AN ABSTRACT OF THE THESIS OF
Ching-Ling Wei
in
Oceanography
for the degree of
presented
Master of Science
18 April, 1985
Title: Carbonate Chemistry of the North Pacific Ocean
Abstract approved:
Redacted for privacy
Dr. Ricardo M. Pytkowicz
Redacted for privacy
Dr. Chen-Tung K. Chen
This thesis covers a wide area of the North Pacific
Ocean, latitudinally from 100 to 55°N and longitudinally
from Japan to the U.S. coast. With two longitudinal cruises
as the main data sources and selected literature data sets
as supplementary data source, the distribution of physical
and chemical properties in the North Pacific Ocean is
presented.
AOU, pH, total CO2, and nutrients (nitrate and
phosphate) are interrelated parameters. Along our two
longitudinal sections they all show a core structure
underlying the salinity minimum layer. From these oxidation
related parameters we obtain a conclusion that the
subsurface water of the eastern North Pacific Ocean is older
than that of the western North Pacific Ocean. Alkalinity,
calcium, and silicate show a monotonically increasing trend
with depth to the deepest sampling depth.
The alkalinity data can be used as a water mass tracer.
Different water masses which show their own mixing trends
can be identified when the correlation of normalinzed
alkalinity with temperature is scrutinized. The vertical
distribution of the normalized alkalinity shows a maximum
core at a depth of about 2500 m in the .North Pacific Ocean.
Not only the calcium carbonate dissolution but also the
circulation in the deep and bottom layer plays a role in
this normalized alkalinity maximum core.
Our analysis of carbonate data shows that about 257. of
the increase in total inorganic CO2 of deep water, after
leaving from the Southern Ocean to the North Pacific, is
contributed by inorganic CaCO3 dissolution. There is no
significant difference of inorganic carbon/organic carbon
ratio between our two sections. However, it was found that
the eastern section has a higher total TCO2 input than that
of the western section.
The degree of saturation with respect to calcite and
aragonite was calculated from all the available data sets.
Mr. Ahmed Rushdi and Professor R.M. Pytkowicz are working on
the effect of Mg on the formation and the properties of
magnesian calcites. There are inetastable forms of these
compounds which may have been interpreted as stable ones in
solubility determinations. Some revision of degree of
saturation may be called for in the future.
Four selected crosssections, three longitudinal and
one latitudinal, show that a large volume of the North
Pacific is undersaturated with respect to CaCO3. The
saturation horizon generally shows a shoaling from the west
to the east and from the south to the north in the North
Pacific Ocean. It was found that lysocline falls on a depth
much deeper (about 2500 m deeper) than the saturation
horizon of calcite and several hundred meters shallower than
the CCD (calcium carbonate compensation depth) depth. Our
results support the kinetic point of view on the CaCO3
dissolution mechanisms.
The direct approach on the fossil fuel increase signal
in seawater is adopted in this thesis. A new set of
preformed equations of alkalinity and total CO2 was obtained
from a more updated GEOSECS data. Our results show that the
penetration depth of fossil fuel CO2 is strongly related to
the surface oceanographic circulation. The shallowest
penetration depth is less than 300 m found in the eastern
equatorial region where the upwelling prevails and the
deepest penetration depth is deeper than 2000 m off Japan
where an interaction of Oyashio and Kuroshio currents is
found.
Carbonate Chemistry of the North Pacific Ocean
by
ChingLing Wei
A THESIS
submitted to
Oregon State University
in partial fulfillment of
the requirements for the
degree of
Master of Science
Completed April 18, 1985
Commencement June 1986
APPROVED:
Redacted for privacy
Professor of Oceanography in Charge of Major
Redacted for privacy
Asso. Prof. of ct1ceanography in Charge of Major
Redacted for privacy
Dean
f the Colie
Oceanography
Redacted for privacy
Dean of t
Graduate r'chool
Date thesis is presented
April 18, 1985
Typed by Ching-Ling Wei for Ching-Ling Wei
ACKNOWLEDGMENTS
I have been very fortunate through these years at OSU
in having so many kind and helpful friends. I would like to
show these people my sincere appreciation by acknowledging
them here. Prof. R. M. Pytkowicz has always been a respected
advisor.
He not only gives valuable suggestions but also
offers tremendous understanding. Without his spiritual
support and encouragement I would not have been able to
complete this thesis.
I also acknowledge Dr. Arthur Chen's advise in my
research work for the first two years as a Masters student
and his pH and TA measurements and collection of Ca samples
on the Miller Freeman 1981 cruise. E. Olson's pH and TA
measurements on the Discoverer 1982 cruise is warmly
appreciated. Permission from Dr. R. Feely and Dr. J. Cline
for Dr. Chen and E. Olson to participate in the Miller
Freeman and Discoverer cruises and their providing
hydrographic data plus the calcium data from the Discoverer
cruise are acknowledged.
I warmly appreciate my discussions with Celeste Correia
and her corrections of my English throughout this thesis.
Celeste's encouragement and friendship are extremely
important to me. Sincere thanks is extended to Dr. Michael
R. Rodman who wrote some of the sections of the hydrography.
I also owe many thanks to Michael for his help in all
aspects. Dr. R. Collier's comment on this thesis is deeply
appreciated.
My appreciatioft is also extended to the faculty members
in Chemical Oceanography who inspired my interest in
oceanography. I would like to say thank you to my fellow
students - Marta, Jim, Orese, Harry, Laurel, and Ahmed - who
went through the highly demanding classes with me and gave
me indelible memories.
Without my parents' support I would not have been able
to come here. My wife, Cheryl, deserves my deepest gratitude
for her love and patience.
I acknowledge the financial support of the U.S
Department of Energy (81 EV1O611 and 19X-89608C) and
National Science Foundation (OCE 80-18770 and 82-15053)
which allowed me to accomplish this study.
TABLE OF CONTENTS
Page
1.
Introduction
2.
Data sources
3.
4.
A.
Outline of main data sources
B.
Methods of data analysis
C.
Literature data sources
D.
Abbreviations
1
5
Hydrography
A.
Background
B.
Inference from this work
Results and discussions
11
32
Chemistry data along 150°W and 165°!
32
(a) AOU and pH
32
(b) Alkalinity data
50
(c) Total CO2
62
(d) Calcium data
65
(e) Nutrients data
73
B.
Calciumalkalinity relationship
81
C.
Inorganic carbon/organic carbon ratio
88
D.
Degree of CaCO3 Saturation, lysocline, and CCD
95
(a) Method of calculation of degree of saturation
95
(b) Degree of saturation of the surface water
96
A.
(c) Vertical distribution of degree of saturation
102
(d) Relationship among C, lysocline, and CCD
111
E. Anthropogenic fossil fuel CO2
118
(a) Method of calculation
118
(b) Preformed equations
118
(c) Penetration depth along 165°E and along 150°W
124
(d) Penetration pattern in the North Pacific Ocean
132
5. Conclusions
135
6. References
138
LIST OF FIGURES
Page
1
Locations of the stations where chemistry data
are available. The solid lines connecting the
stations indicate the cross-sections discussed
in the text. The abbreviations used for the
expeditions are - ENP(Miller Freeman), WNP
(Discoverer), GS (GEOSECS), IP (INDOPAC).
2
Schematic diagram of water masses distribution
of the North Pacific Ocean.
3
12
Distribution of salinity with depth along the
165°E cross-section (Data from Cline, 1982).
5
10
Distribution of salinity with depth along the
150°W cross-section (Data from Feely, 1981).
4
4
13
Distribution of potential temperature with
depth along the 15O°W cross-section (Data from
Feely, 1981).
6
17
Distribution of temperature with depth along
the 165°E cross-section (Data from Cline,
18
1982).
7
Distribution of potential density (6) with
depth along the 150°W cross-section (Data from
Feely, 1981).
8
Distribution of
21
with depth along the 165°E
cross-section (Data from Cline, 1982).
22
9
8-s correlatidn for ENP and selected GS data
(Data from Feely,. 1981) and related GS data.
10
Page
24
Expand S-S correlation for ENP data (Data from
Feely, 1981).
25
11
9-S correlation for selected GS data.
26
12
8-S correlation for WNP data (Data from Cline,
27
1982).
13
Expanded 0-S correlation for WNP (Data from
Cline, 1982) and selected KH75-4 data.
14
28
Distribution of apparent oxygen utilization
(A0U) with depth along the 150°W cross-section
(Data from Feely, 1981).
15
Distribution of AOU with d9 along the 150°W
cross-section (Data from Feely, 1981).
16
Distribution of AOU with depth along the 165°E
cross-section (Data from Cline, 1982).
17
Distribution of AOU with
6
41
Correlation of S with pH(25°C) for the ENP
stations.
21
39
Distribution of p11(25°C) with depth along the
165°E cross-section.
20
38
Distribution of p11(25°C) with depth along the
150°W cross-section.
19
37
along the 165°E
cross-section (Data from Cline, 1982).
18
35
Correlation of B with p11(25°C) for the WNP
42
Page
stations.
22
43
Correlation of S with p11(25°C) for the ENP
stations.
23
45
Correlation of S with p11(25°C) for the WNP
stations.
24
46
Correlation of AOU with pH(25°C) for the ENP
stations.
25
48
Correlation of AOIJ with p11(25°C) for the
WNP
stations.
26
49
Distribution of NTA with depth along the 150°W
cross-section.
27
51
Distribution of NTA with depth along the 165°E
cross-section.
28
53
Correlation of 9 with NTA for ENP and selected
CS stations. The triangles represent surface
data and the squares represent salinity
minimum data.
29
Correlation of 8 with NTA for
WNP
and selected
CS stations. The triangles represent surface
data and the squares represent salinity
minimum data.
55
30
Correlation of S with NTA for ENP stations.
60
31
Correlation of S with NTA for
61
32
Distribution of NTCO2 with depth along the
15o°w cross-section.
WNP
stations.
63
Page
Liz.
33
Distribution of NTCO2 with depth along the
64
1 65°E cross-section.
34
Distribution of NCa with depth along the 150°W
66
cross-section.
35
Correlation of S with NCa for ENP stations.
36
Correlation of 9 with NCa for ENP stations.
68
The triangles represent surface data and
squares represent salinity minimum data
37
Correlation of S with NCa for WNP stations
71
(Data from Peely, 1982).
38
69
Correlation of 8 with NCa for WNP stations.
The triangles represent àurface data and
squares represent salinity minimum data (Data
72
from Feely, 1982).
39
Distribution of NO3 with depth along the 150°W
cross-section (Data from Feely, 1981).
40
74
Distribution of PO4 with depth along the 150°W
cross-section (Data from Feely, 1981).
41
Distribution of Si02 with depth along the
150°W cross-section (Data from Feely, 1981).
42
Distribution of NO3 with depth along the 165°E
cross-section (Data from Cline,
43
1982).
78
Distribution of PO4 with depth along the 165°E
cross-section (Data from Cline, 1982).
44
76
Distribution of Si02 with depth along the
79
Page
165°E cross-section (Data from Cline, 1982).
45
80
Correlation of NCa with NTA for the ENP data.
The regression line shown in the figure is
NCA8996+0.52OxNTA for the samples below 200
83
m.
46
Correlation of NCa with
NPTA1[-(TA+NO3+PO4)x35/S] for the ENP data.
The regression line shown in the figure is
NCa-9167+0.442xNPTA1 for the samples below 200
85
47
Correlation of NCa with
NPTA2[-(TA+17xAOU/138)x35/S] for the EN? data.
The regression line shown in the figure is
NCa9137+0.456xNPTA2 for the samples below 20.0
87
m.
48
Vertical Distribution of inorganic
carbon/organic carbon (IC/OC) ratio for EN?
90
(a) and WNP (b) data.
49
Correlation of IC with OC for the ENP data.
Dashed lines represent the total input of
92
TCO2 (IC+0C.).
50
Correlation of IC with OC for the WNP data.
Dashed lines represent the total input of
TCO2 (-IC+oC).
51
Correlation of surface
with temperature
Pag
for all the available data. The line is a
97
rough fit by eyes.
52
Correlation of surface
with temperature
a
for all the available data. The line is a
98
rough fit by eyes.
53
in the North
Distribution of surface
100
Pacific Ocean.
54
Distribution of surface CTa in the North
101
Pacific Ocean.
55
Vertical distribution of
c
along the 35°N
cross-section. Dashed line represents CCD from
103
Berger et al. (1976).
56
Vertical distribution of
a
along the 35°N
104
cross-section.
57
along the 165°E
Vertical distribution of
106
cros s-section
58
Vertical distribution of
1a along the 165°E
107
cross-section.
59
Vertical distribution of Q
along the 180°
cross-section. Dashed line represents CCD from
109
Berger et al. (1976).
60
Vertical distribution of
a
along the 1800
110
cross-section.
61
Vertical distribution of
cross-section.
C).
along the 150°W
112
Page
62
Vertical distribution of
a
along the 15O°W
113
cros ssection.
63
Vertical distribution of Q, CO
and NTA at
GS233.
64
116
Correlation of surface NTA and NTCO2 with
temperature for the GS data in the Southern
Ocean.
65
120
Correlation of surface NTA and NTCO2 with
temperature for the CS data in the North
Pacific Ocean. The triangles represent GS
stations located in the Northwest North
Pcific. The crosses represent CS stations
located in the equatorial regions. The solid
dots represent the others.
66
122
Correlation of [O.5ITA+TCO2] with AOU for the
deep water of the selected GS Pacific
stations.
67
Distribution of excess CO2 along the 150°W
crosssection (taken from Chen, 1982).
68
127
Distribution of tritium along the 35°N. This
figure is reproduced from Fine et al. (1981).
70
126
Distribution of Excess CO2 along the 165°E
LJsssection.
69
125
Distribution of tritium at selected CS
stations extending from 150 to 55°N between
129
Page
165°W and 170°E (see Fig.1). This figure is
reproduced from Fine et al. (1981).
130
71 Penetration depth of the fossil fuel CO2 in the
North Pacific Ocean.
133
Carbonate Chemistry of the North Pacific Ocean
1. Introduction:
The carbonate system in natural waters is one of the
most complex topics in oceanography. The system has long
interested many oceanographers from various fields since all
three subspheres of the earth (biosphere, lithosphere, and
hydrosphere) play an important role in the carbonate
syst em.
More recently, the fate of fossil fuel CO2 in the ocean
has promoted interest in the study of carbonate chemistry in
the oceans. With the refined determinations for the various
dissociation constants of carbonic and boric acids in
seawater and the sophisticated instrumentation for measuring
the parameters, our knowledge of the CO2carbonate system of
the ocean has rapidly increased.
The extensive oceanographic investigation has been
carried out in the North Pacific Ocean. Those extensive
studies, especially in physical oceanography, have provided
an excellent knowledge as a foundation for any other type of
study such as that of the CO2carbonace system.
The purpose of this thesis is to present the results of
carbonate chemistry, carbonaterelated chemistry, and the
relationship of carbonates and CO2 with geochemical and
physical mechanisms in the North Pacific Ocean. In chapter
2, the sources of data used in this thesis is introduced. tn
chapter 3, the hydrographic background is provided before
2
the main body of the thesis.
In Chapter 4, the carbonate
chemistry and related topics are scrutinized as separate
units.
Many figures used in this thesis were generated for a
forthcoming Department of Energy report (Chen, Wei, Rodman
and Olson, in preparation).
3
2. Data sources:
2.A. Outline of main data source:
The principal data set used in this study consists of
two separate NOAA's Miller Freeman (Feely, 1981) and
Discoverer (Cline, 1982) cruises. They were in the North
Pacific, roughly at longitude along 150°W from June to July
1981 and along 165°E from May to June 1982. Along with
carbonate chemistry data, the standard oceanographic values
of temperature, salinity, oxygen, and nutrients were
measured (Feely, 1981; Cline, 1982).
