet, which has a large
dilution capacity. However, low nitrate concentrations at the surface of
Carr Inlet suggest that enrichment of surface waters could further stimulate
primary production. Also, the flushing time for Carr Inlet is much longer
than the flushing time for any other southern embayment (URS 1986b).
Discharges at the head of Carr Inlet might, therefore, have a greater
ecological impact because the retention time is greater at the head than at
the mouth of the inlet.
5-186
-------
Trends in Water Quality--
Although problems caused by changing station locations and data sources
limited data interpretation for some of the study areas, some general
conclusions may be drawn from the data collected in the southern sound. This
section summarizes the interpretation of the statistical data from
Tables 5.11 and 5.12. The most informative data for detecting temporal
trends were data on physical conditions and dissolved oxygen concentrations.
The data on phosphate concentrations were more useful than the data on
nitrate concentrations because the phosphate data were collected over a
longer time period and were less variable (Appendix E). Data relevant to
evaluating algal growth were sparse, while pollutant data were very
informative in study areas where known problems were monitored.
Physical Conditions—Peelining salinity values were detected in the Carr
Inlet, Nisqually Reach, and Totten Inlet study areas, although these results
were not unequivocal in the Totten Inlet site. Other salinity trends
appeared to have been artifacts of changing stations and data sources.
Increased water temperature values were detected in the Carr Inlet and
Nisqually Reach sites. These increases apparently were due to the cool
temperatures recorded in the early 1950s and 1930s '(see Figure 5.1). Data
for the other southern sound study areas were not collected until after this
cool period, so no major trends in water temperature values were apparent for
these areas.
Dissolved Oxygen—There was some evidence that increases in dissolved
oxygen concentrations occurred at every study area in the southern sound.
Very high concentrations of dissolved oxygen were detected in 1986 at all
the southern sound stations. These 1986 elevations in dissolved oxygen
concentrations may have been related to intense algal blooms (see below).
However, in Budd Inlet the apparent increases may have been influenced by
changes in station location and data sources. In Oakland Bay, increased
concentrations were partially attributable to reduced contamination from
sulfite waste liquor.
5-187
-------
Nutrients—Nitrate data are generally available only since the 1970s.
The only temporal trend detected was a negative slope in the regression for
surface data from Carr Inlet. This decline may have been caused by increased
algal abundance. Changes in phosphate concentrations were detected in all
the southern sound study areas except Nisqually Reach.
The changes consisted of long-term decreases and recent increases. The
long-term decreases appear to have resulted from a few high phosphate
concentrations (e.g., over 2 ug-at/L at 30-m depth) that were recorded in
the 1950s. The analytical techniques used to measure phosphate
concentrations were reasonably good in the 1950s, but the data were fairly
sparse. Thus, the few high values from the 1950s exerted a strong effect on
the statistical analyses in all the southern sound study areas except
Nisqually Reach where data are available from the 1930s. The only
statistically significant (PO.05) recent temporal trends in phosphate
concentrations detected in the southern sound are increases that have
occurred since 1975 at the surface and 10-m depth in the Oakland Bay study
area. An increase in phosphate concentrations since 1973 was detected at
the Budd Inlet study area, with a statistical probability of P=0.08.
Indicators of Phvtoplankton Growth--Carr Inlet is the only study area
in the southern sound for which evidence was found that suggested that algal
densities have changed systematically. Since 1977, Secchi disk depths and
nitrate concentrations have decreased in the Carr Inlet study area, while
values of percent dissolved oxygen saturation at the surface have increased.
Elevated values of percent dissolved oxygen saturation at the surface
were detected during 1986 at all study areas located in the southern sound.
These high concentrations of dissolved oxygen may have been caused by
intense algal blooms that occurred during 1986. The highest dissolved
oxygen saturation values of 1986 occurred during May in the Carr Inlet and
Nisqually Reach study areas. Unfortunately, dissolved oxygen data from May
1986 were not available for the Budd Inlet, Totten Inlet, and Oakland Bay
study areas. It cannot be determined whether the highest dissolved oxygen
saturation values occurred simultaneously in all the southern sound study
5-188
-------
areas. Nonetheless, dissolved oxygen concentrations were generally high in
the southern sound during 1986.
With the exception of the Carr Inlet study area, values of Secchi disk
depth and percent dissolved oxygen saturation at the surface were not
significantly correlated (P>0.3) in the southern sound study areas.
Therefore, with the exception of the Carr Inlet study area, variation in
Secchi disk depth was not closely associated with variation in algal
abundance. Recent increases in Secchi disk depth were detected in the Budd
Inlet and Oakland Bay study areas. The increase in Budd Inlet was driven,
in part, by changes in station location and data sources, and by a few very
high values recorded in the 1980s. The recent increases in Oakland Bay
brought the Secchi disk readings back to levels recorded in the 1950s, but
the cause of these increases cannot be determined from the available data.
Pollutants—The only significant change (P<0.05) in sulfite waste
liquor concentrations was a sharp decline in Oakland Bay that coincided with
the closure of the ITT-Rayonier sulfite pulp mill in 1957. The other
southern sound study areas lacked nearby sources of sulfite waste liquor.
Declines in counts of fecal coliform bacteria were detected in the Carr
Inlet, Nisqually Reach, and Budd Inlet study areas. In the Budd Inlet site,
the fecal coliform data may have been influenced by improvements in the
sewage treatment facilities and by changes in sampling station locations.
The decline in Nisqually Reach appears to have been driven by a high value
detected in 1978, at the beginning of the fecal coliform data set for this
site. The source of this contamination probably was storm runoff carried in
the Nisqually River. No cause of the decline observed in Carr Inlet is
apparent as the contaminated water samples collected early in the surveys of
this area did not appear to have been collected during, or shortly after,
storms.
HOOD CANAL
Hood Canal is defined herein as the portion of Puget Sound west of
Admiralty Inlet and south of Tala Point, including Dabob Bay, The Great Bend
5-189
-------
(where southern Hood Canal bends to the east and north), and Lynch Cove
(see Figure 2.1). Hood Canal is generally narrow (roughly 3 km) and deep.
The area around the Hood Canal is primarily rural. The region contains
15 percent of the surface area, 15 percent of the volume, 16 percent of the
shoreline, and 14 percent of the tidelands of all of Puget Sound south of
Admiralty Inlet (Burns 1985).
Circulation is less complex in Hood Canal than in the rest of Puget
Sound because of the relatively simple shape of the shoreline and the
absence of large islands that could constrain the flow of water. Several
small rivers flow into Hood Canal (see Table 2.1). The lack of vigorous
circulation in Hood Canal allows well-developed density stratification to
persist along most of its length. Density stratification is particularly
well-developed during the summer, when solar heating of surface water
reinforces salinity stratification (Collias et al. 1974). A 50-m deep sill
approximately 15 km south of the entrance to Hood Canal restricts the
circulation of seawater at depth. The deepest portion of Hood Canal (up to
approximately 200 m) extends from Dabob Bay south to The Great Bend. The
mouth of Dabob Bay has a sill at approximately 120-m depth. East of The
Great Bend the basin is less than 50-m deep, and it becomes progressively
shallower approaching Lynch Cove.
Three of the study areas in this characterization study are located in
Hood Canal: Dabob Bay, Mid-Hood Canal, and South Hood Canal. Station
locations are shown in Figure 5.107. Data sources are given in Table 5.13.
The algal bloom seasons for the study sites are given in Table 5.14.
Histograms summarizing the water quality variables are given in Figures
5.108-5.112. Back-up tables of the summary data given in Appendix E. The
ANOVAs comparing the water quality variables before and after 1973 are
summarized in Table 5.15. Long-term and recent regressions are summarized
in Table 5.16.
Based on the percent dissolved oxygen saturation at the surface, algal
blooms were most prevalent in Hood Canal in April through July. However,
photosynthetic rates remained fairly high through late summer (Table 5.14).
5-190
-------
47: 50
47' 40
47 30
47' 20
DAB 624
HCBMe
DABS22
DAB KM
HCB643
HCB544
HCBB46
HCB003
Figure 5.107. Locations of study areas and sampling stations in Hood Canal.
5-191
-------
TABLE 5.13. SAMPLING STATION NUMBERS, DATA SOURCES, AND
SAMPLING PERIODS FOR THE STUDY AREAS IN HOOD CANAL
Study Area
Dabob Bay
Mid-Hood Canal
South Hood Canal
Station
DAB522
DAB524
DAB526
DAB536
HCB002
HCB543
HCB544
HCB545
HCB003
LCH550
LCH552
HCB004
Data
Source
uwa
UW
UW
UW
Ecology
UW
UW
UW
Ecology
UW
UW
Ecology
Sampling Period
1952, 1960, 1965-66
1949-50, 1952-63, 1965-66
1952, 1960, 1962, 1965
1962, 1965
1968-70, 1976, 1978-86
1932-33, 1939, 1952, 1966
1933, 1952-63, 1965-67
1933, 1939, 1952-54, 1966
1968-70, 1975-86
1952-63, 1965-66
1952-63, 1965-66
1968-70, 1975-86
a UW = University of Washington.
5-192
-------
TABLE 5.14. ALGAL BLOOM SEASONS FOR HOOD CANAL STUDY AREAS,
AS DEFINED BY MONTHLY MEAN AND STANDARD ERROR OF PERCENT
DISSOLVED OXYGEN SATURATION IN SURFACE WATER
Percent Dissolved
Month
April
May
June
July
August
September
Dabob
120 +/-
122 +/-
117 +/-
123 +/-
113 +/-
114 +/-
Bay ,
4a
3a
2a
8a
2
3
Mid-Hood
106
120
117
114
110
107
+/-
V-
v-
V-
V-
+/-
Oxyqen Saturation
Canal
6
3a
3a
3a
3
3
South Hood
116 +/-
117 +/-
112 +/-
115 +/-
109 +/-
100 +/-
Canal
3a
2a
2a
3a
3
4
a Months included in the algal bloom season.
5-193
-------
381
e ie 34 e le 30 e
DEPTH (m)
36
DABOB
BAY
MID-HOOD
CANAL
STUDY AREA
SOUTH HOOD
CANAL
6 16 36
DABOB
BAY
e ie 36 e ie 36
DEPTH (m)
MID-HOOD
CANAL
STUDY AREA
SOUTH HOOD
CANAL
Figure 5.108. Mean salinity and water temperature values in the Hood Canal study
areas during the algal bloom season.
5-194
-------
z
UJ
o
>-
X
°s
Q 0)
s~
o
(A
V)
5
11
19
9
8
7
6
5
4
3
a
i
9
39 9 19 39
DEPTH (m)
DABOB
BAY
MID-HOOD
CANAL
STUDY AREA
e ie ae
SOUTH HOOO
CANAL
e ie 39 e ie 39 9 19 39
DEPTH (m)
DABOB
BAY
MID-HOOD
CANAL
STUDY AREA
SOUTH HOOD
CANAL
Figure 5.109. Mean concentrations of dissolved oxygen and dissolved inorganic nitrate
in the Hood Canal study areas during the algal bloom season.
5-195
-------
Ill
e 10 30 e le ae e te 30
DEPTH (m)
DABOB MID-HOOD
BAY CANAL
STUDY AREA
SOUTH HOOD
CANAL
coi
Q.^
trS
x
o
0 10 30 0 10 30 0 10 30
DEPTH (m)
DABOB MID-HOOD
BAY CANAL
STUDY AREA
SOUTH HOOD
CANAL
Figure 5.110. Mean concentrations of dissolved orthophosphate and chlorophyll a in
the Hood Canal study areas during the algal bloom season.
5-196
-------
M
%i
o
DISSOLVE
SATURATI
e
1
I
I
DABOB SOUTH HC
BAY CANAL
MID-HOOD
CANAL
STUDY AREA
OUJ g
LUG e
e
1
DABOB
BAY
SOUTH HOOD
CANAL
MID-HOOD
CANAL
STUDY AREA
Figure 5.111. Mean percent dissolved oxygen saturation at the surface and Secchi disk
depth in the Hood Canal study areas during the algal bloom season.
5-197
-------
c
o
V)
Stpg*
CQ 0)
"O
~ c
OJ —
0)
Q.
e
0 10 30 e le 30
DEPTH (m)
DABOB MID-HOOD
BAY CANAL
STUDY AREA
SOUTH HOOD
CANAL
OiuI3o a)
_l U./-S < 0.
0 10 30 0 10 30 0 10 30
DEPTH (m)
DABOB MID-HOOD
BAY CANAL
STUDY AREA
SOUTH HOOD
CANAL
Figure 5.112. Log of geometric mean concentrations of sulfite waste liquor and fecal coliform
bacteria in the Hood Canal study areas during the algal bloom season.
5-198
-------
TABLE 5.15. NET CHANGE AND PERCENT CHANGE IN THE MEAN VALUES OF WATER QUALITY
VARIABLES IN HOOD CANAL, BASED ON ANOVA COMPARISONS OF DATA
TAKEN BEFORE 1973 WITH DATA TAKEN FROM 1973 TO 1986
Depth
(m)
0
10
30
0
10
30
0
10
30
0
10
30
0
10
30
0
10
30
Dabob Bay Mid-Hood Canal South Hood Canal
Change Change Chanae
Net
NSa
-0.51
-0.63
NS
+1.02
+0.80
NS
+1.49
+1.16
nab
na
na
NS
-0.81
-1*;16
na
na
na
Percent Net Percent Net Percent
Salinity (ppt)
-1.58 6.1
1.8 -0.87 3.0 -0
2.1 -1.10 3.7 -0
Water Temperature (° C)
NS
9.9 +1.35 12.9 +1
9.1 +0.97 10.8
Dissolved Oxygen (mg/L)
NS
16.1 +1.69 20.4 +1
17.6 +1.41 23.5 +1
Nitrate (ug-at/L)
na
na
na
Phosphate (ug-at/L)
NS -0
46.6 -0.50 30.9 -0
46.7 -0.45 19.4 -0
Chlorophyll a (ug/L)
na
na
na
NS
.84
.64
NS
.32
NS
NS
.63
.47
na
na
na
.28
.80
.64
-
na
na
na
2.9
2.2
13.6
25.1
44.5
27.7
30.4
19.0
Dissolved Oxygen Saturation (Percent)
0
10
30
NS
+19.96
+14.89
NS
20.1 +23.03 25.9 +21
21.7 +17.57 28.0 +16
NS
.19
.63
30.7
48.1
na
Seechi Disk Depth (m)
na
NS
Sulfite Waste Liquor (Pearl Benson Index)
0
10
30
0
10
30
na
na
na
Fecal
na
na
na
na
na
na
Coliform Bacteria (No./100 mL)
na
na
na
NS
na
na
na
na
na
a NS = The pre-1973 and 1973-1986 values were not significantly different at
P<0.05, based on a nonparametric one-way ANOVA.
b na Results of the statistical test were not available because of a lack
of data.
5-199
-------
TABLE 5.16. SLOPES OF STATISTICALLY SIGNIFICANT LONG-TERM AND
RECENT REGRESSIONS OF WATER QUALITY VARIABLES AS A
FUNCTION OF YEAR FOR HOOD CANAL
Depth
(m)
0
10
30
0
10
30
0
10
30
0
10
30
0
10
30
0
10
30
0
10
30
Slopes
Dabob Bay Mid-Hood Canal
Long-term Recent Long-term Recent
Salinity (ppt)
NSa NS -0.057 NS
-0.017 NS -0.029 NS
-0.025 NS -0.041 -0.189
Water Temperature (° C)
NS NS NS NS
0.039 NS 0.047 NS
0.036 NS 0.037 NS
Dissolved Oxygen (mg/L)
NS NS NS NS
0.058 NS 0.058 0.282
0.040 NS 0.047 NS
Nitrate (ug-at/L)
nab NS na NS
na NS na NS
na NS na NS
Phosphate (ug-at/L)
NS NS NS NS
-0.029 NS NS NS
-0.032 NS NS NS
Chlorophyll a (ug/L)
na NS na na
na NS na na
na NS na na
Dissolved Oxygen Saturation (Percent)
NS NS NS NS
0.773 NS 0.805 3.456
0.534 NS 0.598 NS
Secchi Disk Depth (m)
na NS na NS
South Hood Canal
Long-term Recent
NS
-0.038
-0.026
NS
0.062
0.027
NS
0.093
0.063
na
na
na
-0.011
-0.029
-0.023
na
na
na
NS
1.161
0.709
NS
NS
NS
NS
NS
NS
NS
NS
0.417
NS
NS
NS
-0.662
NS
NS
NS
na
na
na
NS
4.809
NS
NS
Sulfite Waste Liquorc (Pearl Benson Index)
0
10
30
0
10
30
na na na na
na na na na
na na na na
Fecal Coliform Bacteriad (No./lOO mL)
na -0.018 na NS
na na na na
na na na na
na
na
na
NS
na
na
na
na
na
NS
na
na
a NS Not significant at P<0.05.
na Results of the statistical test were not available because of a lack
of data.
c Data were subjected to a log(X+1) transformation for the regressions.
Data were subjected to a log transformation for the regressions.
5-200
-------
Algal blooms appeared to haveibeen most intense in Dabob Bay, although dif-
ferences among the three study sites were not large.
Dabob Bay ,
The study area is located of'f Pulali Point, about 6.5 km north of the
mouth of Dabob Bay (Figure 5.107). Dabob Bay is a large embayment on the
western side of Hood Canal. Itjis approximately 185 m deep, although a
120-m deep sill at Pulali Point (inhibits deeper circulation north of the
i
study area. The Big Quilcene and little Quilcene Rivers are local sources of
i
fresh water (Table 2.1). Class ;AA water quality standards apply to the
Dabob Bay site. There are no majo\r urban influences in the area. However,
the sill and the resultant sluggish circulation at depth make the deep water
in Dabob Bay prone to low dissolved oxygen concentrations (Collias et al.
1974). \
Environmental Conditions in the Study Areja—
i
Mean salinity and water temperature values during the algal bloom period
are depicted in Figure 5.108. Data ar^ available from 1950 through 1986.
Depth gradients of salinity and water temperature were well developed, with
lower salinity values and higher temperature values recorded at the surface.
The mean salinity value was approximately 1.8 ppt lower at the surface than
at 10-m depth. The thermal gradient between the mean temperatures at the
surface and at 10-m depth was approximately 3.2° C, which is larger than the
thermal gradient in all other sites except! Bellingham Bay and the other Hood
Canal sites. Negative correlations between salinity and water temperature
values (Appendix F) suggest that salinity values were lower during warmer
weather, presumably because snowmelt lowered salinity values in the late
spring and early summer. Density stratification and heating of low salinity
surface waters, both of which are conditions that would be conducive to the
development of intense algal blooms, were prominent features of the water
column at the Dabob Bay study area.
The depth gradients for dissolved oxygen concentrations were quite
distinct from those in study areas outside Hood Canal (Figure 5.109, see also
5-201
-------
Figures 5.6, 5.22, 5.65). In Dabob Bay, the difference between the mean
dissolved oxygen concentrations at the surface and at 10-m depth was only
0.4 mg/L, while the difference between the mean dissolved oxygen
concentrations at 10 and 30-m depth was approximately 2.8 mg/L (Appendix E).
In most of the other deep study areas, the mean dissolved oxygen
concentration at 10-m depth was roughly half the difference between the mean
concentrations at the surface and at 30-m depth. The similarity between
dissolved oxygen concentrations at the surface and at 10-m depth in Dabob
Bay may have been due to similarity in the amounts of dissolved oxygen that
are produced by photosynthesis at the surface and at 10-m depth (see
discussion of chlorophyll a below).
Depth gradients for nutrient concentrations were highly developed in the
Dabob Bay study area (Figures 5.109 and 5.110). Extremely low concentrations
of nutrients, often below the analytical detection limits, were recorded at
the surface. Mean nitrate concentration at the surface was less than one-
tenth the mean concentration at 30-m. The depth gradient in phosphate
concentration also was substantial, as the mean concentration at the surface
was approximately one-third of the mean concentration at 30-m depth. The low
concentrations at the surface suggest that algal production near the surface
could be limited by low nutrient concentrations more frequently in the Dabob
Bay study area [and other Hood Canal study areas (see below)] than in the
other study areas in Puget Sound.
Significant negative correlations (P<0.05, scaled with the Bonferroni
inequality) between nutrient concentrations and both dissolved oxygen
concentrations and water temperature values were not found for the surface
waters, as would be expected in a highly stratified system subject to algal
blooms (Appendix F). However, significant negative correlations were found
between these variables and nutrient concentrations at depths of 10-m and
30-m. The lack of significant correlations at the surface may have been
caused by the insensitivity of the analytical methods used to analyze the
water samples for nutrient concentrations. Because the nutrient
concentrations at the surface typically were below the detection limits,
variations in nutrient concentrations caused by fluctuations in algal blooms
may not have been detected. Therefore, correlations between nutrient
5-202
-------
concentrations and the other variables at the surface would not have been
found.
The vertical distribution of chlorophyll a differed between the Dabob
Bay and Point Jefferson areas. The mean concentration of chlorophyll a was
higher at the surface than at 10-m depth at the Point Jefferson site. This
relationship was reversed at the Dabob Bay site. (Point Jefferson was the
only other site in the characterization study for which a substantial amount
of chlorophyll a data are available for both the surface and 10-m depth.)
This result suggests that maximum phytoplankton densities occurred deeper in
the water column at the Dabob Bay site than at the Point Jefferson site. The
lack of vertical mixing in Dabob Bay may allow diatoms, which are non-motile
but constitute the major type of phytoplankton causing many of the blooms in
spring and early summer, to sink out of surface waters.
Because Dabob Bay is highly stratified and has very sluggish circula-
tion, it would be expected to be highly productive. However, based on the
data for percent dissolved oxygen saturation at the surface, the intensity
of algal blooms in the Dabob Bay study area appeared to have been relatively
high (Figure 5.111), but not as high as the intensity of algal blooms in the
Sinclair Inlet and Carr Inlet areas (Figures 5.24 and 5.67). As discussed
above, it appears that low nutrient concentrations near the surface may have
limited algal production in the Dabob Bay study area.
The mean Secchi disk depth in the Dabob Bay study area (6.0 m) was
among the highest observed in any of the characterization study areas. This
observation supports the interpretation that productivity in the surface
waters of Dabob Bay was low. Secchi disk depth was negatively correlated
(P<0.05 scaled with the Bonferroni inequality) with the values of three
variables at the surface: chlorophyll a concentration, dissolved oxygen
concentration, and percent dissolved oxygen saturation (Appendix F). [The
negative correlation between Secchi disk depth and percent dissolved oxygen
saturation was not quite significant (P=0.09), scaled with the Bonferroni
inequality.] Therefore, the Secchi disk measurements appeared to reflect
turbidity in the near-surface water caused by phytoplankton rather than by
other suspended material. However, the mean concentration of chlorophyll a
5-203
-------
was nearly twice as high at 10-m depth (4.1 ug/L) than at the surface
(2.4 ug/L). Because the mean Secchi disk depth was only 6.0 m, a
considerable amount of phytoplankton must have existed below the Secchi disk
depth. Unfortunately, variation in algal density in the deeper layer was not
reflected by variation in the Secchi disk depth data.
Geometric means of the concentrations of sulfite waste liquor and
fecal coliform bacteria were low (near the detection limits) in the Dabob
Bay site (Figure 5.112). These results are reasonable, as there are no
large sources of these contaminants near the Dabob Bay site.
Water Quality Trends in the Study Area--
A summary of comparisons between water quality data collected before and
after 1973 is given in Table 5.15. Slopes from statistically significant
long-term and recent regressions of the water quality data against year are
given in Table 5.16.
| Physical Conditions—Plots of salinity and water temperature values by
year are shown in Figures 5.113-5.115. No significant (PO.05) temporal
changes in salinity were detected at the surface. However, at depths of
10- and 30-m, mean salinity values for the period 1976-1986 were lower than
mean salinity values from 1950 to 1966 (Table 5.15). Significant declines
(PO.05) in salinity values since 1950 were also detected at 10- and 30-m
depths in the regressions of salinity data against year (Table 5.16).
Similar regressions for the period of 1976 through 1986 were not
statistically significant (P>0.05). The reason(s) for the apparent
decreases in subsurface salinity are unknown, but inputs of high salinity
oceanic water may have declined. Rainfall data from the Seattle-Tacoma
International Airport suggest that rainfall has generally decreased in the
area since the 1950s (Figure 5.2), which would be expected to cause
increases, rather than decreases, in salinity values over time.
The decreased salinity values appear to have been real phenomena that
occurred in the field. Possible effects of changes in station locations and
data sources on the data appear to have been minimal. The Ecology station
5-204
-------
30-
Q.
a
z
irt
ie
" T 1
1959 1955
I960 1965
1970
YEAR
1975 1989 1985 1999
40 •
30
a.
a.
E a0
z
ie
ANNUAL MEAN
J STANDARD ERROR
0 INDIVIDUAL OBSERVATION
—- SIGNIFICANT REGRESSION LINE
(P < 0.05)
1956 1955 I960 1965
1970
YEAR
1975 1980 1985 1990
Figure 5.113. Salinity values at the surface and at 10-m depth in the Dabob Bay study
area during the algal bloom season.
5-205
-------
40
30
a
a
±20
z
ie
e
ANNUAL MEAN
J STANDARD ERROR
0 INDIVIDUAL OBSERVATION
--- SIGNIFICANT REGRESSION LINE
(P<0.05)
1950 1955 1960 1965 1970 1975 1980 1985 1990
YEAR
1990
Figure 5.114. Salinity values at 30-m depth and water temperatures at the surface in the
Dabob Bay study area during the algal bloom season.
5-206
-------
ANNUAL MEAN
I STANDARD ERROR
0 INDIVIDUAL OBSERVATION
—- SIGNIFICANT REGRESSION LINE
(P<0.05)
1959 1955 I960 1965 1970 1975 1980 1985
YEAR
1990
1950 1955
1985 1990
Figure 5.115. Water temperatures at 10- and 30-m depths in the Dabob Bay study area
during the algal bloom season.
5-207
-------
sampled since the mid-1970s is located near the center of the group of
University of Washington stations sampled earlier, and no substantial
freshwater source exists in the area that might have distorted the salinity
pattern when station locations were changed. Contour maps of salinity
values plotted by Collias et al. (1974) also do not suggest that changes in
station locations would have introduced apparent declines in salinity values
into the data. Salinity values depicted in Collias et al. (1974) near the
Ecology site did not appear to be systematically lower than the salinities
either north or south of the Pulali Point area. Also, as discussed in
Chapter 4 and Appendix D, salinity determinations by the University of
Washington and Ecology did not differ systematically.
Water temperatures at 10- and 30-m depths apparently have increased in
the Dabob Bay study area since the 1950s (Tables 5.15 and 5.16). These
increases are similar to the increases in air temperature values that have
been recorded at the Seattle-Tacoma International Airport (Figure 5.1),
which showed that the period 1948-1955 was relatively cool. Based on data
compiled by Collias et al. (1974), changes in station location do not appear
to have introduced apparent increases in water temperature values into the
data. Only long-term increases in water temperature values, which were
derived from early University of Washington data and recent Ecology data
(Table 5.13, Figure 5.107), were statistically significant (PO.05). This
observation raises the possibility that differences in analytical techniques
between University of Washington and Ecology could have introduced apparent
increased values into the data. Measurements of water temperature by the
University of Washington were lower than the measurements by Ecology at the
site where both agencies sampled during the same period of time (Chapter 4
and Appendix D). Hence, differences in the data sources also could have
contributed to the apparent increases in water temperature. In summary,
water temperatures at depth appear to have increased in the Dabob Bay study
area since the 1950s, but the validity of those increases may be suspect.
Dissolved Oxygen—Plots of dissolved oxygen concentrations by year are
shown in Figures 5.116 and 5.117. There was no evidence that the Class AA
water quality standard (see Table 4.2) was violated in surface waters. One
violation at 10-m depth was recorded in 1983. Many violations at 30-m depth
5-208
-------
ANNUAL MEAN
I STANDARD ERROR
0 INDIVIDUAL OBSERVATION
—- SIGNIFICANT REGRESSION LINE
(P < 0.05)
1950 1955
1975 1980 1985 1990
o o
1950 1955 1960 1965
1970
YEAR
1975 1980 1985 1990
Figure 5.116. Concentrations of dissolved oxygen at the surface and at 10-m depth in
the Dabob Bay study area during the algal bloom season.
5-209
-------
ae •
19-
18"
17
16 1
oil4
I 13
oil
x10
Q »
2 8
6
Q 5
4
3
2
1
0
ANNUAL MEAN
I STANDARD ERROR
0 INDIVIDUAL OBSERVATION
SIGNIFICANT REGRESSION LINE
(P < 0.05)
1950 1955 1960
1965
1970
YEAR
—i 1 1 r
1975 1980 1985 1990
40-
z
o
ct
o
z
o
0
1950 1955 1960 1965
1970
YEAR
1975 1980 1985 1990
Figure 5.117. Concentrations of dissolved oxygen at 30-m depth and dissolved inorganic
nitrate at the surface in the Dabob Bay study area during the algal bloom
season.
5-210
-------
were recorded. The mean dissolved oxygen concentration at this depth was
only 0.2 mg/L above the standard. Significant long-term increases (P<0.05)
in dissolved oxygen concentrations were found at 10- and 30-m depths
(Tables 5.15 and 5.16). Increases since 1976 were not significant. Based
on data compiled by Collias et al. (1974) and on the comparisons of dissolved
oxygen data collected by the University of Washington and Ecology (Chapter 4
and Appendix D), changes in station locations and sources of data do not
appear to be likely explanations for the apparent increases in dissolved
oxygen concentrations at depth in the Dabob Bay study area. Although it was
not possible to determine why dissolved oxygen concentrations increased since
the 1950s at 10- and 30-m depths, one explanation is that photosynthetic
rates may have increased at these depths (see following discussion of
indicators of phytoplankton growth).
Nutrients — Plots of nitrate concentrations by year are shown in
Figures 5.117 and 5.118. Because data are only available since 1976,
comparisons of data collected before and after 1973, and long-term
regressions could not be performed. No statistically significant (PO.05)
temporal trends were detected in nitrate concentrations, although the
negative slope of the regression by year was nearly significant (P=0.12) at
30-m depth. The sensitivity of these statistical analyses is questionable
for surface waters because many of the observations were below the analytical
detection limits.
No statistically significant changes in phosphate concentrations were
detected at the surface. However, statistically significant (P<0.05)
declines were detected at 10- and 30-m depths (Figures 5.119 and 5.120,
Tables 5.15 and 5.16). In general, the values recorded between 1953 and
1961 were higher than the values recorded between 1976 and 1986. Changes in
station locations do not appear to have contributed to these apparent
declines in phosphate concentrations because phosphate profiles near the
Dabob Bay study area apparently do not vary systematically with location
(Collias et al. 1974). Although it was not possible to assess the
possibility that differences in analytical techniques over time could have
influenced the apparent decreases, data collected since 1953 were probably
reasonably accurate (Appendix A).
5-211
-------
40
cc
z
o
0
V)
0
ANNUAL MEAN
I STANDARD ERROR
0 INDIVIDUAL OBSERVATION
---- SGNIFICANT REGRESSION LINE
(P < 0.05)
\
1
1950 1955 I960 1965
1
1970
YEAR
1975
198e 1985 1996
<
K
Z
ct
o
z
a
j/i
a
1950 1955 I960 1965 1970 1975 1980 1985 1990
YEAR
Figure 5.118. Concentrations of dissolved inorganic nitrate at 10- and 30-m depths in
the Dabob Bay study area during the algal bloom season.
5-212
-------
X
i/i
o
i
a.
O
O
a
LJ
1950
ANNUAL MEAN
I STANDARD ERROR
O INDIVIDUAL OBSERVATION
SIGNIFICANT REGRESSION LINE
(P < 0.05)
1955 I960 1965
1970 1975 1980 1985 1990
YEAR
°4
Ol
3
LJ
t-
I3
I/)
O
a.
o
o
o
Ul
>
_l
O ,
in I
t/>
O
o o
1950 1955 1960 1965 1970 1975 1980 1985 1990
YEAR
Figure 5.119. Concentrations of dissolved orthophosphate at the surface and at 10-m
depth in the Dabob Bay study area during the algal bloom season.
5-213
-------
o>
3
I3
(/)
o
Q.
O
o
o
LJ
81
to
O
ANNUAL MEAN
STANDARD ERROR
INDIVIDUAL OBSERVATION
SIGNIFICANT REGRESSION LINE
(P<0.05)
o o
e
1958 195S I960 1965
—i—
1979
YEAR
1975 1980 1985 1996
01
Q.
g
3
195« 1955 196* 1965
197*
YEAR
1975 19OT 1985 1999
Figure 5.120. Concentrations of dissolved orthophosphate at 30-m depth and chloro-
phyll 2 at the surface in the Dabob Bay study area during the algal bloom
season.
5-214
-------
It was not possible to determine why phosphate concentrations declined
since the 1950s at 10- and 30-m depths. One explanation is that rates of
consumption of phosphate by phytoplankton could have increased (see following
discussion of indicators of phytoplankton growth). Alternatively, oceanic
inputs of phosphate may have declined.
Indicators of Phvtoplankton Growth—Chlorophyll a data from 1979
through 1986 are plotted by year in Figures 5.120 and 5.121. No temporal
changes were evident. Percent dissolved oxygen saturation at the surface and
Secchi disk depth are plotted by year in Figure 5.122. No statistically
significant changes were detected for either variable. However, the
transparency of the water column is high in the Dabob Bay area (mean Secchi
depth was 6.0 m) and statistically significant increases in percent
dissolved oxygen saturation at depth since 1950 were found (Tables 5.15
and 5.16) (10-m depth: slope=+0.77 percent/yr, P=0.0003; 30-m depth:
slope=+0.53 percent/yr, P=0.001). Regressions of percent dissolved oxygen
saturation at depth since 1973 were not statistically significant (P>0.05).
The long-term increases in oxygen saturation at depth suggest that increasing
photosynthetic rates may have influenced oxygen concentrations. The long-
term declines in phosphate concentrations at 10- and 30-m depths support this
interpretation because these declines may be attributable to increased rates
of nutrient uptake by phytoplankton at these depths. Unfortunately, the
chlorophyll a data only covered the most recent 7 yr, which was not
sufficient to detect long-term trends.
Pollutants—Too few data on concentrations of sulfite waste liquor are
available to warrant analysis. However, large pulp mills do not exist in the
area. A statistically significant (P<0.05) decline in the concentrations
of fecal coliform bacteria was detected at the surface (Figure 5.123). This
trend appears to have been driven by a few low values that were reported
from 1976 through 1980. All values since 1981 were at the detection limit
(i.e., an overall decline was apparent in the data). All the concentrations
were well below Class AA water quality standards.
5-215
-------
30-
01
a.
o
a:
o
ANNUAL MEAN
J STANDARD ERROR
0 INDIVIDUAL OBSERVATION
—- SIGNIFICANT REGRESSION LINE
(P < 0.05)
1959 1955 1969 1965
1970
YEAR
1975 19M 1985 1990
30-
01
d
o.
o
1959 1955 1969 1965
1979
YEAR
1975 1980 1985 1999
Figure 5.121. Concentrations of chlorophyll a at 10- and 30-m depths in the Dabob Bay
study area during the algal bloom season.
5-216
-------
3001
V
a
200
z
LJ
o
0
LJ
o
1001
0-1
ANNUAL MEAN
I STANDARD ERROR
0 INDIVIDUAL OBSERVATION
SIGNIFICANT REGRESSION LINE
(P •: 0.05)
1959 1955 I960 1965
1970
YEAR
1975 1980 1985
1990
161
14
IS
|ie
Q.
U
Q 8
*
-------
oc
UJ
5
<
CD
3
IS
o _
O *
< o
oz
LU-
LL
LL
O
8
1
ANNUAL MEAN
I STANDARD ERROR
0 INDIVIDUAL OBSERVATION
—- SIGNIFICANT REGRESSION LINE
(P < 0.05)
VTiv»« ».».>
1950 1955 I960 1965 1976 1975 1986 1985 1996
YEAR
Figure 5.123. Log of concentrations of fecal coliform bacteria at the surface in the
Dabob Bay study area during the algal bloom season.
5-218
-------
Mid-Hood Canal
The study area includes the region between Hamma Hamma River and Hood
Point (see Figure 5.107). Class AA water quality standards apply in the
region, which lacks major urban influences. The largest source of fresh
water to the study area is the Skokomish River, the two outlets of which
combine to be the seventh largest river flowing into Puget Sound (see
Table 2.1), The Skokomish River contributes approximately 4 percent of the
total freshwater input to the sound. Other sources of freshwater are the
Dosewallips, Duckabush, and Hamma Hamma Rivers, each of which contributes
approximately 1 percent of the freshwater flow entering Puget Sound (see
Table 2.1). The Mid-Hood canal study area is narrow and averages about
165 m in depth. Circulation is sluggish because the volumes of freshwater
inputs and tidal flows, which are the major forces driving water movements
in Hood Canal, are small relative to the total volume of the system.
Environmental Conditions in the Study Area--
Mean salinity and water temperature values during the algal bloom period
are depicted in Figure 5.108. Data are available from 1952 through 1986.
Depth gradients of salinity and water temperature were well developed. The
mean salinity value was 3.3 ppt lower at the surface than at 10-m depth.
The mean water temperature value was 3.7° C higher at the surface than at
10-m depth. The salinity gradient was somewhat greater at the Mid-Hood
Canal site than at the Dabob Bay site. Lower mean surface salinity values
at the Mid-Hood Canal site (25.0 ppt vs. 26.6 ppt at the Dabob Bay site)
probably reflect the closer proximity to substantial freshwater sources. The
thermal gradient in the Mid-Hood Canal site was very steep. This gradient
is partly attributable to high surface temperatures. The mean water
temperature at the surface was 14.9° C, which was the highest surface mean
temperature observed in any study site except at Totten Inlet. The magnitude
of the density stratification in the Mid-Hood Canal study area suggests that
physical factors in the area would be conducive to the development of
intense algal blooms.
5-219
-------
The vertical distributions of concentrations of dissolved oxygen and
nutrients resembled those in the Dabob Bay study area (Figures 5.109 and
5.110). Mean dissolved oxygen concentration was highest at the surface
(10.1 mg/L), and relatively small differences between mean dissolved oxygen
concentrations at the surface and at 10-m depth were observed (0.8 mg/L).
Nutrient concentrations at the surface were frequently below analytical
detection limits. The depth gradient in nitrate concentrations was very
substantial. Nitrate concentrations at the surface averaged less than
6 percent of the nitrate concentrations at 30-m depth. Phosphate
concentrations at the surface averaged approximately one-third of the
phosphate concentrations at 30-m depth. As in the Dabob Bay study area, the
low concentrations of nutrients at the surface suggest that algal production
at the surface frequently could be limited by low nutrient concentrations.
Based on the percent dissolved oxygen saturation at the surface, the
intensity of algal blooms in the Mid-Hood Canal study area was relatively
high (Figure 5.111). However, algal bloom intensity was slightly lower at
this site than at the Dabob Bay site. The mean percent dissolved oxygen
saturation at the surface was 116 percent in the Mid-Hood Canal site and
120 percent in the Dabob Bay site. Too few data on the concentrations of
chlorophyll a were available to warrant interpretation (Appendix E).
Although the mean Secchi disk depth was high (6.0 m) at the Mid-Hood
Canal site, the only statistically significant (P<0.05, scaled with the
Bonferroni inequality) correlation with Secchi disk depth was a positive
correlation with surface water temperature values. This correlation
indicates that water clarity was high when the water column was warm. Such
conditions are typically conducive to the development of algal blooms.
However, such a scenario probably would lead to a negative correlation
between Secchi disk depth and temperature values, which is contrary to the
observed pattern. The cause of the positive correlation between Secchi
disk depth and water temperature values cannot be determined from the
available data. However, contributing factors could include high turbidity
during periods of cool water, such as occurs during the early spring.
Another contributing factor could be that low turbidity is associated with
warm water, presumably because suspended sediments and/or phytoplankton
5-220
-------
densities were low near the surface during the warmer months. Low
phytoplankton densities could occur near the surface in Hood Canal, even
during the bloom season, if, as in the Dabob Bay study area, maximum algal
densities were well below the surface. Also, growth rates at the surface
could have been limited by low nutrient concentrations. Depth profiles of
the composition of suspended material, algal density, and rates of primary
productivity would be useful for assessing the relative importance of these
factors as determinants of water clarity.
The absence of statistically significant (P<0.05, scaled with the
Bonferroni .inequality) correlations between nutrient concentrations and
other variables at the surface probably was due to a lack of analytical
sensitivity in the laboratory analyses of nutrient concentrations. The
analytical detection limits probably were not sufficiently low to allow
detection of most of the variation in the nutrient concentrations at the
surface. As in the Dabob Bay study area, nutrient concentrations were
higher at depth, and the negative correlations at 10-m depth between
nutrient concentrations and both dissolved oxygen concentrations and water
temperature values were statistically significant (P<0.05, scaled with the
Bonferroni inequality). These relationships would be expected at a
stratified site where algal blooms well below the surface waxed and waned in
intensity.
Geometric means of the concentrations of sulfite waste liquor and fecal
coliform bacteria were near analytical detection limits in the mid-Hood
Canal study area (Figure 5.112). These results are reasonable because there
are no large sources of these contaminants near the study area.
Water Quality Trends in the Study Area--
A summary of comparisons between water quality data collected before and
after 1973 is given in Table 5.15. Slopes from statistically significant
long-term and recent regressions of the water quality data against year are
given in Table 5.16.
5-221
-------
Physical Conditions — Plots of salinity and water temperature values by
year are shown in Figures 5.124-5.126. Statistically significant (PO.05)
declines in salinity values were detected at the surface and at 10- and 30-m
depths (Tables 5.15 and 5.16). At 30-m depth, the long-term decline appears
to have been driven by the recent decline. Although changes in station
location and data sources could have influenced the data, it appears that
these declines in salinity were real phenomena. Changes in station location
would have introduced an apparent decline in salinity into the data. The
station sampled recently (Ecology's Station HCB003) is closer to the
Skokomish River than are the stations sampled earlier (University of
Washington's Stations HCB543, HCB544, HCB545) (see Table 5.13 and
Figure 5.107). However, salinity profiles generally are flat in the
Mid-Hood Canal region (Collias et al. 1974). Moreover, salinity values at
the three University of Washington stations did not differ significantly,
although the University of Washington stations were farther from each other
than the Ecology station was from University of Washington's Station HCB545.
Station HCB545 was the University of Washington station located closest to
the Skokomish River. Thus, changes in station location do not seem to have
introduced the apparent salinity declines into the data. Data compatibility
checks (discussed in Chapter 4 and Appendix D) did not indicate that salinity
determinations by the University of Washington and Ecology differed
systematically from each other. Therefore, the interpretation that salinity
values in the Mid-Hood Canal study area have declined since the early 1950s
appears credible.
As was noted for the Dabob Bay study area, water temperature values at
10- and 30-m depths have increased since the early 1950s (Tables 5.15
and 5.16). These changes generally coincided with the changes in the air
temperature data from Seattle-Tacoma International Airport, which indicated
that a cool period existed during much of the 1950s. Because horizontal
temperature profiles in the study area do not vary systematically with
location along the central axis of Hood Canal (Collias et al. 1974),
artifacts caused by changes in station location do not appear to have
introduced the apparent temperature increases into the data. As discussed
above for Dabob Bay, differences in analytical techniques between the
University of Washington and Ecology could have influenced the apparent
5-222
-------
a
a.
^x
f 80
j
10
10
ei
I960 1955 I960 1965 1979 197S 1986 1985 1996
YEAR
40-
30"
ae-
«••
6 •
IS
^A A^ A J
156 1955 1966 1965 1976
ft * o
o o ° o
ANNUAL MEAN
J STANDARD ERROR
0 INDIVIDUAL OBSERVATION
— - SIGNIFICANT REGRESSION LINE
(P < 0.05)
I'll
1975 1986 1985 19!
YEAR
Figure 5.124. Salinity values at the surface and at 10-m depth in the Mid-Hood Canal
study area during the algal bloom season.
5-223
-------
40"
30-
a
a.
z
<
10
*"»-
ANNUAL MEAN
I STANDARD ERROR
0 INDIVIDUAL OBSERVATION
—- SIGNIFICANT REGRESSION LINE
(P < 0.05)
1950 1955 1960 1965 1970 1975 1980 1985 1999
YEAR
24
23
22
21
°*19
?18
I"
u'6
§15
13
12
11
10
9
8
1950 1955 1960 1965 1970 1975 1980 1985 1990
YEAR
Figure 5.125. Salinity values at 30-m depth and water temperatures at the surface in
the Mid-Hood Canal study area during the algal bloom season.
5-224
-------
24
23"
aa
ai
|l9
f 17-
|l5-
< H
£ 13"
2 12-
" 11
10-
9
8-
7-
1<
24
23
22
21
^20
S19H
? 18
I;';
|15-
o:
u 12-
*~ 11
10
9
8
7 -
IS
Figure J
ANNUAL MEAN
I STANDARD ERROR
0 INDIVIDUAL OBSERVATION
---- SIGNIFICANT REGRESSION LINE
(P < 0.05)
O 0
0 fi
" 0 ° L[
L? y^ .f ywHfrj
'^Y i^* *y
o
> i i i i 1 1 1 r
>S0 1955 1960 1965 1970 1975 1980 1985 1990
YEAR
o
o o
t ffi^v^J
t^ ^^^ ^4^11^"""" "•TO; ^o 11
o
50 1955 1960 1965 1970 1975 1980 1985 1990
YEAR
5.126. Water temperatures at 10- and 30-m depths in the Mid-Hood Canal
study area during the algal bloom season.
5-225
-------
increases in the water temperature data. However, the most likely
interpretation is that the apparent increases in water temperature at
10- and 30-m depths were real phenomena.
Dissolved Oxygen — Plots of dissolved oxygen concentrations by year are
shown in Figures 5.127 and 5.128. The Class AA water quality standard (see
Table 4.2) was violated once in surface waters and sporadically at 10-m
depth. Concentrations at 30-m depth were usually below the standard. The
mean dissolved oxygen concentration at this depth was 0.2 mg/L below the
standard. As was observed in the Dabob Bay study area, dissolved oxygen
concentrations at 10- and 30-m depths appear to have increased (Tables 5.15
and 5.16). At 10-m depth, the recent trend appears to have driven the
long-term trend. The likely influence of the changes in station locations
over time would have been to artificially decrease dissolved oxygen
concentrations at depth. This increase would have resulted because a tongue
of water with low concentrations of dissolved oxygen often extends northward
from Lynch Cove into the southern portion of the Mid-Hood Canal study area
(Collias et al. 1974), and because the recent data were collected at the
southern-most station included in the data set. Because the observed
increases in dissolved oxygen concentrations were contrary to the decreases
that might have been introduced into the data by the changes in station
location, station changes probably did not introduce artificial changes in
dissolved oxygen concentration into the data.
Nutrients—Plots of nitrate concentrations by year are shown in
Figures 5.128 and 5.129. Because data are only available since 1977,
comparisons of data collected before and after 1973, and long-term
regressions by year could not be performed. The recent regressions of
nitrate concentrations by year were not statistically significant (P>0.05)
(Table 5.16). However, negative slopes were found at all three depths, and
the slopes were nearly significant for data collected at 10-m depth
(slope=-0.76 ug-at/L/yr, P=0.10) and 30-m depth (slope=-0.83 ug-at/L/yr,
P=0.054). These apparent declines in nitrate concentration do not appear to
have been artifacts of changes in station location or analytical technique
because all the nitrate data came from the same station and agency.
5-226
-------
20
19
IS
17
16
"oi 14
113
o 11
Q 9
2 8
3 7
«" c
in 6
3
a
i
e
ANNUAL MEAN
I STANDARD ERROR
0 INDIVIDUAL OBSERVATION
SIGNIFICANT REGRESSION LINE
(P < 0.05)
1950 1955 I960 1965
1970
YEAR
1975 1980 1985
1990
20
19
18-
17-
16
^oi 14
113
5121
oil
x 10
Q 9
> 8
o 71
ul c
c/1 6
Q 5
4"
3"
2"
1 :
0-
1950 1955 1960 1965 1970 1975 1980 1985 1990
YEAR
Figure 5.127. Concentrations of dissolved oxygen at the surface and at 10-m depth
in the Mid-Hood Canal study area during the algal bloom season.
5-227
-------
20
19
18
17
161
MS
14
13
12
111
10
9
8
7
6
5
4
3
a
1
0
ANNUAL MEAN
STANDARD ERROR
INDIVIDUAL OBSERVATION
SIGNIFICANT REGRESSION LINE
(P < 0.05)
o o
1950 1955 I960 1965 1970 1975 1989 1985 1999
YEAR
a
^
z
o
ct
o
z
o
UJ
o
(ft
0
1950 1955 1960 1965
1979
YEAR
1975 1989 1985 1999
Figure 5.128. Concentrations of dissolved oxygen at 30-m depth and dissolved
inorganic nitrate at the surface in the Mid-Hood Canal study area
during the algal bloom season.
5-228
-------
40 '
~ V
_J
15
1
J30"
UJ
<
1-
z
7 CV
£
o
o:
O
Z
^
o
u
o
VI
V)
5
0-
ANNU
I STAN(
O INDIV
SIGNI
(P<0
O
0
I
t
\
1\
o\
1
<
o
i i i 1 1 1
1950 1955 1960 1965 1970 1975
YEAR
ALMEAN
)AHD ERROR
DUAL OBSERVATION
=ICANT REGRESSION LINE
05)
O
0 0
0 0
f 0 0
1
t
s A
^^ /n
w
oiollcto
— i 1 r
1980 1985 1990
<
^
Z
O
a:
O
Z
o
in
in
1950 1955 1960 1965 1970 1975 1980 1985 1990
YEAR
Figure 5.129. Concentrations of dissolved inorganic nitrate at 10- and 30-m depths in
the Mid-Hood Canal study area during the algal bloom season.
5-229
-------
Plots of phosphate concentrations by year are shown in Figures 5.130
and 5.131. Comparisons with the nonparametric ANOVA of mean phosphate
concentrations recorded before and after 1973 indicated that the recent
concentrations were significantly (P<0.05) lower at both depths (Table 5.15).
The regressions of phosphate concentration by year had negative slopes at
both depths, although neither slope was significant (P=0.1) (Table 5.16).
Phosphate concentrations did not change significantly (PX).6) at the
surface. As was discussed previously, changes in station location and data
sources do not appear to have introduced apparent changes into the phosphate
data. As was discussed for Dabob Bay, increase in nutrient consumption by
phytoplankton or decreases in oceanic inputs may have contributed to the
decreases in phosphate concentrations.
Indicators of Phvtoplankton Growth—No statistically significant
temporal trends were detected for indicators of phytoplankton growth.
However, long-term data were limited to the percent dissolved oxygen
saturation at the surface. Oxygen saturation data for the surface are
plotted in Figure 5.131. Data on Secchi disk depth date back to 1977
(Figure 5.132). However, the transparency of the water column was high in
the Mid-Hood Canal region (mean Secchi disk depth was 6.0. m), and long-term
increases in percent dissolved oxygen saturation were found at depth (10-m
depth: slope=+0.81 percent/yr, P=0.0001; 30-m depth: slope=+0.61, P=0.0009).
The recent increase (since 1977) in percent dissolved oxygen saturation at
10-m depth (slope=+3.46 percent/yr, P=0.006) appears to have driven the
long-term trend at this depth, but the recent regression at 30-m depth was
not statistically significant (P>0.6).
Because nitrate concentrations have declined and oxygen saturation
percentages have increased at 10-m depth since 1977, it appears that
photosynthetic activity and phytoplankton abundance have increased at this
depth. Unfortunately data on chlorophyll a concentrations are not available
to test this hypothesis. No explanation is available for the long-term
increase in percent dissolved oxygen saturation at 30-m depth. However,
changes in photosynthetic activity at depth might have influenced the data.
5-230
-------
VI
O
a.
O
o
a
UJ
Si
in
ANNUAL MEAN
I STANDARD ERROR
0 INDIVIDUAL OBSERVATION
SIGNIFICANT REGRESSION LINE
(P < 0.05)
Ol
3
Ul
V)
o
a.
O
O
a
1
in 1
£
o
1950 1955 196e 1965 1979 1975 1988 1985 1990
YEAR
5
1958 1955 I960 1965 1978 1975 1980 1985 1990
YEAR
Figure 5.130. Concentrations of dissolved orthophosphate at the surface and at 10-m
depth in the Mid-Hood Canal study area during the algal bloom season.
5-231
-------
X
Ol
3
o
a.
O
o
Q
Ul
_l
O
ANNUAL MEAN
I STANDARD ERROR
O INDIVIDUAL OBSERVATION
—- SIGNIFICANT REGRESSION LINE
(P < 0.05)
1950
1955 1963 1965
1970
YEAR
1975
1980 1985 1990
300-
e
a
§ 200
tr
D
I—
in
O
>-
O 100
a
o
i/i
IS)
o
o o
o o
1950 1955 1960 1965
1970
YEAR
1975 1980 1985
1990
Figure 5.131. Concentrations of dissolved orthophosphate at 30-m depth and percent
dissolved oxygen saturation at the surface in the Mid-Hood Canal study
area during the algal bloom season.
5-232
-------
?
X
111
Q
W
Q
X
O
o
HI
(A
lt>
14 '
la-
ic-
s'
6 •
4
2
0
ANNUAL MEAN
I STANDARD ERROR
0 INDIVIDUAL OBSERVATION
---- SIGNIFICANT REGRESSION LINE
(P < 0.05)
°0
If
«_J4\o
o I
0
i i ~i i 1 1 1 —
1959 1955 1969 1965 1979 1975 1989
YEAR
o
I $
.0 4
ff^&r* * \
' T O •
00 «
1 1
1985 19
c
m
o
o
u
fr
LL
U.
O
o
1959 1955 1969 1965 1979 1975 1989 1985 1999
YEAR
Figure 5.132. Secchi disk depth and log of concentrations of fecal coliform bacteria
at the surface in the Mid-Hood Canal study area during the algal bloom
season.
5-233
-------
Pollutants—The only data available for pollutants included in the.
characterization study are data on concentrations of fecal coliform bacteria
at the surface (Figure 5.132). No significant temporal trend was detected
in the fecal coliform data.
South Hood Canal
The study area is located in a rural area near Sisters Point. It is
approximately 6.5 km east of The Great Bend in Hood Canal (see Figure 5.107)
and 18 km west of Lynch Cove, the head of Hood Canal. It is the study area
in Hood Canal most removed from oceanic influences. Class AA water quality
standards apply in the region. There are no major population centers near
the study area, but there are many summer homes along the shoreline. The
study area is roughly 35-40 m deep. Two rivers, the Skokomish and the
Tahuya, flow into Hood Canal near Sisters Point. Combined, these rivers
contribute approximately 5 percent of the total freshwater flow into Puget
Sound (see Table 2.1). Circulation below the surface is sluggish in the
study site, and the area is prone to episodes of low dissolved oxygen
concentrations in sub-surface water, particularly in late summer. Bottom
water usually is replaced only annually in late summer or early autumn
(Collias et al. 1974).
Environmental Conditions in the Study Area--
Mean salinity and water temperature values during the algal bloom season
are shown in Figure 5.108. Data are available from 1952 through 1986.
Depth gradients of these two variables in the South Hood Canal area were the
largest of all the Hood Canal sites, with low salinity values and high
temperatures having been recorded at the surface. The mean salinity value
was 4.8 ppt lower at the surface than at 10-m depth. The mean water
temperature value was 3.9° C higher at the surface than at 10-m depth. The
South Hood Canal study area can be characterized as having very substantial
density stratification and very low rates of vertical mixing and
circulation. These physical conditions would be expected to be highly
conducive to the development of algal blooms and, as a consequence, low
dissolved oxygen concentrations at depth.
5-234
-------
Given the physical environment in the South Hood Canal study area, the
depth distribution of dissolved oxygen concentrations at this site was
distinct from the depth distributions observed at the Dabob Bay and Mid-Hood
Canal sites (Figure 5.109). The surface concentrations of dissolved oxygen
were similar in the three areas (10.1-10.5 mg/L), but the dissolved oxygen
concentrations were much lower at 10-m and 30-m depth in the South Hood
Canal area. At the South Hood Canal study area, the mean dissolved oxygen
concentration at 10-m depth was only 74 percent of the mean value at 10-m
depth at the Dabob Bay study area. At 30-m depth, the mean concentration of
dissolved oxygen was only 60 percent of the mean value at this depth in the
Dabob Bay site. The lower dissolved oxygen concentrations at 10- and 30-m
depths in the South Hood Canal study area were probably influenced by the
very low dissolved oxygen concentrations in the deep source water. Other
surveys of Hood Canal have also reported very low dissolved oxygen
concentrations in the deep waters in this area during spring and summer
(e.g., Collias et al. 1974).
Although the vertical distributions of nutrients at the South Hood
Canal site were similar to those in the Dabob Bay and Mid-Hood Canal sites,
the gradients were more extreme in South Hood Canal (Figures 5.109
and 5.110). Surface concentrations of nitrate and -phosphate at the South
Hood site were very low, typically below the analytical detection limits.