Data from the Miller Freeman consisted of hydrographic
casts of 10 to 15 sampling intervals down to roughly 2500 m.
The Discoverer carried Out the sampling to a deeper depth
down to roughly 3000 m. Measurement of pH is available for 9
stations of the Miller Freeman and for all the stations of
the Discoverer. Alkalinity data is available for all the
Miller Freeman and Discoverer expeditions. Calcium data is
available for all the stations of the Miller Freeman and 8
stations of the Discoverer (Feely, 1982).
The tracks for both cruises are shown in Fig.1. The
cruise tracks for supplemental data can also be found in the
figure. The crosssections, indicated by solid lines, will
be discussed in the text.
$40_____
_____$60
160 ________$00
14th!
120W
_ _
.1
V J
( iL
it
31*
LI,
I
"0
T
oIl
a,
33IlI
I
g
41
s..
11
50 3111$
10
I
a
A
I
A
SO
1$
4 A
50
41
A
A
40
,
A
S1
ill
\
Oil
1
Sil
II
Oil
I
Sill
.
0flip
4
IS
2(
13I
\
511*
I0
Ill)
5133
Sill
A IIIQOPAG
$ o
II
Sn,
.315
U ---1(1135-4
0
0
-.
$4
$60
$00
$40
60
l2W
V V
LocaiLOfl60 the stations where chemistry data are available. The solid lines
connecting the stations indicate the cross-sections discuNSC(I in the text. The
ENP (Hi tier Freeman), WNP(Discoverer)
abbreviations used for the expeditions are
Fi.1
Cs (CEOSECS ) ,
IP (1 NDOPAC)
5
2.B. Methods of data analysis:
The analytical procedures in measuring the pH and
alkalinity used in Miller Freeman expedition were described
fully in Chen (1982). Same procedures were used for 1982
Discoverer by Olson and for 1984 INDIVAT 1 by Chen and Wei.
The pH was measured at 25+0.02°C. Calibration was
carried with NBS 4.004 and 7.415 buffers. Alkalinity
analysis were performed using the potentiometric titration
method of Dyssen and Sillen (1967), modified by Edm9nd
(1970). Calcium samples were taken by Chen on the Miller
Freeman cruise and were measured by Mrs. Quan Wu at Oregon
State University by using the complexometric titration
method (Tsunogai et al., 1968b; Olson and Chen, 1981).
Calcium data of Discoverer cruise was obtained by Feely
(1982).
2.C. Literature data source:
Two other data sources supplemented the two cruises,
Miller Freeman and Discoverer data. The major one was the
Pacific GEOSECS (Geochemical Ocean Sections Study) data from
1973-1974. Undoubtedly, the extensive GEOSECS expedition was
a milestone in the study of the CO2carbonate system. It
provided a large data base with high precision. Broecker and
6
Takahashi (1978) suggested that a correction of -15 )ieq/kg
of GEOSECS
titratiometric total CO2 data of the Pacific
Ocean is needed to achieve the internal consistency of the
total alkalinity, total CO2 and PCO2 data. The revised
carbonate data (Takahashi et al., 1980) will be used for the
following discussion. The precision for the GEOSECS
alkalinity and total CO2 data was estimated to be ±0.2% and
0.5% (ld), respectively, for the Pacific Ocean data based
upon the deep water (>4000 in) sample measurement (Takahashi
et al., 1981). An updated equation of state (Millero and
Poisson, 1981) was used to recalculate density of seawater
from all the GEOSECS depth, salinity, and temperature data.
The station network is rather coarse but the number of
levels sampled and of parameters measured is extensive.
The INDOPAC data (1978) along 35°N provided a high
quality close station spacing to couple with our two N-S
sections. Some Japanese data (Hattori, 1977) from the
Northwest Pacific Ocean (KH75-4) was also used.
In order for a data set to be used for supplemental
coverage, it had to have the full array of water chemistry
or at least as extensive an array as our cruises. The data
generally had to be recent in order to match the quality of
the presently obtained chemistry data. In some areas where
there was poor station coverage such as the Gulf of Alaska
7
these criteria were relaxed.
The carbonate data of the Miller Freeman and Discoverer
cruises were compared with the literature data sets. It was
found that both Miller Freeman and Discoverer data sets and
GEOSECS Pacific data are consistent with each other in
oxygen, alkalinity, and total CO2. The calculated total CO2
of the Miller Freeman and Discoverer cruises is also
consistent with the GEOSECS measured total CO2 data.
However, INDOPAC alkalinity data seems systematically lower
than the other data sets by 10-15 ,ieq/kg. The INDOPAC
alkalinity data are adjusted to agree with our data sets
when used.
2.D. Abbreviations used for data sets:
For the purpose of convenience the abbreviations listed
below are used in this thesis for the following discussion:
GS - GEOSECS expedition
IP - INDOPAC expedition
ENP - Eastern cross-section along 150°W from
Miller Freeman 1981
WNP - Western cross-section along 165°E from
Discoverer 1982
8
3.Hydrography:
3.A. Background:
Our study covers an area north of 1O°N to 55°N, in
which two large-scale gyres appear in the surface
wind-driven layer. There exist good reviews of these gyres
in literature (Sverdrup et a]., 1942; Tully, 1964; Masuzawa,
1972). These two gyres, subtropical and subarctic, play an
important role in our study area.
A simple description is presented. The clockwise
subtropical gyre, of which the north edge lies in the west
wind drift centered at about 40°N, is bounded to the north
by North Pacific Current, to the east by Californi.a Current,
to the south by North Equatorial Current, and to the west by
Kuroshio Current. The subtropical gyre exhibits westward
intensification and a sluggish eastern equatorward flow. The
anti-clockwise subarctic gyre has a smaller size than the
subtropical one. The western boundary current, Oyashio,
serves as a west boundary for the gyre. An eastward
Subarctic Current forms the southern portion of the gyre.
Much of the water of the Subarctic Current flows north to be
recognized as the Alaska Current off the coast of Canada.
The Alaska Current flows westward along the Aleutian
Islands. This current is called Alaska Stream (Dodimead,
9
Favorite, and Hirano, 1963; Favorite, 1967; Park et al.,
1968).
Below the layer of the surface circulation system a
salinity minimum water called North Pacific Intermediate
Water (NPIW) is found. The origin and distribution of the
NPIW has been extensively discussed by Reid (1965). It
appears to circulate in a clockwise gyre and its relatively
high oxygen content indicates frequent replenishment.
The deep water of the North Pacific, underlying the
NPIW, is characterized by very uniform properties. The
source, path, and modification of the deep and bottom water
of the North Pacific can be reviewed from the existing
literatures (Knauss, 1962; Reid and Lynn, 1971; Edmond,
Chung and Sclater, 1971; Mantyla, 1975; Mantyla and Reid,
1983). In summary, the bottom water flows from the Southwest
Pacific Basin to the Central Pacific Basin through the
Samoan Passage near 100S, 169°W. Part of this south origin
water spreads from the Central Pacific Basin to the north
through the Wake Island Passage into the Northwest Pacific
Basin and to the east into the Northeast Pacific Basin.
A schematic diagram shown in Fig.2 serves as a summary
for the oceanographic circulation and water masses found in
the North Pacific Ocean. In the surface wind-driven layer, a
water mass called North Pacific Surface Water (NPSW) can be
Frontal tone
0
ao
tO
NPST
40
31)
NPSW
p.
NPSA
JLw
E
1000
1000
4-
NPDW
30001-
40001-
Fig2
NPBW
Schematic diagram of water masses distribution of the North Pacific Ocean.
0
11
divided, by a frontal zone, into North Pacific Subarctic
Water (NPSA) and North Pacific Subtropical Water (NPST)
according to the characteristic physical properties.
Underlying the NPSW, a water mass called NPIW characterized
by a salinity minimum is found. Two deep water masses, North
Pacific Deep Water (NPDW) and North Pacific Bottom Water
(NPBW), are found below the NPIW.
3.8. Inferences from this work:
Vertical crosssections of the measured variables were
prepared for both longitudinal sections occupied by the
Miller Freeman (Feely, 1981) and Discoverer (C].ine, 1982)
data.
Salinity
Fig.3 and 4 show the vertical distribution of salinity
(S) for the ENP (Feely, 1981) and WNP (dine, 1982) data,
respectively. The depth distributionof the S for both ENP
and WNP sections shows a deepening in the Central Pacific
and shallowing approaching the equator. This results from
the general surface convergence of water towards the center
of the anticyclonic gyre and thus deepens the isopleths in
midlatitudes. In the equatorial regions upwelling results
which carries properties closer to the sea surface.
The vertical section of' salinity for the ENP section
12
150
200
250
30°
35°
40°
45'
l..at.
I
22
Sta.
r
19
I
I
18
io
14 13
50
I
I
a
I
6
4
55°N
I
2
''N34.2 _..<
-
34.4
E
34.5.
.
_& - -
2000
S, Id3
3°c
Fig.3
Distribution of salinity with depth along the 150°W
cross-section (Data from Feely, 1981).
13
at.
II
I
i
$
St..
I
I
I
I
I
I
II
is
IN
SI
I
3
II
II
Is
I
I
1
ZI
'SI
E
34.4
13
S
'WI
s, i
Is"
34 C
Fig.4
Distribution of salinity with depth along the 165°E
crosssection (Data from Cline, 1982).
14
(Fig.3) shows the development of a shallow salinity maximum
extending southward from the vicinity of the subtropical
front where evaporation is high (Tsuchiya, 1968). Its depth
is at approximately 100 m and salinities exceed 34.8 x
WNP data also shows a shallow salinity maximum extending
southward from about 25°N to about 18°N (Fig.4). This
Smaximum is not as clear as the ENP section presented above
but shows more saline characteristics (saltier than
35.2x103).
The salinity section of ENP along 150°W shows a broad
frontal feature from 35.00 x iO
to 33.25 x icr3 and
extending from 31°N to 39°N (Fig.3). Thus, on the basis of
surface salinity, the subarctic front appears to be located
near 39°N and the subtropical front near 31°N.
Salinity structure along the WNP section (Fig. 4) is
more complicated than that of the ENP section. A highly
advective region is shown by a zigzag salinity contours
from 35°N to 40°N down to a depth of 1500 m. The deepening
of salinity contours at this region and the upwelling in the
equatorial and in the northern North Pacific make the
salinity structure show an overall concave pattern along the
crosssection. Two sharp horizontal salinity gradients in
the surface layer are found respectively at 43°N and 37°N.
They were identified as the subarctic front at the north and
15
as the subtropical fronts at the south, respectively
(Dodimead at al., 1963; Masuzawa, 1972).
Two shallower salinity minimums have been investigated
in the Northeast Pacific by Reid (1973) which have an
analogous origin.
Using STD data, Kenyon (1978) has found
the shallow salinity minimum in the upper 200 m to be
extensive and continuous in the eastern Pacific from 125°W
and 175°W along 35°N. It is likely that the fresher
near-surface waters north of 40°N contain these layers but
the sampling interval for our data was not sufficiently
detailed to delineate them.
On both longitudinal sections the predominant
north/south feature is the low salinity deep stratum which
is situated near the surface in the north. It deepens and
appears as a salinity minimum tongue at about 600 m at
mid-latitudes, then shallows towards the equator.
The origin and distribution of the deep salinity
minimum has been extensively discussed by Reid (1965) who
found that it originated in the northwest Pacific north of
45°N in the Sea of Qkhotsk and in the vicinity of Kamchatka
Peninsula. Reid (1965) has shown the salinity minimum water
mass originates from winter time convection and subsequent
vertical diffusion to the density levels associated with
this deep S-mm. These density levels are nearest the
16
surface in the northwest Pacific. Winter convective
overturning and vertical diffusion subsequently maintain the
salinity minimum. However, the minimum may result from the
formation of warmer and more saline waters above it in
summer.
In the ENP section of 150°W, the major salinity-minimum
core (values from about 33.75 to 34.20 x 10
N to S)
begins to appear near 45°N and is associated with the 26.8
surface at a depth of about 300 m. It deepens southward
to about 500 m and then.rises to near 400 m at 15°N (Fig.3).
In the WNP section of 165°E, the clear salinity minimum core
associated with
centered at depth of 400 m at 40°N
and deepens into a depth of 700 m at 37°N is shown in Fig.4.
This S-minimum core becomes shallower into a depth of 300 m
at 20°N. The salinity at this tongue-like core increases
from 33.8x103 to 34.2x103 to the south.
Temperature
Fig.5 shows the vertical distribution of potential
temperature (e) for the ENP data (Feely, 1981). Fig.6 shows
the vertical distribution of temperature (T) for the WNP
data (Cline, 1982).
The overall trend to the isotherm on Fig.5 is a general
upward slope to the north so that for any level shallower
17
15°
I
20
I
$9
30
25
40
35
45
6
$413
io
8
50
55°N
I
I
I
I
6
4
2
S
a
200
eloc
a-.- - -
-- - - - - -
.- - - -
--
300
Fig.5
Distribution of potential temperature with depth
along the 150°W crosssection (Data from Feely, 1981).
iS
as
as
IS
ii
4$
1
Ii
II
a
U,
- -
,
A
/
- - -
t\/__/-
I$SI
I
Fig.6
Distribution of temperature with depth along the
(Data from dine, 1982).
165°E cross-section
19
than roughly 2000 m the temperature is usually less to the
north, which is consistent with the accepted view of heat
input in the equatorial regions of the oceans (Pickard,
1979).
There are exceptions to this general inclination. South
of about 20°N the isotherms warmer than about 10°C shallow
towards the equator which is typical for isopleths which
upwell in equatorial waters (Tsuchiya, 1968; Craig et al.,
1981).
In Fig.5, two regions of maximum surface temperature
gradients are noticeable. A southerly gradient near 35°N
centered on about the 17°C isotherm corresponds to the
subtropical front; and a northerly gradient near 43°N
centered on the 11°C corresponding to the subarctic front.
Hence the temperature transitional zone is shifted several
degrees northward compared to the salinity front.
Fig.6 shows a more complicated isotherm structure for
the WNP section than that of the ENP section. The isotherm
between 35°N and 40°N shows a zig-zag like structure to a
depth of about 1500 m. This results from the active
interaction of two major western boundary currents (Oyashio
and Kuroshio) which have large differences in temperature.
North of 40°N the isotherms have a general upward slope to
the north. South of 20°N two different general inclinations
4.]
are observed. The isotherms warmer than 8°C shallow toward
the equator, which is the result of equatorial upwelling.
Density
The potential density
6e
is calculated from depth, S,
and T for ENP (Peely, 1981) and WNP (Cline, 1982) data and
is plotted with depth (Figs.7 and 8).
In the midlatitude the isopycnals show a deepening
relative to stations in the north or the south. This
deepening shifts northward with depth so that the deepening
of the 26.00 d9 isopycual occurs near 26°N and the 27.50
isopycnal occurs near 41°N for the ENP section. The
potential density structure forthe WNP section is similar
to that of the ENP section. However, the shifting of
deepening sites of the isopycnal is not as clear as the ENP
(Fig.7) for the WNP (Fig.8) section.
Fig.8 shows 26.0 69 contour has a steep slope at about
37°N. This water corresponds to the Kuroshio Extension and
was proposed by Masuzawa (1972) to be the Subtropical
Boundary. It is important to note that WNP2O was located at
175°E (Fig.1) where the 26.8 6
isopycnal does not outcrop.
Note the change in contour intervals for densities greater
than 27.00 6. The densest water existing on these
crosssections is 27.74.
21
IS'
Ste.
r
Z2
20
9
25'
16
30
40
35'
10
1413
6
45
6
50'
4
55N
2
2e.O537i
T::
ZT.O
a
£
a
0.
F4'I'I']
Fig.7
Distribution of potential density (6) with depth
along the 150°W cross-section (Data from Feely, 1981).
30!