Although low nutrient concentrations frequently could have limited algal
growth in the surface water, nutrient concentrations at 10- and 30-m depths
in South Hood Canal were the highest of all the Hood Canal sites. Mean
nitrate concentrations were 13.0 ug-at/L and 27.0 ug-at/L at 10- and 30-m
depths, respectively. Mean phosphate concentrations were 2.1 ug-at/L and
2.9 ug-at/L at 10- and 30-m depths, respectively.
Based on the mean percent dissolved oxygen saturation at the surface
(115 percent), the intensity of the algal blooms at the South Hood Canal
site was high. However, it was slightly lower than the intensities of the
blooms at the Mid-Hood Canal and Dabob Bay sites (Figure 5.111). No
chlorophyll a data are available for the South Hood Canal site. The water
was more turbid in the South Hood Canal study area than it was in the other
5-235
-------
two Hood Canal study sites, as the mean value of Secchi disk depth (4.8 m)
was lowest in the South Hood Canal study area (Figure 5.111). However, the
difference in turbidity may not have been due to phytoplankton densities
because indicators of phytoplankton growth did not exhibit higher values at
the South Hood site than at the Mid-Hood Canal or Dabob Bay sites. The lack
of significant correlations (P<0.05, scaled with the Bonferroni inequality)
between Secchi disk depth and either surface dissolved oxygen concentration
or surface percent dissolved oxygen saturation (Appendix E) suggests that
the changes in turbidity were too variable to yield a reliable indication of
phytoplankton density.
As discussed for the Dabob Bay and Mid-Hood Canal sites, the lack of
correlations between nutrient concentrations and the values of the other
variables at the surface probably was due to the lack of analytical
sensitivity in the laboratory analyses of nutrient concentrations. Like the
other two sites on Hood Canal, nutrient concentrations were higher at depth.
As would be expected at a stratified site where algal blooms waxed and waned
in intensity, statistically significant (PO.05, scaled with the Bonferroni
inequality) negative correlations were found at 10-m depth between nutrient
concentrations and both dissolved oxygen concentrations and water temperature
values. Thus, as in Dabob Bay, much of the phytoplankton biomass probably
occurred well below the surface.
Geometric means of the concentrations of sulfite waste liquor and fecal
coliform bacteria were near analytical detection limits in the South Hood
Canal area (Figure 5.112), presumably because the area is relatively
undeveloped.
Water Quality Trends in the Study Area--
A summary of comparisons between water quality data collected before and
after 1973 is given in Table 5.15. Slopes from statistically significant
long-term and recent regressions of the water quality data by year are given
in Table 5.16.
5-236
-------
Physical Conditions—Pints of salinity and water temperature values by
year are shown in Figures 5.133-5.135. No statistically significant
changes in salinity values were detected for surface water. However,
salinity values at 10- and 30-m depths apparently have declined since 1952.
Changes in salinity values since 1976 were not statistically significant
(PXK05).
It does not appear that the above changes in salinity values were
artifacts of changes in station location and data sources. However, this
possibility cannot be thoroughly evaluated. Although horizontal salinity
gradient was recorded in the area (Collias et al. 1974), the Ecology station
sampled since 1968 was located approximately half way between the two
University of Washington stations sampled from 1952 through 1966
(Figure 5.107, Table 5.13). Therefore, the horizontal salinity gradient
probably did not affect the average data values. As was discussed for Dabob
Bay, differences in the analytical methods used to determine salinity by the
University of Washington and Ecology do not appear to explain the apparent
changes in salinity values (Chapter 4 and Appendix D). In summary, salinity
values at 10- and 30-m depths in the South Hood Canal study area appear to
have declined. The cause of the salinity declines is not known, but
decreased inputs of oceanic water may have been involved.
Water temperature values at 10- and 30-m depths appear to have increased
(Tables 5.15 and 5.16). As was discussed for Dabob Bay, these increases may
have been caused by the increased air temperatures, as were detected at the
Seattle-Tacoma International Airport (see Figure 5.1). Mean annual air
temperatures were cool from 1948 through 1955. Although water temperatures
in Hood Canal are often higher near the head of Lynch Cove (Collias et al.
1974), changes in station locations over time probably did not introduce
these apparent increases into the data. As discussed for Dabob Bay,
differences in analytical techniques between the University of Washington
and Ecology might have contributed to the apparent increases in water
temperatures (see Chapter 4 and Appendix D). Thus, it appears that water
temperatures have increased in the South Hood Canal study area, although
changes in station location and data sources over time may have contributed
to these apparent changes.
5-237
-------
40"
30
a
a
Eae
z
_l
I/)
e
1950 1955 I960 1965 1970 1975 1980 1985 1990
YEAR
40'
a
a
Eae-
0i
ANNUAL MEAN
I STANDARD ERROR
0 INDIVIDUAL OBSERVATION
SIGNIFICANT REGRESSION LINE
(P •: 0.05)
1950 1955 1960 1965 1970 1975 1980 1985 1990
YEAR
Figure 5.133. Salinity values at the surface and at 10-m depth in the South Hood Canal
study area during the algal bloom season.
5-238
-------
40
30
a.
a
v»x
£20
z
10-
0
- ANNUAL MEAN
J STANDARD ERHOH
0 INDIVIDUAL OBSERVATION
--- SIGNIFICANT REGRESSION LINE
(P < 0.05)
1950 1955 1960 1965
1970
YEAR
1975 1980 1985
1990
24-
23"
22
21
SIS
£ 13
5 12"
*~ 11
10;
9;
8
7-
1950 1955 1960 1965 1970
YEAR
1975 1980 1985 1990
Figure 5.1 34. Salinity values at 30-m depth and water temperatures at the surface in the
South Hood Canal study area during the algal bloom season.
5-239
-------
24
23"
22"
21
201
19
18
17
16
15
14
13
12
11
10
9
8
7
ANNUAL MEAN
J STANDARD ERROR
0 INDIVIDUAL OBSERVATION
---- SIGNIFICANT REGRESSION LINE
(P < 0.05)
1950 1955 I960 1965 1976 1975 1980 1985 1990
YEAR
24
23'
22-
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
1950 1955 1960 1965 1970 1975 1980 1985
YEAR
1990
Figure 5.135. Water temperatures at 10- and 30-m depths in the South Hood Canal study
area during the algal bloom season.
5-240
-------
Dissolved Qxvaen--P1ots of dissolved oxygen concentrations by year are
shown in Figures 5.136 and 5.137. There was no evidence that the Class AA
water quality standard (see Table 4.2) was violated in surface waters.
Concentrations below the standard occurred occasionally at 10-m depth. The
frequency of violation at 10-m depth was higher than at the Mid-Hood Canal
study area. Concentrations were usually below the standard at 30-m depth.
The mean dissolved oxygen concentration at this depth was 2.7 mg/L below the
standard. Statistically significant (P<0.05) increases in dissolved oxygen
concentrations were found at 10- and 30-m depths. The increase since 1976
was statistically significant only at 10-m depth (Tables 5.15 and 5.16).
The long-term increase at 10-m depth appears to have been driven in large
part by the recent increase. The recent increase at 10-m was not an
artifact because the same sampling station and data source was used since
1976. The long-term increase at 30-m depth was detected statistically.
However, much of this trend appears to have been driven by some very low
values reported in 1952, the first year from which data were obtained.
Thus, the most important change detected for dissolved oxygen concentrations
at the South Hood Canal study area was an increase at 10-m depth since 1976.
Nutrients—Plots of nitrate concentrations by year are shown in
Figures 5.137 and 5.138. Because data are available only since 1976,
comparisons of data collected before and after 1973, and long-term
regressions of the data by year could not be performed. No statistically
significant (PO.05) changes in nitrate concentrations were detected at the
surface, but, as at the Dabob Bay and Mid-Hood Canal sites, the analytical
methods probably were not sufficiently sensitive to detect nitrate
concentrations reliably at the ambient surface concentrations. A
statistically significant (PO.05) decrease in nitrate concentrations was
detected at 30-m depth (Table 5.16); at 10-m depth the decline was nearly
significant (P=0.07). These decreases were not artifacts of changes in
station location or data sources because the nitrate data were all collected
at the same sampling station by Ecology.
Statistically significant long-term decreases in phosphate
concentrations were detected at the surface and at 10- and 30-m depths
5-241
-------
20
19
18
17"
16
y->. . — .
l!5:
oil
gie-
o 9-
LJ O •
> »
£ 7-
5 5
3
2
1
0"
ANNUAL MEAN
J STANDARD ERROR
0 INDIVIDUAL OBSERVATION
— - SIGNIFICANT REGRESSION LINE
(P < 0.05)
o — — —
0 ° ^O °
« I* ° « ° ° ? T 8 ° JJ ° 1
\lf»\t \ » t $ IJHIM ,A*V^
t; TV ^^-^H^tlf o|»
o i * ° o
" T 1 1 1 1 1 I I 1
1950 1955 1960 1965 1970 1975 1980 1985 1990
YEAR
20
19-
18-
17-
16-
CM5-
^oi 14
£ 1 1
~ 1£
0 11
gie
o 9
2 8
^ 7
m /•
0 g 0 0 ° *^
° xf^ 1. -f/^' '
" " sT\ ~ ~ \ to$^ ^***a* H O
<^* o ° \y * ° 8 o o
o y o
o 1
o
1 1 1 1 1 1 1 1 1
1950 1955 1960 1965 1970 1975 1980 1985 1990
YEAR
Figure 5.136. Concentrations of dissolved oxygen at the surface and at 10-m depth in the
South Hood Canal study area during the algal bloom season.
5-242
-------
£0
19
18
17"
16
13
oil
gie
a 9
U o -I
> 8 1
T-
6"
0 5
4 -
3
2
1
0
ANNUAL MEAN
I STANDARD ERROR
0 INDIVIDUAL OBSERVATION
---- SIGNIFICANT REGRESSION LINE
(P c 0.05)
1 1 1 1 1 1 1 1
1950 1955 1960 1965 1970 1975 1980 1985
YEAR
1998
cr
o
CE
O
Z
o
LJ
t/\
o
1950 1955 1960 1965 1970 1975 1980 1985 1990
YEAR
Figure 5.137. Concentrations of dissolved oxygen at 30-m depth and dissolved inorganic
nitrate at the surface in the South Hood Canal study area during the algaf
bloom season.
5-243
-------
40
30
CK
Z
O
DC
O
O
in
0
ANNUAL MEAN
J STANDARD ERROR
0 INDIVIDUAL OBSERVATION
SIGNIFICANT REGRESSION LINE
(P < 0.05)
000000 o
1950 1955 I960
1965 1970 1975 1989 1985 1990
YEAR
o:
20
o:
O
Z
o
UJ
o
1/1
tf)
o
I 7 I I I I I I ^
1950 1955 1960 1965 1970 1975 1980 1985 1990
YEAR
Figure 5.138. Concentrations of dissolved inorganic nitrate at 10- and 30-m depths in
the South Hood Canal study area during the algal bloom season.
5-244
-------
(Tables 5.15 and 5.16, Figures 5.139 and 5.140). Changes in phosphate
concentrations since 1976 were not statistically significant. The values
recorded from the early portion of the data set generally were higher than
the values recorded from the recent portion of the data set. As discussed
above, changes in station locations do not appear to have introduced these
decreases into the data, although phosphate concentrations in Hood Canal are
often higher close to the head of Lynch Cove (Collias et al. 1974). As
discussed for Dabob Bay (see also Chapter 4 and Appendix D), changes in
analytical techniques probably had little effect on the phosphate data.
The apparent long-term declines in phosphate concentrations in South Hood
Canal appear to have been real phenomena. The cause of these phosphate
declines may have been increased photosynthesis (see below) or decreased
oceanic inputs.
Indicators of Phvtoplankton Growth—Data on chlorophyll a concentrations
are not available. Percent dissolved oxygen saturation at the surface and
Secchi disk depth are plotted by year in Figures 5.140 and 5.141. No
statistically significant (PO.05) temporal trends were detected for either
variable. As was noted for the Dabob Bay and Mid-Hood Canal areas, the
percent dissolved oxygen saturation has increased at depth since 1952 (10-m
depth: slope=+1.16 percent/yr, P=0.0001; 30-m depth: slope=+0.71 percent/yr,
P=0.0003). Although the long-term increase at 10-m depth appears to have
been driven by the recent increase (since 1976) at 10-m depth (slope=+4.81,
P=0.002), the recent changes in data values at the surface and 30-m depth
were not significant. These results suggest that photosynthetic rates and
algal abundances near 10-m depth may have increased since 1976.
Pollutants—The data for the concentration of sulfite waste liquor were
not analyzed because only a few points were available (Appendix E). Concen-
trations of fecal coliform bacteria since 1976 are plotted by year in
Figure 5.141. No significant temporal trends were evident. Many values
were at the detection limit.
5-245
-------
o>
3
o
I
Q.
O
O
Q
t/i
o
0
ANNUAL MEAN
I STANDARD ERROR
0 INDIVIDUAL OBSERVATION
---- SIGNIFICANT REGRESSION LINE
(P < 0.05)
1950 1955 1966 1965
1970
YEAR
1975 1989 1985 1990
o
Q.
O
O
a
1-
yt
o
e
1950 1955 1960 1965 1970
YEAR
1 I I T
1975 1980 1985 1990
Figure 5.139. Concentrations of dissolved orthophosphate at the surface and at 10-m
depth in the South Hood Canal study area during the algal bloom season.
5-246
-------
5 '
Ol
J
£3
i/i
o
a.
o
o
o
0-L,
1950
1955 I960 1965
1978
YEAR
1975 1980 1985 1990
300
2 200
1/1
z
u
>-
O 100
a
o
to
ANNUAL MEAN
J STANDARD ERROR
0 INDIVIDUAL OBSERVATION
SIGNIFICANT REGRESSION LINE
(P < 0.05)
1950 1955 1960 1965 1970 1975 1980 1985 1990
YEAR
Figure 5.140. Concentrations of dissolved orthophosphate at 30-m depth and percent
dissolved oxygen saturation at the surface in the South Hood Canal study
area during the algal bloom season.
5-247
-------
I
£
Ul
Q
^
w
Q
I
U
O
LU
(0
16"
•
14"
12'
10
8
6
4
a
0
ANNUAL MEAN
J STANDARD ERROR
0 INDIVIDUAL OBSERVATION
— SIGNIFICANT REGRESSION LINE
(P < 0.05)
O
0
1 1
o
° !^W ?
o / w *
/ y T
rt/i , V O 1
t 1 1 r Hi
* ^T~T! *
f 1 0 O 0
i 1 0 0
o
1959 1955 I960 1965 1979 1975
YEAR
1980
1985 1999
P
o
f
8*
21
Uj^
u.
u.
O
O
O
1959 1955 1960 1965 1970 1975 1980 1985 1990
YEAR
Figure 5.141. Secchi disk depth and log of concentrations of fecal coliform bacteria at the
surface in the South Hood Canal study area during the algal bloom season.
5-248
-------
Summary of Results for Hood Canal
Major findings for Hood Canal are provided in this section. Environ-
mental conditions in the study areas are summarized and compared. A brief
assessment of the sensitivity of the Hood Canal study areas to pollution is
provided. Trends in water quality are also summarized.
Environmental Conditions--
Depth gradients of salinity and water temperature values were well
developed in all the study areas on Hood Canal. Salinity gradients were
steepest in the South Hood Canal study area, the study area closest to
substantial sources of fresh water. The mean salinity value was 4.8 ppt
lower at the surface than at 10-m depth at this site. Thermal gradients
also were steepest in the South Hood Canal study area. The mean water
temperature value was 3.9° C higher at the surface than at 10-m depth at
this site. These results suggest that the rates of vertical mixing and
circulation are low throughout Hood Canal, and are lowest in the South Hood
Canal area.
Substantial depth gradients in dissolved oxygen concentrations were
well developed in all three Hood Canal study areas. Mean surface dissolved
oxygen concentrations (10.1-10.5 mg/L) were similar in all three areas.
Dissolved oxygen concentrations at 10-m depth were nearly as high as the
concentrations were at the surface in the Dabob Bay and Mid-Hood Canal study
areas, suggesting that photosynthetic rates tended to be high at depth in
those two areas. Low dissolved oxygen concentrations at depth were most
prevalent in the South Hood Canal site. The mean dissolved oxygen
concentration at 30-m depth was only 4.3 mg/L (46 percent saturation) at
this site.
Depth gradients of nutrient concentrations were highly developed in all
three study areas. All three study areas exhibited very low nutrient con-
centrations at the surface (e.g., mean nitrate concentrations <2 ug at/L).
The depth gradients for nitrate concentrations were particularly steep.
Mean nitrate concentrations at the surface typically were less than
5-249
-------
10 percent of the mean concentrations at 30-m depth. Because vertical mixing
rates are low and because the photic zone tends to be deep in Hood Canal,
the low nitrate concentrations at the surface suggest that nutrients could
limit the production of phytoplankton in near-surface waters. In general,
nutrient concentrations were highest at depth in the South Hood Canal site.
Lower nutrient concentrations were found at depth in the more northern study
sites. For example, the mean nitrate concentration at 30-m depth was
27.0 ug-at/L in the South Hood Canal site, 22.3 ug-at/L in the Mid-Hood Canal
site, and 20.5 ug-at/L in the Dabob Bay site.
The intensity of algal blooms (determined by the percent dissolved
oxygen saturation at the surface) was high in all three Hood Canal sites.
The blooms were most intense in the Dabob Bay area (mean surface dissolved
oxygen saturation was 120 percent) and were least intense in the South Hood
Canal area (mean surface dissolved oxygen saturation was 115 percent). The
data for Secchi disk depth indicate that the clarity of the surface water in
the Dabob Bay and Mid-Hood Canal study areas was relatively high (mean
Secchi disk depths were 6.0 m), and that the surface water was more turbid
in the South Hood study area (mean Secchi disk depth was 4.8 m). The
relative contributions of phytoplankton and other suspended particulate
material to these turbidity patterns is unknown.
The chlorophyll a data from Dabob Bay indicate that the highest
concentration of chlorophyll a occurred well below the surface. By infer-
ence, high chlorophyll a concentrations below the surface also may have
existed in the Mid-Hood and South Hood areas. The occurrence of low surface
concentrations of chlorophyll a in the Dabob Bay study area supports the
hypothesis that low nutrient concentrations may limit the growth of
phytoplankton in the surface waters of Hood Canal.
The concentrations of sulfite waste liquor and fecal coliform bacteria
were near analytical detection limits in all of the Hood Canal study areas.
Because the region is relatively rural and undeveloped, major impacts from
pollutants would not be expected (Singleton, L., 30 November 1987, personal
communication; Tarr, M., 30 November 1987, personal communication).
5-250
-------
Sensitivity to Nutrient Enrichment--
Because of limitations in the capacity to export or assimilate
pollutants without deleterious ecological effects, all three Hood Canal
study sites appear to be sensitive to inputs of excess nutrients. Nitrate
concentrations were very low in the surface waters of the sites, which
suggests that nitrogen inputs to surface waters would be rapidly used by
growing phytoplankton. Because flushing rates in the study areas are low,
export rates for pollutants would also be low. The potential for
deleterious impacts of pollutants is probably highest in the South Hood
Canal site, which is the shallowest and least flushed of the study areas.
Inputs of small amounts of nutrients to deep water in northern or
central Hood Canal (i.e., below the photic zone and pycnocline) might not
have a substantial impact on phytoplankton growth. Although flushing rates
are low, the volume of deep water along much of the length of Hood Canal
could dilute pollutant inputs at depth. Vertical mixing rates are low in
the system, which suggests that nutrients discharged to deep water might not
reach the photic zone during the bloom season. Based on the available
information, it would seem likely that nutrients discharged at depth to
northern or central Hood Canal would be exported during the autumnal
replacement of deep water that occurs in Puget Sound.
Trends in Water Quality--
The three study areas in Hood Canal exhibited similar patterns of
change in water quality. Information in the following discussion is based
on the material in Tables 5.15 and 5.16. The similarities in the changes
detected at the three sites suggest that changes in station location may not
have affected the data substantially. However, no definitive analysis of
this hypothesis was possible. Because the same data sources were used at
all three sites, the influences of changes in the data sources over time
probably were the same in all three sites. Thus, the potential impact of
changes in data sources cannot be evaluated by comparisons among the study
areas.
5-251
-------
Physical Conditions--Sub-surface salinity values appear to have declined
and sub-surface water temperature values appear to have increased since the
early 1950s in all three Hood Canal study areas. Changes in station
locations and data sources do not appear to have contributed to the salinity
decreases. However, changes in the data sources may have contributed to the
apparent temperature increases. The cause(s) of the decreases in salinity
values in Hood Canal are unknown, but may involve decreased inputs of
oceanic water. The temperature increases generally coincided with the
trends in air temperature at the Seattle-Tacoma International Airport, which
indicate that the early 1950s was a relatively cool period.
These changes in the physical conditions in Hood Canal suggest that
hydrographic factors have evolved in the area since the 1950s. The absence
of changes in salinity and water temperature values at the surface suggests
that the physical factors affecting the surface water were distinct from the
physical factors affecting the water at depth. The extreme density
stratification in Hood Canal may allow changes at depth to occur
independently of changes at the surface.
Dissolved Oxygen—There was no evidence for substantial changes in
dissolved oxygen concentrations Jn the surface waters, nor was there a
substantial number of violations of the Class AA water quality standard (see
Table 4.2) for dissolved oxygen. The frequency of violations increased with
depth at all study areas as well as with distance from the mouth of Hood
Canal. However, dissolved oxygen concentrations have increased steadily at
10- and 30-m depths in all three Hood Canal study areas. There have been no
major changes in discharges to Hood Canal during the study period that could
explain these increases (Singleton, L., 30 November 1987, personal
communication; Tarr, M., 30 November 1987, personal communication). As with
the physical conditions discussed above, the absence of changes in dissolved
oxygen concentrations at the surface may indicate that the surface waters
were responding to different environmental factors than were the waters at
depth.
Nutrients—Nutrient concentrations in all three Hood Canal study areas
appear to have declined at 10- and 30-m depths. Nitrate data are only
5-252
-------
available back to the late 1970s. Most of the statistical tests for
detecting changes in nitrate concentrations were not significant (P>0.05).
However, substantial negative slopes of nitrate concentrations by year were
detected at 10- and 30-m depths at all three sites [e.g., the smallest slope
was -0.65 ug-at/L/yr (P=0.12) at 30-m depth]. At all three sites, phosphate
concentrations at 10- and 30-m depths declined significantly (P<0.05) since
the early 1950s. However, phosphate concentrations have not changed
significantly in the three sites since the late 1970s. These declines in
nutrient concentrations may be attributed to increased photosynthesis (see
below). The long-term declines in phosphate concentrations may also have
been influenced by decreased oceanic inputs.
Indicators of Phvtoplankton Growth—Statistically significant changes
in the indicators of phytoplankton growth were not found at any of the Hood
Canal sites. The variables used as these indicators either were not
suitable for detecting trends at depth, or they contained limited data.
Secchi disk depth and percent dissolved oxygen saturation at the surface do
not provide information about productivity at depth.
Increases in the percent dissolved oxygen saturation and decreases in
nutrient concentrations at 10- and 30-m depths suggest that photosynthetic
rates and algal abundances have increased at depth. The geographic gradient
in the percent dissolved oxygen saturation at 10-m depth suggests that the
greatest changes were detected in South Hood Canal and that the smallest
changes were detected in Dabob Bay. Hence, the physical factor(s) causing
the apparent changes in photosynthetic activity were most influential in
South Hood Canal.
Pollutants—The only statistically significant change in the concentra-
tions of sulfite waste liquor or fecal coliform bacteria was a decline in
fecal coliform bacteria in the surface waters of the Dabob Bay study area.
However, fecal coliform concentrations always were low in the study areas,
and never approached the Class AA water quality standard. Thus, the three
5-253
-------
Hood Canal study areas were not substantially impacted by either sulfite
waste liquor or fecal coliform bacteria.
5-254
-------
CHAPTER 6. SUMMARY AND RECOMMENDATIONS
The trend analyses for the 13 study areas and a brief assessment of the
sensitivity of the study areas to nutrient enrichment are summarized in this
chapter. Recommendations are also given regarding the implementation of
environmental monitoring programs.
SUMMARY OF WATER QUALITY TRENDS IN PUGET SOUND
Although problems caused by changes in station locations and data
sources limited data interpretation in some areas (e.g., Port Gardner, Budd
Inlet, Oakland Bay), numerous trends in the water quality of Puget Sound were
observed. Results of the study are summarized in Table 6.1. The informa-
tion in Table 6.1 was derived from the interpretations provided in Chapter 5.
Some statistically significant results that appeared to have been artifacts
of changes in station locations or data sources are omitted from Table 6.1.
Several limitations in the data sets used in this study may have
adversely affected the sensitivity of the analyses. Most of the sampling
stations were located offshore, removed from the influences of local onshore
pollutant sources. Data typically consisted of only monthly samples without
replication. Several sources of variation that could have strongly in-
fluenced the data, (e.g., time of day and stage of tide during which
samples were collected) were not controlled during the sampling. In
addition, because long-term data from below 30-m depth were only available
at the Point Jefferson study area, changes in dissolved oxygen concent-
rations at depth could not be assessed for the other study areas. Readers
are cautioned that trends observed in each study area only apply to the
immediate vicinity of the sampling stations (i.e., conditions nearby may
have been different).
6-1
-------
TABLE 6.1. SUMMARY OF WATER QUALITY TRENDS IN PUGET SOUND3
Study
Area
Bellingham
Bay
Port
Gardner
Point
Jefferson
Sinclair
Inlet
City Water-
way
Carr
Inlet
Nisqually
Reach
Budd
Inlet
Totten
Inlet
Oakland
Bay
Dabob
Bay
Mid- Hood
Canal
South Hood
Canal
Depth
(m)
0
10
30
0
10
0
10
30
100
150
200
0
10
0
10
0
10
30
0
10
0
10
0
10
0
10
0
10
30
0
10
30
0
10
30
Sal in. b
L R
Od
0
. 0
• •
0 +
- 0
0
0 -
0 .
0 0
. 0
0 0
. 0
0
0
- 0
- 0
0 .
0 .
0
0 0
. 0
. 0
0 0
- 0
0
0
0
- -
0 0
0
- 0
Water
Temp.
L R
. +
0 0
• +
• •
0 -
+ 0
+ 0
0 0
0 0
0 .
0 0
. 0
0 0
. 0
+ 0
«• 0
+ 0
* 0
+ 0
0 0
0 0
0 0
0 0
0 0
. 0
0 0
+ 0
+ 0
0 0
» 0
+ 0
0 0
+ 0
+ 0
Diss.
Oxygen
L R
0 0
0 0
. 0
0 0
0 0
0 -
0 0
0 0
0 0
0 0
• •
0 0
. 0
0 0
. 0
0 +
0 +
+ +
0 +
+ 0
• •
0 0
0 »
+ 0
. p
0 0
+ 0
+ 0
0 0
+ +
+ 0
0 0
•f +
f 0
D i ss .
Nitrate
L R
North
. 0
. 0
. 0
Central
. 0
•
. ,
m t
, ,
. ,
• •
. 0
. 0
. 0
• +
South
. 0
. 0
. 0
. 0
. 0
. 0
. 0
.0
0 0
. 0
Hood
. 0
. 0
. 0
. 0
f
•
. 0
f
•
Diss.
Phos.
L R
Sound
0 0
0 +
. 0
Sound
- 0
- +
. .
f f
. .
. .
• •
. 0
• •*•
. +
. +
Sound
0 0
0
- 0
0 0
0 0
. 0
• +
0
0
- +
• +
Canal
0 0
- 0
- 0
0 0
0
0
- 0
- 0
0
Diss. OXy. Seech i Fecal col.
Chi. a Satur. Depth SWLC Bacteria
LR LR LR LR LR
00 00
.0
00
00 .0-0 . +
. .
0 . - - 0 +
0 . . .
• • • • •• •• ••
.. . . . • • . ••
.. .. .. .. ..
. .
00 .00. .0
. .
.0 00 .0 .0
.0
0 + 0
. . •• •• • . ..
. .
0 + . 0 . .
• •
00
• •
00 000. .0
. .
0 . +0 0 » - . .0
• •
.0 00 . 0 . .
.0 «• 0
.0 +0
00 .0 . . .0
+ +
+ 0 . . . '.
00 00 . . 00
+ +
+ 0 '. '. '. '. '. '.
The trends depicted in this table were derived from interpretations in the text and not directly from
statistical tables. Some results that were statistically significant (P<0.05) were omitted from this table
because they appeared to be artifacts of changes in data sources. Also, the recent trends in phosphate
concentrations at 10-m depth in the Budd Inlet site and the recent trends in nitrate concentrations at 10-
and 30-m depths in the Mid-Hood Canal site and at 10-m depth at the South Hood Canal site were not quite
statistically significant (P<0.10). However, because those trends were judged to be credible, they were
included in the table.
6-2
-------
TABLE 6.1 (Continued)
L Long-term trend based on all the data available from a given study area.
R Recent trend based on all the data available from a given study area from 1973 to 1986.
c SWL = Sulfite waste liquor.
d 0 = No trend.
= Declining trend.
+ Increasing trend.
Trend cannot be determined because of ambiguity in the results or a lack of data.
6-3
-------
Physical Conditions
Except for a slight increase in salinity values detected since 1973 at
the surface in the Point Jefferson study area, salinity values either
declined or did not change significantly in the study areas. Decreased
salinity values recorded for the Bellingham Bay, Carr Inlet, Totten Inlet,
Dabob Bay, Mid-Hood Canal, and South Hood Canal study areas occurred
gradually, having begun with relatively high salinity values in the early
1950s. The gradual salinity declines in the Nisqually Reach and Point
Jefferson study areas date back to the early 1930s. Changes in salinity
values were not detected in the Sinclair Inlet and City Waterway study
areas. However, data collection at these two sites did not begin until the
late 1960s, resulting in a shorter data record for detection of a gradual
trend.
Factors underlying the declines in salinity values are not known. One
explanation involves decreasing inputs of high salinity water at depth from
the Strait of Juan de Fuca. Declines in salinity values were detected at
100- and 150-m depths at the Point Jefferson study area, the only study area
for which data collected at depths greater than 30-m were available. Point
Jefferson is close to Admiralty Inlet, through which the deep water from the
Strait of Juan de Fuca must pass to reach most of Puget Sound. Moreover,
most of the other declines in salinity values that were detected in the
study occurred at 10- and 30-m depths, not at the surface. The salinity of
deeper water may have a greater influence over the salinities at 10- and
30-m depths than over the salinity at the surface. (It must be emphasized
that the foregoing discussion merely presents a plausible hypothesis for the
salinity declines, and that information to test the hypothesis was not
available to this project.)
The available information on rainfall and runoff does not appear to
explain the observed declines in salinity values. Rainfall data recorded at
the Seattle-Tacoma International Airport showed that a wet period occurred
during the early 1950s. However, no statistically significant changes in
total runoff to Puget Sound were detected between 1930 and 1978 (see
Figure 5.2). The effect of changes in rainfall would have been to increase
6-4
-------
salinity values since the 1950s, which is contrary to the observed decrea-
ses. Of possible importance is that the salinity declines typically were not
detected at the surface, the portion of the water column most directly
influenced by rainfall and runoff.
Changes in water temperature values generally coincided with climatic
changes. Water temperatures in the study areas either increased or did not
change significantly, except for declines in surface water temperature
values in the Bellingham Bay and Point Jefferson study areas. The Carr
Inlet, Dabob Bay, Mid-Hood Canal, and South Hood Canal data sets began in
the early 1950s, which was a cool period in Seattle (see Figure 5.2).
Similarly, long-term increases in water temperature values were detected in
the Point Jefferson and Nisqually Reach study areas. The data sets analyzed
for these two areas began in the early 1930s, which also was a relatively
cool period (NOAA 1985). The Bellingham Bay study area was unusual in that
a long-term decline in water temperature values was detected. However, the
first year of data collection for this site was 1958, which happened to be
an unusually warm year (see Figure 5.2).
Water temperatures in most of the study areas where significant changes
were not detected (Sinclair Inlet, City Waterway, Budd Inlet, Totten Inlet,
and Oakland Bay) also may have been influenced by climate. Data collection
did not begin in these sites when climatic conditions were markedly dif-
ferent from recent conditions. Therefore, the absence of changes in water
temperature values in these areas does not preclude a climatic influence
over water temperature.
Dissolved Oxygen
Dissolved oxygen concentrations in the study areas generally have
increased or have not changed significantly during the study period. Very
low dissolved oxygen concentrations were rarely observed. Except for the
Point Jefferson study area, which is unlikely to have low dissolved oxygen
concentrations at depth because of high rates of circulation, dissolved
oxygen data were collected only from the top 10 or 30 m of the water column.
Low dissolved oxygen concentrations in near-bottom waters could have
6-5
-------
occurred in all the study areas except Point Jefferson without being
detected in this study.
Increased dissolved oxygen concentrations were observed in the southern
sound study areas. Dissolved oxygen data in these study areas were strongly
influenced by very high values observed in 1986, the last year from which
data were obtained. The most recent points in these data sets had the
highest values, inducing a positive slope to the regressions of dissolved
oxygen concentrations by year. Although limitations of the available data
preclude definitive interpretations, these high dissolved oxygen concentra-
tions of 1986 appear to have been caused by intense algal blooms.
In the Hood Canal study areas, increased dissolved oxygen concentra-
tions appear to have occurred more gradually than did the increased concen-
trations recorded in the southern sound sites. Unlike the southern sound
sites, unusually high dissolved oxygen concentrations were not observed in
1986 in the Hood Canal sites.
With the exception of surface waters in the Point Jefferson study area,
none of the study areas in the northern sound and central sound study areas
exhibited significant temporal changes in dissolved oxygen concentrations.
Unusually high dissolved oxygen concentrations were not observed in these
regions during 1986.
Discharges of oxygen-demanding wastes from pulp mills only influenced
dissolved oxygen concentrations significantly in the Oakland Bay study area.
The high dissolved oxygen concentrations recorded in 1986 contributed to the
increase detected in dissolved oxygen concentrations at this site. However,
a few very low values also were detected early in the data set. Because
these early low values coincided with high concentrations of sulfite waste
liquor, pulp mill discharges probably were the causal agent. The pulp mill
in the Oakland Bay area closed in 1957. Extremely low dissolved oxygen
concentrations have not been found since that time.
Dissolved oxygen concentrations in the Bellingham Bay and City Waterway
study areas were not markedly influenced by changes in the discharges of
6-6
-------
nearby pulp mills. These study areas are less exposed to the local pulp
mill discharge plumes than was the Oakland Bay study area. Also, dilution
of the effluent in these areas probably is more effective than in Oakland
Bay, which is shallow and poorly flushed. However, changes in dissolved
oxygen concentrations close to the discharge points of the pulp mills in
Bellingham Bay and City Waterway could have occurred without having been
detected at the study sites.
The influence of pulp mill discharges on dissolved oxygen concentra-
tions in the Port Gardner study area could not be determined because the
proximity of the sampling stations to the discharge points of the local pulp
mills varied greatly over time.
Nutrients
With the exception of the study areas in Port Gardner, Carr Inlet, and
Hood Canal, changes in nitrate concentrations do not appear to have resulted
from well-developed temporal trends. The recent increase in nitrate
concentrations detected statistically at 10-m depth in the City Waterway
study area appears to be attributable to erratic fluctuations. Data for
this site include a few low values near the beginning of the data set and a
few high values near the end of the data set.
Data on nitrate concentrations in the Port Gardner, Carr Inlet, and Hood
Canal study areas are only available since the mid-1970s. The decline in
the Carr Inlet study area may have been caused by increased nutrient uptake
by algae (see below). Substantial decreases in nitrate concentrations were
detected in the South Hood Canal study area. Decreases also were detected
in the Mid-Hood Canal and Dabob Bay study areas, although the declines were
not well developed in the Dabob Bay site. The factor affecting the nitrate
concentrations in Hood Canal [apparently algal blooms (see below)] was
probably most influential in southern Hood Canal. No explanation is
available for the decline in nitrate concentrations in the Port Gardner study
area.
6-7
-------
Temporal changes in phosphate concentrations were apparent in 11 of the
12 study areas from which phosphate data were available. Statistically
significant (PO.05) long-term decreases (since the 1950s) were detected in
seven of the nine study areas from which long-term data are available.
However, no long-term increases or recent decreases were detected. Recent
increases (since the mid-1970s) were detected at the surface and/or at 10-m
depth at six study areas. Five of the six recent increases were statisti-
cally significant at PO.05. The significance level at the sixth site (Budd
Inlet) was P=0.08. Both long-term decreases and recent increases were
detected in the Port Gardner and Oakland Bay study areas.
The cause(s) of the widespread decreases in phosphate concentrations
since the 1950s are unknown. Because declines occurred in both urban and
rural study areas, anthropogenic influences do not explain these results.
One explanation involves decreased inputs of phosphate in oceanic water from
the Strait of Juan de Fuca, but this hypothesis could not be tested in this
study. Although it was not possible to calibrate the analytical techniques
used in the 1950s with those used more recently, the older techniques
generally were accurate (Appendix A).
The recent increases in phosphate concentrations all occurred in urban
study areas. No evidence of recent changes in phosphate concentrations was
found in any rural study area (Figure 6.1). The absence of detectable
changes in phosphate concentrations in rural study areas suggests that
local anthropogenic factors may have influenced phosphate concentrations in
the urban study areas.
Although changes in numerous factors (e.g., sewage discharges, urban
runoff) may have influenced the phosphate data in the urban study areas,
three of the urban increases may be attributable at least in part to known
anthropogenic factors. Because sulfite waste liquor removes dissolved
orthophosphate from seawater solution (Westley and Tarr 1978), reductions in
the discharges of sulfite waste liquor by the local pulp mills during the
1970s may have contributed to the increased phosphate concentrations in
these two areas. In another case, phosphoric acid has been added to the
effluent discharged by the kraft pulp mill near the City Waterway study area
6-8
-------
en
^
UJ
<
>•
D
I—
(A
Z
<
CD
DC
3
(0
<
UJ
DC
<
>
Q
3
H
0)
_l
<
OC
3
DC
Bellingham Bay
Budd Inlet
City Waterway
Oakland Bay
Port Gardner
Sinclair Inlet
Carr Inlet |
Dabob Bay
W///////////////////////M
'///////////
///////////
'/////////////////////////////////////////)(
'///////////////////////////////,
'//////////////////////////////I
//////////////////////////////A
W//////////////////////h
South Hood Canal
Nisqually Reach
Totten Inlet
I 1
\
I
1
^
1 1 1 1 1 1 1 1 1 1
-0.02 0 0.02 0.04 0.06 0.08 0.10
RATE OF CHANGE OF PHOSPHATE
CONCENTRATIONS SINCE 1973
(u.g at/L/yr)
STATISTICALLY SIGNIFICANT (P<0.05)
STATISTICALLY SIGNIFICANT (P<0.10)
NOT STATISTICALLY SIGNIFICANT
Figure 6.1. Rates of change of phosphate concentrations during the algal bloom seasons
in urban and rural study areas since 1973.
6-9
-------
since 1977. This addition may have resulted in increased phosphate
concentrations in City Waterway (Henry, C., 17 November 1987, personal
communication). The recent increases in phosphate concentrations that were
detected in the remaining urban areas (Sinclair Inlet, Budd Inlet, Oakland
Bay) do not appear to be attributable to known anthropogenic factors.
Indicators of Phvtoplankton Growth
Few systematic changes were evident in the values of the variables used
in this study to indicate phytoplankton growth. No changes were detected in
chlorophyll a concentrations, although relatively few data are available.
This general lack of detected change can be attributed in part to inadequate
sampling frequency. The typical duration of an algal bloom in Puget Sound
is on the order of days. The monthly samplings used in most of the data
sources included in this study do not provide sufficient temporal resolution
to assess algal bloom dynamics effectively.
Some changes in algal abundance apparently were detected. In the Carr
Inlet study area, a decline in Secchi disk depth and an increase in percent
dissolved oxygen saturation at the surface suggest that algal densities may
have increased in this area. Tn the Point Jefferson study area, increased
Secchi disk depths and decreased values of surface percent dissolved oxygen
saturation suggest that phytoplankton concentrations have declined.
However, this decline appears to have been due to erratic fluctuations in
phytoplankton abundance, rather than to a systematic trend.
With the exception of the study areas located on Hood Canal and the
Carr Inlet and Point Jefferson study areas (discussed above), the changes in
surface percent dissolved oxygen saturation and Secchi disk depth that were
detected do not suggest that substantial changes in phytoplankton abundance
have occurred. Increased percentages of dissolved oxygen saturation at the
surface were detected at the Nisqually Reach and Oakland Bay study areas, and
increased Secchi disk depths were detected at the Oakland Bay study area.
The increase in surface percent dissolved oxygen saturation at the Nisqually
Reach study area occurred since 1977, and may be attributable statistically
to the high dissolved oxygen concentrations observed throughout the southern
6-10
-------
sound in 1986. The increase in percent dissolved oxygen saturation at the
surface at the Oakland Bay study area occurred since 1956, and may be
attributable to both the low dissolved oxygen concentrations caused by
sulfite waste liquor in 1956-57 and the high dissolved oxygen concentrati-
ons observed in the southern sound in 1986. The increased Secchi disk depths
at the Oakland Bay study area appear to have been the result of erratic
fluctuations, rather than a systematic trend.
In the Hood Canal study areas, statistically significant changes were
not detected in the values of the standard indicators of phytoplankton
growth (chlorophyll a concentration, percent dissolved oxygen saturation at
the surface, Secchi disk depth). However, these indicators do not provide
sufficient information to characterize phytoplankton abundance in Hood
Canal. Average chlorophyll a and nutrient concentrations were higher at
10-m depth (well below mean Secchi disk depth) than at the surface.
Although the chlorophyll a data set from the Dabob Bay study area included
data from depths of 10 and 30 m, data collection only began in 1979, while
the apparent changes in phytoplankton abundance that were detected in Hood
Canal (see below) occurred earlier. Therefore, the only standard indicator
variables with data from the period during which increases in phytoplankton
abundance apparently occurred were percent dissolved oxygen saturation at
the surface and Secchi disk depth. Unfortunately, these variables provide
information only about conditions near the surface. Maximum phytoplankton
abundances in Hood Canal are probably well below the surface.
Changes in the values of additional variables that may respond to
phytoplankton abundances in the Hood Canal study areas suggest that increases
in phytoplankton abundance may have occurred below the surface. At 10- and
30-m depths, values of percent dissolved oxygen saturation have increased,
while concentrations of phosphate (long-term) and nitrate (recently) have
declined. These chemical changes suggest that photosynthetic activity has
increased at depth. The data discussed previously on water clarity and
depth distributions of nutrients and chlorophyll a concentrations demonstrate
that substantial rates of photosynthesis probably occur well below the
surface. Unfortunately, data to confirm the possible increase in phyto-
6-11
-------
plankton abundance at depth in Hood Canal (e.g., uninterrupted data on
chlorophyll a concentrations since the 1950s) are not available.
Pollutants
Concentrations of sulfite waste liquor and fecal coliform bacteria
either declined or did not change greatly in the various study areas.
Concentrations of sulfite waste liquor declined in all four study areas
located near pulp mills (Bellingham Bay, Port Gardner, City Waterway, and
Oakland Bay). The sulfite waste liquor decline in Oakland Bay coincided with
the closure of the local pulp mill. The declines in the other sites
generally coincided with upgrades in the effluent treatment procedures used
by nearby pulp mills.
Several changes in the concentrations of fecal coliform bacteria may be
attributable to changes in point or nonpoint sources. The decline in the
concentration of fecal coliform bacteria in the Bellingham Bay study area
coincided with improvements in the sewage treatment facilities and with
closures of combined sewer overflows in the area. The increase in the
concentrations of fecal coliform bacteria in the Port Gardner study area
probably was due to an increase in the abundance of the bacterium, Kleb-
siella. which is discharged in large amounts from the secondary treatment
facilities of sulfite pulp mills.
The decline in the concentrations of fecal coliform bacteria in the
Nisqually Reach study area probably was due to one high value that was
detected in 1978, at the beginning of the data set for this variable. Low
concentrations of fecal coliform bacteria were detected in the Nisqually
Reach site after 1978, so a declining trend was found. Because the high
1978 value came from a sample collected near the end of a heavy rainstorm,
the bacteria probably are attributable to runoff into the Nisqually River
drainage basin.
No explanations are apparent for the remaining changes detected in the
concentrations of fecal coliform bacteria. Declines in the relatively rural
Carr Inlet and Dabob Bay study areas may have resulted from the slightly
6-12
-------
higher concentrations that were observed near the beginnings of the data
sets in those two areas. These early, slightly higher concentrations were
followed primarily by values at the detection limit. The elevated concentra-
tions were not associated with storm events or changes in point source
discharges. It is uncertain whether these statistical declines in the Carr
Inlet and Dabob Bay study areas were real phenomena.
SENSITIVITY TO NUTRIENT ENRICHMENT
Sensitivity of an estuary to nutrient enrichment depends on nutrient
inputs and physical factors. Urban study areas probably have the highest
potential for receiving large inputs of nutrients because of the presence of
large human populations. The urban study areas in this project are Bel ling-
ham Bay, Port Gardner, Sinclair Inlet, City Waterway, Budd Inlet, and
Oakland Bay. (Elliott Bay, which was not included in this study, also is
adjacent to a large population.) Strong density stratification and low
flushing rates tend to promote algal blooms and limit export rates of excess
nutrients. These factors affect several of the study areas, including
Sinclair Inlet, Budd Inlet, Totten Inlet, Oakland Bay, Dabob Bay, Mid-Hood
Canal, and South Hood Canal.
The following study areas appear to be most sensitive to nutrient
enrichment, due to both proximity to urban populations and to physical
factors:
• Sinclair Inlet
• Budd Inlet
• Oakland Bay
• South Hood Canal.
The Sinclair Inlet, Budd Inlet, and Oakland Bay study areas are adjacent to
cities. Increases in phosphate concentrations since 1973 were detected in
all three of these areas. The available evidence does not indicate that
6-13
-------
algal blooms have increased in intensity in response to these increases in
phosphate concentrations. However, as discussed previously, sampling
frequency was insufficient to assess algal bloom dynamics effectively. South
Hood Canal is also highly vulnerable to nutrient enrichment because of
physical factors. It may have significant .inputs of nutrients during the
summer, due to the presence of numerous summer homes.
RECOMMENDATIONS FOR ENVIRONMENTAL MONITORING IN PUGET SOUND
The water quality characterization study involved analysis of data
originally collected for a variety of purposes by several independent groups
of researchers. A retrospective study of existing data provides a unique
opportunity to assess the historical and existing studies from the perspec-
tive of a trends analysis. The following institutional and technical
recommendations are based on the results of this water quality characteriza-
tion study and the comments of the characterization work group and other
peer reviewers.
Institutional Recommendations
1. One organization should oversee all water quality monitoring in
Puget Sound to maximize the compatibility of field techniques, laboratory
techniques, and database formats, and to coordinate geographic coverage.
Use of the protocols recommended by PSEP (U.S. EPA 1986a) would standardize
the field and laboratory techniques of monitoring programs used in Puget
Sound.
2. Changes in field and laboratory techniques should be fully
documented. New techniques should be calibrated with old techniques. These
steps will facilitate future trend analyses as technology evolves.
6-14
-------
Technical'Recommendations
Monitoring Program Design--
1. The goals of the monitoring program should be stated quantitatively
before the study design is developed (e.g., how much change in dissolved
oxygen concentrations should be detectable over a given time period?).
2. The allocation of sampling effort should be assessed statistically,
using existing data, as an initial step in developing the sampling design.
The following points should be considered during development of the sampling
design.
• Assess the influences of known sources of variation (e.g.,
time of day, stage of tide) on the water quality variables of
interest in particular types of locations (e.g., open main
channel, enclosed embayment).
• Design monitoring programs to reduce the impact of sources of
variation that are not of interest. For example, dissolved
oxygen concentrations are strongly influenced by predictable
diel, tidal, and fortnightly variation. Fortnightly sampling
of the water during a particular window of time would
minimize the impact of these factors (e.g., collecting
samples near the noon high tide) because sampling would
always occur at the same time of day, stage of tide, and
phase of the semi-lunar tidal cycle.
t Use statistical power analysis to compare the ability of
alternative study designs to detect a given amount of
environmental change.
3. The influence of physical factors on water quality should be
addressed to improve understanding of ecosystem function and to permit
comparisons of the influences of natural and anthropogenic factors on water
quality. Data acquisition should reflect the temporal scale of variation
6-15
-------
for each variable of interest. Potentially important physical variables are
listed below:
t Oceanographic data (e.g., sampling through the entire water
column to estimate extrinsic oceanic inputs into a locality)
• Climatic data (e.g., air temperature, wind velocity, input
rates for runoff). These data may be available from NOAA and
USGS.
4. The monitoring program should include embayments with limited
flushing and mixing (e.g., Budd Inlet, Sinclair Inlet) as high priority areas
because water quality is most sensitive to anthropogenic degradation in such
areas. Water quality changes in areas with high flushing rates and rapid
currents (e.g., West Point, Point Jefferson, Nisqually Reach) are more
difficult to detect because contaminants do not accumulate in such areas.
5. Sampling stations should be located close enough to large local
contaminant sources to be able to detect a likely change (e.g., effect of
improvements in sewage treatment on ambient nutrient concentrations).
6. Monitoring for low dissolved oxygen concentrations at depth should
focus on sites, depths, and periods where and when low dissolved oxygen is
likely to occur (i.e., in near-bottom waters of poorly flushed areas in
late summer).
7. Intensities of algal blooms should be monitored frequently during
periods when blooms are likely to occur in each particular locality.
Because algal blooms wax and wane over a period of only a few days, the
time between consecutive samplings should be less than a few days (e.g.,
daily). Blooms are prominent during spring and summer, but the seasonal
occurrence of algal blooms differs among sites.
8. Changes in phytoplankton communities should be monitored by
measurement of chlorophyll a concentrations and species identification.
Changes in species composition of the phytoplankton community may have
6-16
-------
ecological consequences (e.g., changes in the quality of the phytoplankton
as a food source for filter-feeding organisms) even if the concentration of
chlorophyll a does not change.
9. Sampling for contaminants discharged episodically (e.g., discharges
from combined sewer overflows and pulp mills) should coincide with discharge
events. Infrequent sampling scheduled at random with respect to discharge
events will detect only very large changes.
10. In poorly flushed embayments where nutrients may limit algal growth
during algal blooms, the detection of changes in nutrient concentrations
requires frequent sampling and analytical detection limits that are lower
than those generally used in the existing monitoring programs for Puget
Sound. Lower analytical detection limits might be achieved simply by
collecting larger sample volumes.
11. A microbiological test is needed to distinguish between bacterial
contamination from sewage or agricultural runoff, which represents a risk of
possible exposure to human pathogens, and bacterial contamination from
Klebsiella. Although this organism is detected in standard tests for fecal
coliform bacteria, its principal source is the effluent from secondary
treatment ponds of sulfite pulp mills. Therefore, violations of water
quality standards (and the subsequent closure of shellfish beds) attributed
to contamination by sewage may actually be caused by exposure to secondary
pulp mill effluent. Although not well studied in Puget Sound, the health
risk caused by environmental contamination from Klebsiella appears to be
low. Possible applicable microbiological tests include screening for
Escherichia coli or enterococci (Singleton, L., 24 September 1987, personal
communication).
12. Future monitoring programs should include some sampling stations
where a long-term historical record of water quality already exists. This
strategy would allow that changes in water quality over time would be
documented with a record that extends as far back in time as possible. Many
such stations were used in this study.
6-17
-------
Specific Technical Issues--
1. Water quality should be monitored closely in Budd Inlet to determine
whether nitrogen removal by the LOTT sewage treatment plant is successful in
reducing the intensity of algal blooms.
2. The Secchi disk should to be replaced with, or be supplemented by,
more quantitative measures of specific water column variables (e.g.,
suspended particulate matter, concentration of chlorophyll a, depth of the
photic zone). Secchi disk depth data cannot distinguish between turbidity
caused by suspended particulate material, and turbidity caused by phytoplank-
ton. The Secchi disk can only provide measurements of turbidity in the
upper-most portion of the water column, while maximum phytoplankton densities
may occur below the Secchi disk depth.
3. Monitoring of variables affected by a pycnocline (e.g., salinity,
temperature, concentrations of dissolved oxygen, chlorophyll a, and
nutrients) should include determination of the depth of the pycnocline.
Sampling would then be done above and below the pycnocline, as well as at
the depths normally sampled. This procedure would reduce the variability in
the water quality data caused by sampling only at a given depth when the
depth of the pycnocline fluctuates.
4. Some variables must be sampled near the surface (e.g., photosyn-
thesis rate), but because many variables change rapidly very close to the
surface (e.g., temperature, salinity, dissolved oxygen, nutrient concentra-
tions), sampling at 1-m depth, rather than right at the surface, may provide
more representative data. Also, sampling at 1-m depth may reduce scatter in
the data and avoid possible artifacts caused by surface contaminants.
6-18
-------
CHAPTER 7. REFERENCES
Anderson, J., A. Copping, T. Jagielo, J. Postel, W. Peterson, B. Dumbauld,
G. Heron, R. Hood, and M. Strom. 1984. Volume III. Section 4. Water column
ecology, pp. 1-248. In: Renton Sewage Treatment Plant Project: Seahurst
Baseline Study. Q.J. Stober and K.K. Chew (eds). Final Report. Prepared
for the Municipality of Metropolitan Seattle, Seattle, WA.
Baker, D. 29 October 1987. Personal Communication (phone by Dr. Stephen K.
Brown). City Engineering Office, Bremerton, WA.
Baker, D. 13 November 1987. Personal Communication (phone by Dr. Stephen
K. Brown). City Engineering Office, Bremerton, WA.
Barnes, H. 1959. Apparatus and methods of oceanography. Part 1: Chemical.
Interscience Publishers, Inc., New York, NY. 341 pp.
Bechtel, T. 22 March 1988. Personal Communication (phone by Dr. Stephen
K. Brown). Scott Pulp Mill, Everett, WA.
Bellingham, City of. 1984. Application for variance from secondary
treatment requirements under Section 301(h), Clean Water Act. Submitted to
U.S. Environmental Protection Agency. Prepared by CH2MHILL, Bellevue, WA.
Burns, R.E. 1985. The shape and form of Puget Sound. Washington Sea Grant
Program and University of Washington Press, Seattle, WA. 100 pp.
Chasan, D.J. 1981. The water link: a history of Puget Sound as a resource.
Puget Sound Books and University of Washington Press, Seattle, WA. 179 pp.
Collias, E.E. 1970. Index to physical and chemical oceanographic data of
Puget Sound and its approaches, 1932-1966. Washington Sea Grant Publ. No.
70-4. State of Washington, Department of Natural Resources, Olymp.ia, WA.
823 pp.
Collias, E.E., N. McGary, and C.A. Barnes. 1974. Atlas of physical and
chemical properties of Puget Sound and its approaches. Washington Sea Grant
Publ. No. 74-1. University of Washington Press, Seattle, WA. 235 pp.
Collias, E.E., and J.H. Lincoln. 1977. A study of the nutrients in the main
basin of Puget Sound. Final Report M77-2. Prepared for the Municipality of
Metropolitan Seattle, Seattle, WA. 151 pp.
Cunningham, R. 22 May 1987. Personal Communication (phone by Dr. Stephen
K. Brown). Washington Department of Ecology, Olympia, WA.
Dalseg, R. 17 September 1987. Personal Communication (phone by Dr. Stephen
K. Brown). Municipality of Metropolitan Seattle, Seattle, WA.
7-1
-------
Dames & Moore. 1981. Preliminary draft physical oceanography technical
report for Commencement Bay studies. 108 pp. + appendices.
Duxbury, A.C. 1975. Orthophosphate and dissolved oxygen in Puget Sound.
Limnol. Oceanogr. 20:270-274.
Duxbury, A.C. 22 May 1987. Personal Communication (phone by Dr. Stephen K.
Brown). Washington Sea Grant Program, Seattle, WA.
Duxbury, A.C. 15 October 1987. Personal Communication (phone by Dr.
Stephen K. Brown). Washington Sea Grant Program, Seattle, WA.
Duxbury, A.C. 3 December 1987. Personal Communication (phone by Dr.
Stephen K. Brown). Washington Sea Grant Program, Seattle, WA.
Ebbesmeyer, C.C., and C.A. Barnes. 1980. Control of a fjord basin's
dynamics by tidal mixing in embracing sill zones. Est. Coast. Mar. Sci.
11:311-330.
Evans-Hamilton, Inc. and D.R. Systems, Inc. 1987. Puget Sound environmental
atlas. Prepared for U.S. Environmental Protection Agency, U.S. Army Corps
of Engineers, and Puget Sound Water Quality Authority, Seattle, WA. Evans-
Hamilton, Inc., Seattle, WA and D.R. Systems, Inc., Nanaimo, BC.
Federal Water Pollution Control Administration and Washington State Pollution
Control Commission. 1967. Pollutional effects of pulp and paper mill
wastes in Puget Sound. A report on studies conducted by the Washington
State Enforcement Project. 474 pp.
Felicetta, V.F., and J.L. McCarthy. 1963. Spent sulfite liquor: X. The
Pearl Benson, or nitroso, method for the estimation of spent sulfite liquor
concentration in waters. Tappi 46(6):337-346.