25°
20°
15°
6
5
3
Sta rL3o
8
.
9
12
5th1
I
I
I
I
I__al'
450
40°
35°
$4
$6
$8
20
__!-;
20.0
:....r..__.._21.0
000
3000J_
Fig.8
:
:
Distribution of
with depth along the 165°E cross-section
(Data from Cline, 1982).
1'.)
23
9s diagram
Figs.9 and 10 show a full scale and expanded scale ENP
0/S distribution. Fig.11 shows the eis distribution for
selected GEOSECS stations in the North Pacific. The 8S
correlation for the WNP data is shown in Fig.12 and the
expanded es correlation for the same stations is shown in
Fig.13. The 8S relationships for all the stations differ
from each other in the upper layer, but converge to the same
8S relationship at greater depths.
The stations nearest to each other on the two cruises
are GS204 (31°22'N, 150°2'W) and ENPI3 (32°.1'N, 150°2.I'W).
The 9/S correlation of GS204 was also plotted in Fig.9. Both
ENPI3 and GS204 show similar 9/5 structure but ENP13 is
somewhat saltier above the salinity minimum and has an
approximately 1°C warmer salinity minimum.
It is noticeable in both Figs.9 and 12 that the els of
surface mixed layer shows a general conservative mixed
between two water types, NPSA and NPST. The water between
the mixing layer and salinity minimum layer for the
southerly stations of both ENP and WNP data also show a
linear mixing structure.
WNP2O, analogous to Edmond's (197i) SEVENTOW7178
station, shows the linear 8S region extends from the
surface to the deepest sampling depth. The intensive
'I
F
24
22
20
I6
'
1i1
''4
1t6
''9
22
GS204
$6
11'
I'
12
$0
e
6
4
2
I
3 2.0
I
d a
33.0
S. $0'
Fig.9 9S corre1atjo
and related CS data.
for ENP and selected CS data (Data
from Feely, 1981)
I"
9
U
8
7
6
0
0
5
£NP
4
oZ
04
06
£8
.22
(
340
35.0
SI i-
Fig.10
Expand 9-S correlation for ENP data (Data from
Feely, 1981).
t\)
UI
26
p.
Iv
GS
p 219
00
9
A218
0
214
a
a
o1
0202
A
.224
A
£ 225
B
A
7
0 212
I
+ 22!
I
0
6
0
a
a
S
-
+
a
0
++
00I.0
0
OU
4
A
a0
d
so
4A
30
'32.0
34.0
S. I0'
Fig.11
eS correlation for selected GS data.
35.0
I
I
32.5
33.0
I)
0
33.5
34.0
345
350
355
S. 10'
Fig.12
0-S correlation for WNP data (Data from Cline,
1982).
-4
28
I
to
I-
4
0
9
/
'V
/
/
/
/
/
8
0, q
it.
,1
,/
/
/
/
/s
/
/1
/
/
/
d\
/
7
/0
/
/
/
/
/
/
6
/
/
/
/
____,
/
0
/
/
/
\/
CT
'I
I/I
__
7
-/
/\
//
'O
/
0
\f-.
/
it.
F
/
'
k
/°\
/ \
\
C-
WNP5
0-5
z-5 '-6
3
0-6
-0-8
£ 9
' 2
-
//
' 14
0 16
-0-lB
1-20
,
'S
wJ
Is
\a__s'I
I
32.5
/
33.0
I
33.5
34.0
34.5
S ,
Fig.13
Expanded e-S correlation for WNP (Data from Cline,
1982) and selected 1(1175-4 data.
U
29
upwelling may be the main mechanism for this observation.
A potential temperature minimum at Sa33.2x103 is
observed at WNPI.8. In order to show the source of this
temperature minimum core, 9-S of three Japanese KH7S-4
Rakuho Maru stations are used (Fig.13).
KH75-4-7 and
K1175-4-6 located near the Kamchatka Strait, one in and one
Out of the Bering Sea (Fig.!). WNPI8 and two Japanese
stations all show a thermocline, which is believed to be
caused by summer heating of the surface layer. The
temperature minimum core is also found for these two
stations. This core falls on the isopycnal surface of
d926.5. This temperature minimum water is proposed to be
the remnant Bering Sea winter water. The core of temperature
minimum is similar to the dichothermal temperature in the
Bering Sea (Tully, 1964). The dichotherma). structure results
from existence of the residue of the winter cooled water
lying in between the upper seasonal and deep permanant
halocline. The winter remnant water may flow out from the
Bering Sea along the isopycual d9 surface of 26.5. The
linkage between this temperature minimum water and salinity
minimum water found in more southerly stations needs to be
studied further.
In the GEOSECS network samples the northwestern
Pacific, a temperature-minimum structure is seen at GS218
which is not present on the 150°W ENP section. The most
northerly stations (Fig.9) show a general continuous
increase of salinity with decreasing temperature (and hence
depth) below the mixed layer. Proceeding southward the
salinityminimum layer develops and becomes most pronounced
in the central North Pacific region of the section at ENP13,
14, 16,
19. The most southerly, ENP22, shows practically no
reduced salinity signature in the strata containing the
salinity minimum on the stations further north. This is
especially noticeable in Fig.1O which shows expanded e-s
correlation for ENP data. A deep salinity minimum is also
noticeable in Fig.12. Three most northerly stations,
WNP16,18,20 show no salinity minimum core property. Like ENP
data, the salinity minimum core is diminished for the
northerly stations, WNP16,18,20.
The deepwater at temperatures below 1.5°C is quite
uniform (Figs.1O and 13) and this is particularly evident
for the GEOSECS data (Fig.11) where sampling depth extends
to the bottom. However between the salinity minimum strata
at roughly 5°C to 6°C and the temperature 1.5°C (depthlSOO
m )
there are two separate distinct curves to the 9/S
distribution. This is seen in both the Miller Freeman and
Discoverer data (Figs.9 and 12) particularly on the expanded
scale relationship (Figs.1O and 13) and in the GEOSECS data
31
(Fig.11). The northerly stations with no salinity minimum,
or showing the beginning of the salinity minimum
(ENP2,4,6,8,1O; WNP16,18,20, and GS221, 219, 218), are
significantly fresher on any given isotherm.
32
4. Results arid discussions:
4.A. Chemistry data along 150°W and 165°E
The results of carbonate data and related chemistry
data for the ENP and WNP expedition are shown by
longitudinal sections. Following the discussion of the
various chemistry data along the two cross-sections, the
correlation of 8 and S with various carbonate parameters
will be used to scrutinize the distribution of the
parameters.
4.A.(a) AOU and pH:
The oxygen distribution for ENP is similar to that made
by the others (Smetanin, 1962; Postma, 1964; Reid, 1965;
Duedall and Coote, 1972). The presence of oxygen minimum is
a reflection of the North Pacific circulation system where
oxygen decreases downstream from its aeration source and
replenishment is primarily by vertical diffusion (Reid,
1965).
Apparent Oxygen Utilization (AOU) is a rough measure of
the amount of oxygen consumed by oxidation since the waters
left their surface source. It is calculated by the
difference between the solubility of oxygen at the potential
temperature and salinity, and the measured oxygen
concentration. We assume that a parcel of seawater was at
equilibrium with the atmosphere before it sank. Hence, a
33
negative AOUof seawater means oversaturation with respect
to the atmospheric oxygen while the positive AOU means
undersaturation.
As shown in Fig.14 which shows the vertical
distribution of AOU along the ENP section, most surface
water is oversaturated with respect to atmospheric oxygen.
The 100% saturation horizon outcrops between ENP2 and ENP4.
Like that described by Postma (1964), the ocean shifts from
an oxygen source into an oxygen sink at about 55°N. AOU's
were found to be useful to study the direction of water
motion in the North Pacific Ocean (Pytkowicz and Kester,
1966; Park, 1967; Alvarez-Borrego and Park, 1971). Fig.15
shows the distribution of AOU with de along the ENP section.
This figure shows a diapycnal character at densities less
than 27.3. The maximum AOU enters along 27.4 de from the
north as a greater than 300 ,.iM/kg tongue. The corresponding
depth is about 1200 m (Fig.14). Continuing south to 15°N the
maximum is greater than 280 )1M/kg and along the same itrata.
Reid and Mantyla (1978) show that the low oxygen water in
this depth range
32.0, i.e. density reference
adiabatically to 1000 db which is approximately 6e
27.35)
enters as a tongue from the east roughly between 45°N and
30°N. Thus the motion, or rather the direction of older
waters, for the water at this level is towards the west for
$50
20°
10°
Lal.
250
30°
I
I
26
22
Sa.&
19
16
35°
450
40°
I
$0
$4 13
500
55°N
I
t_J6
8
4
2
_j_._
0
290
250
iiiijiiii
:300
C
4.-
\
a.
Si
280
2OOOTT>NiZBO
250
230
220
'
AOU, jimoI/kg
3000'
Fig.14
Distribution of apparent oxygen utilization (AOU) with depth along the
15O°W crossBeCttOfl (Data from Feely,
1981).
4:-
35
Lat.
Sta.
24.0,
i15°
22
2C?
11
3F?
2
tI
14 13
32
4
8
10
5
6
+
5N
4
2
4
4
'
AOU, jmoI/kg
25.0
\.
.'
66
26.0
27.
.----.
Fig.15
260 -------
------
240
Distribution of AOU with 6
along the 150°W
crosssection (Data from Feely, 1981).
36
most of this cross-section.
Figs.16 and 17 show the distribution of AOU with depth
and 6e' respectively, along the 165°E cross-section. The AOU
distribution of WNP is similar to that of the ENP section
except several points. The AOU value for the water below a
depth of 1000 m of WNP (Fig.16) seems to be lower than that
of ENP (Fig.14). According to Pytkowicz and Kester (1966),
the gradients of AOU indicates the relative ages of the
waters. Thus, the water below 1000 m for EN? is proposed to
be older than that of WNP. The carbon-14 data in the North
Pacific also support the older age of the deep water of the
eastern North Pacific Ocean than the western North Pacific
Ocean (Fiadeiro, 1982). There is another difference between
the AOU distribution of ENP and of WNP. A sharp AOU vertical
gradient for the water above 300 m is found in the north of
WNP section. There is no identical phenomenon found in the
ENP section. An active upwelling process must exist at
WNP2O. This is similar to the findings of Dodimead et al.
(1963) and Park (1967).
Distribution of pH(25.°C) with depth for ENP data along
150°W is shown in Pig.18. pH distribution has a similar
pattern with AOU (Fig.14). A tongue-like minimum core of pH
extends equatorward. The 7.5 contour reaches to 25°N. The
axis of pH minimum deepens at 40-SO°N to a depth of about
15°
20°
25°
3d
350
I
L...at
3
Sta____
6
5
89
450
400
N
I
12
14
I
16
18
0
280
200
I/kg
Fig.16
220
-_ 210
Distribution of AOIJ with depth along the 165°E cross-section
(Data from Cline, 1982).
L.
20
15°
Lat.
20°
25°
35°
30°
50°N
45°
40°
I
I
I
Sti
6
8
9
14
I
tO
I
AOu,
ol / kg
:'
25.0
:.°
L.
Oe
20.0.
I 01
27.0
t
1.
300
2I0
Fig.17
Distribution of AOU
crosssection
:
with
°
aiong
(Data from Cline, 1982).
tne ion.
38
100
I (IL
I
200
150
I
I
250
300
I
400
35°
450
500
55°N
I
Sb.
0o0
£
4V
2000
3000
Fig.18
Distribution of pH(25°C) with depth along the 150°W
crosssection.
40
1200
in
and becomes shallower equatorwardly to a depth of 800
m at 25°N and poleward to a depth of 700 m. The surface pH
value decreases northward, which is the result of a high
partial pressure of CO2 because of higher solubility of CO2
in the colder north region.
The pH(25°C) data of the WNP section are plotted in
Fig.19. This section is very similar to the ENP section. The
surface pH value increases from north to south. The isopleth
in surface layer shows a concave pattern. A clear pH minimum
core centered at 1000
in
extends equatorward.
Fig.20 shows correlation of e and p11(25°C) for the ENP
cruise. All the pH data from EN? and WNP cruises has a
minimum. From this figure, we can see the difference of pH
minimum between ENP16,19 and the others. The pH minimum of
former stations is about 7.5, while the pH minimum of the
other stations is about 0.05 pH units lower than two
southernmost stations. The minimum p11 value of two
southernmost stations are lying on about 4°C while the pH
minimums of the other stations are lying on about 3.5°C.
This difference depicts a different history of the pH
minimum waters of these two groups of stations. The pH
minimum water of northerly group stations is older, colder,
and more corrosive than that of ENP16,19. Fig.21 shows the
same correlation for the WNP stations. The two sections, EN?
15°
20°
25°
I
I__at'
3
5
30°
I
6
89
35°
450
40°
I
12
I
14
16
18
50°N
I
20
DO0
a-
V
a
1.58
7.80 -
7.85
pH (25°C)
Fig.19
\-__x
7.87
Distribution of pH(25°C) with depth along the 165°E
cross-section.
8.
3-
I
I
0
8 .2-
0
0
0
060
£
A
V
8.
S
U
0
V
0
£
-
0
0
0
4
IT
In7,
ENP
('J
x
I
0
EL7 .87
g-4
0-6
0
AB
v'0
A3
4
7...
40
7 .6-
0
v
o
U00Vo
U
£
7
.
.
0-14
o_16
-
-
-
0-19
-I'
I
5
I
I
10
IS
I
20
25
o,oc
Fig.20
Correlation of 8 with p11(25°C) for the ENP
stations.
4:-
8.4
.
8.3
0
8.2
,4
$9,
0
£0
gn
V
0
8.I
0
)c
a
a
8.0
0
0
I0
LA
V
8
7.9
('I
I
WNP
I- 3
7.8
x- 5
0 6
7.7
8
$ 'x
7.6
A 9
j1Px
'4
7.5
a
.
'-12
I4
0
goo°
0IS
7.4
o-20
7.3
0
5
tO
15
20
25
e, °c
Fig.21
Correlation of 0 with pH(25°C) for the WNP stations.
44
and WNP, have similar 8-pH correlations. Two southernmost
stations, WNP3,5, show the same characteristics as ENP16,19.
However, WNP3,5 have higher pH minimum values of about 7.6
than that of ENP16,19. Hence, the water at the pH minimum
layer of ENP is older than that of WNP. The northernmost
station, WNP2O, shows a relatively homogeneous temperature
layer with a large range of pH value (from 7.35 to 7.67).
Park (1968a,b) also found this phenomenon at near-by site.
He attributed this large variation of p11 value to be mostly
due to the organic decomposition and that an upwelling
mechanism brings this cool and acidic water to the surface
layer.
Figs.22 and 23 show the S-pH(25°C) correlation for ENP
and WNP data respectively. The S-pH correlation shows better
mixing characteristics for surface water than that which was
shown in B-p11 correlation. The important factors that
control the pH value for surface water are temperature,
salinity, biological activity, TA, and saturation relative
to the partial pressure of CO2 in the atmosphere. For the
WNP cruise, surface p11(25°C) increases from 7.95 at the
northern latitude to 8.30 at the southern latitude. For the
ENP cruise, surface pH(25°C) increases from 8.00 at the
north to 8.25 at the south. The surface pH values for those
stations located in between the two southernmost and
U
S
U
0.
0
7
7
7
I
7.
32:5
33.0
s,Io-3
Fig.22
Corre1aton of S with pH(25°C) for the ENP stations.
U'
I
I
8.
S
Q*
,,..Q_)e
w,.v
£
\
/(
\
8.
0
\
o-o
\
vIo
I')
(\1
\
a-
7.
7.i
7.
N
4
%_oOo/ .)
£ $4
-I6
o-18
u-20
7.
I
32.5
33.0
33.5
3'I.O
3-I.3
S. I0
Fig.23
Correlation of S with pH(25°C) for the WNP stations.