Greenberg, A.E., R.R. Trussell, and L.S# Clesceri (eds). 1985. Standard
methods for the examination of water and wastewater. 16th edition.
American Public Health Association, Washington, DC. 1,268 pp.
Hayward, A. 24 July 1987. Personal Communication (phone by Dr. Stephen K.
Brown). Municipality of Metropolitan Seattle, Seattle, WA.
Heinle, D.R., C.F. D'Elia, J.L. Taft, J.S. Wilson, M. Cole-Jones, A.B.
Caplins, and L.E. Cronin. 1980. Historical review of water quality and
climatic data from Chesapeake Bay with emphasis on effects of enrichment.
Final Report. U.S. Environmental Protection Agency, Chesapeake Bay Program
Grant Number R806189010. Publ. No. 84. Chesapeake Bay Research Consortium,
Inc., Annapolis, MD. 128 pp. + appendices.
Henry, C. 17 November 1987. Personal Communication (phone by Dr. Stephen
K. Brown). Simpson-Tacoma Kraft Mill, Tacoma, WA
Johnson, B. 21 July 1987. Personal Communication (phone by Dr. Stephen K.
Brown). Washington Department of Ecology, Olympia, WA.
7-2
-------
Jones & Stokes Associates, Inc. 1984. Water quality management in Puget
Sound: managing for long-term cumulative effects. Final Report. Prepared
for U.S. Environmental Protection Agency Region X, Seattle, WA. Jones &
Stokes Associates, Inc., Bellevue, WA. 148 pp. + appendices.
Knittel, M.D. 1975. Occurrence of Klebsiella pneumoniae in surface waters.
Appl. Microbiol. 29:595-597. (not seen).
Krafft, W. 23 July 1987. Personal Communication (phone by Dr. Stephen K.
Brown). Washington Department of Ecology, Manchester, WA.
Krafft, W. 29 July 1987. Personal Communication (phone by Dr. Stephen K.
Brown). Washington Department of Ecology, Manchester, WA.
Kruger, D.M. 1979. Effects of point source discharges and other inputs on
water quality- in Budd Inlet, Washington. Report No. DOE 79-11. Washington
Department of Ecology, Olympia, WA.
Lehman, K. 20 May 1987. Personal Communication (phone by Dr. Stephen K.
Brown). Municipality of Metropolitan Seattle, Seattle, WA.
Lincoln, J.H., and E.E. Collias. 1975. An oceanographic study of the Port
Orchard system. Final Report M75-102. Submitted to URS Company, Seattle,
WA. University of Washington, Department of Oceanography, Seattle, WA.
Loehr, L. 21 July 1987. Personal Communication (phone by Dr. Stephen K.
Brown). Northwest Pulp and Paper Association, Seattle, WA.
Matsuda, R. 24 July 1987. Personal Communication (phone by Dr. Stephen K.
Brown). Municipality of Metropolitan Seattle, Seattle, WA.
Morris, A.W., and J.P. Riley. 1963. The determination of nitrate in sea
water. Anal. Chim. Acta 29:272-279.
Municipality of Metropolitan Seattle. 1969. Metro — the first ten years.
Municipality of Metropolitan Seattle, Seattle, WA. 39 pp.
Municipality of Metropolitan Seattle. 1986. Water quality data and
technical information guide. 8th edition. Municipality of Metropolitan
Seattle, Seattle, WA.
Murphy, J., and J.P. Riley. 1962. A modified single solution method for the
determination of phosphate in natural waters. Anal. Chim. Acta 12:162-176.
National Oceanic and Atmospheric Administration. 1945-85. Local climatolo-
gical data reports (for the years 1945 through 1985). NOAA, National
Climatic Data Center, Asheville, NC.
7-3
-------
National Oceanic and Atmospheric Administration. 1984a. Synthesis of
current measurements in Puget Sound, Washington. Volume 2: Indices of mass
and energy inputs into Puget Sound: runoff, air temperature, wind, and sea
level. NOAA Tech. Memo. NOS QMS 4. NOAA, Rockville, MD. 45 pp. +
appendices.
National Oceanic and Atmospheric Administration. 1984b. Synthesis of
current measurements in Puget Sound, Washington. Volume 1: Index to
current measurements made in Puget Sound from 1908-1980, with daily and
record averages for selected measurements. NOAA Tech. Memo. NOS QMS 3.
NOAA, Rockville, MD. 38 pp. + appendices.
National Oceanic and Atmospheric Administration. 1985. Temporal trends in
selected environmental parameters monitored in Puget Sound. NOAA Tech.
Memo. NOS OMA 19. NOAA, Rockville, MD. 166 pp.
National Oceanic and Atmospheric Administration. 1986a. Oxygen depletion
in Long Island Sound. Draft Report. Prepared for the U.S. Environmental
Protection Agency Region I, under U.S. EPA/NOAA Interagency Agreement No.
DW-13931613-01 and Battelle Contract No. 68-03-3319. NOAA, Rockville, MD.
85 pp. + appendices.
National Oceanic and Atmospheric Administration. 1986b. Dynamics of
Commencement Bay and approaches. NOAA Tech. Mem. NOS OMA 24. NOAA,
Rockville, MD. 79 pp.
National Oceanic and Atmospheric Administration. 1987. Contaminant
transport from Elliott and Commencement Bays. Draft Final Report. Submitted
to the U.S. Environmental Protection Agency, Seattle, WA. Prepared by the
Pacific Marine Environmental Laboratory, Seattle, WA. 258 pp.
Neilson, B.J., and L.E. Cronin. 1981. Estuaries and nutrients. Humana
Press, Inc., Clifton, NJ. 643 pp.
Olcay, N. 1959. Oceanographic conditions near the head of southern Puget
Sound, August 1957 through September 1958. M.S. Thesis. University of
Washington, Seattle, WA. 59 pp. + appendices.
Poppe, J. 9 November 1987. Personal Communication (phone by Dr. Stephen K.
Brown). Bremerton Waste Water Utility, Bremerton, WA.
Preisendorfer, R.W. 1986. Secchi disk science: visual optics of natural
waters. Limnol. Oceanogr. 31(5):909-926.
Puget Sound Water Quality Authority. 1986a. Draft 1987 Puget Sound water
quality management plan and environmental impact statement. Draft Report.
Puget Sound Water Quality Authority, Seattle, WA.
Puget Sound Water Quality Authority. 1986b. State of the sound 1986. Final
Report. Prepared by Puget Sound Water Quality Authority with the assist-
ance of Entrance Engineers, Inc., Seattle, WA.
7-4
-------
Richards, F.A., and T.G. Thompson. 1952. The estimation and characteriza-
tion of plankton populations by pigment analysis. II. A spectrographic
method for the estimation of plankton pigments. J. Mar. Res. 2:156-172.
Riley, J.P. 1975. Analytical chemistry of sea water, pp. 193-514. In:
Chemical Oceanography. J.P. Riley and G. Skirrow (eds). Academic Press,
Inc., New York, NY.
Robinson, M.G., and L.N. Brown. 1983. A recurrent red tide in a British
Columbia coastal lagoon. Can. J. Fish. Aquat. Sci. 40:2135-2143.
SAS. 1985. SAS user's guide: statistics, version 5 edition. SAS Institute,
Inc., Cary, NC. 956 pp.
Shapiro and Associates. 1983. Inventory of wetland resources and evaluation
of wetland management in western Washington. Final Report prepared for
Washington Department of Ecology, Olympia, WA. Shapiro and Associates,
Inc., Seattle, WA. 102 pp.
Singleton, L. 7 August 1987. Personal Communication (phone by Dr. Stephen
K. Brown). Washington Department of Ecology, Olympia, WA.
Singleton, L. 22 September 1987. Personal Communication (phone by Dr.
Stephen K. Brown). Washington Department of Ecology, Olympia, WA.
Singleton, L. 24 September 1987. Personal Communication (phone by Dr.
Stephen K. Brown). Washington Department of Ecology, Olympia, WA.
Singleton, L. 20 October 1987. Personal Communication (phone by Dr.
Stephen K. Brown). Washington Department of Ecology, Olympia, WA.
Singleton, L. 30 November 1987. Personal Communication (phone by Dr.
Stephen K. Brown). Washington Department of Ecology, Olympia, WA.
Snedecor, G.W., and W.G. Cochran. 1980. Statistical methods. 7th edition.
Iowa University Press, Ames, IA. p. 507.
Steel, R.G.D., and J.H. Torrie. 1960. Principles and procedures of statis-
tics with special reference to the biological sciences. McGraw-Hill Book
Co., Inc., New York, NY. 481 pp.
Storm, P.C. 1981. A literature review of the bacterium, Klebsiella, spp.
Final Report. Prepared for the U.S. Army Corps of Engineers, Environmental
Resources Section, Seattle, WA. 21 pp.
Strickland, J.D.H., and T.R. Parsons. 1972. A practical manual of seawater
analysis. Bull. 167. 2nd edition. Fish. Res. Bd. Canada, Ottawa, Canada.
310 pp.
Strickland, R.M. 1983. The fertile fjord: plankton in Puget Sound. Puget
Sound Books and University of Washington Press, Seattle, WA. 145 pp.
7-5
-------
Struck, P. 9 November 1987. Personal Communication (phone by Dr. Stephen
K. Brown). Kitsap County Health Department, Bremerton, WA.
Tacoma, City of. 1983a. Section 301(h) waiver: application for modification
of secondary treatment requirements for the Central Waste Treatment Plant.
Submitted to U.S. Environmental Protection Agency, Seattle, WA. Parametrix,
Inc., Bellevue, WA. 224 pp. + appendices.
Tacoma, City of. 1983b. Section 301(h) waiver: application for modification
of secondary treatment requirements for the North End Waste Treatment Plant.
Submitted to U.S. Environmental Protection Agency, Seattle, WA. Prepared
by Parametrix, Inc., Bellevue, WA. 232 pp. + appendices.
Tarr, M.A. 22 May 1987. Personal Communication (phone by Dr. Stephen K.
Brown). Washington Department of Fisheries, Brinnon, WA.
Tarr, M.A. 30 November 1987. Personal Communication (phone by Dr. Stephen
K. Brown). Washington Department of Fisheries, Brinnon, WA.
Tetra Tech, Inc. 1985. Commencement Bay nearshore/tideflats remedial
investigation. Final Report. Prepared for Washington Department of
Ecology, Olympia, WA, and U.S. Environmental Protection Agency, Seattle,
WA. Tetra Tech, Inc., Bellevue, WA.
Tetra Tech, Inc. 1986. Puget Sound monitoring program: a proposed plan.
Draft Report. Prepared for the U.S. Environmental Protection Agency,
Seattle, WA. Tetra Tech, Inc., Bellevue, WA. 104 pp. + appendices.
Thomas, K. 27 October 1987. Personal Communication (phone by Dr. Stephen
K. Brown). Department of Public Works, Bellingham, WA.
Uchida, B. 28 July 1987. Personal Communication (phone by Dr. Stephen K.
Brown). Municipality of Metropolitan Seattle, Seattle, WA.
URS, Inc. 1986a. Comprehensive circulation and water quality study of Budd
Inlet. Final Report. Prepared for the Washington Department of Ecology,
Olympia, WA. URS, Inc., Seattle, WA.
URS, Inc. 1986b. Circulation and flushing in southern Puget Sound. Final
Report. Prepared for the Washington Department of Ecology, Olympia, WA.
URS, Inc., Seattle, WA. 58 pp. + appendices.
U.S. Environmental Protection Agency. 1984. Water quality dependent water
uses in Puget Sound. EPA 910/9-83-118a. Final Report. U.S. EPA Contract
No. 68-01-6348, Work Assignment No. 29. Prepared by ORB Associates,
Bellevue, WA.
U.S. Environmental Protection Agency. 1986a. Recommended protocols for
measuring selected environmental variables in Puget Sound. U.S. EPA Region
X, Seattle, WA.
7-6
-------
U.S. Environmental Agency. 1986b. Reconnaissance survey of eight bays in
Puget Sound. Final Report. U.S. EPA Contract No. DE-AC06-76RLO 1830 and
Interagency Agreement No. DW89930272-01-1. Prepared by Battelle Pacific
Northwest Laboratory, Sequim, WA. 231 pp. + appendices.
U.S. Geological Survey. 1985. Streamflow statistics and drainage-basin
characteristics for the Puget Sound region, Washington. Volume I. Western
and southern Puget Sound. USGS Open-File Report 84-144-A. USGS, Tacoma,
WA. 330 pp.
Waddell, D. 20 May 1987. Personal Communication (phone by Dr. Stephen K.
Brown). Municipality of Metropolitan Seattle, Seattle, WA.
Washington Department of Ecology. 1976. Ecological baseline and monitoring
study for Port Gardner and adjacent waters. A summary report for the years
1972 through 1975. Washington Department of Ecology, Olympia, Wa.
Weiss, R.F. 1970. The solubility of nitrogen, oxygen, and argon in water
and seawater. Deep Sea Res. 17:721-735.
Westley, R.E. 1957a. Physical and chemical data south Puget Sound hydro-
graphic trips 1956. Wash. Dep. Fish. Hydrographic Data I(l):l-48.
Westley, R.E. 1957b. Physical and chemical data north Puget Sound hydro-
graphic trips 1956 and 1957. Wash. Dep. Fish. Hydrographic Data II(6):1-21
+ appendix.
Westley, R.E. 1958. Physical and chemical data south Puget Sound hydro-
graphic trips 1957. Wash. Dep. Fish. Hydrographic Data II(5):l-56.
Westley, R.E., and M.A. Tarr. 1959. Physical and chemical data north Puget
Sound hydrographic trips 1958. Wash. Dep. Fish. Hydrographic Data
III(l):l-46 + appendices.
Westley, R.E., and M.A. Tarr. 1960. Physical and chemical data north Puget
Sound hydrographic trips 1959. Wash. Dep. Fish. Hydrographic Data
III(2):l-22 + appendices.
Westley, R.E., and M.A. Tarr. 1978. Some effects of sulfite waste liquor
(SWL) on seawater. Wash. Dep. Fish. Progress Rep. No. 67. 31 pp.
Winter, D.F., K. Banse, and G.C. Anderson. 1975. The dynamics of phyto-
plankton blooms in Puget Sound, a fjord in the northwestern United States.
Mar. Biol. 29:139-176.
Zar, J.H. 1974. Biostatistical analysis. Prentice-Hall, Inc., Englewood
Cliffs, NJ. 620 pp.
7-7
-------
APPENDIX A
HISTORY OF ANALYTICAL TECHNIQUES USED IN
WATER QUALITY STUDIES IN PUGET SOUND
-------
HISTORY OF ANALYTICAL TECHNIQUES USED IN
WATER QUALITY STUDIES IN PUGET SOUND
VALIDITY OF HISTORICAL TECHNIQUES
The analytical techniques historically used in the Puget Sound water
quality studies that were selected as data sources for this characterization
project are summarized below. The validity of the techniques is also
assessed. A time line of the major changes in techniques for these water
quality studies is also presented.
Salinity
Early salinity determinations in Puget Sound were made using the Knudsen
method. This method involves precipitation of halides with silver nitrate.
Potassium chromate is used as a titration endpoint indicator. The quality
of data produced with this method is excellent; reported accuracy can be as
high as 0.01 ppt.
More recently, salinity determinations have been made using various
methods involving measurements of conductivity or refraction. Conductivity-
based methods typically are highly reliable [e.g., Riley (1975) cites a
precision of 0.003 percent]. Refraction-based measurements, made with a
salinometer, are probably less reliable than conductivity-based measurements.
However, agencies using salinometers typically calibrate their instruments
with titration- or conductivity-based measurements of salinity standards.
Thus, data produced from salinometers were deemed of sufficient quality for
this project.
Water Temperature
Historically, reversing thermometers or laboratory thermometers in
water bottles have been used for water quality investigations of Puget
Sound. The former method is probably superior because the data are obtained
in situ. However, either method is adequate for this investigation. More
recent work has generally involved various types of in situ electronic
thermometers, which probably yield higher quality data.
Dissolved Oxygen
Most of the determinations of dissolved oxygen in Puget Sound monitoring
programs have either been based on the Winkler method or have been calibrated
to it. This method uses a series of chemical reactions that ultimately
liberate an amount of iodine from reagents added to a water sample that is
equal to the amount of dissolved oxygen in that water sample. The amount of
iodine is then measured by titration. This method is quite reliable for
unpolluted water. Strickland and Parsons (1972) estimate the precision of
A-l
-------
ship-board Winkler determinations as +/- 0.1 mg/L for a single determination
over a wide range of concentrations.
The presence of certain pollutants in Puget Sound may cause errors in
Winkler determinations of dissolved oxygen. Barnes (1959) states that
chlorinated effluents and sulfite-containing wastes from pulp mills can
cause analytical errors in Winkler titrations by interfering with iodine
production. Moreover, sulfites are oxidized by the oxygen dissolved in the
receiving water, which lowers ambient dissolved oxygen concentrations.
Barnes (1959) suggests a modification of the Winkler method for water
containing such pollutants. This modification involves pre-treatment of the
samples with hypochlorite to oxidize the pollutants. Apparently, such a
modification has not been included in any of the routine monitoring programs
conducted in Puget Sound (Cunningham, R., 22 May 1987, personal communi-
cation; Duxbury, A., 22 May 1987, personal communication; Tarr, M., 22 May
1987, personal communication). According to Cunningham (22 May 1987,
personal communication), analytical errors caused by sulfite waste liquor in
some areas of Puget Sound can reduce the apparent dissolved oxygen concen-
tration by about 0.7 mg/L. This problem reduces the reliability of dissolved
oxygen determinations in areas heavily contaminated by wastes from sulfite
pulp mills. Study areas investigated in the characterization project
potentially affected by sulfite wastes include Bellingham Bay, Port Gardner,
and Oakland Bay. The greatest impact of this problem is on data from the
1950s, when discharges of sulfite wastes were the highest.
More recently, electronic oxygen probes have been used in water quality
studies of Puget Sound. Such methods are less accurate than Winkler
titrations, but are more efficient. In most cases, the oxygen probes are
calibrated against Winkler titrations. Although electronic methods are less
susceptible to interference by pollutants (Riley 1975), proper equilibration
of the oxygen probe is needed when recording depth profiles. Generally,
competent use of oxygen probes can provide data of sufficient quality for
use in this project.
Nitrate
Colorimetry was one of the earliest methods for measuring nitrate in
seawater. It involved the use of reduced strychnine. Color instability was
a substantial problem. Also, because the colors were measured visually in
Messier Tubes, differences in the visual capacities of the investigators may
have affected the data (Riley 1975). Since the mid-1960s, methods of
measuring nitrate in seawater have depended on the reduction of nitrate to
nitrite. In the presence of sulphanilamide, the nitrite is converted to a
highly colored dye (Morris and Riley 1963). A correction for the amount of
nitrite originally in the water sample is made by subtracting the original
amount from the total after the conversion of the nitrate to nitrite. By
the mid-1970s, auto-analyzers were reliably used. Because no nitrate data
from before the mid-1960s were available for the areas studied in the
characterization project, all the available nitrate data were deemed
acceptable.
A-2
-------
Phosphate
Concentrations of dissolved orthophosphate are measured in seawater
using a filtration technique that removes particulates and produces of a
phosphomolybdenum blue complex in the filtered water. Prior to 1953,
dissolved phosphate was measured visually with Messier Tubes in studies of
Puget Sound. The early assays also suffered from problems with color
instability and salt errors. Since 1953, various spectrophotometers
and auto-analyzers have been used. In the mid-1960s, modifications developed
by Murphy and Riley (1962) to improve color stability were widely adopted.
Phosphate data from the mid-1960s and beyond are the most reliable, and
phosphate data from before 1953 are the least reliable. However, because the
measurement of phosphate in seawater has a long history in Puget Sound, the
limited amount of early phosphate data were retained, at least for the sake
of comparison.
Chlorophyll a
Reliable measurements of chlorophyll in seawater date from the work of
Richards and Thompson (1952). The Richards and Thompson (1952) method
involves the extraction of phytoplankton pigments and measurement of light
extinction by the extracted pigments at three wavelengths. Because studies
through the mid-1960s did not include a correction for phaeophytins (chloro-
phyll degradation products), chlorophyll data from before the mid-1960s
probably overestimate chlorophyll concentrations by about 25 percent (Heinle
et al. 1980). It is not possible to calculate a correction factor retrospec-
tively without actual data on phaeophytin concentrations. Therefore, none
of the small amount of chlorophyll data that exist for Puget Sound from
before this period were used in the water quality characterization project.
Since the mid-1970s, fluorometric methods sometimes have been used for
chlorophyll determinations in Puget Sound. Such data are somewhat less
reliable than absorbance-based data, but are of adequate quality for this
project.
Secchi Disk Depth
Secchi disk depth is a measurement of water transparency. It provides
an estimate of the amount of particulate matter in the water, as well as an
estimate of the depth of the photic zone. The Secchi disk is a white disk,
usually 30 cm in diameter. However, Ecology uses a 20-cm diameter Secchi
disk (Singleton, L., 22 September 1987, personal communication). The Secchi
disk is lowered into the water until it disappears from view. Readings are
affected by the visual acuity of the observer and several factors that
affect the light field in the water. These factors include the height of
the sun above the horizon, refraction caused by surface waves, and ship
shadows. The diameter of the Secchi disk only affects Secchi depth slightly
[e.g., increasing the diameter of a Secchi disk from 43 to 60 cm increases
the Secchi depth approximately 1 percent (Preisendorfer 1986, p. 923)]. The
effects of these and related factors are relatively small. Use of Secchi
depths is well-accepted, as long as the data are interpreted only as a
simple index of water clarity (Preisendorfer 1986).
A-3
-------
Fecal Coliform Bacteria
Historically, the concentration of fecal coliform bacteria in Puget
Sound was measured by the most probable number (MPN) method (Greenberg et al.
1985). The MPN method involves a 48-h incubation of a series of dilutions
of a water sample in a culture medium. Fecal coliform bacteria are detected
by the presence of gas bubbles that are produced by bacterial metabolism.
The precision of the MPN method is low, and, especially at low concentrations
of bacteria, this method tends to yield overestimates of the actual bacterial
concentration.
By the mid-1970s, the membrane filtration method for measuring the
concentration of fecal coliform bacteria was widely adopted by the agencies
working in Puget Sound. The membrane filtration method involves filtration
of a water sample through a membrane filter (pore size=0.45 urn). The
filtered material is incubated for 24 h in a culture medium, after which
fecal coliform colonies are counted under a dissecting microscope.
The membrane filtration method is both more accurate and more precise
than the MPN method (Greenberg et al. 1985). Moreover, the two methods
do not yield compatible results. Only fecal coliform data produced by the
membrane filtration method were used in the water quality characterization
project.
Sulfite Waste Liquor
Measurement of sulfite waste liquor in Puget Sound has been based on
the Pearl Benson Index, which dates back to 1940. This method detects the
lignin sulfonates in sulfite waste liquor through the formation of highly
colored quinone oxime derivatives (Barnes et al. 1963; Felicetta and McCarthy
1963). Lignin sulfonates are chemically stable waste products of the
pulping process. The intensity of the color in a treated sample is propor-
tional to the amount of lignin sulfonates present in the original sample.
Other lignins and tannins are also detected by the Pearl Benson Index, which
could cause problems with erroneously high values in some samples. However,
the errors caused by this problem probably are too small to be a substantial
concern for samples heavily contaminated by pulp mill discharges.
Prior to 1953, values of the Pearl Benson Index were determined by
visual comparisons of treated samples to treated standards, probably using
Nessler tubes. After 1953, various spectrophotometers were used. The
earlier data are less reliable and were not analyzed in the characteri-
zation project.
TIME COURSE OF ANALYTICAL TECHNIQUES
A time line of the changes in analytical techniques in the monitoring
programs used as data sources for this project is given in Table A-l. This
information was obtained from reading reports produced during the original
studies and from interviews with the^participating scientists.
A-4
-------
TABLE A-l. TIME LINE OF ANALYTICAL TECHNIQUES USED IN THE
ANALYSIS OF WATER QUALITY IN PUGET SOUND
Organization
Temperature
UW
WDF
Ecology
Ecology
Metro
Salinity
UW
WDF
Ecology
Years Used
1932-1987
1975-1985
1956-1987
1965-1970
1970-1972
1973-1987
1965-1967
1967-1982
1982-1987
1932-1960
1960-1984
1975-1985
1984-1987
1956-1987
1965-1970
1970-1972
1973-1985
Analytical Method Used
Deep sea reversing thermometers
In situ CTD and STD
Lab thermometer in (WDF) water
bottle
Lab thermometer (?)
No data
Electric thermometer on D.O.
probe
Lab thermometer in bottle
Deep sea reversing thermometer
In situ CTD
Knudsen titration
Precision salinity bridge
In situ CTD and STD
Auto-Analyzer
Knudsen titration
Salinometer calibrated by
Knudsen titration
No data
Salinometer calibrated in
laboratory by an STD
A-5
-------
TABLE A-l. (Continued)
Organization
Ecology (Continued)
Metro
Dissolved Oxygen
UW
WDF
Ecology
Years Used
1986
1987
1965-1968
1968-1971
1971-1972
1972-1982
1982-1987
1932-1975
1975-1986
1956-1986
1965-1970
1970-1972
1973-1987
Analytical Method Used
Knudsen titration
Hydrometer and titration
Knudsen titration
Laboratory salinometer
In situ CTD
Laboratory salinometer
In situ CTD
Winkler titration
Carpenter modification of
Winkler titration
Winkler titration
Winkler titration
No data
In situ D.O. probe calibrated
Metro 1965-1982
1982-1983
1983-1987
Dissolved Orthophosphate
UW 1932-1953
1953-1962
1962-1970
against Winkler titration
Winkler titration
In situ D.O. probe
Winkler titration
Thompson and Robinson Nessler
Tubes
Thompson and Robinson Spectro-
photometer
Murphy and Riley Spectro-
photometer
A-6
-------
TABLE A-l. (Continued)
Organization
Years Used
Analytical Method Used
UW (Continued)
WDF
Ecology
Metro
Nitrate
UW
WDF
1970-1987
1956-1962
1962-1973
1973-1987
1965-1970
1970-1972
1973-1987
1966-1971
1972-1987
1932-1953
1953-1963
1963-1970
1970-1987
1956-1966
1966-1973
1973-1987
Hager, Gordon and Park
Auto-Analyzer
Thompson and Robinson Spectro-
photometer
Murphy and Riley Spectrophoto-
meter
Murphy and Riley Photometer
Standard Methods for Water and
Wastewater Manual, American
Waterworks Assoc.
No data
U.S. EPA Manual 365.1 Auto-
Analyzer techniques
Strickland and Parsons
No marine data
Thompson and Robinson Messier
Tubes
Thompson and Robinson Spectro-
photometer
Morris and Riley Spectrophoto-
meter
Armstrong, Stearns and
Strickland Auto-Analyzer
Thompson and Robinson Spectro-
photometer
Morris and Riley Spectrophoto-
meter
Morris and Riley Photometer
A-7
-------
TABLE A-l. (Continued)
Organization
Years Used
Analytical Method Used
Ecology
Metro
Chlorophyll a
UW
WDF
Ecology
1965-1970
1970-1972
1973-1987
1965-1969
1969-1971
1971-1975
Metro
1952-1970
1970-1987
1956-1970
1970-1973
1973-1987
1965-1970
1973-1987
1966-1968
Standard Methods for Water and
Wastewater Manual, American
Waterworks Assoc.
No data
U.S. EPA Manual NO. 365.1 Auto-
Analyzer
Strickland and Parsons
buffered hydrazine
Non-buffered hydrazine FWPCA
Manual
Strickland and Parsons Cadmium
reduction
Richards and Thompson
Spectrophotometer
Strickland and Parsons
Fluorometer Auto-Analyzer
Richards and Thompson Spectro-
photometer
Strickland and Parsons
Spectrophotometer
Strickland and Parsons
Photometer
Strickland and Parsons
Spectrophotometer
Strickland and Parsons
Fluorometer Auto-Analyzer
Creitz and Richards
A-8
-------
TABLE A-l. (Continued)
Organization
Years Used
Analytical Method Used
Metro (Continued)
1968-1987
Strickland and Parsons
Fecal Coliform Bacteria
Ecology 1973-1987
Metro
Sulfite Waste Liquor
UW
WDF
Ecology
1967-1976
1977-1987
1940-1953
1953-1967
7-1953
1953-1971
1967-1984
Standard Methods for Water and
Wastewater Membrane Filtration
Standard Methods for Water and
Wastewater Most Probable Number
Standard Methods for Water and
Wastewater Membrane Filtration
Pearl Benson Index Nessler
Tubes (?)
Pearl Benson Index Spectro-
photometer
Pearl Benson Index Nessler
Tubes (?)