47
northernmost stations fall on the linear trend connected by
two end members. This linear trend cannot be explained
theoretically. Such an increasing pH trend from high to low
latitude was also observed by Park (1966) in the region of
the Northeastern Pacific Ocean. Wider pH range simply due to
wider T range.
Since pH is not a conservative property, it cannot be
used as a parameter to study the mixing phenomenon. However,
the coil-like pattern of S-pR plots (Figs.22 and 23) shows a
characteristic water-type property. The NPSW with a linear
S-pH correlation, NPIW with salinity minimum layer for
northerly stations, NPDW with pH minimum value, and NPEW as
the end of increased pH trend, make up of this coil-like
pattern.
Park (1968a, 1968b) discussed two factors, AOU and
CaCO3 dissolution, which affect the vertical distribution of
pH value. The pH(25°C) and AOU correlation is ploted for
both ENP and WNP data (Figs.24 and 25). ENP pH data
apparently shows better precision than that of WNP. The pH
minimum generally coincides with the AOU maximum. However,
WNP2O shows a significant difference between AOtJ maximum and
pH minimum (Fig.25). Again, this phenomenon is subject to
the complicated involvement of upwelling which may have the
atmospheric oxygen entering the upwelled oxygen-depleted
48
I
I
I
ENP
8.
%
a
8 .2
o6
aB
8 .1-
,I0
I3
A
8 .0
c 14
0
0
0
-
::::
c'J
.8
°a
7
l:
7
8
000
7
-
7
I
IOU
200
300
AOU, pmot/kg
Fig.24
Correlation of AOU with pH(25°C) for the ENP
stations.
49
0
a
I',
c'j
3-
0
100
200
300
AQU, j.rno/kg
Fig.25
Correlation of AOU with p11(25°C) for the WNP
stations.
50
water. We found a slope of 2.5 pH(pH unit)/AOU(mmol/kg) for
the waters below the maximum AOU layer. This slope is higher
than the slope obtained by Park (1968b) but still falls in
the range of uncertainty. Our shallow sampling depth (3000
m) may cause uncertainty of the overall trend for the deep
water. The slope for the water above AOU maximum layer
varies from 1.7 to 2.1 depending on the location of the
station. Because of the relatively fast exchange rate of
oxygen between seawater and the atmosphere, the slope of
surface layer is not reliable. Our data covered a wider area
which includes oceanographic subarctic and subtropical
region than that of Park (1968b). This wide coverage also
makes our pli-AOU correlation not as unique as that obtained
by Park. However, the average slope of 2.1 LpH/i.AOU is close
to the calculated slope of 2.0 ApH/AOU obtained by Park
(1968b). The lag between two trends, deep water and shallow
water, is mainly caused by calcium carbonate dissolution.
The schematic arrows are shown indicating the effect of
CaCO3 dissolution and organism decomposition.
4.A.(b) Alkalinity data:
The vertical distribution of normalized alkalinity
(NTA-alkalinityx35/S) along 150°W section is shown in
Fig.26. The NTA generally increases with depth. A shallow
15°
Lot.-
20°
25°
35
3cf
I I.
s1
'1
I,'
...
5c?
I
55°N
I
'L4_
23I'\(
24O
2430
S. D
iiiiii
11)
45
I
....--..-.-.-.-.--..-.
a
4cf
I
I
::::II..
200
2470
I
NTA, jieq/kg
Fig.26
Distribution of NTA with depth along the 15O°W
CtOSssectjon.
Ui
52
minimum layer is observed at the region of 37°-50°N. A low
NTA pool is found in between the two fronts.
Fig.27 shows distribution of NTA for WNP data along
165°E. NTA increases monotonically with depth. This is
consistent with the work of Park (1967). The surface NTA
decrease southward from the subarctic region to the
subtropic region. A low NTA pool is found lying between 20°N
and 30°N. The NTA contour lines are pushed downward at 20°N.
This may be a clue of
NPDW
intrusion. The alkalinity
distribution in the Pacific was found to have a maximum at a
depth of about 2500 m (Takahashi et al., 1981; Fiadeiro,
1980). The sampling depth for the WNP expedition, except
WNP12, is too shallow to see the maximum. However, a maximum
core at a depth of 2500-3000 m is observed at WNPI2.
The correlation of potential temperature and normalized
alkalinity data is plotted (Figs.28 and 29) for both ENP and
WNP sections and selected stations of the GEOSECS
expedition. The purpose of adding GEOSECS stations near our
two sections is to show the better correlation by increasing
the sample size and to find the NTA distribution for bottom
water. Early in 1956, alkalinity was used as a parameter for
water mass analysis (Koczy,1956). From Figs.28 and 29, we
can identify the different trends representing five water
masses; NPSW, North Pacific Central Water (NPCW), NPIW,
$50
Lat
200
3
25°
I
I
5
30°
6
8
40°
35°
I
I
9
$2
45°
I
14
I
$6
$8
50°N
I
20
-c
1
C
1
Fig.27
Distribution of NTA with depth along the 165°E cross-section.
Lfl
4
Iz
NP(
24
0
5
10
0 ,
Fig.28
Correlation of 0 with NTA for ENP and selected CS stations. The triangles
represent surface data and the squares represent salinity minimum
data.
a
4
I-
z.
NP
5
10
IS
20
25
e ,°c
Fig.29
Correlation of 0 with NTA for WNP and selected CS stations. The triangles
represent surface data and the squares represent salinity minimum
data.
Ui
56
NPDW, and NPBW along the water column. The linear lines in
Figs.28 and 29 are roughly drawn by eyes.
a. NPSW- The NTA distribution of surface water is
complicated by physical conditions, biological activity (to
less extent), river runoff, and oceanographical circulation.
The surface NTA is denoted by triangles in Figs.28 and 29.
Although the data are scattered, a rough linear line is
plotted in Figs.28 and 29. This line represents an increase
of 4 seq/kg for a decrease of 1°C surface temperature. NTA
of the subarctic and of the subtropical serve as two ends of
the linear line.
Low temperature, high NTA water is mainly
from the Gulf of Alaska. Both in Figs.28 and 29, there is a
high NTA water for temperatures higher than 25°C. The cause
of this anomaly may be related to the equatorial upwelling
which contains high alkalinity upwelled from subsurface
layer (Fiadeiro, 1980). The 25°C break is consistent with
the work of Chen and Pytkowicz (1979), which was based on
GEOSECS data.
b. NPCW- Below the NPSW, a layer of water defined as
NPCW is identified. The e-NTA correlation of this layer is
also scattered, especially for the ENP data, and shows
generally a smaller NTA value than NPSW and NPIW. More
scattering of ENP data than WNP data of this water may
result from the general circulation pattern which has a
57
westward intensification and a broad, sluggish eastern flow
(Mazuzawa, 1972). The NPCW of the Eastern North Pacific
receives more influence of other water masses which have
different NTA values to NPCW. In other words, NPCW for WNP
retains more original properties and receives less influence
from the other water masses. A tentative linear trend is
drawn on these waters. The extent of variations of NTA with
temperature seems narrower than that of NPSW. The increased
trend of NTA is about 3 peq/kg/°C.
c. NPIW- There is a well-known salinity minimum in the
North Pacific Ocean as discussed earlier. This salinity
minimum core represents NPIW and is believed to be formed in
the western subarctic ocean (Reid, 1965) and circulates
roughly in a clockwise pattern (Sverdrup. et al, 1942). The
NTA at salinity minimum layer is denoted by squares in
Figs.28 and 29. Fig.29 shows that there is a linear
correlation between S and NTA. This line is also drawn on
Fig.28 on which correlation seems weak. This trend has a
steeper slope than overlying NPCW and NPSW, which means a
higher rate of "input" of alkalinity (about 5.5 peq/kg°C).
Generally, the NTA of this water has higher NTA than that of
NPCW and lower than that of NPSW. This causes a minimum NTA
layer in NPCW down the water column. However, this minimum
is diminished for the water in the Subarctic region. The
58
mechanism causing this shallow alkalinity is unknown.
d. NPDW The NTA in NPDW is characterized by a sharp
NTA increase (12 )leq/kg/°C). The NTA value of WNP is lower
than that of ENP. Fiadeiro (1980) also found a higher
alkalinity for the Eastern North Pacific. tn Shiller and
Gieskes' (1980) 35°N crosssection, a higher alkalinity
value also was observed in the Eastern region. However, the
increase rate vs potential temperature does not show
significant difference for the two sets of data. The sharp
NTA increase is because this water is undersaturated with
respect to calcium carbonate (the saturation value vs
calcite and aragonite will be discussed later). The
dissolution of calcareous materials must occur. The increase
of pressure also favors the dissolution of calcareous
mineral. The eNTA trends of NPDW and overlying NPIW cross
at about 6°C for ENP data and at about 5°C for WNP data.
e. NPBW NPSW trend cannot be found if only the data of
our two sections is used. Thus, the NTA of several GEOSECS
stations near our two sections is used to see the
distribution of NTA in the bottom layer. The bottom water
has potential temperature of about 1.1°C. The NTA decreases
dramatically toward the sea bottom. The circulation pattern
in the deep North Pacific causes this maximum. Fiadeiro
(1980) suggested the effect of vertical eddy diffusion from
59
the deep water which contains high alkalinity, not of local
sources or flux from the bottom. The relationship among NTA
distribution, oceanographic circulation, and calcite degree
of saturation will be discussed in a later section.
The correlation of salinity with NTA for ENP and WNP
stations is shown in Figs.30 and 31, respectively. S-NTA
correlation is more informative for the surface mixing but
is less informative for the subsurface mixing. The S-NTA
correlation is interestingly T-shaped with a horizontal line
representing surface water and a vertical line representing
subsurface water. The same finding was observed by Koczy
(1956). The knot connecting the two trends is NPIW which has
an average salinity of 34.0x103 and NTA of about 2370
ieq/kg. "Horizontal" means a small variation in NTA and
"vertical" means a sharp increase of NTA. The fresh, high
alkalinity water is subarctic water and saline, low
alkalinity water is subtropical water. The transition zone
is located between the two ends and has NTA value between.
Beneath the core of salinity minimum, the salinity and NTA
increase monotonically to our deepest sampling layer (about
3000 in). A rough linear relationship can be seen on these
subsurface S-NTA data. The deviation of S-NTA trend for
those stations south of 25°N (ENP16,19,22,26, WNP3,5,6) is
caused by the salinity difference. At this layer, southerly
a'
32.5
Fig.30
33.0
.
33.5
LO
.au
Correlation of S with NTA for EN? stations
I
I
0
230(0
9
a
£
0 A.
S
'.o
0
o
235(-
$
8
A
0
*
X4
1,9
*
0
S
a
0
0
a-
*
8°
0
o9
.- 3
*- 5
0
24O0-
o-6
A
24O-
325
33.0
B,
II
34.0
33.5
34.5
0- 8
'-12
'-14
° 16
0- 18
20
35.0
355
5, I0
Fig.31
Correlation of S with NTA for WNP stations.
0'
62
stations have higher salinity than that of the others.
4.A.(c) Total CO2:
The total inorganic CO2 for both of our ENP and WNP
sections was calculated from the pH and alkalinity data. For
150°W section only ENP2-19 has Total CO2 because pH data is
not available for the other ENP stations. The total CO2 for
ENP and WNP was calculated from pH(25°C) and TA by using the
first and second dissociation constants of carbonic acid of
Mehrbach et al. (1973) and dissociation constant of boric
acid of Lyman (1957). Fig.32 shows vertical distribution of
normalized total CO2 (NTCO2TCO2x35/S) along 150°W. The
distribution is very similar to the distribution of AO!J and
pH. Since the pH and NTCO2 are strongly related to the
production and consumption of the organism in the ocean, the
similar contour pattern of the AOU, NTCO2 and pH contours is
not surprising. A high NTCO2 core centered at the depth of
about 1200 m extends equatorward. The highest NTCO2 is found
at a depth of 1200 m at about 43°N. The surface NTCO2 which
is strongly affected by the biological activity and surface
temperature shows a decreasing trend from the north to the
south.
Same NTCO2 contours are constructed for WNP data along
165°E (Fig.33). This section is very similar to the ENP
200
100
Lal.
Sb.
25°
19
350
16
1413
45°
40°
10
8
---
:::: r
..
:
50°
I
I
a
I
I
22
4-
30°
6
4
5N
I
2
I
.
2430 __!_.__-
1000.E
4-
0.
4)
0
2000
2400
.
NTCO2
,
pmol/kg
3000'
Fig.32
Distribution of NTCO2 with depth along the 150°W crosssection.
a'
150
250
20°
I
I__al.1
3
30°
I
6
5
SIa4
1
89
350
4cf
I
45°
I
12
14
50°N
I
16
18
I
20
z000
I'N_,2
340
4d)
2860
2340
N.
NTCO2, jimol/kg
Fig.33
Distribution of NTCO2 with depth along the 165°E cross-section.
a'
65
section. It is noted that the maximum NTCO2 core is lying at
a depth of several hundred meters deeper than that of the
AOU maximum core. This gives an impression that the
dissolution of inorganic calcareous material also contribute
the increase of TCO2.
4.A.(d) Calcium data:
Another important carbonate parameter, calcium
concentration, is plotted in Fig.34. The specific calcium
(calcium/chiorinity) is known to be useful for studying the
water mass circulation, mixing and diffusion in the Pacific
Ocean (Tsunogai et al.,1968a; Sagi, 1969; Tsunogai et al.,
1973; Horibe et al., 1974; Shiller and Gieskes, 1980).. The
normalized calcium (NCa-Cax35/S) we use here has the same
meaning as that of specific calcium. The NCa increases
monotonically from the surface to the deepest sampling
depth. In the northern region, the surface NCa decreases
southward. A high NCa value of more than 10280 )lmol/kg is
found at the deep layer of the Northeast Pacific Ocean,
which is similar to Tsunogai's (1973) result. The water here
represents the oldest of the deep waters of the world oceans
(Knauss, 1962; Mantyla, 1975; Mantyla and Reid, 1983). The
calcium content in this water is higher as a result .of
CaCO3 dissolution. A dome of high NCa is found in the deep
100
15°
200
25°
30°
35
2219161413
I
Lal.
Sta.6
4O'
45°
500
I
I
55°N
I
230
C
4I)
.
270
NCaI0000, JrnoI/ kg
3000'
Fig.34
Distribution of NCa with depth along the iso°w cross-section.
0'
a'
67
water north of 40°N.
Salinity and NCA for ENP data is shown in Fig.35. The
surface NCa shows scattered patterns on which the clear
correlation can not be found. Tsunogai et al. (1968a) found
a linear correlation between chlorinity and specific calcium
in the region of the Western North Pacific Ocean where the
western boundary currents prevail. The lack of clear linear
correlation may be due to the complicated biologic uptake
and river runoff at the Eastern North Pacific. However,
in
the region of the Gulf of Alaska the surface S-NCa
correlation shows a linear mixing process. The overall trend
of S-NCa relation in surface water is increasing calcium
with decreasing salinity. Shiller and Gieskes (1980) also
found a high NCa in those surface waters of northern origin.
In contrast to the surface layer, subsurface water has an
increasing NCa trend with increasing salinity. Like
alkalinity data, the S-NCa combination of NPIW which has a
uniform NCa value of about 10230 pmol/kg, seems to be a
connecting knot between two trends, surface and subsurface,
of S-NCa correlation. The deviation of two southernmost
stations, ENP22 and ENP26, is due to the high salinity in
the region of these stations.
Fig.36 (9-NCa correlation) shows a similar pattern as
S-NTA correlation for the same stations but more scattered
0
0
UI
0
E
0
0
z
102
E NP
2
-IO o-19
ii- 4 £-I3 -22
- 6 0-14
-26
- B 0-16 £35 0-41
32.5
33.0
33.5
34.0
34.5
35.0
35.5
SI
Fig.35
Correlation of S with NCa for ENP stations.