Pearl Benson Index Spectrophoto-
meter
Pearl Benson Index Spectrophoto-
meter
A-9
-------
PHOSPHATE
American Public Health Association. 1971. Standard methods for the
examination of water and wastewater (13th Edition).
Murphy, J., and J.P. Riley. 1962. A modified single solution method for the
determination of phosphate in natural waters. Anal. Chim. Acta 12:162-176.
Redfield, A.C., H.P- Smith, and B.H. Ketchum. 1937. The cycle of organic
phosphorus in the Gulf of Maine. Biol. Bull. Woods Hole 73:421-423.
Robinson, R.J., and T.G. Thompson. 1948. The determination of phosphorus in
seawater. J. Mar. Res. 7:33-39.
Whiteledge, T., S.C. Malloy, C.J. Patton, and C.D. Wirick. 1981. Automated
nutrient analysis in seawater. Report BNL-51398. Brookhaven National
Laboratory, Upton, N.Y.
NITRATE
Chon, D.T.-W., and R.J. Robinson. 1953. Polarographic determination of
nitrate in seawater. J. Mar. Res. 12:1-12.
FWPCA. 1969. Manual of analytical techniques for the National Eutrophi-
cation Research Program.
Morris and J.P. Riley. 1963. Anal. Chim. Acta 29:272.
Mull in, J.B., and J.P. Riley. 1955. The spectorphotometric determination of
nitrate in natural waters with particular reference to seawater. Anal.
Chim. Acta 12:464-480.
Zwicker, B.M.G., and R.J. Robinson. 1944. The photometric determination
of nitrate in seawater with a strichnidine reagent. J. Mar. Res. 5:214-232.
CHLOROPHYLL a
Creitz, G.I., and F.A. Richards. 1955. The estimation and characterization
of plankton populations by pigment analysis. III. A note on the use of
"millipore" membrane filters in the estimation of plankton pigments. J.
Mar. Res. 14(3):211-216.
Parsons, T.R., Y. Maita, and C.M. Lalli. 1984. A manual of chemical and
biological methods for seawater analysis. Pergamon Press.
Richards, F.A., and T.G. Thompson. 1952. The estimation and character-
ization of plankton populations by pigment analysis. II. A spectrographic
method for the estimation of plankton pigments. J. Mar. Res. 2:156-172.
Strickland, J.D.H., and T.R. Parsons. 1968. A practical handbook of
seawater analysis. Fish. Res. Board of Canada. Bull. 167 Section IV. 3.
Ottawa, Canada.
A-ll
-------
FECAL COLIFORM BACTERIA
Greenberg, A.E., R.R. Trussell, and L.S. Clesceri, (eds). 1985. Standard
methods for the examination of water and wastewater (sixteenth edition).
American Public Health Association, Washington, DC. 1268 pp.
SULFITE WASTE LIQUOR
Barnes, C.A., E.E. Collias, V.F. Felicetta, 0. Goldschmid, B.F. Hrutfiord,
A. Livingston, J.L. McCarthy, G.L. Toombs, M. Waldichuk, and R. Westley.
1963. A standardized Pearl Benson, or nitroso, method recommended for the
estimation of spent sulfite liquor or sulfite waste liquor concentration in
waters. Tech. Assoc. Pulp and Paper Industry 46(6):337-346.
Felicetta, V.F., and J.L. McCarthy. 1963. Spent sulfite liquor: X. The
Pearl Benson, or nitroso, method for the estimation of spent sulfite liquor
concentration in waters. Tech. Assoc. Pulp and Paper Industry 46(6) -.347-350.
A-12
-------
APPENDIX B
SUMMARY OF DATA SET QUALITY ASSURANCE REVIEWS
-------
INTRODUCTION
Results of the quality assurance reviews for the data sets included in
the characterization study are summarized below. The corrections described
herein were made only in the data file analyzed in the characterization
study and not in the original files belonging to the agencies that provided
the data. Therefore, these errors may still exist in their files.
UNIVERSITY OF WASHINGTON
Several major problems were encountered during the quality assurance
review of the University of Washington's data set. Although much of the
data were in STORET prior to the initiation of the characterization project,
some data had never been entered into STORET and had to be read into a
computer file. The data were read from existing keypunched cards made
available by E.E. Collias. When a hard copy of the new data entered from
these cards was examined, the letters "B" or "V" sometimes occurred as part
of the depth value. These letters apparently were an artifact of the
obsolete STORET format that was used at the time the data were originally
punched on cards. The letters appeared only in data from continuation cards
for particular records. The space in the data field occupied by these
letters was dropped from the characterization database.
Other problems included errors in data coding and units of measurement.
Data coding errors included the occasional addition of 8 h to the time of
sample collection and reporting nitrite concentrations in the nitrate data
field. Because the time of day was not analyzed in the characterization
study, except to check that Secchi disk depth readings were taken only during
daylight hours, the additional 8 h did not affect the project. The erroneous
nitrate data were dropped from the characterization database. The units of
measurement reported in the University of Washington documentation (Collias
1970) differed from the units encoded in the STORET file for sample depth,
phosphate concentration, and dissolved oxygen concentration. Corrections
for sample depth required multiplying values measured in feet by 0.305 to
obtain values measured in meters. Corrections for dissolved oxygen
concentration required changing values measured in mg-at/L to values
measured in mg/L. Phosphate concentrations were reported to be encoded in
mg/L, but were actually encoded in tenths of mg/L. Correction required
multiplying the original values by 10.
WASHINGTON DEPARTMENT OF ECOLOGY
Some problems were encountered during the quality assurance review of
Ecology's data set. The units for sample depth were given as meters in the
documentation, but the depths were in feet in the STORET file. These values
were simply converted by multiplication by the necessary scaling factor,
0.305. Extremely high phosphate concentrations were reported from some
sites in August 1985. These concentrations were caused by an apparent
laboratory error (Krafft, W., 23 July 1987, personal communication); these
B-l
-------
phosphate data were dropped. Data below detection limits were reported as
the detection limit, accompanied in a separate data field by a STORE! code
that identified such points as having concentrations below the detection
limit. However, a few points with undetected values for concentrations of
nitrate and fecal coliform bacteria in the mid-1970s were reported as
zeroes. These values were changed to the appropriate detection limit
values.
WASHINGTON DEPARTMENT OF FISHERIES
No problems were encountered during the quality assurance review of
these data sets.
METRO
The quality assurance review of Metro's data set uncovered two signifi-
cant problems. Some Secchi disk depth values were unreasonably high. Metro
field personnel indicated that Secchi disk depths of 10 m are common in the
region included in this study (i.e., Point Jefferson), but that substantially
higher readings are not credible (Waddell, D., 20 May 1987, personal
communication). However, approximately 6 percent of the Secchi disk depth
values were above 15 m, with values running as high as 72 m. In a table of
the frequency distribution of the Secchi depth values for this region, the
values above 15 m inexplicably jumped from one figure to the right of the
decimal to two figures to the right of the decimal. Because this shift
suggested an error in data coding, all Secchi depths greater than 15 m were
discarded.
The other problem encountered in Metro's data set involved an analytical
anomaly. An oxygen probe was used to measure dissolved oxygen concentrations
in the water column, starting in September 1982. However, the probe was not
properly equilibrated at each depth for measurements taken between September
1982 and January 1983 (Lehman, K.( 20 May 1987, personal communication).
Various corrections in the field procedures were instituted from February
1983 to September 1983, except that no corrections were used during April
1983. Starting in October 1983, Metro's database contains only Winkler
dissolved oxygen determinations. Therefore, the dissolved oxygen data for
the periods September 1982 to January 1983 and April 1983 were discarded.
CLIMATIC DATA
The quality of the climatic data is high. The locations of the
monitoring stations have not changed greatly over the periods of the
observations. The weather data are validated by the National Climatic Data
Center staff before the reports are published. The runoff estimates, which
are based on gaging data from seven stations, were compared with more
detailed runoff estimates based on data from 22 gaging stations for the
period of 1970-1975 (NOAA 1984a). The mean difference in the two estimates
was only 3.3 percent, suggesting that the seven gaging stations used for the
1930-1978 runoff estimates provided reasonably good data.
B-2
-------
The quality of the computer file containing the climatic data is also
high. Both the weather and the runoff data were entered manually into a
file for the characterization project. A hard copy of the data was validated
by comparison with the original data reports.
B-3
-------
APPENDIX C
SOURCES OF PUGET SOUND WATER QUALITY DATA
-------
INSTITUTION: U.S. Navy - Bangor
CONTACT: Mr. Rick Spencer, U.S. Naval Submarine Base - Bangor, Bldg. 1101,
Code 8622, Bremerton, WA 98315; Phone (206) 396-4192.
DATA DESCRIPTION; Monitoring program to evaluate the impact of naval activi-
ties on the water quality of Hood Canal. Water samples taken for trace
metals (chromium, copper, iron, lead, mercury, silver, zinc, nickel),
nutrients (ammonia, nitrates, nitrites, Kjeldahl nitrogen, orthophosphate),
total organic carbon, pH, salinity, temperature, and dissolved oxygen.
Secchi disk readings taken concurrently with water sampling.
LOCATION; Twenty sites in Hood Canal and Dabob Bay ranging from 47° 43'
46"N to 47° 46' 29" N and 122° 42' 10" W to 122° 46' 77" W.
PERIOD/FREQUENCY; 1974 to present and ongoing. All 20 sites sampled twice
per year in summer and winter.
DATA FORMAT; Raw data files. All data sent to Naval Energy and Environ-
mental Support Activities, Port Heuneme, CA.
PRIORITY: Low because of frequency of observation and not in the area
chosen for analyses.
DATA EVALUATION; Unique because of area covered and scope of sample types.
Nominal or low importance to trends study unless Ecology data are bad. These
data are not in chosen area.
INSTITUTION: Washington Department of Natural Resources
CONTACT; Mr. Tom Mumford, Research and Development Center, Washington
Department of Natural Resources, Olympia, WA 98504; Phone (206) 753-3703.
DATA DESCRIPTION: Hydrographic and chemical data from surface waters
(temperature, salinity, phosphorus, nitrates, nitrites, and ammonia).
LOCATION; See below.
PERIOD/FREQUENCY;
Budd Inlet - 1979-1980. Daily monitoring.
Squaxin Island - Fall 1982 to Spring 1983. Daily monitoring.
Harstene Island - Fall 1982 to Spring 1983. Daily monitoring.
McNeil Island - Fall of 1982 to present and ongoing. Temperature and
salinity daily, nutrients weekly.
C-l
-------
DATA FORMAT; Raw data files with portions also available on magnetic tape.
PRIORITY; Medium for project, especially Budd Inlet,
DATA EVALUATION; Unique to sample areas - quality control good.
INSTITUTION; Washington Department of Fisheries
CONTACT; Mr. Stan Hammer, Fox Island Net Pens, 335 Island Blvd., Fox
Island, WA 98333; Phone (206) 857-4324.
DATA DESCRIPTION; Temperature and dissolved oxygen measurements taken to
protect salmon rearing operations.
LOCATION; Fox Island.
PERIOD/FREQUENCY; Mid-1970s to present and ongoing. Temperature measure-
ments taken daily. Dissolved oxygen samples taken daily during critical
periods, generally June and July.
DATA FORMAT; Raw data files.
PRIORITY; Low for project.
DATA EVALUATION: Probably inconsistent quality control; low importance to
areas being used for analysis.
INSTITUTION; Padilla Bay National Estuarine Sanctuary.
CONTACT; Mr. Terry Stevens, Padilla Bay National Estuarine Sanctuary. 1043
Bay View-Edison Rd., Mount Vernon, WA; Phone (206) 428-1558.
COMMENTS; Padilla Bay Estuarine Sanctuary does not fund research nor
perform any environmental data gathering activities independent of other
institutions or government agencies. However, they provide facilities for
research conducted under the auspices of other agencies. A summary of past
research conducted in the Padilla Bay area is given below.
Type of Sampling Aoencvfs) Date Investigator
Sulfite Waste 1946 Fish & Wildlife 1946 Saxton-Young
(Water Quality) Ser., WDF
Pulp Mill Pollution WA Water Pollution 1950 G. Orlob -
& Oyster Culture Commission A. Neale
C-2
-------
Sulfite Waste Liquor
Pollution Fidalgo &
Padilla Bays
Industrial Waste
(Water Quality)
Oyster
(Water Quality)
Oyster
(Water Quality)
Eelgrass
Benthic Forams of
Samish & Padilla Bays
Prelim. Inventory of
Biota of Padilla Bay
Swinomish Channel
Maint. Dredging
Effects on Biota of
Fidalgo Bay due to
Navigation Channel
Swinomish Channel
Dredged Material
Reuse Study
Investigation of Tidal
Soils of Padilla Bay
Subtidal Benthic Comm-
munities and Density
of Petroleum-Degrading
Bacteria in Padilla Bay
Southwest Padilla Bay
Tidelands Environ.
Impact Assessment
Physical, Chemical,
& Biological charac-
teristics of Padilla Bay
Trace Metals in
Ecosystem of
Padilla Bay
WA Water Pollution
Commission
Pollution Control
Commission
Pollution Control
Commission
WDF
WDG/Funded by Fish
& Wildlife Serv.
WWU M.S. Thesis
WA Dept. of Game
U.S. Army Corps of
Engineers EIS
U.S. Army Corps of
Engineers
Skagit Co. Planning
Dept.
WSU
WWU M.S. Thesis
1948
1957
1952
1950
1971-75
1973
1976
1976
1977
1980
1980
1982
WWU
U of W
WWU M.S. Thesis
C-3
1983
1984
1985
W. Saxton
A. Neale
A. Neale
Orlob-Neale-
Lindsay
B. Jeffrey
D. Scott
R. Jeffrey
D. Turner
J. Barreca
Huxley College
R. Wissman
L. Antrim
-------
Padilla Bay
Dungeness Crab
Habitat Study
Definitions of
Origins & Fates of
Organic Nitrogen in
Padilla Bay Food Webs
Padilla Bay Base-
Water Quality Record
Intertidal Benthos
Subtidal-Eelgrass
Benthos
Beach Seine (fish)
Marine Birds
Marine Birds
Marine Birds
Marine Mammals
Land Use/Land Cover
Drift Sectors
Inventory of Com-
pilation of Biota (Data)
Inventory of Com-
pilation of Biota (Data)
U of W
U of W
1986
1986
WWU
WWU Huxley College
Funded by Ecology
WWU Huxley College
Funded by Ecology
WWU Huxley College
Funded by Ecology
WDG + funded by
U.S.F.W.S
John Graham Co.
Funded by ACOE
U.W. funded by EPA
through NOAA (MESA)
NMFS funded by NOAA
(MESA)
WDG funded by OCZM
through WDOE
John Norman Assoc.
funded through WDOE
WWU Huxley College
WDF, WDG
WDG
1986
1974-75,
1979
1976
1974-75
1965-79
1977-78
1978-79
1977-79
1978
1977
1976
1977
P. Dinnel
R. McMillan
D. Armstrong
R. Wissman
P. Cassidyline
G. McKeen
Webber-Smith
Webber-Smith
Webber-Smith
Webber-Smith
Peters-Richter
Manuwal-Wahl
R. Everitt
R. Albright
J. Norman
B. Jeffrey
Sweeney
LOCATION; University of Washington
CONTACT; Dr. Carl Lorenzen, University of Washington, Dept. of Oceanography
Seattle, WA 98195; unavailable.
C-4
-------
DATA DESCRIPTION: Research directed toward understanding phytoplankton
dynamics and seasonal variability. Measurements made of phytoplankton
pigment concentrations and primary productivity throughout a vertical
profile to a depth of 100 m.
LOCATION; Dabob Bay.
PERIOD/FREQUENCY: 1975-1985. A single site occupied at monthly intervals.
DATA FORMAT; Raw data files.
PRIORITY; Medium for chlorophyll a data in Dabob Bay.
DATA EVALUATION; Data hard to access. Advice of Dr. Lorenzen unavailable.
INSTITUTION: University of Washington
CONTACT; Dr. Jerry Stober, Fisheries Research Institute, College of
Fisheries, University of Washington, Seattle, WA 98195; Phone (206) 543-9041.
COMMENTS; In an effort to evaluate the potential ecological impacts of a
nuclear power plant, a multidisciplinary study of the fisheries and marine
ecology of northern Skagit Bay in the vicinity of Kiket Island was undertaken
by the Fisheries Research Institute. Because of the diverse data collected,
each component of the research is considered individually in the following
summaries. All of the reports cited in these sections can be found in:
Stober, Q.J., and E.O. Salo. 1973. Ecological studies of the proposed
Kiket Island nuclear power site. University of Washington College of
Fisheries, Fisheries Research Institute, FRI-UW-7304. Final Report.
Submitted to Snohomish County P.U.D. and Seattle City Light.
PRIORITY: Low.
DATA EVALUATION: Unique for Skagit Flats area and high density data for
1 yr; not applicable for present trends study because of location.
LOCATION: University of Washington
CONTACT: Dr. Jerry Stober, Fisheries Research Institute, College of Fisher-
ies, University of Washington, Seattle, WA 98195; Phone (206) 543-9041.
DATA DESCRIPTION; Hydrographic data (temperature, salinity, turbidity,
dissolved oxygen) from surface and bottom waters.
LOCATION; Similk Bay, North Skagit Bay, Swinomish Channel.
C-5
-------
PERIOD/FREQUENCY; 1970-1972. Continuous record of temperature at surface,
3 m, and bottom. Grid sampling of surface waters, March-July 1970; March-
May 1971; and March-August 1972.
DATA FORMAT: Stober, Q.J., S.J. Walden, and D.T. Griggs. Seasonal water
quality in North Skagit Bay. In: Stober et al. (1973). Chap. 4, pp. 7-34.
PRIORITY: Low.
DATA EVALUATIONS; See previous comments.
INSTITUTION; University of Washington.
CONTACT; Dr. Jerry Stober, Fisheries Research Institute, College of Fisher-
ies, University of Washington, Seattle WA 98195; Phone (206) 543-9041.
DATA DESCRIPTION; Investigation of temporal and spatial distribution and
abundance of ichthyoplankton. Two replicate vertical plankton hauls taken
from both bottom to surface and 5 m to surface. Nansen casts for temperature
and salinity taken at each station prior to zooplankton sampling.
LOCATION; Northern Skagit Bay.
PERIOD/FREQUENCY; January 1971 through April 1972 with sampling intervals
spaced 1 wk to 1 mo apart. Some stations repeated as frequently as twice
per cruise.
DATA FORMAT; Blackburn, J.E. Pelagic eggs and larval fish of Skagit Bay.
In: Stober et al. (1973). Chap. 6, pp. 71-118.
PRIORITY; Low.
DATA EVALUATION; See previous comments.
INSTITUTION; University of Puget Sound.
CONTACT: Dr. Eric Lindgren, University of Puget Sound, 1500 N. Warner,
Tacoma, WA 98416; Phone (206) 765-3121.
DATA DESCRIPTION: Hydrographic data on surface waters (temperature,
salinity, dissolved oxygen, turbidity, pH).
LOCATION; Tacoma Narrows.
PERDIO/FREQUENCY; 1973 to present and ongoing. Samples taken annually
every fall and occasionally in spring:
DATA FORMAT; Student reports.
C-6
-------
COMMENT: Data collected by students as part of an introductory oceano-
graphy class. Inexperience of students makes the data highly suspect.
PRIORITY; Low.
DATA EVALUATION: Infrequent sampling but long time span. Data hard to
access and may have quality control problems.
INSTITUTION: Shoreline Community College.
CONTACT; Mr. Jack Serwold and Mr. Bob Harman, Shoreline Community College,
16101 Greenwood Avenue N., Seattle, WA 98133; Phone (206) 546-4101.
DATA DESCRIPTION; Species composition and abundances of benthic diatoms,
foraminifera, and macroinvertebrates collected using 0.1 m2 Van Veen grab
sampler. Concurrent Secchi disk readings and temperature and salinity
measurements at 1 and 3 m. Sampling has recently included a plankton
sample at 3-m depth.
LOCATION; Approximately 2,000 sites throughout Puget Sound, primarily in
the Nisqually Delta,, central basin, and northern sound. Samples generally
taken at 1, 5, 10, 20 fathoms and in the deep areas of each region.
PERIOD/FREQUENCY: Nearly all work to date has been done as single surveys
with only occasional resampling of specific sites. Sampling periods are as
follows:
Central basin: 1974-1978
Central basin north of Edmonds: 1981
Commencement Bay: 1980-1981
Everett-Port Susan: 1978-1979
Nisqually Delta: 1982
Northern Saratoga Passage-Skagit Bay: 1984
DATA FORMAT: Raw data files.
COMMENTS: The level of analysis of the benthic samples is dependent on
taxonomic groups. Molluscs have been identified to species; polychaetes and
other groups have generally been identified only to higher taxa.
PRIORITY; Low.
DATA EVALUATION; Sample locations not repeated through time. Data difficult
to access.
C-7
-------
INSTITUTION: Olympic Community College
CONTACT; Dr. Don Seavy, Olympic Community College, 16th and Chester,
Bremerton, WA 98310; Phone (206) 478-4557.
DATA DESCRIPTION; Measurements of surface water temperature, salinity, pH,
and dissolved oxygen along with concurrent zooplankton samples.
LOCATION; Several stations within Sinclair Inlet.
PERIOD/FREQUENCY; 1977 to present and ongoing. Monthly samples but lacking
the summer months.
DATA FORMAT; Raw data files.
COMMENTS; Zooplankton samples only partially worked up but available for
further analysis. Much of the hydrological data has been forwarded to Alan
Mearns, NOAA.
PRIORITY; Low.
DATA EVALUATION; Geographically limited; time span good for area covered;
data difficult to access.
LOCATION: Highline School District.
CONTACT; Mr. Lauren Rice, Marine Technology Dept., 18010 8th Avenue
S.,Seattle, WA 98148, Phone (206) 433-2524.
DATA DESCRIPTION; Vertical profiles of temperature, salinity, and dissolved
oxygen.
LOCATION: Shilshole.
PERIOD/FREQUENCY; 1975 to present and ongoing, annually each May.
DATA FORMA: Raw data files.
PRIORITY; Low; infrequently sampled.
DATA EVALUATION: This group no doubt has nearshore data near Fully Point in
the south main basin study area. Data difficult to / access and may have
problems in quality control; limited time coverage.
LOCATION; Battelle Northwest
CONTACT: Dr. Jack Anderson, Battelle Pacific Northwest Division, Marine Res-
earch Laboratory, Route 5, Box 1000, Sequim, WA 98382; Phone (206) 683-4151.
C-8
-------
COMMENTS; During 1972-1974 Battelle Northwest was involved in an extensive
baseline study involving both chemical and biological surveys, prior to
operation of the ARCO refinery at Cherry Point. This study represents a
potentially valuable database for any future monitoring efforts in the
Strait of Georgia, but is still considered proprietary data by ARCO.
PRIORITY: Low.
DATA EVALUATION: Outside of Bellingham Bay study region; data difficult to
access.
INSTITUTION; Seattle Aquarium
CONTACT; Mr. Bill Bruin, Seattle Aquarium, Pier 59, Seattle, WA 98101;
Phone (206) 625-4358.
DATA DESCRIPTION; Hydrographic and water quality measurements of aquarium
intake water (temperature, salinity, pH, turbidity, total coliform, dissolved
oxygen). Intake located 80 ft below surface.
LOCATION: Elliott Bay.
PERIOD/FREQUENCY; 1977 to present and ongoing. Data collected intermit-
tently in 1977. Since 1978, temperature, salinity, pH, and turbidity have
been collected daily, total coliform and dissolved oxygen on a weekly basis.
DATA FORMAT: Raw data files.
PRIORITY; Low.
DATA EVALUATION; Data have good time span but are not within a study region
of the sound included in the characterization study.
INSTITUTION; Point Defiance Zoo and Aquarium.
CONTACT: Mr. John Rupp, Pt. Defiance Zoo and Aquarium, N 54th Street and N
Pearl, Tacoma, WA 98407; Phone (206) 592-5223.
DATA DESCRIPTION; Hydrographic measurements on aquarium intake water
(temperature, salinity, dissolved oxygen, pH). Intake located 15-20 ft
below surface.
LOCATION; Point Defiance.
PERIOD/FREQUENCY; 1982 to present and ongoing. Sampling at irregular inter-
vals but approximately on a monthly basis. Greatest sampling frequency in
winter and spring.
C-9
-------
DATA FORMAT; Raw data files.
PRIORITY: Low.
DATA EVALUATION; Sampling location is outside of study areas.
INSTITUTION; Domsea Farms, Inc.
CONTACT; Mr. Mike Gardner, Domsea Farms, Inc., 4398 West Old Bel fair
Highway, Bremerton, WA 98312; Phone (206) 479-9941.
DATA DESCRIPTION; Dissolved oxygen measurement of surface waters to protect
salmon rearing operations.
LOCATION; Fort Ward (Bainbridge Island) and Orchard Point.
PERIOD/FREQUENCY; 1975 to 1978. Monitoring on an irregular basis only when
there is cause for concern. Most samples taken during fall months.
DATA FORMAT; Raw data files.
PRIORITY; Low.
DATA EVALUATION; Outside of study and not continuous in time or over the
annual cycle.
INSTITUTION; Sundquist Laboratory.
CONTACT; Mr. Paul Cassidy, Sundquist Laboratory, 1900 Shannon Point Avenue,
Anacortes, WA 98221; Phone (206) 293-6800.
DATA DESCRIPTION; Hydrographic data of surface waters (temperature, pH,
turbidity, dissolved oxygen, total alkalinity, carbonate alkalinity,
dissolved 0)2, and salinity).
LOCATION; Shannon Point, Anacortes.
PERIOD/FREQUENCY; 1974 - present and ongoing (temperature, pH, dissolved
oxygen, turbidity). 1977 to present and ongoing (total and carbonate
alkalinity, C02, salinity). Sampling was daily but currently approximately
three times per week. Special study of Padilla Bay in 1985 for 1-yr period.
Samples taken several times per month.
DATA FORMAT; Raw data files.
PRIORITY; Low for Shannon Pt., high for Padilla Bay locations, if used.
C-10
-------
DATA EVALUATION; Monthly data. Data presentation poor but analytical
methods probably good. Can be accessed on 5.25-in floppy disk.
INSTITUTION; Tulalip Tribes.
CONTACT; Mr. Dave Somers, Tulalip Tribe, 7600 Totem Beach Road, Marysville,
WA 98370; Phone (206) 653-4588.
DATA DESCRIPTION; Parametrix, Inc. was contracted to conduct a baseline
survey of the water quality and fisheries resources of Tulalip Bay in
preparation for expansion of a salmonid hatchery operation. A wide variety
of parameters were measured in the surface waters of the bay, including
general physical and chemical properties, nutrients, coliforms, trace
metals, and synthetic organics.
LOCATION: Tulalip Bay, four stations.
PERIOD/FREQUENCY; General physical/chemical properties, nutrients, and
microbial analyses: April 13 to June 27, 1979; weekly sampling frequency.
Metals and synthetic organics: April 18 to June 27, 1979; sampling every
third week,
DATA FORMAT; Campbell, R.F., and D.E. Weitkamp. 1979. Water quality and
nearshore fish investigations in Tulalip Bay, Washington, 1979. Prepared by
Parametrix, Inc. for the Tulalip Tribes, Marysville, WA.
PRIORITY; Low.
DATA EVALUATION; Surface values, good mix of variables, but not in study
areas; limited time span.
INSTITUTION: Post Point Sewage Treatment Plant, Bellingham, WA.
CONTACT: Operator on Duty; (206) 676-6977; or Gary Hess, same number.
DATA DESCRIPTION; Not well known, but must be water quality data including
conventionals; data results from study by CH2M HILL in Bellingham's effort
to secure secondary treatment waiver (1982-1983).
LOCATION; Bellingham and Sammish Bays.
PRIORITY: Medium for Bellingham Bay.
DATA EVALUATION; Data unevaluated.
C-ll
-------
INSTITUTION: Evans-Hamilton
CONTACT: Jeff Cox, Evans-Hamilton, 6302 21st Northeast, Seattle, WA 98105;
Phone (206) 525-5268.
DATA DESCRIPTION: Full suite water parameter data, Metro Se^hurst Study,
1982-83.
LOCATION; Main Basin (south).
PRIORITY: Medium for south half of Main Basin; time covered too short.
DATA EVALUATION: This may be the only source of water quality data from
Metro Seahurst Study; Metro claims their data tapes are unreadable; Evans-
Hamilton's data tapes are edited and would cost money to access.
INSTITUTION: University of Washington.
CONTACT; Jim Postel, School of Oceanography, WB-10, University of Washing-
ton.
DATA DESCRIPTION; Chlorophyll a and water property data; Metro Seahurst
Study, 1983-1983.
LOCATION; Main Basin (south).
PRIORITY: Medium for south half of Main Basin; time covered too short.
DATA EVALUATION: Published report by A. Copping, J. Postel, and J. Anderson;
data should be good but may be hard to access by computer; must check with
Postel for confirmation.
INSTITUTION: University of Washington,
CONTACT; Jan Downs, School of Oceanography, WB-10, University of Washington;
Phone (206) 543-9658.
DATA DESCRIPTION: Chlorophyll a and water property data; monthly collected
by Carl Lorenzen.
LOCATION: Dabob Bay region.
DATA FORMAT: Data logs and notebooks; a little published; some compiled by
J. Downs.
PRIORITY: Medium for Dabob Bay area; sampling not done in characterization
study areas.
C-12
-------
DATA EVALUATION; Good data but study sites far from Ecology station; 1979-
early 1980's; data hard to access by computer; Lorenzen not capable of
assistance.
INSTITUTION; University of Washington.
CONTACT; Bruce Frost, School of Oceanography, WB-10, University of Washing-
ton.
DATA DESCRIPTION; Chlorophyll a and water property data; weekly 1979-1980,
monthly 1982, 1984, 1985; includes nutrients.
LOCATION; Dabob Bay region.
DATA FORMAT; Data logs and some published.
PRIORITY; Medium for Dabob Bay region; sampling not done in characterization
study areas.
DATA EVALUATION; Good data coverage through time; hard to access by
computer.
INSTITUTION; University of Washington.
CONTACT; George Anderson, retired, School of Oceanography, WB-10, University
of Washington; or Jim Postel, same address, (206) 543-6141.
DATA DESCRIPTION; Chlorophyll a data, Main Basin of Puget Sound, and
productivity data 1964-67.
LOCATION; Main Basin of Puget Sound.
DATA FORMAT; Printed and published data; nine-track tape available through
J. Postel.
PRIORITY; Medium for Main Basin area; major source of chlorophyll a data
for this area; time covered too short.
DATA EVALUATION; Unique source but data may be hard to access.
INSTITUTION: University of Washington.
CONTACT; Willis K. Peterson, School of Oceanography, WB-10, University of
Washington; Phone (206) 543-6141.
C-13
-------
DATA DESCRIPTION: Chlorophyll a data, Main Basin of Puget Sound, and
productivity in 1975.
LOCATION: Main Basin of Puget Sound.
DATA FORMAT: Published report: Phytoplankton Production and Standing Stick
in the Main Basin of Puget Sound. Puget Sound Interim Studies, 1977.
PRIORITY: Medium for Main Basin area; significant source of Chlorophyll a
data; time covered too short.
DATA EVALUATION; Unique study but data may be hard to access.
INSTITUTION; Municipality of Metropolitan Seattle.
CONTACT: Rich Tomlinson or Ray Dalseg, Metro Water Quality Lab; Phone (206)
684-2313.
DATA DESCRIPTION; Salinity, temperature, dissolved oxygen, some nutrients,
chlorophyll a, special purpose surveys in Main Basin of Puget Sound; 1966 -
present; some monthly.
LOCATION; Main Basin Puget Sound.
DATA FORMAT; Computerized data in STORET.
PRIORITY; High for Main Basin. Required to extend time line for Point
Jefferson Station.
DATA EVALUATION; Unique source but doubtful quality control in dissolved
oxygen due to field techniques and continual change in equipment and
technologies.
INSTITUTION; University of Washington.
CONTACT; Eugene E. Collias, 4318 First Avenue NE, Seattle, WA 98105; phone
(206) 633-5570.
DATA DESCRIPTION; Salinity, temperature, dissolved oxygen, nutrients,
Secchi disk; Puget Sound Interim Studies, Metro's Nutrient Budget of Puget
Sound (1976).
LOCATION; Main Basin Puget Sound.
DATA FORMAT; Computer card storage and tape.
PRIORITY; High for location, types of parameters, and the project.
C-14
-------
DATA EVALUATION: Good, University of Washington data to extend time series
at Point Jefferson.
INSTITUTION: Washington Department of Ecology.
CONTACT; Merley McCall, Chief Chemist, Manchester Lab, (206) 442-0370.
DESCRIPTION; Salinity, temperature, dissolved oxygen, nutrients,
chlorophyll a, and others; routine surveys 1967 to present, widely spaced
Puget Sound stations about once each month (except winter); 0, 10, 30-m
depths only.
LOCATION: Stations throughout Puget Sound.
DATA FORMAT: Computerized data in STORET.
PRIORITY: High throughout Puget Sound; required to extend time series from
earlier work.
DATA EVALUATION; Lack of continuity and changing lab practices may affect
data.
LOCATION; Marine Science Center.
CONTACT; Mr. James Kolb, Marine Science Center, 17771 Fjord Drive NE,
Poulsbo, WA 98370; Phone (206) 779-5549.
DATA DESCRIPTION: Salinity, temperature, dissolved oxygen, pH, weather.
LOCATION; Liberty Bay, fall-winter 1971-72, twice monthly spring 1973.
DATA FORMAT: Report log sheets by P. Maloney and M. J. Delk.
PRIORITY; High for location, low for project.
DATA EVALUATION; Unique because area involved; low importance to study;
directed study by high school students.
INSTITUTION; Washington State Department of Fisheries.
CONTACT; Marvin Tarr, Point Whitney Lab, Brinnon, WA; (206) 796-4601.
DATA DESCRIPTION: Salinity, temperature, dissolved oxygen, nutrients,
chlorophyll a; 1950s and 1970s data from shellfish producing areas.
C-15
-------
LOCATION: Much in small embayments of southern sound and in Bellingham and
Dabob.Bays.
DATA FORMAT; Log sheets and printed reports.
PRIORITY: High for location, types of parameters, and the project.
DATA EVALUATION; Good data through time in areas of sound but subject to
changes in chemical techniques.
INSTITUTION; University of Washington.
CONTACT: Fish-Ocean Library, School of Oceanography.
DATA DESCRIPTION; M.S. and Ph.D. theses containing chlorophyll a and water
property data.
LOCATION; Main Basin and Dabob Bay regions.
DATA FORMAT; Variable, some data appendices.
PRIORITY; Medium for project.
DATA EVALUATION; See list of selected theses given below. These will be
very difficult to extract data from because most of these are interpretations
of the data.
C-16
-------
CHLOROPHYLL a SOURCES
John P. Bavlon, Spring Changes in Phytoplankton Abundance in a Deep Estuary,
Hood Canal, Washington, Journal of Marine Research, Vol. 17, 1958, p. 53-67.
(Chlorophyll a and productivity estimates for four periods in 1953. Data are
in the publication.)
Robert Theodore Cooney, PhD Dissertation, Zooplankton and Micronekton
Associated with a Diffuse Sound-Scattering Layer in Puget Sound, Washington,
1971. Fish Ocean Library GC/7/Th 19248 (see Figure 8 for plotted values, 3
stations).
Jed Hirota, MS Thesis, Use of Free-Floating Polyethylene Cylinders on
Studies of Puget Sound Phytoplankton Ecology, 1967. (Some chlorophyll and
productivity estimates) Fish Ocean Library. GC/7/TH16457.
Hans Julian Hartmann, MS Thesis, Release and Assimilation of Dissolved
Organic Carbon by Natural Marine Phytoplankton Populations, 1974. (Some
Chlorophyll a and C14 productivity values)-
Jerry David Larrance, MS Thesis, A Method for Determining Volume of
Phytoplankton in a Study of Detrital Chlorophyll a, 1964. Fish-Ocean Library
551.46, Th 13399.
Willis K. Peterson, Serena Campbell, Phytoplankton Production and Standing
Stock in the Main Basin of Puget Sound, 1977. Metro Interim Studies.
Fish-Ocean Library, QK 192 C34 1977.
Robert Munson, The Horizontal Distribution of Phytoplankton in a Bloom in
Puget Sound, May 1969, Non-Thesis MS degree report, 1970. (available from
Karl Bause, University of Washington; has data for 1967, salinity, temper-
ature, dissolved oxygen, phosphate, nitrate, Secchi depth, Chlorophyll a).
Mark David Ohman, PhD Dissertation, 1983. The Effects of Predation and
Resource Limitation on the Copepod Pseudocalnus sp. in Dabob Bay, a
Temperate Fjord. Fish Ocean Library.. GS/7/Th 31369.
Jeffrey Albert Runge, PhD Dissertation, 1981. Egg Production of Calanus
pacificus Brodsky and its Relationship to Seasonal Changes in Phytoplankton
Availability. Fish Ocean Library, GC/7/Th 29440.
Frank Randolph Shuman, PhD Dissertation, 1978, The Fate of Phytoplankton
Chlorophyll in the Euphotic Zone, Washington Coastal Waters. Fish Ocean
Library, GS/7/Th 26441. (Ci4 and Chlorophyll a values, Dabob Bay, 1975-
1976).
C-17
-------
Andrea Copping, PhD Dissertation, 1982, The Distribution and Passage of
Organic Matter in the Marine Food Web, Using Nitrogen as a Tracer. Fish
Ocean Library, GC/7/Th 30233.
Nicholas A. Welschmeyer, PhD Dissertation, 1982. The Dynamics of Phytoplank-
ton Pigments: Implications for Zooplankton Grazing and Phytoplankton
Growth, Dabob Bay. Fish Ocean Library, GC/7/Th 29691.
Jeffrey Reinge, 1985. Relationship of Egg Production of Calanus pacificus
to Seasonal Changes in Phytoplankton Availability in Puget Sound, Washing-
ton. Limnology and Oceanography, 30(2), pp. 382-396.
C-18
-------
APPENDIX D
COMPARABILITY OF DATA FROM DIFFERENT SOURCES
AT STATIONS WITH OVERLAPPING SAMPLING PERIODS
-------
TABLE D-l. DATA COMPARISONS BETWEEN WASHINGTON DEPARTMENT OF FISHERIES
AND WASHINGTON DEPARTMENT OF ECOLOGY
Variable
Dissolved
Oxygen
Salinity
Water
Temperature
Agency
WDFa
Ecology"
WDF
Ecolgy
WDF
Ecolgy
WDF
Ecology
WDF
Ecology
WDF
Ecology
WDF
Ecology
WDF
Ecology
WDF
Ecology
WDF
Ecology
WDF
Ecology
WDF
Ecology
Number of
Season Observations Mean
Spring
Summer
Autumn
Winter
Spring
Summer
Autumn
Winter
Spring
Summer
Autumn
Winter
22
3
29
3
20
1
10
3
22
3
29
3
20
1
10
3
20
2
31
3
20
1
8
3
9.06
9.70
7.37
6.17
7.18
6.50
9.64
8.83
23.41
23.67
27.02
26.37
25.53
25.80
18.81
21.73
12.76
11.10
18.05
19.60
11.38
13.70
6.83
5.70
Standard
Error p
0.16 --c
0.23
0.19
0.48
0.18
0.08
1.45
0.53
1.92
0.19
1.66
0.52
0.88
1.75
0.69
0.50
0.27
1.19
0.49
0.45
2.07
a Washington Department of Fisheries, Station 23, Oakland Bay.
b Washington Department of Ecology, Station OAK004, Oakland Bay.
c -- = Not statistically significant (P>0.05).
D-l
-------
TABLE D-2. DATA COMPARISONS BETWEEN METRO
AND UNIVERSITY OF WASHINGTON
Variable
Dissolved
Oxygen
Salinity
Water
Temperature
Agency
Metro3
uwb
Metro
UW
Metro
UW
Metro
UW
Metro
UW
Metro
UW
Metro
UW
Metro
UW
Metro
UW
Metro
UW
Metro
UW
Metro
UW
Number of
Season Observations Mean
Spring
Summer
Autumn
Winter
Spring
Summer
Autumn
Winter
Spring
Summer
Autumn
Winter
97
26
100
14
15
11
6
10
97
34
98
15
15
11
6
10
112
33
108
15
19
12
6
10
11.18
11.00
9.65
8.81
7.09
6.99
9.37
9.05
26.50
27.42
27.80
28.75
29.35
30.14
27.19
27.09
10.05
10.80
13.48
13.61
11.96
11.17
7.35
7.60
Standard
Error p
0.18 --c
0.31
0.17
0.55
0.11
0.16
0.14
0.18
0.21 *d
0.18
0.22 *
0.48
0.34 *
0.11
0.76
0.65
0.17 **e
0.21
0.13
0.32
0.16
0.30
0.16
0.25
a Metro, Station KSBP01, Point Jefferson.
b University of Washington, Station PSB305, Point Jefferson.
c -- = Not statistically significant (P>0.05).
d * = Statistically significant (P<0.05).
e ** = Statistically significant (P<0.01).
D-2
-------
TABLE D-3. DATA COMPARISONS BETWEEN UNIVERSITY OF WASHINGTON
AND WASHINGTON DEPARTMENT OF ECOLOGY
Variable
Dissolved
Oxygen
Salinity
Water
Temperature
Agency
uwa
Ecology^
UW
Ecology
UW
Ecology
UW
Ecology
UW
Ecology
UW
Ecology
UW
Ecology
UW
Ecology
UW
Ecology
UW
Ecology
UW
Ecology
UW
Ecology
Number of
Season Observations Mean
Spring
Summer
Autumn
Winter
Spring
Summer
Autumn
Winter
Spring
Summer
Autumn
Winter
7
3
4
2
5
2
3
3
7
3
4
2
5
2
3
2
7
3
4
2
5
2
3
3
9.32
11.10
8.43
8.50
6.98
5.75
8.78
8.50
28.62
26.57
29.80
27.30
30.48
31.80
28.99
30.70
9.66
10.63
13.04
13.15
11.01
11.55
8.33
8.53
Standard
Error p
0.38 **c
0.12
0.57 --d
0.90
0.18 *e
0.05
0.52
0.45
0.39 *
0.47
0.19 *
1.20
0.09
4.20
0.53
1.40
0.37
0.37
0.32
0.15
0.57
0.65
0.15
0.78
a University of Washington, Station PSB318, Alki Point.
b Washington Department of Ecology, Stati9n PSB002, Alki Point.
c ** = Statistically significant (P<0.01). ~^
d -- = Not statistically significant (P>0.05).
e * = Statistically significant (P<0.05).
D-3
-------
APPENDIX E
DESCRIPTIVE STATISTICS FOR WATER QUALITY VARIABLES
-------
TABLE E-l. BELLINGHAM BAY STUDY AREA
Variable
Salinity
Water
Temperature
Dissolved
Oxygen
Dissolved
Oxygen
Saturation
Nitratea
Phosphate3
Chloro-
phyll a
Secchi Depth
Sulfite
Waste
Liquor
Units
ppt
°C
mg/L
Percent
ug-at/L
ug-at/L
ug/L
m
Pearl
Benson
Index
Fecal3 No./lOO mL
Col i form
Bacteria
Depth (m)
0
10
30
0
10
30
0
10
30
0
10
30
0
10
30
0
10
30
0
10
30
--
0
10
30
0
10
30
No. Obs.
103
52
33
102
52
32
99
48
28
98
46
26
35
35
35
42
35
35
0
0
0
24
100
51
33
35
8
8
Standard
Mean Error
24.24
28.44
28.50
14.52
11.08
9.65
10.14
8.95
7.56
116.11
97.96
80.47
3.88
10.08
14.91
0.67
1.11
1.47
__
--
--
3.10
22.09b
4.40b
2.86b-
2.44b
1.30b
1.73b
0.42
0.21
0.41
0.25
0.23
0.22
0.17
0.23
0.26
1.94
2.71
2.69
0.94
1.19
1.12
0.08
0.09
0.11
__
--
--
0.31
1.16
1.17
1.25
1.25
1.13
1.44
Coeff.
Variation
(Percent)
17.7
5.5
8.3
17.2
14.7
12.6
16.5
17.6
18.0
16.5
18.8
17.0
144.2
69.8
44.6
74.8
49.0
46.3
__
--
--
49.8
11.7
38.2
69.6
88.4
81.3
92.3
3 The database contains values that are the actual analytical detection
limits for samples that did not contain detectable amounts of nitrate,
phosphate, or fecal coliform bacteria. Therefore, the means presented in
Appendix E may overestimate the actual means, particularly for depths and
locations where the value of the variable in question typically was at or
near the analytical detection limit.
b Geometric mean.
E-l
-------
TABLE E-2. PORT GARDNER STUDY AREA
Variable
Salinity
Water
Temperature
Dissolved
Oxygen
Dissolved
Oxygen
Saturation
Nitrate3
Phosphate3
Chloro-
phyll a
Secchi Depth
Sulfite
Waste
Liquor
Units Depth (m)
ppt
°C
mg/L
Percent
ug-at/L
ug-at/L
ug/L
m
Pearl
Benson
Index
Fecal3 No. /100 mL
Col i form
Bacteria
0
10
30
0
10
30
0
10
30
0
10
30
0
10
30
0
10
30
0
10
30
--
0
10
30
0
10
30
No. Obs.
85
47
22
85
47
22
85
47
22
85
47
22
24
24
0
43
41
17
0
0
0
18
63
25
3
25
9
0
Standard
Mean Error
19.77
28.00
29.13
11.93
9.64
8.64
10.07
8.87
7.46
106.33
94.12
76.93
5.98
18.23
--
0.68
1.63
2.28
--
--
--
2.74
17. 33?
6.44b
21.97b
22.87J>
2.19b
--
0.54
0.12
0.07
0.26
0.20
0.15
0.18
0.17
0.14
1.84
1.94
1.48
1.37
1.35
--
0.06
0.09
0.09
--
--
--
0.34
1.17
1.33
3.85
1.54
1.29
--
Coeff.
Variation
(Percent)
25.4
3.0
1.2
20.2
14.2
8.2
16.4
12.9
8.9
15.9
14.2
9.0
112.7
36.1
--
59.3
33.9
16.4
--
--
—
53.1
11.0
37.1
48.5
32.6
58.6
--
3 The database contains values that are the actual analytical detection
limits for samples that did not contain detectable amounts of nitrate,
phosphate, or fecal coliform bacteria. Therefore, the means presented in
Appendix E may overestimate the actual means, particularly for depths and
locations where the value of the variable in question typically was at or
near the analytical detection limit.
Geometric mean.
E-2
-------
TABLE E-3. POINT JEFFERSON STUDY AREA
Variable Units
Salinity ppt
Water °C
Temperature
Dissolved mg/L
Oxygen
Dissolved Percent
Oxygen
Saturation
Phosphate3 ug-at/L
Chloro- ug/L
phyll a
Seech i Depth m
Depth (m)
0
10
30
100
150
200
0
10
30
100
150
200
0
10
30
100
150
200
0
10
30
100
150
200
0
10
30
100
150
200
0
10
30
100
150
200
--
No. Obs.
394
224
201
73
58
38
544
239
219
79
60
36
406
238
217
82
64
37
406
238
216
82
63
37
29
21
19
6
6
7
184
22
1
0
0
0
495
Mean
27.31
28.35
29.13
29.79
29.92
30.12
12.26
11.39
10.38
9.86
9.83
9.57
10.97
9.73
8.26
7.48
7.32
7.08
119.44
103.89
86.93
76.82
75.08
74.00
1.41
1.70
2.08
2.03
2.08
2.11
5.59
1.64
0.36
--
--
--
4.69
Standard
Error
0.09
0.07
0.05
0.06
0.07
0.04
0.08
0.11
0.09
0.13
0.12
0.17
0.08
0.07
0.05
0.08
0.09
0.10
1.06
1.01
0.63
1.14
1.33
0.97
0.12
0.10
0.09
0.05
0.05
0.05
0.31
0.42
--
--
--
--
0.10
Coeff.
Variation
(Percent)
6.6
3.6
2.4
1.6
1.7
0.7
14.6
14.3
12.6
11.5
9:8
10.5
15.3
11.7
9.1
9.9
9.5
8.6
17.8
15.1
10.6
13.4
14.0
7.9
44.9
27.5
18.5
5.6
6.0
6.0
74.7
121.0
--
--
--
--
47.4
a The database contains values that are the actual analytical detection
limits for samples that did not contain detectable amounts of nitrate,
phosphate, or fecal coliform bacteria. Therefore, the means presented in
Appendix E may overestimate the actual means, particularly for depths and
locations where the value of the variable in question typically was at or
near the analytical detection limit.
-------
TABLE E-4. SINCLAIR INLET STUDY AREA
Variable
Salinity
Water
Temperature
Dissolved
Oxygen
Dissolved
Oxygen
Saturation
Nitrate3
Phosphate3
Chloro-
phyll a
Secchi Depth
Sulfite
Waste
Liquor
Units
ppt
°C
mg/L
Percent
ug-at/L
ug-at/L
ug/L
m
Pearl
Benson
Index
Fecal3 No./lOO mL
Col i form
Bacteria
Depth (m)
0
10
30
0
10
30
0
10
30
0
10
30
0
10
30
0
10
30
0
10
30
--
0
10
30
0
10
30
No. Obs.
43
36
0
44
37
0
40
34
0
38
32
0
38
38
0
38
38
0
0
0
0
28
11
6
0
39
6
0
Standard
Mean Error
27.96
28.40
--
14.54.
12.84
—
11.34
8.88
—
133.98
102.16
--
2.72
8.06
--
0.93
1.28
--
__
--
--
3.53
3.60b
2.71b
--
1.92b
3.52b
--
0.15
0.11
--
0.35
0.30
--
0.30
0.24
--
3.88
2.71
--
0.59
0.70
--
0.08
0.09
--
__
--
--
0.26
1.37
1.58
--
1.26
1.69
--
Coeff.
Variation
(Percent)
3.6
2.3
--
16.1
14.1
--
16.9
15.8
--
17.8
15.0
--
134.0
53.9
--
56.3
45.5
--
__
--
--
38.9
44.5
63.3
--
128.5
57.7
--
3 The database contains values that are the actual analytical detection
limits for samples that did not contain detectable amounts of nitrate,
phosphate, or fecal coliform bacteria. Therefore, the means presented in
Appendix E may overestimate the actual means, particularly for depths and
locations where the value of the variable in question, typically was at or
near the analytical detection limit.
Geometric mean.
E-4
-------
TABLE E-5. CITY WATERWAY STUDY AREA
Variable
Salinity
Water
Temperature
Dissolved
Oxygen
Dissolved
Oxygen
Saturation
Nitrate3
Phosphate3
Chloro-
phyll a
Secchi Depth
Sulfite
Waste
Liquor
Units
ppt
°C
mg/L
Percent
ug-at/L
ug-at/L
ug/L
m
Pearl
Benson
Index
Fecal3 No./lOO mL
Coli form
Bacteria
Depth (m)
0
10
30
0
10
30
0
10
30
0
10
30
0
10
30
0
10
30
0
10
30
--
0
10
30
0
10
30
No. Obs.
44
35
0
42
33
0
37
29
0
37
29
0
36
35
0
36
35
0
13
13
0
25
40
30
1
36
7
0
Standard
Mean Error
23.33
28.43
--
12.57
10.97
--
9.44
9.08
--
104.53
100.71
--
14.14
15.97
--
1.41
1.44
--
4.76
1.78
—
2.89
4.74*>
2'43u
5.00b
13.79b
1.57b
™ ~
0.44
0.09
--
0.27
0.26
--
0.30
0.25
--
3.42
2.75
--
0.93
0.86
--
0.11
0.12
--
1.57
0.41
—
0.25
1.21
1.19
--
1.34
1.19
™ ™
Coeff.
Variation
(Percent)
12.4
1.8
--
13.9
13.5
--
19.4
14.6
--
19.9
14.7
--
39.3
32.0
--
47.5
48.2
--
118.5
82.8
—
43.9
39.9
62.0
--
28.2
69.3
~ ™
3 The database contains values that are the actual analytical detection
limits for samples that did not contain detectable amounts of nitrate, phos-
phate, or fecal col i form bacteria. Therefore, the means presented in
Appendix E may overestimate the actual mdans, particularly for depths and
locations where the value of the variable in question typically was at or
near the analytical detection limit.
Geometric mean.
E-5
-------
TABLE E-6. CARR INLET STUDY AREA
Variable
Salinity
Water
Temperature
Dissolved
Oxygen
Dissolved
Oxygen
Saturation
Nitrate3
Phosphate3
Chloro-
phyll a
Secchi Depth
Sulfite
Waste
Liquor
Fecal3 No
Col i form
Bacteria
Units
ppt
°C
mg/L
Percent
ug-at/L
ug-at/L
ug/L
m
Pearl
Benson
Index
./100 mL
Depth (m)
0
10
30
0
10
30
0
10
30
0
10
30
0
10
30
0
10
30
0
10
30
--
0
10
30
0
10
30
No. Obs,
86
84
74
87
84
74
80
77
68
79
76
67
50
36
36
69
68
64
10
10
10
31
5
2
2
37
0
0
Standard
Mean Error
28
28
28
13
11
10
11
9
8
128
105
91
3
10
14
0
1
1
2
5
0
6
3
0
2
1
.58
.80
.84
.88
.54
.86
.14
.56
.41
.47
.33
.36
.61
.65
.63
.89
.36
.57
.20
.18
.99
.22
.62^
.00?
.24b
.11
__
--
0
0
0
0
0
0
0
0
0
2
2
1
0
0
0
0
0
0
0
1
0
0
1
2
1
.06
.05
.05
.28
.17
.18
.24
.20
.16
.63
.03
.59
.83
.75
.75
.05
.06
.06
.87
.51
.29
.50
.38
.24
.05
__
--
Coeff.
Variation
(Percent)
2
1
1
19
13
14
19
18
15
18
16
14
140
42
30
50
34
31
125
92
91
44
34
78
93
_
-
.0
.5
.5
.1
.3
.5
.4
.1
.3
.2
.8
.2
.0
.1
.6
.7
.5
.2
.6
.2
.8
.4
.6
.5
.1
_
-
3 The database contains values that are the actual analytical detection
limits for samples that did not contain detectable amounts of nitrate,
phosphate, or fecal coliform bacteria. Therefore, the means presented in
Appendix E may overestimate the actual means, particularly for depths and
locations where the value of the variable in question typically was at or
near the analytical detection limit.
Geometric mean.
E-6
-------
TABLE E-7. NISQUALLY REACH STUDY AREA
Variable
Salinity
Water
Temperature
Dissolved
Oxygen
Dissolved
Oxygen
Saturation
Nitrate3
Units
ppt
°C
mg/L
Percent
ug-at/L
Phosphate3 ug-at/L
Chloro-
phyll a
Secchi Depth
Sulfite
Waste
Liquor
ug/L
m
Pearl
Benson
Index
Fecal3 No./lOO mL
Col i form
Bacteria
Depth (m)
0
10
30
0
10
30
0
10
30
0
10
30
0
10
30
0
10
30
0
10
30
--
0
10
30
0
10
30
No. Obs.
84
57
19
85
57
18
78
51
19
77
50
18
37
36
0
61
51
13
0
0
0
31
5
2
0
37
0
0
Standard
Mean Error
26
28
29
12
11
10
9
8
8
105
100
87
10
12
1
1
1
6
2
2
1
.42
.67
.10
.64
.87
.94
.49
.97
.10
.61
.16
.11
.79
.89
--
.38
.50
.76
—
--
—
.51
.04b
.24b
--
.47b
--
--
0
0
0
0
0
0
0
0
0
1
1
1
0
0
0
0
0
0
1
2
1
.38
.07
.08
.22
.23
.35
.16
.14
.17
.85
.68
.50
.64
.77
--
.06
.06
.10
—
--
--
.48
.55
.24
--
.12
--
--
Coeff.
Variation
(Percent)
13
1
1
16
14
13
14
11
9
15
11
7
36
35
-
31
29
21
_
-
-
41
74
78
-
83
-
-
.2
.8
.2
.3
1
• V
.7
.5
.3
.1
.4
.9
.3
.3
.9
-
.1
.9
.4
_
-
-
.2
.6
.4
-
.1
-
-
3 The database contains values that are the actual analytical detection
limits for samples that did not contain detectable amounts of nitrate,
phosphate, or fecal coliform bacteria. Therefore, the means presented in
Appendix E may overestimate the actual means, particularly for depths and
locations where the value of the variable in question typically was at or
near the analytical detection limit.
Geometric mean.
E-7
-------
TABLE E-8. BUDD INLET STUDY AREA
Variable
Salinity
Water
Temperature
Dissolved
Oxygen
Dissolved
Oxygen
Saturation
Nitrate3
Phosphate9
Chloro-
phyll a
Secchi Depth
Sulfite
Waste
Liquor
Units
ppt
°C
mg/L
Percent
ug-at/L
ug-at/L
ug/L
m
Pearl
Benson
Index
Fecal3 No./lOO mL
Col i form
Bacteria
Depth (m)
0
10
30
0
10
30
0
10
30
0
10
30
0
10
30
0
10
30
0
10
30
--
0
10
30
0
10
30
No. Obs.