0'
I
I
S
10150
NF'DW
S
I
£
'
I
S
A
I
N
A
A
NPIW
A
S
I
A
I
A
S
II
0
E
A
I
IL*
A
.
A
S
0
LA
A
A
I
$0200
S
I
I
I
£
$0250
i:.'.
;Ir,
Ill S
I
S
I
I..
/5
10300
0
5
tO
$5
e,°c
20
25
30
Fig.36
Correlation of 0 with NCa for ENP stations. The
triangles represent surface
data and squares represent salinity minimum data.
0'
70
especially for the surface water (denoted by triangles). The
NCa value increases with decreasing temperature. A break of
slope for 8-NCa relationship is coincident with e-NCa
combination of NPIW (denoted by squares) which also shows a
slight increasing NCa trend with decreasing e. NPDW shows a
higher slope which means the dissolution of CaCO3 mineral
prevails.
Feely's (1982) calcium data along 165°E is used here.
Figs.37 and 38 show the correlation between S-NCa and e-NCa,
respectively. Both figures have similar patterns as ENP's
pattern (Figs.35 and 36). Surface NCa mixes conservatively
with increasing NCa when salinity decreases. Subsurface
water has the opposite trend of surface water: NCa increase
with increasing salinity. However, the difference of overall
deep water NCa of ENP and WNP seems too large to account for
the difference of NTA of the two sets of data. Considering
the subsurface samples, EN? has higher NTA than WNP's NTA by
about 30 peq/kg. If only CaCO3 dissolution is assumed to
cause this difference, then the upper limit of difference in
NCa should be 15 j.imol/kg. However, the observed difference
seems to exceed 25 pmol/kg for NCa data. The possibility of
systematic error between the two sets of calcium data may
not be eliminated. In the 35°N cross-section of Shiller and
Gieskes (1980) there is no significant difference in calcium
I
I
10150
o__0
E 10200
10250
o.00..
V
0
33.0
33.5
34.0
a-16
0
u-20
34.5
35.0
35.5
S,
Fig.37
Correlation of S with NCa for WNP stations (Data from Feely, 1982).
I-
[I,JDhsJ
£
S
.
10150
NPSW
£
I
S
S
£
I
S
.4.
a'
S
.
10200
e
0a
I
a
£
2
$55,.
Ii
S
I
10250
5l:
%$
j.s..
S
S
.
S
S
a
I
S
S
S
10300
___________________I
10
Fig.38
L
15
e, °c
1
25
I
Correlation of 8 with NCa for WNP stations. The triangles represent
surface data and squares represent salinity minimum data
(Data from Feely, 1982).
73
concentration between the Eastern North Pacific Ocean and
the Western North Pacific Ocean.
4.A.(e) Nutrient data:
It has long been known that the oxygen minimum water
originates in the deep water of the northern North Pacific
and is associated with nutrient maximum (Sverdrup et al.
1942). Figs.39 and 40 show the vertical distribution of
nitrate and phosphate respectively along the 150°W section.
Both figures show a maximum core, the axis of which is
centered at about 1000 m. The isopleths of both nitrate and
phosphate in the surface layer deepen at the midlatitude
and shallow at the highlatitude and at the equatorial
region. Silicate, however, is not so uniformly associated
with oxygen as the other nutrients. The maximum does not
appear on this section (Fig.41). It is located below the
oxygen minimum and is quite broad with values only slightly
less than the maximum found just above the bottom. This is
shown quite well on plates 13 and 29 of the CEOSECS Pacific
Atlas (Craig et al., 1981). The eastwest vertical sections
along 35°N (Kenyon, 1983) show the silicate maximum at 150°W
is at about 2500 m whereas the oxygen minimum is at about
1000 m and the nutrient of both phosphate and nitrate are
near 1000 m. Between ENPI3 and ENP14, the silicate signal
marked by a depression of the isopleths extends from the
15°
Lot.
I
200
I
I
25°
I
30°
I
350
I
400
I
45°
I
500
I
55°N
I -
Ski.
Ililsis]
0.
2000
3000'
Fig.39
Distribution of
NO3 with depth
along the iso°w
cross-section
(Data
trom Feely, 1981).
-J
(00
Lal.
(50
20°
25°
300
35°
40°
45°
50°
55°N
Sm.
[SI$ISi
E
a)
ID
2000
000l
Fig.40
Distribution of PO4 with depth along tiLe iso°w cross-section (Data
from Feely, 1981).
-.4
100
Lot.
15°
I
20°
I
a,'
25°
I
..
30°
.
40°
35°
.-
I
45°
50°
5N
--
S to.
-C
a'U
2000
3000'
Fig.41
Distribution of Sf02 with depth along the 150°W cross-section (Data
from Feely, 1981).
0'
77
surface to the deepest samping levels. A similar feature was
noted at these stations for temperature and salinity but not
to the depths indicated by silicate. Were it not for these
two stations the general trend of the isopleths would be
relatively flat and quite different from the AOU trend which
roughly shows a concave downward bow to the isopleths. North
of these two stations the silicate values are higher along a
particular isopycnal than south of these stations.
Figs.42-44 show the vertical distribution of nitrate,
phosphate and silicate of WNP data. Again, a tongue-like
high nitrate and phosphate core, lying at the AOU maximum
layer, extends from north to south. The transition from
subarctic to subtropic region is marked by surfac.e
nutrients. From salinity cross-section (Fig.4), the
subarctic front is lying between WNP14 and WNP16. Across
this front, from WNP16 to WNPI4, phosphate decreases from
0.85 to 0.10, nitrate from 9.3 to 1.1, and silicate from
17.42 to 4.3 )lmol/kg respectively. This characteristic is
consistent with the finding of Park (1968c). Like the ENP
section, the WNP cross-section of silicate shows a different
pattern with an increasing trend with depth, and it does not
show a maximum coreat the AOU maximum core. Again, this is
not surprising because silicate is not remineralized along
with the organic matter (Edmond, 1974).
15°
Latl
200
I
3
25°
5
400
35°
I
I
Sta..
300
6
89
s
45°
I
12
14
50°N
I
(6
(8
20
i
a
a
a)
Fig.42
Distribution of NO3 with depth along the 165°E
from Cline, 1982).
cross-section (Data
15°
LaI.I
200
25°
I
3
350
30°
I
5
6
8
40°
45°
I
9
12
I
14
16
IA
500 N
I
20
C
4
C
Flg.43
Distribution of PO4 with depth along the 165°E cross-section (Data
from Cline, 1982).
'.0
15°
200
I
I__at
3
25°
I
I
5
Sb. i
30°
6
89
350
400
I
45°
I
12
$
14
I
16
18
50°N
I
20
i_
Lu
E
Fig.44
Distribution of Si02 with depth along the 165°E crosssection (Data
from Cline, 1982).
03
0
81
4.8. Alkalinity-Calcium relationship:
The similar pattern between the e-NTA/e-NCa and
S-NTA/S-NCa presented in previous sections is not surprising
since both calcium and alkalinity are related to the
dissolution of CaCO3 mineral. Several researchers have paid
attention to the correlation between these two parameters
(Tsunogai et al., 1973; Horibe et al., 1974; Brewer et al.,
1975; Shiller and Cieskes, 1980; Tsunogai and Watanabe,
1981; Kanarnori and Ikegami, 1982; Ikegami and Kanamori,
1983). Tsunogai et al. (1973) used calcium and alkalinity
data of 1(1170-1 and 1(1170-2 to study the TA-CA relationship
along 170°W. They found (2TA-Ca)/Cl is not constant and
proposed a possibility of sources of calcium other than the
dissolution of calcium carbonate. Horibe et al. (1974) found
a correlation coefficient between NCa and Normalized
carbonate alkalinity to be about 0.5. Brewer et al. (1975)
used surface alkalinity and calcium as a reference and
correlated the difference of calcium and alkalinity between
measured values and surface reference values (both were
normalized to 35x103 salinity). They found the slope is
larger than the theoretical 0.5 slope and proposed potential
alkalinity to compensate the effect caused by decomposition
of organic matter to the titration alkalinity. Tsunogai and
82
Watanabe (1981) applied this model to KH78-3, and KH-75-4
data in the northern North Pacific and Bering Sea and
received a slope close to 0.5. Kananiori and Ikegami (1982)
and Ikegami and Kanamori (1983) used nitrate concentration
as a index of the effect of organic decomposition to
alkalinity (PAaTA+1.26NO3) to correlate their KH8O-2 and
KH82-1 data obtained from the western North Pacific. They
also obtained a slope of 0.5 between NCa and normalized PA.
In our study, both the nutrients and oxygen are used as an
index for the purpose of estimating the organic
decomposition effect to the titration alkalinity.
Fig.45 shows the correlation of NTA and NCA for all the
ENP sample. Unlike Tsunogai and Watanabe's (1981) results,
the surface data of our ENP stations scattered. NTA/NCa
scattering near surface is probably real because the data
covered a large area. Shiller and Gieskes (1980) also found
a large variation of NCa/NTA ratio from 0.4-0.8 for the
INDOPAC mixed-layer samples. Hence, the assumption that
preformed calcium value for the surface water and subsurface
water is similar, which is suggested by Tsunogai and
Watanabe (1981), should be studied further. The least square
method is used to fit data points below 200 m depth for the
ENP data. However, the NTA/NCa correlation has no physical
meaning. The NTA/NCa correlation consists of variation in
83
10300
S
S
: :;
. .1...
$
10250
.
I
S
.
S.,
a
02
I
C
S
10200
IjT
.
I
I
S.
S
II
S
S
S
S
10150
S
2300
2400
2500
NTA, peq/kg
Fig.45
Correlation of NCa with NTA for the ENP data. The
regression line shown in the figure is NCA8996+O.52OXNTA
for the samples below 200 m.
84
preformed NTA and NCa which may not be entirely related to
CaCO3 dissolution. The correlation slope is 0.520 + 0.007
for NTA-NCA correlation. The number, 0.007, represents the
range of the slope at the 95% confidence interval. Our
slope, 0.520, is much lower than that observed by Horibe et
a)..
(0.734), Brewer et al. (0.688), and Tsunogai et a)..
(0.74).
Fig.46 correlates normalized potential alkalinity
[NPTA1-(TA+NO3+PO4)x35/S] and NCa for the ENP samples. The
correlation coefficient for the samples below 200 m is 0.442
+ 0.042. The number, 0.042, represents the range of the
slope at 95% confidence interval. The potential alkalinity
proposed by Brewer et al. (1975) cannot be confirmed from
our data since both of the correlation slopes of NCa-NTA and
NCa-NPTAI do not agree with the theoretical value of 0.5
within the limit of uncertainty. Although we know the
protonation due to the organic decomposition on the
titration alkalinity plays an important role, the amount of
alkalinity that is changed by this protonation is not known.
As pointed out by Brewer et al. (1975), the preformed values
for nitrate and phosphate were ignored in the potential
alkalinity model, which means an overestimated potential
alkalinity may be obtained.
Instead of nutrient data, AOU may serve as a better
85
10300
S
S
'2'
S.
S
10250
S
.
0
E
S
.
t.
S
S
.
,
5
.'
S
I
I,
S
z
I0200
e
S
_______
S
.
.
II
S
5S
S
S
.
.
S
S
S
.
.
S
S
10150
S
2300
2400
2500
NPTAI, jieq/kg
Fig.46
Correlation of NCa with NPTAIN(TA+NO3+PO4)x35/S]
for the ENP data. The regression line shown in the figure is
NCa9167+0.442xNPTAI for the samples below 200 m.
86
index to estimate the degree of protonation from organic
matter on the alkalinity. It has been discussed by Chen and
Pytkowicz (1979), Chen and Millero (1979), and Chen et al.
(1981) that 138 mole of oxygen, assuming
Redfield-Katchum-Richards ratio is correct, will be consumed
to decompose 1 mole of organic matter and reduce the
alkalinity by 17 eq. The model is followed to calculate
preformed alkalinity. The normalized potential alkalinity
obtained by AOU correction (NPTA2(TA+17/138)xA0Ux35/S] is
plotted against NCa. The result is shown in Fig.47. A little
improvement is obtained compared to NPTAI-NCa correlation.
The least square fit for the samples below 200 m shows the
correlation slope being 0.456 + 0.042. The number, 0.042,
represents the range of the slope at 95¼ confidence
interval.
In summary, variation of the preformed calcium and
alkalinity concentrations needs to be considered in order to
properly investigate the influences of CaCO3 dissolution and
organic matter decomposition on the measured calcium and
alkalinity values. This calls for accurate calcium and
alkalinity data in the formation region of the subsurface
waters.
87
10300
S
t.
10250
S
S
.
S
S
4
S.
S
S
.
S
S
IU
S
..
10200
S S
S
S
.5
I
/
7
S
S
S
10150
S
2300
2400
NPTA2,
Fig.47
2500
ueq/kg
Correlation of NCa with NPTA2[(TA+17xAOU/138)x35/S]
for the ENP data. The regression line shown in the figure is
NCa-9137+O.456xNPTA2 for the samples below 200 m.
88
4.C. Inorganic Carbon/Organic Carbon ratio:
Li et al. (1969) studied the sources of the excess CO2
in the deep water of the Pacific and Atlantic Oceans. They
concluded that about 20% of the excess CO2 in the deep water
is derived from the inorganic CaCO3 dissolution. Edmond
(1974) used a ratio of shell-derived to tissue-derived CO2
in the deep water, without eliminating the effect of fossil
fuel CO2 increase, to study the mechanism of TCO2 input.
Chen et al. (1981) improved the method of Edmond's (1974)
model to calculate inorganic carbon/organic carbon ratio of
the South Pacific. They reached the result that about 12% of
the total CO2 input into the deep waters of the South
Pacific is contributed from CaCO3 dissolution. The same
technique is adopted from Chen et al. (1981) but new
preformed equations for TA and TCO2 are used. The preformed
equations used here are obtained from revised GEOSECS data
(Takahashi et al., 1980). The preformed equations will be
discussed in the section on fossil fuel CO2 signal. Only the
resulting equations are shown here.
TA°(jieq/kg)-2384-4.2xe (+9)
TCO(pmol/kg)-2219-11xG (±16)
The inorganic carbon/organic carbon ratio is
represented by IC/OC and it was calculated by the following
equations:
89
Ic/oc - (0.16038(TCo2+4O)+TA]/[2(ATCO2+40)-TA]
where
£TCO2-TCO2(measured)-TCOxS/35
TATA(measured)-TA°xS / 35
The estimated amount of anthropogenic fossil fuel CO2
signals is 40 imol/kg total CO2. Chen et al. (1981) pointed
Out the importance of the contribution of the fossil fuel
CO2 for the IC/OC ratio. Here we use 40 jimol/kg as the
anthropogenic fossil fuel CO2 contribution to the preformed
value of TCO2. An underestimate of 0.05-0.07 in IC/OC ratio
will be observed if the fossil fuel signal is ignored in
calculation. Several GS stations which are close to our two
cross-sections are also used to calculated IC/OC ratio. The
results are consistent between the data sets.
The distribution of IC/OC with depth for the samples
below 500 m is shown in Fig.48 for ENP, WNP, and selected GS
stations. Unlike the South Pacific, there is rio sharp
decrease in the bottom layer. This is expected because of
the Benthic Front (Craig et al., 1972) does not extend into
the North Pacific Ocean. The IC/OC ratio increases markedly
from 0.1 to 0.35 from 500 a to 2500 a. The IC/OC ratio shows
a constant value below 2500 a. This reflects the high
degradation rates of organic tissue in upper water. Edmond
(1974) found a ratio of 0.26 at a water depth of 3000 m in
the Northeast Pacific, which is lower than our value of
90
IC/0C
0.2
0.1
0.4 0 0.1
0.3
(bJ
(a]
&;
0
0.4
0.3
0.2
a
0
IL 0
A 0
x
a0
0
(.1.1.]
a
a
a
a
V
a
U
E
a
WNP
'U
a
3000
a
0
ENP
-2
I
x-4
0 6
I
0-6
. $
I
a
'- $
VIa
I
'- 10
400
.