85
60
0
85
59
0
80
59
0
79
59
0
61
58
0
65
57
0
0
0
0
50
47
24
0
63
18
0
Standard
Mean Error
25.90
27.95
--
15.61
12.96
—
9.62
7.93
--
114.62
90.54
--
1.95
4.81
--
1.47
1.51
--
_ _
--
--
3.11
3.89^
2.83b
--
4.18b
2.88b
--
0.34
0.26
--
0.32
0.23
--
0.25
0.31
—
3.26
3.31
--
0.29
0.52
--
0.09
0.10
--
_ _
--
--
0.21
1.11
1.22
--
1.26
1.39
--
Coeff.
Variation
(Percent)
12.2
7.1
--
18.7
13.5
—
23.2
29.6
--
25.3
28.1
--
115.1
87.8
--
51.3
51.7
--
_ —
--
--
48.6
32.5
52.7
--
100.3
79.6
—
a The database contains values that are the actual analytical detection
limits for samples that did not contain detectable amounts of nitrate, phos-
phate, or fecal coliform bacteria. Therefore, the means presented in
Appendix E may overestimate the actual means, particularly for depths and
locations where the value of the variable in question typically was at or
near the analytical detection limit.
b Geometric mean.
E-8
-------
TABLE E-9. TOTTEN INLET STUDY AREA
Variable
Sal inity
Water
Temperature
Dissolved
Oxygen
Dissolved
Oxygen
Saturation
Nitrate3
Units
ppt
°C
mg/L
Percent
ug-at/L
Phosphate3 ug-at/L
Chloro-
phyll a
Secchi Depth
Sulfite
Waste
Liquor
Fecal3 No
Col i form
Bacteria
ug/L
m
Pearl
Benson
Index
./100 mL
Depth (m)
0
10
30
0
10
30
0
10
30
0
10
30
0
10
30
0
10
30
0
10
30
--
0
10
30
0
10
30
No. Obs.
80
43
0
79
41
0
64
40
0
63
39
0
31
31
0
47
40
0
0
0
0
43
41
7
0
33
0
0
Standard
Mean Error
27.91
28.03
--
15.45
14.47
--
10.03
10.15
--
120.16
119.23
--
1.77
2.53
--
1.31
1.37
--
__
--
--
4.26
2.20b
3i31b
--
1.04b
--
--
0.08
0.10
--
0.22
0.29
--
0.21
0.21
--
2.44
2.51
--
0.38
0.43
--
0.06
0.07
--
__
--
--
0.23
1.10
1.12
--
1.03
--
--
Coeff.
Variation
(Percent)
2.5
2.4
--
12.7
12.9
--
16.4
13.3
--
16.1
13.2
--
119.8
94.0
--
33.2
33.0
--
__
--
--
35.3
53.0
31.4
--
97.1
--
--
3 The database contains values that are the actual analytical detection
limits ,for samples that did not contain detectable amounts of nitrate,
phosphate, or fecal coliform bacteria. Therefore, the means presented in
Appendix E may overestimate the actual means, particularly for depths and
locations where the value of the variable in question typically was at or
near the analytical detection limit.
Geometric mean.
E-9
-------
TABLE E-10. OAKLAND BAY STUDY AREA
Variable
Salinity
Water
Temperature
Dissolved
Oxygen
Dissolved
Oxygen
Saturation
Nitrate3
Units Depth (m)
ppt
°C
mg/L
Percent
ug-at/L
Phosphate3 ug-at/L
Chloro-
phyll a
Secchi Depth
Sulfite
Waste
Liquor
ug/L
m
Pearl
Benson
Index
Fecal3 No./lOO mL
Col i form
Bacteria
0
10
30
0
10
30
0
10
30
0
10
30
0
10
30
0
10
30
0
10
30
--
0
10
30
0
10
30
No. Obs.
104
28
0
102
28
0
87
27
0
81
26
0
43
29
0
38
29
0
21
0
0
53
56
3
0
29
3
0
Standard
Mean Error
22
25
13
12
8
9
92
110
4
5
1
1
2
2
16
9
3
1
.13
.45
--
.55
.82
--
.38
.85
--
.44
.99
—
.62
.81
--
.19
.12
--
.31
--
--
.84
.56b
!32b
--
.10b
!59b
--
0
0
0
0
0
0
2
3
0
1
0
0
0
0
1
1
1
1
.42
.25
--
.27
.47
--
.25
.28
--
.97
.43
--
.66
.08
--
.07
.07
--
.17
--
--
.10
.25
.45
--
.25
.26
--
Coeff.
Variation
(Percent)
19.
5.
-
20.
19.
~
27.
14.
"•
28.
15.
-
93.
99.
-
36.
34.
-
34.
--
--
24.
19.
12.
--
61.
67.
--
3
2
-
4
3
~
7
7
—
9
7
-
0
9
-
8
0
-
5
6
5
8
5
5
3 The database contains values that are the actual analytical detection
limits for samples that did not contain detectable amounts of nitrate,
phosphate, or fecal coliform bacteria. Therefore, the means presented in
Appendix E may overestimate the actual means, particularly for depths and
locations where the value of the variable in question typically was at or
near the analytical detection limit.
b Geometric mean.
E-10
-------
TABLE E-ll. DABOB BAY STUDY AREA
__^ " - — — —
Variable
Salinity
Water
Temperature
Dissolved
Oxygen
Dissolved
Oxygen
Saturation
Nitrate3
Phosphate3
Chloro-
phyll a
Secchi Depth
Sulfite
Waste
Liquor
Fecal3 No
Coliform
Bacteria
Units
PPt
°C
mg/L
Percent
ug-at/L
ug-at/L
ug/L
m
Pearl
Benson
Index
./100 mL
Depth (m)
0
10
30
0
10
30
0
10
30
0
10
30
0
10
30
0
10
30
0
10
30
--
0
10
30
0
10
30
No. Obs.
66
61
61
65
60
60
64
58
58
63
58
58
32
33
33
48
49
49
24
24
23
28
6
2
2
34
2
2
Standard
Mean Error
26.63
28.43
29.25
14.03
10.86
9.26
10.45
10.04
7.22
119.96
109.85
76.51
1.92
5.69
20.46
0.70
1.19
1.94
2.36
4.12
1.33
6.01
1.26b
1.00b
2.24b
1.09b
2.'oOb
2.00b
0.19
0.10
0.08
0.37
0.22
0.13
0.20
0.23
0.19
2.19
2.74
2.09
0.60
1.16
1.15
0.05
0.09
0.09
0.57
0.84
0.32
0.36
1.16
—
2.24
1.04
--
--
Coeff.
Variation
(Percent)
5.7
2.7
2.2
21.3
15.6
11.0
15.2
17.1
20.2
14.5
19.0
20.8
176.2
117.4
32.2
53.8
54.9
32.2
119.4
99.5
116.4
31.9
83.9
—
78.5
94.2
--
--
3 The database contains values that are the actual analytical detection
limits for samples that did not contain detectable amounts of nitrate,
phosphate, or fecal coliform bacteria. Therefore, the means presented in
Appendix E may overestimate the actual means, particularly for depths and
locations where the value of the variable in question typically was at or
near the analytical detection limit.
b Geometric mean.
E-ll
-------
TABLE E-12. MID-HOOD CANAL STUDY AREA
Variable
Salinity
Water
Temperature
Dissolved
Oxygen
Dissolved
Oxygen
Saturation
Nitrate3
Phosphate3
Chloro-
phyll a
Secchi Depth
Sulfite
Waste
Liquor
Fecal3 No
Col i form
Bacteria
Units
ppt
°C
mg/L
Percent
ug-at/L
ug-at/L
ug/L
m
Pearl
Benson
Index
./100 mL
Depth (m)
0
10
30
0
10
30
0
10
30
0
10
30
0
10
30
0
10
30
0
10
30
--
0
10
30
0
10
30
No. Obs,
55
52
52
55
52
52
52
49
49
52
49
49
29
29
29
41
41
39
1
1
1
25
5
3
3
29
0
0
Standard
Mean Error
25.03.
28.30
29.04
14.94
11.22
9.57
10.06
9.27
6.82
116.47
102.02
72.69
1.43
6.55
22.28
0.73
1.27
2.01
2.70
3.10
1.00
6.01
1.90b
1.71b
0.00b
1.14b
__
--
0.30
0.13
0.15
0.35
0.21
0.15
0.18
0.24
0.22
1.78
2.93
2.46
0.33
1.32
1.24
0.07
0.10
0.10
_ _
--
--
0.27
1.48
1,71
0.00
1.06
__
--
Coeff.
Variation
(Percent)
9.0
3.4
3.7
17.6
13.8
11.5
12.8
17.9
23.0
11.2
20.1
23.7
125.4
108.9
30.0
56.9
48.6
30.3
_ _
--
--
22.5
73.6
85.1
0.0
91.1
__
"
3 The database contains values that are the actual analytical detection
limits for samples that did not contain detectable amounts of nitrate,
phosphate, or fecal col i form bacteria. Therefore, the means presented in
Appendix E may overestimate the actual means, particularly for depths and
locations where the value of the variable in question typically was at or
near the analytical detection limit.
Geometric mean.
E-12
-------
TABLE E-13. SOUTH HOOD CANAL STUDY AREA
Variable
Salinity
Water
Temperature
Dissolved
Oxygen
Dissolved
Oxygen
Saturation
Nitrate3
Phosphate3
Chloro-
phyll a
Seech i Depth
Sulfite
Waste
Liquor
Fecal3 No
Col i form
Bacteria
Units
ppt
°C
mg/L
Percent
ug-at/L
ug-at/L
ug/L
m
Pearl
Benson
Index
./100 mL
Depth (m)
0
10
30
0
10
30
0
10
30
0
10
30
0
10
30
0
10
30
0
10
30
--
0
10
30
0
10
30
No. Obs.
84
74
59
82
74
59
81
71
57
79
71
57
40
40
40
66
62
54
0
0
0
33
11
7
7
40
3
3
Standard
Mean Error
23.56
28.32
29.19
14.38
10.44
9.46
10.14
7.39
4.31
115.21
80.31
45.93
1.75
12.97
26.98
0.84
2.12
2.90
—
--
--
4.79
2.30J
3.43°
2.48b
1.48b
2. 00b
2.00b
0.23
0.12
0.07
0.35
0.19
0.14
0.14
0.31
0.22
1.34
3.53
2.32
0.74
1.53
0.85
0.06
0.11
0.11
-_
--
—
0.28
1.36
1.39
1.60
1.12
--
"• ••
Coeff.
Variation
(Percent)
8.8
3.5
1.8
21.9
15.7
11.0
12.2
35.1
38.5
10.3
37.1
38.1
266.6
74.7
20.0
56.9
43.6
27.6
—
--
—
32.9
68.8
40.5
78.8
84.2
..
— —
3 The database contains values that are the actual analytical detection
limits for samples that did not contain detectable amounts of nitrate,
phosphate, or fecal coliform bacteria. Therefore, the means presented in
Appendix E may overestimate the actual means, particularly for depths and
locations where the value of the variable in question typically was at or
near the analytical detection limit.
b Geometric mean.
E-13
-------
APPENDIX F
SUMMARY OF CORRELATION COEFFICIENTS BETWEEN
WATER QUALITY VARIABLES
-------
TABLE F-l. PEARSON PRODUCT-MOMENT CORRELATION COEEFFICIENTS
BETWEEN WATER QUALITY VAIRABLES IN THE BELLINGHAM BAY STUDY AREA
Depth: 0 m
Water Dissolved
Temp. Oxygen
Salinity -0.28a 0.31
Water Temp. -0.31
Diss. Oxygen
Nitrate
Phosphate
Dissolved Oxygen Saturation
Nitrate
nsb
ns
ns
Phosphate
ns
ns
ns
0.52
Diss.
Oxy.
Sat.
0.38
ns
0.96
ns
ns
Secchi
Disk
Depth
ns
ns
ns
ns
0.61
ns
Depth: 10 m
Water
Temp.
Dissolved
Oxygen
Nitrate
Phosphate
Salinity
Water Temp.
Diss. Oxygen
Nitrate
ns
ns
ns
ns
-0.71
ns
ns
ns
ns
ns
Depth: 30 m
Salinity
Water Temp.
Diss. Oxygen
Nitrate
Water
Temp.
ns
Dissolved
Oxygen
ns
ns
Nitrate
0.62
ns
ns
Phosphate
ns
0.55
ns
ns
a Numerical table entries are statistically significant (P<0.05 scaled with
the Bonferroni inequality) correlation coefficients.
b ns = Not statistically significant (P>0.05 scaled with the Bonferroni
inequality).
F-l
-------
TABLE F-2. PEARSON PRODUCT-MOMENT CORRELATION COEFFICIENTS BETWEEN
WATER QUALITY VARIABLES IN THE PORT GARDNER STUDY AREA
Depth: 0 m
Water Dissolved
Temp. Oxygen
Salinity -0.53a nsb
Water Temp. ns
Diss. Oxygen
Nitrate
Phosphate
Dissolved Oxygen Saturation
Nitrate
ns
-0.60
ns
Phosphate
ns
-0.47
ns
0.53
Diss.
Oxy.
Sat.
ns
ns
0.96
ns
ns
Secchi
Disk
Depth
ns
ns
ns
ns
0.68
ns
Water
Temp.
Depth: 10 m
Dissolved
Oxygen
Nitrate Phosphate
Salinity
Water Temp.
Diss. Oxygen
Nitrate
ns
ns
ns
ns
-0.60
ns
0.49
ns
ns
ns
Water
Temp.
Depth: 30 m
Dissolved
Oxygen
Phosphate
Salinity
Water Temp.
Diss. Oxygen
ns
ns
ns
ns
ns
ns
a Numerical table entries are statistically significant (P<0.05 scaled with
the Bonferroni inequality) correlation coefficients.
b ns = Not statistically significant (P>0.05 scaled with the Bonferroni
inequality).
F-2
-------
TABLE F-3. PEARSON PRODUCT-MOMENT CORRELATION COEFFICIENTS BETWEEN
WATER QUALITY VARIABLES IN THE POINT JEFFERSON STUDY AREA
Water Dissolved
Temp. Oxygen
Salinity -0.33a -0.30
Water Temp.
Diss. Oxygen
Phosphate
Dissolved Oxygen
Chlor. a
0.15
Saturation
Depth: 0 m
Phosphate
0.60
-0.59
-0.78
Diss.
Oxy.
Sat.
-0.28
0.35
0.92
-0.81
Chlor. a
nsb
ns
0.54
ns
0.50
Secchi
Disk
Depth
0.20
ns
-0.59
ns
-0.59
-0.58
Depth: 10 m
Salinity
Water Temp.
Diss. Oxygen
Water
Temp.
-0.24
Dissolved
Oxygen
-0.39
ns
Phosphate
ns
ns
-0.63
Depth: 30 m
Salinity
Water Temp.
Diss. Oxygen
Water
Temp.
ns
Dissolved
Oxygen
-0.20
-0.39
Phosphate
ns
ns
ns
Depth: 100 m
Salinity
Water Temp.
Water
Temp.
ns
Dissolved
Oxygen
ns
-0.52
F-3
-------
TABLE F-3. (Continued)
Depth: 150 m
Salinity
Water Temp.
Water
Temp.
ns
Dissolved
Oxygen
-0.40
-0.38
Salinity
Water Temp.
Water
Temp.
Depth: 200 m
Dissolved
Oxygen
ns
ns
-0.56
a Numerical table entries are statistically significant (P<0.05 scaled with
the Bonferroni inequality) correlation coefficients.
b ns = Not statistically significant (P>0.05 scaled with the Bonferroni
inequality).
F-4
-------
TABLE F-4. PEARSON PRODUCT-MOMENT CORRELATION COEEFICIENTS BETWEEN
WATER QUALITY VARIABLES IN THE SINCLAIR INLET STUDY AREA
Water
Temp.
Salinity nsa
Water Temp.
Diss. Oxygen
Nitrate
Phosphate
Dissolved
Oxygen
ns
ns
Depth:
Nitrate
ns
ns
-0.46b
0 m
Phosphate
ns
ns
ns
ns
Diss.
Oxy.
Sat.
ns
ns
0.96
-0.61
ns
Dissolved Oxygen Saturation
Secchi
Disk
Depth
ns
ns
ns
ns
ns
ns
Depth: 10 m
Water Dissolved
Temp. Oxygen Nitrate Phosphate
Salinity ns
Water Temp.
Diss. Oxygen
Nitrate
ns
ns
ns
-0.47
ns
ns
ns
ns
ns
a ns = Not statistically significant (P>0.05 scaled with the Bonferroni
inequality).
b Numerical table entries are statistically significant (P<0.05 scaled with
the Bonferroni inequality) correlation coefficients.
F-5
-------
TABLE F-5. PEARSON PRODUCT-MOMENT CORRELATION COEFFICIENTS BETWEEN
WATER QUALITY VARIABLES IN THE CITY WATERWAY STUDY AREA
Depth: 0 m
Water Dissolved
Temp. Oxygen Nitrate Phosphate
Diss. Secchi
Oxy. Disk
Sat. Chi or. a Depth
Salinity
Water Temp.
Diss. Oxygen
Nitrate
Phosphate
-0
Dissolved Oxygen
Chlor. a
.40a
nsb
ns
ns
-0.45
ns
ns
ns
ns
ns
ns
ns
0.98
ns
ns
Saturation
ns
ns
ns
ns
ns
ns
0.52
-0.63
ns
ns
ns
ns
ns
Depth: 10 m
Salinity
Water Temp.
Diss. Oxygen
Nitrate
Phosphate
Water
Temp .
ns
Dissolved
Oxvaen
ns
ns
Nitrate
ns
ns
ns
Phosphate
0.46
ns
ns
ns
Chlor. a
ns
ns
ns
ns
ns
a Numerical table entries are statistically significant (P<0.05 scaled with
the Bonferroni inequality) correlation coefficients.
b ns = Not statistically significant (P>0.05 scaled with the Bonferroni
inequality).
F-6
-------
TABLE F-6. PEARSON PRODUCT-MOMENT CORRELATION COEFFICIENTS BETWEEN
WATER QUALITY VARIABLES IN THE CARR INLET STUDY AREA
Water Dissolved
Temp. Oxygen
Salinity nsa ns
Water Temp. ns
Diss. Oxygen
Nitrate
Phosphate
Dissolved Oxygen Saturation
Chlor. a
Water Dissolved
Temo. Oxvaen
Salinity 0.36 -0.56
Water Temp. -0.48
Diss. Oxygen
Nitrate
Phosphate
Depth:
Nitrate
ns
ns
ns
Depth:
Nitrate
ns
ns
ns
0 m
Phosphate
0.49b
ns
-0.49
ns
10 m
Phosphate
ns
ns
ns
ns
Diss.
Oxy.
Sat. Chlor. a
ns ns
ns ns
0.96 ns
-0.53 ns
-0.56 ns
ns
Chlor. a
ns
ns
ns
ns
ns
Secchi
Disk
Depth
ns
ns
ns
ns
ns
ns
ns
Depth: 30 m
Salinity
Water Temp.
Diss. Oxygen
Nitrate
Phosphate
Water
Temp.
ns
Dissolved
Oxvaen
-0.55
-0.52
Nitrate
ns
ns
ns
Phosphate
0.45
ns
ns
ns
Chlor. a
ns
ns
ns
ns
ns
a ns = Not statistically significant (P>0.05 scaled with the Bonferroni
inequality).
b Numerical table entries are statistically significant (P<0.05 scaled with
the Bonferroni inequality) correlation coefficients.
F-7
-------
TABLE F-7. PEARSON PRODUCT-MOMENT CORRELATION COEEFICIENTS BETWEEN
WATER QUALITY VARIABLES IN THE NISQUALLY REACH STUDY AREA
Water
Temp.
Salinity
Water Temp.
Diss. Oxygen
Nitrate
Phosphate
Dissolved Oxygen
nsa
Dissolved
Oxygen
ns
ns
Depth:
Nitrate
ns
-0.44b
ns
0 m
Phosphate
0.38
ns
ns
ns
Diss.
Oxy.
Sat.
ns
0.31
0.93
ns
ns
Saturation
Secchi
Disk
Depth
0.50
ns
ns
ns
ns
ns
Water
Temp.
Dissolved
Oxygen
Depth: 10 m
Nitrate Phosphate
Salinity
Water Temp.
Diss. Oxygen
Nitrate
ns
ns
ns
ns
ns
ns
ns
ns
ns
ns
Depth: 30 m
Water
Temp.
Salinity 0.61
Water Temp.
Diss. Oxygen
Nitrate
Dissolved
Oxygen
ns
-0.61
Nitrate
ns
ns
ns
Phosphate
ns
ns
ns
ns
a ns = Not statistically significant (P>0.05 scaled with the Bonferroni
inequality).
b Numerical table entries are statistically significant (P<0.05 sclaed with
the Bonferroni inequality) correlation coefficients.
F-8
-------
TABLE F-8. PEARSON PRODUCT-MOMENT CORRELATION COEEFICIENTS BETWEEN
WATER QUALITY VARIABLES IN THE BUDD INLET STUDY AREA
Water
Temp.
Salinity nsa
Water Temp.
Diss. Oxygen
Nitrate
Phosphate
Dissolved
Oxygen
ns
ns
Depth:
Nitrate
ns
-0.38b
ns
0 m
Phosphate
ns
ns
ns
ns
Diss.
Oxy.
Sat.
ns
0.31
0.97
ns
ns
Dissolved Oxygen Saturation
Secchi
Disk
Depth
ns
ns
ns
0.47
ns
ns
Depth: 10 m
Water
Temp.
Salinity ns
Water Temp.
Diss. Oxygen
Nitrate
Dissolved
Oxygen
ns
-0.51
Nitrate
ns
-0.48
0.39
Phosphate
ns
ns
ns
ns
a ns = Not statistically significant (P>0.05 scaled with the Bonferroni
inequality).
b Numerical table entries are statistically significant (P<0.05 scaled with
the Bonferroni inequality) correlation coefficients.
F-9
-------
TABLE F-9. PEARSON PRODUCT-MOMENT CORRELATION COEFFICIENTS BETWEEN
WATER QUALITY VARIABLES IN THE TOTTEN INLET STUDY AREA
Salinity
Water Temp.
Diss. Oxygen
Nitrate
Phosphate
Water
Temp.
0.57a
Dissolved
Oxygen
nsb
ns
Depth:
Nitrate
ns
ns
ns
0 m
Phosphate
0.47
ns
ns
ns
Diss.
Oxy.
Sat.
ns
ns
0.96
ns
ns
Dissolved Oxygen Saturation
Secchi
Disk
Depth
ns
ns
ns
ns
ns
ns
Depth: 10 m
Salinity
Water Temp.
Diss. Oxygen
Nitrate
Water
Temp.
0.77
Dissolved
Oxygen
ns
ns
Nitrate
ns
ns
ns
Phosphate
0.42
0.43
ns
ns
a Numerical table entries are statistically significant (P<0.05 scaled with
the Bonferroni inequality) correlation coefficients.
b ns = Not statistically significant (P>0.05 scaled with the Bonferroni
inequality).
F-10
-------
TABLE F-10. PEARSON PRODUCT-MOMENT CORRELATION COEFFICIENTS BETWEEN
WATER QUALITY VARIABLES IN THE OAKLAND BAY STUDY AREA
Water
Temp.
Sal ini ty
Water Temp.
Diss. Oxygen
Nitrate
Phosphate
Dissolved Oxygen
nsa
Dissolved
Oxygen
ns
-0.33b
Depth:
Nitrate
ns
ns
ns
0 m
Phosphate
ns
ns
ns
ns
Diss.
Oxy.
Sat.
ns
ns
0.97
ns
ns
Saturation
Secchi
Disk
Depth'
ns
ns
ns
ns
0.68
ns
Water
Temp.
Dissolved
Oxygen
Depth: 10 m
Nitrate Phosphate
Salinity 0.72
Water Temp.
Diss, Oxygen
Nitrate
ns
ns
ns
ns
ns
ns
ns
ns
ns
a ns = Not statistically significant (P>0.05 scaled with the Bonferroni
inequality).
b Numerical table entries are statistically significant (P<0.05 scaled with
the Bonferroni inequality) correlation coefficients.
F-ll
-------
TABLE F-ll. PEARSON PRODUCT-MOMENT CORRELATION COEFFICIENTS BETWEEN
WATER QUALITY VARIABLES IN THE DABOB BAY STUDY AREA
Sal ini ty
Water Temp.
Diss. Oxygen
Nitrate
Phosphate
Water
Temp.
-0.383
Dissolved
Oxygen
nsb
-0.38
Depth: 0 m
Nitrate Phosphate
ns ns
ns ns
ns ns
0.69
Diss.
Oxy.
Sat.
ns
ns
0.92
ns
ns
Dissolved Oxygen Saturation
Chlor. a
Chlor.
ns
ns
0.67
ns
ns
0.57
Secchi
Disk
a Depth
ns
ns
-0.53
ns
ns
ns
-0.53
Depth: 10 m
Salinity
Water Temp.
Diss. Oxygen
Nitrate
Phosphate
Water
Temp.
-0.34
Dissolved
Oxygen
-0.39
ns
Nitrate
ns
-0.49
-0.60
Phosphate
ns
ns
-0.69
0.54
Chlor.
ns
ns
ns
ns
ns
a
Depth: 30 m
Salinity
Water Temp.
Diss. Oxygen
Nitrate
Phosphate
Water
Temp.
-0.34
Dissolved
Oxygen
ns
ns
Nitrate
ns
ns
-0.46
Phosphate
ns
ns
-0.47
0.44
Chlor. a
ns
ns
ns
ns
ns
a Numerical table entries are statistically significant (P<0.05 scaled with
the Bonferroni inequality) correlation coefficients. ;
b ns = Not statistically significant (P>0.05 scaled with the Bonferroni
inequality).
F-12
-------
TABLE F-12. PEARSON PRODUCT-MOMENT CORRELATION COEFFICIENTS BETWEEN
WATER QUALITY VARIABLES IN THE MID-HOOD CANAL STUDY AREA
Water
Temp.
Salinity
Water Temp.
Diss. Oxygen
Nitrate
Phosphate
Dissolved Oxygen
nsa
Dissolved
Oxygen
ns
-0.56b
Depth:
Nitrate
ns
-0.50
ns
0 m
Phosphate
ns
ns
ns
ns
Diss.
Oxy.
Sat.
ns
ns
0.92
ns
ns
Saturation
Secchi
Disk
Depth
ns
0.55
ns
ns
ns
ns
Water
Temp.
Dissolved
Oxygen
Depth: 10 m
Nitrate Phosphate
Salinity
Water Temp.
Diss. Oxygen
Nitrate
ns
-0.57
0.52
ns
-0.70
-0.76
0.41
-0.52
-0.49
0.52
Depth: 30 m
Water
Temp.
Salinity -0.45
Water Temp.
Diss. Oxygen
Nitrate
Dissolved
Oxygen
-0.54
ns
Nitrate
0.61
ns
-0.48
Phosphate
ns
ns
ns
0.50
a ns = Not statistically significant (P>0.05 scaled with the Bonferroni
inequality).
b Numerical table entries are statistically significant (P<0.05 sclaed with
the Bonferroni inequality) correlation coefficients.
F-13
-------
TABLE F-13. PEARSON PRODUCT-MOMENT CORRELATION COEFFICIENTS
BETWEEN WATER QUALITY VARIABLES IN THE SOUTH HOOD CANAL STUDY AREA
Salinity
Water Temp.
Diss. Oxygen
Nitrate
Phosphate
Water
Temp.
nsa
Dissolved
Oxygen
ns
-0.50b
Depth:
Nitrate
ns
ns
ns
0 m
Phosphate
ns
ns
ns
ns
Diss.
Oxy,
Sat,
ns
ns
0.85
ns
n$
Dissolved Oxygen Saturation
Secchi
Disk
Depth
ns
ns
ns
ns
ns
ns
Depth: 10 m
Salinity
Water Temp.
Diss. Oxygen
Nitrate
Water
Temp.
-0.44
Dissolved
Oxygen
-0.53
0.45
Nitrate
0.41
-0.52
-0.62
Phosphate
0.49
ns
-0.47
0.62
Water
Temp.
Dissolved
Oxygen
Depth: 30 m
Nitrate Phosphate
Salinity ns
Water Temp.
Diss. Oxygen
Nitrate
ns
-0.33
ns
ns
ns
ns
ns
-0.39
ns
a ns = Not statistically significant (P>0.05 scaled with the Bonferroni
inequality).
b Numerical table entries are statistically significant (P<0.05 sclaed with
the Bonferroni inequality) correlation coefficients.
F-14
* U.S. GOVERNMENT PRINTING OFFICE 1988; 523-2*3/00453
-------
|