.- 3
V
'-14
a
0-16
a
V
a
I
a-13
o 14
0 18
a 16
0-19
0-20
a
a
a GSZO4
'a
a GS221
eGS2)8
a
a
V
V
U
a
I
Fig.48
Vertical
Distribution of. inorganic carbon/organic
carbon (IC/OC) ratio for ENP (a) and WNP (b) data.
91
0.35. From the discussion above, we can conclude that about
25% increase in TCO2, after leaving from the Southern Ocean,
is contributed by inorganic CaCO3 dissolution below a water
depth of 3000 m.
Figs.49 and 50 show the different ways to express the
distribution of IC/OC ratio. The solid lines represent the
value of IC/OC. The dashed lines represent contours of total
input of TCO2 since the water mass leaves its source region.
Since oxidation of organisms contributes the largest part of
TCO2 increase (about 75% for deep water), the layer of
maximum increase of TCO2 is generally coincident with the
maximum AOU layer. Two southernmost stations of both the ENP
and WNP cruises (ENP16,19, WNP3,5) show the lowest extreme
of TCO2 increase, which is a result of less influence from
the oldest NPDW.
It is also found that ENP stations have higher total
TCO2 input than that of the WNP cruise. This is expected due
to the fact of the older age of the Northeast Pacific Deep
Water. ENP stations show a narrower range of increase and a
higher increase of TCO2 than the WNP stations for the deep
water, which means more homogeneous water was found in the
Eastern North Pacific. WNP18 and WNP2O have unique
characteristics (Fig.50). First, they have a lower IC/OC
ratio at the same isopycnal surface than the other stations.
t
V'8
250
0
0 00
92
01
Z'
00
-Q
6
A
0
0
9.
06
A
0
s...
a
d
150
0
-
0
0
0
E
Ci
0
0
C-)
ENP
'80
oo
ix
A
'
o-6
-
8- $
"00
cP
- Jo
0
50
-
0- 14
70
0- J5
0- 19
o
to
20
30
40
50
60
70 80
I C, jimot/kg
Fig.49
Correlation of IC with OC for the ENP data. Dashed
lines represent the total input of TCO2 (IC+OC).
O
50
_._oc'
ot
0
94
This lower IC/OC ratio is a consequence of lower inorganic
CaCO3 input to TCO2 and/or higher input from organic
decomposition. Second, WNP2O has a thick layer of high TCO2
increase. This is the result of high CO2 content water which
was pumped up to the surface layer through upwelling. We
cannot find an analogous station from the ENP cruise.
The maximum IC is always found lying below the depth of
OC maximum. The IC part of TCO2 increase does not show a
significant increasing trend after IC maximum is achieved.
This could be explained by kinetic control on the
dissolution of CaCO3 minerals. Edmond (1974) also postulated
that the calcareous tests are transported as far as the deep
water before significant dissolution takes place.
e
95
4.D. Degree of saturation of CaCO3, lysocline, and CCD
It is wellknown that the deep water of the North
Pacific Ocean represents the oldest age of the world oceans.
Hence, the concentration of carbonate ion in the North
Pacific Ocean shows the lowest value. Takahashi et al.
(1981) found a low mean C0
concentration in the Pacific
Ocean and attributed this to the low alkalinity/total CO2
ratio of this area. The result of low C0
concentration is a
lower degree of saturation and then a shallower saturation
horizon of the Pacific Ocean than the other regions.
The distribution of the saturation value of seawater
with respect to calcite and aragonite will be presented and
its relationship with compensation depth and lysocline will
be addressed.
4.D.(a) Method of calculation of degree of saturation:
In calculation of degree of saturation of seawater with
respect to calcite and aragonite, the carbonate ion
concentration has to be calculated from the measured
carbonate data. The C0
concentration in seawater at the in
situ temperature and pressure conditions has been computed
using the WNP's and ENP's salinity, temperature, alkalinity,
pH, and calcium data. In addition to our two meridional
sections along 165°E and 15O°W, all the GEOSECS and INDOPAC
96
alkalinity and total CO2 of the North Pacific Ocean are used
to calculate the degree of saturation of seawater. Two
cr088-sections, one longitudinal along roughly 1800 and one
latitudinal along roughly 35°N, are selected from GEOSECS
stations to make the distribution of saturation value
clearer. The effect of pressure on the dissociation
constants for carbonic and boric acids determined by
Culberson et al. (1967) and Culberson (1972) is used for the
computation. Our expression for the saturation states for
calcite and aragonite in seawater is in percent of
saturation:
U - ICP / K'sp x 100%
where
ICP(Ca2)x(CO)
(M/kg)2
K'sp- apparent solubilityproduct for calcite or
aragonit e
The apparent solubility product for calcite and aragonice in
seawater at 1 atm total pressure determined by Ingle et al.
(1973) and by Berner (1976), respectively, are used. In
order to obtain the apparent solubility product at in situ
conditions, the pressure effect on the solubility products
summarized by Culberson (1972) is used.
4.D.(b) Degree of saturation of the surface water:
Figs.51 and 52 show the correlation of temperature with
97
5
tO
IS
20
25
ro
Fig.51
Correlation of surface
with temperature for a11
the available data. The line is a rough fit by eyes.
T, °C
5
Fig.52
10
15
Correlation of surface
20
25
with temperature for all
the available data. The line is a rough fit by eyes.
99
degree of saturation of seawater with respect to calcite
and aragonite
a'
respectively, for the surface
water of the North Pacific Ocean. It is evident that all the
surface waters are CaCO3 oversaturated. The linear
correlation between
c'
a
and temperature is found in
both figures. A high degree of saturation is observed for
high temperature surface seawater and a low degree of
saturation is observed for low temperature surface seawater.
This is consistent with the work of Lyakhin (1968) who
studied the calcium carbonate saturation of Pacific water.
The degree of saturation value is strongly influenced by
TA/TCO2 ratio which also shows a linear dependence to the
surface temperature.
The contours of the surface saturation value are
constructed in Figs.53 and 54 for calcite and aragonite,
respectively. The overall pattern is similar to the
temperature distribution which in turn is influenced by the
surface oceanographic circulation.
fl
and C)a decrease
slightly from west to east. In western subarctic regions low
c
and low 0a are observed which are the characteristics
of the cold Oyashio current. Low
and low
are also
found in cold Gulf of Alaska. A sharp change from north to
south at the region of mixing between the cold Oyashio and
warm Kuroshio around 40°N occurred. A high
and
a
basin
____I4dE
160
180
160
140
$20W
'7
fV72
'\
U
4OO
0
41
,,:
Lb LA A
A
A
A
A
o
21
()C
10
0
>600
\
.0/0
10
0
0
IId'E
Fig.53
.__L160
-
ISO
Distribution of surface
160
140
in the North Pacific Ocean.
I-
0
0
i4dE
160
180
160
(40
12dW
$__
°
300.
A
A
£
A
A
£A A
A
N
a 0/0
(60
Flg.54
180
Distribution of surface
160
a
(40
IZW
in the North Pacific Ocean.
0
102
at the center of the subtropical region is not clear enough
to be defined.
4.D.(c) Vertical distribution of degree of saturation:
Four cross-sections of saturation values of calcite and
aragonite are ploted by using the ENP, WNP, and GEOSECS
data. The selected longitudinal and latitudinal sections are
shown in the map (Fig.1).
(1) Profile along 35°N:
The latitudinal cross-section along approximately 35°N
was selected also by Takahashi (1975). But non-adjusted
TCO2 data was used to calculate saturation value and only
the
was calculated in his paper. The revised carbonate
data (Takahashi at al., 1980) is used to recalculate the
degree of saturation of both calcite and aragonite. The
results are shown in Figs.55 and 56. A very small variation
of saturation value is found in the water column below 1000
in
depth. A shoaling phenomenon of the saturation horizon
from the west to the east, is caused by the general surface
circulation pattern in the North Pacific Ocean. The west
North Pacific has the deepest saturation horizon at a depth
of 1100 In. The intensified western boundary current results
in the deepening of the saturation horizon. A characteristic
minimum
and
1a layer can be seen all the way from coast
to Coast. This minimum layer is relatively narrow and was
103
Long.
GS
0
224 223
120°W
$40'
$60'
225 226 2 4 21.3. 212
L-4500
'o°°
$800
$600
140°E
204 202
20$
L-4
J
z
:
2000
:
:
E
3000.
g
:
a.
U
.
C.
4000
5000
Fig.55
Vertical distribution of (
along the 35°N
crosssection. Dashed line represents CCD from Berger et al.
(1976).
104
160°
Long. 140°E
1400
160°
180°
GS 224 223
C
22
24
226
213
212
120°W
I
I
204 202
201
I
JLJ3Oo-I
I
60
:
2000
E
-
£
30OO
:
0.
C
4000
-------
.1
>r
R
:/%j.
5000
:1.
Fig.56
(60
/: :
Vertical distribution of
cross-section.
:
a
along the 35°N
105
not shown in Takahashi's (1975) profile. The existence of
this layer is strongly related to the oxygen minimum, layer.
The low saturation value found in this layer is believed to
be caused by high oxidation of organic matter, which
releases the carbon dioxide and reduces the carbonate ion
concentration.
(2) Profile along 165°E:
Fig.57 shows distribution of the saturation value of
calcite along 165°E. The saturation horizon deepens from
less than 150 m at the north to deeper than 3000 m at 25°N.
Fig.58 shows the saturation horizon of aragonite onset from
less than 100 m at north, from 750 m at 30°N, and from 500 m
at 15°N. Byrne et al. (1984) studied the dissolution rate of
aragonite particulates collected by freedrifting èediment
trap along the Discoverer cruise. Their findings revealed
that the depth at which high dissolution rate of aragonite
was found is shallower at high latitude than that of at
midlatitude. This is a support of our saturation profile
which shows that the saturation horizon concaves at
midlatitude.
A sharp change from undersaturation to supersaturation
for calcite is found in between WNPS and WNP6. The
oversaturated water at WNP3 and WNP5 may be the result of
the younger age of the seawater at that region. WNP6 defined
the southern limit of NPDW which is low oxygen, high
150
200
I
I._aI.I
3
250
I
5
300
6
8
400
350
I
I
9
450
I
(2
(4
50°N
I
16
(8
I
20
-f
C
4
C
Fig.57
Vertical distribution of Q
along the 165°E cross-section.
0
0'
(50
1.01.1
20°
I
3
25°
I
5
350
3d
I
6
8
450
40°
I
9
(2
(4
I
(6
lB
0
w
I
20
II
C
I
Fig.58
Vertical distribution O
fl
along the 165°E crosssection.
p.-
0
108
nutrients, and acidic water. Such a sharp change of
saturation value from north to south was also observed by
Hawley and Pytkowicz (1969) along their 170°W crosssection.
We do not have deep enough samples at WNP3 and WNPS to see
the distribution of saturation value down to the bottom. A
core of low saturation value is found as a tongue extending
from north to south at a water depth of 700 ui.
It is clear that the saturation horizon is also
affected by the surface circulation. A depression of
saturation contours at 30°N is located slightly south of the
subtropical convergence zone. The contour line of saturation
horizon almost outcrops to the surface at 50°N. Again, an
upwelling must play an important role on this shallow
saturation horizon.
(3) Profile along 180° W:
Takahashi (1975) selected similar but not identical
crosssections. Only a calcite saturation value was
calculated from the original GEOSECS data set in his paper.
As the 165°E crosssection, the contour lines in the surface
layer concave downward at midlatitude (Figs.59 and 60). The
100%
contour and 70%
a
contour deepen from 300 m at
the north to 2200 a at the south. Such a trend may depict
the large scale upwelling phenomenon for the deep water of
the northern region. The cores of low saturation value are
observed in both Figs.59 and 60. Takahashi (1975) did not
109
La1 ION
15°
I
GS
200
25°
30°
I
23
35°
40°
450
214
500
I
I
217
215
219
55°N
L
219
2000"'"
:
4000
/
-
Fig.59
Vertical distribution of 0
along the 1800
crosssection. Dashed line represents CCD fron Eerger et al.
(1976).
110
LQt. I0°N
15°
20°
250
300
I
35
I
400
45°
500
55°N
I
L-
V
Fig.60
Vertical distribution of 0a along the 1800
cross-section.
111
show this minimum core in his paper.
(4) Profile along 150°W:
This data set was presented by Feely and Chen (1982)
who used the estimated calcium concentration to calculate
the ionic product for the data above 1500 a. Figs.61 and 62
show our results of recalculated saturation value using
measured calcium concentration for all the samples taken.
The distribution pattern is similar to the profile along
165°E.
4.D.(d) The relationship among
lysocline, and CCD:
The mechanisms that control the distribution of the
calcium carbonate sediments is not clear. The relationship
among the saturation horizon, lysocline, and calcium
carbonate compensation depth (CCD) has been investigated by
several workers (Li et al., 1969; Edmond and Gieskes, 1970;
Pytkowicz, 1970; Heath and Culberson, 1970; Lisitzin, 1972;
Takahashi, 1975; Broecker and Takahashi, 1978). In order to
examine the above mentioned three properties, we need to
know the distribution of saturation horizon, lysocline, and
CCD. Peterson (1966) and Berger (1967) carried out in situ
experiments of the dissolution of calcite in the Central
North Pacific Ocean (18°49'N 168°31'W). Their work has
contributed knowledge to the in situ carbonate dissolution
study. It was found that at a depth of about 3700 a the rate
of dissolution of calcite increases sharply. The depth is
0
100
Lol.
I
26
S to.
I
22
200
I
19
250
I
300
I
16
4 13
350
450
40°
I
I
tO
I
8
500
55°N
I
6
4
2
0-
2000
Fig.61
Vertical distribution of Q
along the 150°W cross-section.
F'.)
100
Lat.
15°
I
26
Sb. i
22
200
25°
19
300
35°
I
I
16
1413
45
40°
I
50
55°N
I
10
8
6
4
4
200
-C
4II)
04
3000
Fig.62
Vertical distribution of
a
along the 150°w cross-section.
2
114
called lysoc].ine. Unfortunately, this kind of experiment is
scarce for the study to be more extensive. Another indirect
approach of lysocline was carried in laboratory experiments
by Morse and Berner (1972) and Berner and Morse (1974) by
examining the parameter, ApR, the defference between the pH
value for the calcite-seawater equilibrium and that of
seawater pH value. Takahashi (1975) suggested azpH value of
0.08 instead of 0.15 proposed by Morse and Berner (1972) as
an indicator of a sharp increase dissolution rate of calcium
carbonate. According to Berner and Wilde (1972),
pHO.O8 is
equal to the saturation value of 91% fl This value is
adopted in this thesis as a qualitative reference of
lysocline depth. The term "qualitative" is carefully used
because of the vague depth at which Oc91% falls. A large
amount of water in the North Pacific has a calcite
saturation value of 90-100% Q. The scattering of data also
contributes a large enough noise for the c-91% depth to be
poorly defined.
In Figs.55 and 59 the CCD is shown according to the
work of Berger et al. (1976). It can be seen that CCD
usually falls on a depth above the 80%
contour in the
mid-latitude region. Fig.59 shows a shoaling of CCD north of
40°N to a depth where the
90%
c
value is larger than 90%. If
contour was defined as lysocline in the North
Pacific, then CCD generally falls on a depth deeper than
115
lysocline and lysocline generally falls on a depth deeper
than the saturation horizon of calcite.
GS233 is selected for the purpose of discussion of the
relationship between C, lysocline, and CCD. This station
is selected because it is located near the station at which
Peterson's (1966) in situ experiment was carried out. The
distribution of carbonate ion concentration and normalized
alkalinity at GS233 is also shown in Fig.63 down the water
column from surface to sea bottom.
ven though the maximum
uncertainty of determination of the saturation horizon, 750
a, is considered (Plath et al. 1980; Pytkowicz, 1983) the
depth is still much shallower than lysocline and CCD depth.
The depth of lysocline estimated from ApH is consistent with
Peterson's (1966) at a depth of about 3500 a. Berger et
al.'s (1976) CCD at about 4400 a is also shown in the
figure. The difference of depths of saturation horizon,
lysocline, and CCD cannot be explained except by kinetic
arguments.
C0
concentration increases slightly below a depth of
saturation horizon. However, this increase cannot be
attributed only to the dissolution of CaCO3. Not only CaCO3
dissolution but also the effect of physical condition and pH
value on the speciation of carbonic acid affect the
concentration of C0
ion. Nevertheless, the general
increasing trend of C0
concentration and decreasing trend
116
NTA
COj
_(peq/kg)
I
I
jpmoVkg)
200
tOO
0
I
I
I
(1.
2450
2400
2350
2300
2OC)
300
400
(%)
E
'1
Fig.63
S233.
Vertical distribution of
COj, and NTA at
117
of saturation value below the saturation horizon still give
the inference of the importance of solubility product on the
saturation value for the deep water.
NTA is independent of temperature, salinity, and
pressure effect if per weight unit was used. Hence,
NTA
distribution may serve as a better indicator of CaCO3
dissolution than CO
ion concentration. The significance of
a maximum NTA layer is noted here. The maximum NTA layer
generally lying at a depth between 3000-3500 m in the North
Pacific is always shallower than both lysocline and CCD.
This is not expected since the deep lysoc].ine and CCD should
imply more CaCO3 dissolution below these depths, hence, an
increasing trend should be found. The explanation for this
phenomenon is that the rate of dissolution of CaCO3 is not
fast enough to overcome the flushing of bottom water at that
region. The location of GS233 is one of the main passages of
the Pacific Bottom Water with relatively high current speed
of about 10 cm/sec originated from the Southern Ocean
(Edmond et al., 1971). The rate of delivery of low
alkalinity bottom water is higher than the rate of in situ
production from dissolution of CaCO3.
118
4.E. Anthropoginic fossil fuel CO2 signal:
4.E.(a) Method of calculation:
The method of computation and limitation of the model
have been described in detail elsewhere (Chen and Pytkowicz,
1979; Chen and Millero, 1979; Chen, Millero and Pytkowicz,
1982). The following equations summarize the method.
excess co2 a TCO(present) - TC0(old)
- TCO(present) - TCO2(measured) - 0.5
TA(measured) - 0.78 AOU + 0.5 TA°(present)
where all quantities except AOU are normalized to 35x103
salinity.
Since the old deep water are free from the fossil fuel
CO2, the TCO2 for the deep water must be smaller than the
TCO2 of the young surface water from which the preformed
equation is derived. Hence, the excess CO2 for the deep
water has a positive value while the excess CO2 for the
surface water is zero.
In the various mathematical terms of the above equation
the preformed equations of alkalinity and total CO2 (TA and
TCO2) and the coefficient preceding the AOU are most
controversial. Hence the discussion of these quantities will
be emphasized before the results of calculation are
presented.
4.E.(b) Preformed equations and AOU coefficients:
119
The main idea of the model proposed by Chen and Millero
(1979) is: the difference between the TCO2 of the
anthropogenic CO2 free seawater and the TCO2 of its source
seawater which contains the full amount of anthropogenic
CO2 reflects the increase of excess CO2. How to select a
proper preformed value for TA and TCO2 to calculate excess
CO2 is one of the most problematic parts of this model. To
attack this problem the CS data has been used to derive the
preformed equations (Chen and Pytkowicz, 1979). Instead of
using all the surface NTA and NTCO2 of the CS expedition, it
is more suitable to use the surface NTA and NTCO2 of those
waters representing the source of the water under study. It
has been accepted long before that the Southern Ocean are
the only source of deep waters in the Pacific (Sverdrup et
al., 1942; I(nauss, 1962). Hence, surface NTA and NTCO2 for
those GS stations south of 50°S are used to derive the
preformed equations. Fig.64 shows the correlation of surface
NTA and NTCO2 with temperature for those CS stations located
south of 50°S. These stations encompassed all three large
oceans in the southern region. The reason for adding some
stations north of the Antarctic Convergence is to increase
the sample size which covers a wider range of temperature to
get a more reliable regression line. The least-squares fit
of these data result in the following equations:
120
2400
.
2
250
4
z
2300'
h
n.
121
TA°(}ieq/kg)
TCO
2384 - 4.2 x e
(jiaol/kg) - 2219 - 11 x $
Both TA° and TCO
(+ 9)
(+ 16)
are normalized to 35x103 salinity and the
numbers in parentheses give one standard deviation of the
fit for the above equations. These equations are used to
calculate excess CO2 for the water below NPIW. For the water
above NPIW the situation is very complicated because the
suitable preformed TA and TCO2 cannot be found from existing
data. The surface TCO2 in the North Pacific Ocean is very
complicated. As presented in Fig.65 different trends can be
found depending on the geographical location of the
stations. In Fig.65 the surface NTA and NTCO2 are denoted by
different symbols depending on the location of the stations
in the North Pacific Ocean. Fig.65(a) shows all the 9NTA
correlation falls on one linear line. The linear fit for
these data is NTA2407-3.46x8. Hence, the preformed
alkalinity of the North Pacific surface water is always
larger than that of the deep water in the range of
temperature found in the world oceans. Fig.65(b) shows the
correlation of surface NTCO2 with temperature for the North
Pacific CS stations. Unlike 9NTA correlation, three groups
of data - Northwest Pacific, equatorial, and other stations
- can be recognized. None of these trends is coincident with
the preformed TCO2 equation derived from the Southern Ocean
2400
122
230
(a)
.I
2300
e, c
Fig.65
Correlation of surface NTA and NTCO2 with
temperature for the GS data in the North Pacific Ocean. The
triangles represent GS stations located in the Northwest
North Peific. The crosses represent GS stations located in
the equatorial regions. The solid dots represent the
others.
123
data. The cause of these different trends is not known. The
NTCO2 at certain temperatures is different for the samples
found in the different area, especially for high temperature
data. This difference is too large to get a unique preformed
total CO2 equation. Such a different surface e/NTA and
G/NTCO2 correlation between the North Pacific and Southern
Ocean leads to the conclusion that the preformed NTA and
preformed NTCO2 of the surface layer and the deep layer must
be different. If this is true, a mixing problem will arise.
It is feasible to use its own preformed equations for the
surface mixed layer. However, it is difficult to determine
the preformed TA and TCO2 value for those waters that are
subject to mixing from different water masses if the mixing
ratio is unknown. For NPIW it is even more complicated
because the surface NTA and NTCO2 in the northwestern North
Pacific Ocean where the NPIW originates is not available.
However, we can say that the surface excess CO2 should be
zero. Through extrapolation and data examination it is still
possible to see the fossil fuel signal in the North Pacific
Ocean.
The equation for calculating the excess CO2
mentioned above is rearranged:
(TCO(present) - TCO2(measured)] - 0.5 [TA(measured) TA°(present)]
0.78 AOU + excess CO2
124
-
.ATCO2 - 0.5TA
0.78 AOU + excess CO2
Hence, if (ATCO2 - O.5TA) is plotted against AOU a slope
of 0.78 should be obtained and the intercept is the fossil
fuel CO2 increase signal. This has been done by Chen arid
Pytkowicz (1979). By using the same stations in their paper
and newly derived preformed equations (áTCO2 - O.5LTA) is
recalculated and plotted against AOU for the sample below
the AOU maximum layer. The result is shown in Pig.66. The
correlation slope is 0.78 and the intercept is -40 )lmol/kg.
4.E.(c) Penetration depth along 150°W and 165°E
sections:
Figs.67 and 68 show the results of fossil fuel CO2
penetration along the 150°W and 165°E cross-section. The
waters underlying the 40 pmol/kg excess CO2 contour have a
homogeneous excess CO2 value close to 40 jimol/kg, which
means these waters are free from fossil fuel CO2. The
decreasing excess CO2 from deep to surface layer means more
fossil fuel CO2 content In the shallow water than in the
deep water. Chen (1982) has discussed the penetration of
excess CO2 along the ENP section. Re proposed that the
deepening of penetration depth at 30°N (Fig.67) is not a
local phenomenon but is an advected feature derived from the
northwest North Pacific. The WNP section, 500 west of the
125
02
E
I
Fig.66
Correlation of [O.5iTA+ATCO2] with AOU for deep
water of the selected GS Pacific stations.
126
ENP-I9
16
14
13
so
a
6
4
2
0,
5O0F301
.
.
S
5000
.
S
S
pmollkg
I ZOO
30p4
20'N
Fig.67
Distrjbutjo of exce
cross_sj0
40ri
50N
CO2 along the 150°w
(taken froul Chen, 1982).
25°
20°
15°
30°
45°
40°
35°
I
Lo11
6
5
3
I-
S1a.#
5001
8
NN
9
12
I
14
16
18
5dN
I
20
I
:
E
10
40
imoI/kg
[MsI]
Fig.68
Distribution of Excess CO2 along the 165°E crosssection.
t'.)
I
128
ENP, shows a similar form as the ENP section but with
generally deeper penetration depth. The shoaling penetration
from west to east is also seen in tritium distributions
which are also supported by recent atmospheric inputs.
Fig.69 shows the east-west vertical. section along 35°N of
tritium reproduced from Fine, Reid, and Ostlund (1981). The
0.5 TU contour shoals from 900 m at 165°E to 600 in at 150°W.
The zero TU was not shown in their figure. If the zero TU
contour is assumed to be parallel to 100 a underlying 0.5 TU
contour, then the penetration depth from tritium data is
1000
in
at 165°E and 700 a at 150°W. This result is very
consistent with our results for excess CO2 (Figs.67 and
68).
Fig.70 shows the vertical section of tritium, also
reproduced from Fine et al. (1981), from selected GS
stations between 1S°-55°N between 165°W and 170°E (see
Fig.1). This section is located between our two sections.
Similar to our excess CO2 pattern, the 0.2 TU contour shows
an overall concave structure with the deepest penetration at
900 in or deeper at 38°N and the shallowest penetration at
400
in
at 10°N. It is significant to note that the tritium
data shown here are from the GEOSECS expedition which
occurred about ten years earlier than our ENP and WNP
expedition. Therefore, the penetration depth obtained from
2
226
214
213
212
204
202
201
500
:1:
[t1SI
Fig.69
I
60E
180
160W
I
140
Distribution of tritium along the 35°N. This figure is reproduced from
Fine et al. (1981).
I-.
I'.)
229
228
'1
227
214
215 216 217
218 219
U)
Lu
ILd
z
I
I-
0
Iii
IxoI-
I
20
40°N
Fig.70
Distribution of tritium at selected CS stations extending from 15° to
55°N between 165°W and 170°E (see Fig.1). This figure is reproduced from
Fine et al. (1981).
0
131
the tritium data shown here may be shallower than today's
depth of penetration since vertical diffusion and continued
advecion are still bringing the fossil fuel CO2 signal
downward over the last decade. An increasing penetration
depth of tritium has been observed in several regions of the
world oceans (Broecker and Peng, 1982). Sarmiento (1983a,b),
using tritium data from the North Atlantic, estimates that
penetration along isopycnals to depths of 700 m may occur
rapidly, on the order of 10-40 years, and that deeper
penetration occurs on a longer time scale. Jenkins (1980,
1982) also finds renewal rates to be on a decade scale for
the Sargasso Sea using 3He and tritium.
In Pig.68 the deepest excess CO2 signal can be found at
a depth of 1500 m (69a27.5) at 25-30°N where the contours of
the other physical and chemical properties also concave
downward. The 40 )lmol/kg contour north of about 40°N shows a
deeper penetration which is in contrast to the upwelling
phenomenon proposed in previous sections since upwelling
should bring the older subsurface water to the upper layer.
There is a possibility that the excess CO2 signal found in
this layer is attributed to the NPIW which originates from
its aeration source area (see 3.8 for the discussion of the
NPIW). In other words, it is possible that there is some
source for this water mass that is not of Southern Ocean
132
origin. This cannot be proven unless more information is
available in the northwest North Pacific Ocean. A dashed
line is drawn for the northern end of 40 pmol/kg contour to
show the uncertainty of the excess CO2 distribution.
4.E.(d) Penetration depth in the North Pacific Ocean:
The calculation of excess CO2 is subject to relatively
large uncertainty from various sources (Chen and Millero,
1979) and effects the accuracy of the calculation of the
depth of penetration. However, from the more precise data
sets we still can qualitatively estimate the penetration
pattern in the NorthPacific. Fig.71 shows the estimated
penetration depth in the North Pacific Ocean by using all
the available data sets. The overall pattern shows a
similarity with the oceanographic feature. The penetration
depth is generally deeper for those areas of high advection
and is shallower for more stagnant regions and upwelling
areas.
Several prominant features can be found: (1) A general
shoaling trend from west to east is observed; (2) a
deepening of penetration depth at 35-40°N which is
coincident with subtropical convergence; (3) a deep pool at
the western North Pacific, away from Japan, is observed. The
deepening is probably related to the interaction of Kuroshio
I!
140E
leo
lao
L
10
-
lao
"
4500
-
S-----------------
-
S
0
14öE
160
laO
160
140
12JW
Fig.71 Penetration depth of the fossil fuel CO2 in the North Pacific Ocean.
134
and Oyashio currents. The mixing of these currents enhances
the penetration of fossil fuel CO2 into the ocean; (4) the
penetration depth in the equatorial upwelling region is the
shallowest in the North Pacific Ocean. The penetration depth
is less than 300 m, especially in the eastern equator area.
135
5. Conc1usjo
With the emphasis on the chemistry in the North Pacific
Ocean, the following conclusions can be made:
* NTA, NCa, and silicate show a different distribution
than those properties effected by the shallow
decomposition of organic matter (AOL1, pH, NTCO2,
nitrate, phosphate). They generally increase
monotonically with depth. The reason for this is that
these substance are released by dissolution, enhanced
by pressure, while the soft organic matter is oxidized
within the water column. The oxidation occurs primarily
in the intermediate waters which bring a large organic
load from high latitudes.
* Chemistry data shows that the deep water of the eastern
North Pacific is older than that of the western North
Pacific.
* Alkalinty data is useful for the water mass analysis.
Each water mass shows its own variation trend with
temperature.
* Our work of alkalinitycalcium relationship shows that
preformed calcium concentration of the surface water
may not be the same as that of the subsurface water.
* Our carbonate data analysis shows that about a 25%
increase in total inorganic CO2 of deep water, after
136
leaving from the Southern Ocean to the North Pacific
Ocean, is contributed by inorganic CaCO3 dissolution.
* In comparing the Miller Freeman amd Discoverer data,
there is no significant difference of IC/OC ratio.
However, it was found that the eastern section has
higher TCO2 input than that of the western section due
to the older age of the deep water along the eastern
section.
* The saturation horizon of the North Pacific Ocean with
respect to calcite and aragonite generally shallows
from the west to the east and from the south to the
north. This depth was found at a depth much shallower
than that of the lysocline and CCD depth.
* A new set of preformed equation for the excess CO2
calculation for the deep water was obtained from the
revised GS data:
TA°(jieq/kg) - 2384 - 4.2 x 9 (+9)
TCO
(pmol/kg)
2219 - 11 x 9 (+16)
* The penetration depth of fossil fuel CO2 in the North
Pacific Ocean was constructed. The penetration depth is
influenced not only by the vertical diffusion but also
by horizontal advection. Pytnoclines also play a role
on the penetration depth in the region where the
pycnocline is found. The deepest penetration depth is
greater than 2000 m
,
which is found in the western
137
North Pacific Ocean. The shallowest penetration depth
is less than 300 m, which is found in the eastern
equatorial region.
138
6. References
AlvarezBorrego, S. and K. Park (1971). AOU as indicator of
waterflow direction in the Central North Pacific. J.
Oceanogr. Soc. Japan. 27: 142-151.
Berger, W.H. (1967). Porininiferal ooze: solution at depths.
Science. 156: 383-385.
Berger, W.N., C.G. Adelseck, Jr., and L.A. Mayer (1976).
Distribution of carbonate in surface sediments of the
Pacific Ocean. J. Geophys. Res. 81: 2617-2627.
Bernet, R.A. and P. Wilde (1972). Dissolution kinetics of
CaCO3 in seawater 1. Saturation state parameters for
ki'netics calculations. Am. J. Sci. 273: 826-839.
Bernet, R.A. and J. W. Morse (1974). Dissolution kinetics of
CaCO3 in seawater LV. Theory of calcite dissolution. Am.
J. Sci. 274: 108-134.
Bernet, R.A. (1976). The solubility of calcite and aragonite
in seawater at atmospheric pressure and 34.57.. salinity.
Am. J. Sci. 276: 713-730.
8rewer, P.G., G.T.F. Wang, H.P. Bacon and D.W. Spencer
(1975). An oceanic calcium problem? Earth Planet. Sci.
Lett. 26: 81-87.
Broecker, W.S. and T. Takahashi (1978). The relationship
between lysocline depth and in situ carbonate ion
concentration. Deep Sea Res. 25: 65-95.
139
Broecker, W.S. and T.-H. Peng (1982). Tracers in the Sea,
Chapt. 8, Lamont-Doherty Geological Observatory, Columbia
University, Palisades, New York, 690 p.
Byrne, R.H., J.G. Acker, P.R.. Betzer, R.A. Feely and M.H.
Cates (1984) Water column dissolution of aragonite in the
Pacific Ocean. Nature 312: 321-326.
Chen, C.-T. (1982). Oceanic penetration of excess CO2 in a
cross section between Alaska and Hawaii, Geophys. Res.
Lett. 9: 117-119.
Chen, C.-T. and R.M. Pytkowicz (1979). On the total
CO2-titration alkalinity-oxygen system in the Pacific
Ocean. Nature 281: 362-365.
Chen, C.-T. and F. Millero (1979). Gradual increase of
oceanic CO2, Nature 277: 205-206.
Chen, C.-T., R.M. Pytkowicz and E.J. Olson (1981).
Evaluation.of the calcium problem in the South Pacific.
Geochem. J. 16:1-10.
Chen, C.-T., F.J. Millero and R.M. Pytkowicz (1982) Comment
on calculation the oceanic CO2 increase: A need for
caution by A.M. Shiller. J. Geophy. Res. 87: 2083-2085.
Cline, J. (1982). Unpublished data from Discoverer.
May-Jun., 1982. NOAA PMEL.
Craig, H., Y. Chung and M. Fiadeiro (1972). A benthic front
in the South Pacific. Earth Planet. Sci.Lett. 16:
50-65.
140
Craig, H, W.S. Broecker, and 0. Spencer (1981). GEOSECS
Pacific Expedition, Vol.4, Sections & Profiles. National
Science Foundation, 251 pp.
Culberson, C., D.R. Kester, and R.M. Pytkowicz (1967). High
pressure dissociation of carbonic and boric acids in
seawater. Science 157: 56-61
Culberson, C.H. (1972). Processes affect the oceanic
distribution of carbon dioxide. Ph.D. Thesis, Oregon
State Univ., pp. 178.
Dodimead, A.J., F. Favorite and T. Hirano (1963). Salmon of
the North Pacific, Part 11: Review of the oceanography of
the subarctic Pacific region. International North Pacific
Fisheries Comm. Bulletin #13, 195 pp.
Duedall, I. W. and A.R. Coote (1972). Oxygen distribution in
the Pacific Ocean. J. Geophys. Res. 77: 2201-2203.
Dyssen, D. and I.. G. Sillen (1967). Alkalinity and total
carbonate in seawater; a plea for P-T-independent data.
Tellus, 19: 113-121.
Edmond, J. M. (1970). High precision determination of
titration alkalinity and total carbon dioxide of seawater
by potentiometric titration. Deep Sea Res. 17: 737-750.
Edmond, J.M. and J.M.T.M. Gieskes (1970). On the calculation
of the degree of saturation of seawater with respect to
calcium carbonate under in situ conditions. Geochim.
Cosmochim. Acta. 34: 1261-1291.
141
Edmond, J.M., Y. Chung and J.G, Sciater (1971). Pacific
bottom water: Penetration east around Hawaii. J. Geophys.
Res. 76: 8089-8097.
Edmond, J. ti. (1974). On the dissolution of carbonate and
silicate in the deep ocean. Deep Sea Res. 21: 455-479.
Favorite, F. (1967). The Alaska Stream. mt. North Pacific
Fish. Comm. Bull. 21: 1-20.
Feely, R. A. (1981). Unpublished data from Miller Freeman.
Jun.-Jul., 1981. NOAA PMEL.
Feely, R. A. (1982). Unpublished data from Discoverer.
May-Jun., 1982. NOAA PNEL.
Feely, R.A. and C.-T. Chen (1982). The effect of excess CO2
on the calculated calcite and aragonite saturation
horizons in the Northeast Pacific. Gephysic. Res. Lett.
9: 1294-1297.
Fiadeiro, M. (1980). The alkalinity of the deep Pacific.
Earth Planet. Sci. Lett. 49: 499-505.
Fiadeiro, M. (1982). Three dimensional modeling of tracers
in the deep Pacific Ocean: II. Radiocarbon and
circulation. 3. Mar. Res. 40: 537-550.
Fine, R.A., 3. L. Reid and H.G. Ostlund (1981). Circulation
of tritium in the Pacific Ocean. 3. Phys. Oceanogr. 11:
3-14.
Hattori, A. (1977) Preliminary report of the Hakuho Maru
Cruise KH75-4, University of Tokyo, 87 pp.
142
Hawley, J.E. and R.M. Pytkowicz (1969). Solubility of
calcium carbonate in seawater at high pressure and 2°C.
Geochim. Cosmochim. Acta. 33: 1557-1561.
Heath, G.R. and C.H. Culberson (1970). Calcite: Degree of
saturation, rate of dissolution, and the compensation
depth in the deep oceans. Geol. Soc. Am. Bull. 81:
3157-3160.
bribe, 'I., K. Endo and H. Tsubota (1974). Calcium in the
South Pacific and its correlation with carbonate
alkalinity. Earth Planet. Sci. Lett. 23: 136-140.
Ikegami,H. and S. Kanamori (1983).
Calciumalkalinitynitrate relationship in the North
Pacific and the Japan Sea. J. Oceanogr. Soc. Japan. 39:
9-14.
INDOPAC Expedition Data Report (1978). Physical, chemical
and biological data. University of California, Scripps
Institute of Oceanography. StO Ref. 78-21.
Ingle, S.E., C.H. Culberson, J.E. Hawley and R.M. Pytkowicz
(1973). The solubility of calcite in seawater at
atmospheric pressure and 35% salinity. Mar. Chem.
1:
295-307.
Jenkins, W.J. (1980). Tritium and 4He in the Sargasso Sea.
J. Mar. Res. 38: 533-569.
Jenkins, W.J. (1982). On the climate of the subtropical
ocean gyre: Decade time scale variations in water mass
143
renewal in the Sargasso Sea. J. Mar. Res. 40: 265-290.
Kanamori, S. and H. Ikegami (1982). Calcium-alkalinity
relationship in the North Pacific. J. Oceanogr. Sci.
Japan 38: 57-62.
Kenyon, K.E. (1978). The shallow salinity minimum of the
eastern North Pacific in winter. J. Phys. Oceanogr. 8:
1061-1069.
Kenyon, K.E. (1983). Sections along 35°N in the Pacific.
Deep Sea Reg. 30: 349-369.
Knauss, J.A. (1962). On some aspects of the deep circulation
in the Pacific. J. Geophys. Res. 67: 3943-3954.
Koczy, F.F. (1956). The specific alkalinity. Deep Sea Res.
3: 279-288.
Lisitzin, A.P. (1972) Sedimentation in the World Ocean. Soc.
Econ. Paleontol. Mineral. spec. pubi. 17, 218 pp.
Li, Y.-R., T. Takahashi and
W.S. Broecker (1969). Degree of
saturation of CaCO3 in the oceans. J. Geophys. Res. 74:
5507-5525.
Lyakhin, Y.1. (1968). Calcium carbonate saturation of
Pacific water. Oceanology (USSR) 8: 4453.
Lyman, J. (1956). Buffer mechanisms in seawater, Ph.D.
Thesis, Univ. of California, Los Angeles, pp. 196.
Mantyla, A.W. (1975). On the potential temperature in the
abyssal Pacific Ocean. J. Mar. Res. 33: 341-354.
Mantyla, A.W. and J.C. Reid (1983). Abyssal characteristics
144
of the world ocean. Deep Sea Res. 30: 808-833.
Masuzawa, J. (1972). Water charcteristics of the North
Pacific central region. In: Kuroshio, Physical Aspects of
the Japan Current. Stommel, H.S. and K. Yoshida, eds.
Univ. Washington Press, 95-127 pp.
Mehrbach, C., C.H. Culberson, J.E. Hawley, and R. N.
Pytkowicz (1973). Measurement of the apparent
dissociation constants of carbonic acid in seawater at
atmospheric pressure. Limol. Oceanogr. 1.8: 897-907.
Nillero, F.J. and A. Poisson (1981). International
one-atmosphere equation of state of sea-water. Deep Sea.
Res. 28: 625-629.
Morse, J.W. and R.A. Berner (1972). Dissolution kinetics of
calcium carbonate in sea water: II. A kinetic origin for
the lysocline. Am. J. Sci. 272: 840-851.
Olson, E.J. and C.T. Chen (1981) Interference in the
determination of calcium in seawater. Limnol. Oceanogr.
27: 375-380.
Park, K. (1966). Surface pH of the Northeastern Pacific
Ocean. J. Oceanogr. Soc. Kgrea 1: 1-6.
Park, K. (1967). Chemical features of the Subarctic Boundary
near 170 W. J. Fish. Res. Bd. Canada 24: 899-908.
Park, K. (1968a). The processes contributing to the vertical
distribution of apparent pH in the Northeastern Pacific
Ocean. J. Oceanogr. Soc. Korea 3: 1-7.
145
Park, K. (1968b). Seawater hygrogenion concentration:
vertical distribution. Science 162: 357-358.
Park, K. (1968c). Surface phosphate and silicate
distributions in the Northeastern Pacific Ocean. J. Fish.
Res. Bd. Canada 25: 2739-2741.
Park, K., S.W. Hager, J.E. Pirson and D.S. Ball (1968).
Surface phosphate and silicate distributions in the
Northeastern Pacific Ocean. J. Fish. Res. Bd. Canada 25:
2739-2741.
Peterson, M.N.A. (1966). Calcite: rates of dissolution in a
vertical profile in the central Pacific. Science 154:
154 2-1543.
Pickard, G.L. (1979). Water, salt and heat budgets of the
oceans. In: Descriptive Physical Oceanograph. Chapt
5.
3rd ed. Pergmoa Press.
Plath, D., K.S. Johnson, and R.M. Pytkowicz (1980). The
solubility of calcite probably containing magnesium in
seawater. Mar. Cheni.
10: 9-29.
Postma, H. (1964) The exchange of oxygen and carbon dioxide
between the ocean and the atmosphere. Netherlands J. Sea
Bes. 2: 258-283.
Pytkowicz, R.M. and D.R. Kester (1966). Oxygen and phosphate
as indicators for the deep intermediate waters in the
Northeast Pacific Ocean. Deep Sea Res. 13: 373-379.
Pytkowicz, R.M. (1970). On the carbonate compensation depth
146
in the Pacific Ocean. Geochiin. Cosmochim. Acta. 34:
836-83 9.
Pytkowicz, R.M. (1983). Solubility of carbonates and related
processes. In: Equilibrium, Nonequilibriuui and Natural
Waters, Chap. 5, V.11, John Wiley and Sons.
Reid, J.L. (1965). Intermediate waters of the Pacific Ocean.
Johns Hopkins Oceanographic Studies, No.2, 85 pp.
Reid, J.L. (1973). The shallow salinity minimum of the
Pacific Ocean. Deep Sea Res. 20: 51-68.
Reid, J.L. and R.J. Lynn (1971). On the influence of the
Norwagian-Greenland and Weddell Seas upon the bottom
waters of the Indian and Pacific Oceans. Deep Sea Res.
18: 1063-1088.
Sagi, T. (1969). The concentration of calcium and the
calcium chlorosity ratio in the western North Pacific.
The Oceanogrphical Magazine, 21.: 61-66.
Sarmiento, J.L. (1983a). A tritium box model of the North
Atlantic thermocline. J. Phys. Oceanogr. 13: 1269-1274.
Sartniento, J.L. (1983b). A simulation of bomb tritium entry
into the Atlantic Ocean. J. Phys. Oceanogr. 13:
1924-1939.
Shiller, A.M. and J.M. Gieskes (1980). Processes affecting
the oceanic distribution of dissolved calcium and
alkalinity. J. Geophys. Res. 85: 2719-2727.
Smetanin, D.A. (1968). Certain feature of meridional
147
distribution of chemical characteristics of water in the
Pacific Ocean. In: Chemistry of Sea Water. S.V. Bruevich,
ad., 197 pp.
Sverdrup, H.U., M.W. Johnson, and R.H. Fleming (1942). The
Oceans, their physics, chemistry, and general biology.
New York, PrenticeHall, 1087 p.
Takahashi, T (1975). Carbonate chemistry of seawater and the
calcite compensation depth in the oceans. In: Dissolution
of deepsea carbonates, Spec. Pubi. 13, p 159, Cushman
Found, for Foraminiferal. Res. Sliter,W.S., A.W.H. Be, and
W.H. Berger eds. Washington, D.C.
Takahashi, T., W.S. Broecker, A.E. Bainbridge and R.F. Weiss
(1980). Carbonate chemistry of the Atlantic, Pacific and
Indian Oceans: The results of the GEOSECS expeditions,
1972-1978. LamontDoherty Geological Observatory.
Takahashi, T., W. Broecker and A. Bainbridge (1981). The
alkalinity and total carbon dioxide concentration in the
world oceans. In: Carbon Cycle Modeling. Bolin, B. ed.
SCOPE 16, 271-286 pp.
Tsuchiya, M. (1968). Upper waters of the intertropical
Pacific Ocean. Johns Hopkins Oceanographic Studies, No.4,
50 pp.
Tsuaogai, S., it. Mishimora, and S. Nakaya (1968a). Calcium
and mangnesium in seawater and the ratio of calcium to
chlorinity as a tracer of water masses. J. Oceanogr. Soc.
148
Japan 24: 153-159.
Tsunogai, S., M. Nishimura and S. Nakaya (1968b).
Complexometric titration of calcium in the presence of
larger amount of magnesium. Talanta 15: 385-390.
Teunogal, S, E. Matsumoto, K. Kido, Y. Nozaki and A. Hattori
(1973). Two discontinuities in the deep water of
western North Pacific Ocean. Deep Sea Res.
the
20: 527-536.
Tsunogai, S. and ?. Watanabe (1981). Calcium in the North
Pacific water and the effect of organic matter on the
calcium- alkalinity relation. Geochim. J. 15: 95-107.
Tully, J.P. (1964). Oceanographic regions and processes in.
the seasonal zone of the North Pacific Ocean. In: SCudies
on Oceanogrphy. 68-84 pp.