USE OF HISTORICAL SALINITY DATA IN THE
COMPUTATION OF DEPTH CORRECTIONS
DUE TO VARIATIONS IN SOUND SPEED

Luis Maria Cabral Leal de Faria

<%0,

<><

\

NAVAL POSTGRADUATE SCHOOL

Monterey, California

THESIS

USE OF HISTORICAL SALINITY DATA IN THE
COMPUTATION OF DEPTH CORRECTIONS
DUE TO VARIATIONS IN SOUND SPEED

by

Luis Maria Cabral Leal de Faria

September 1980

Th

Th

ssis
ssis

Advisor: R. H. Bourke
Co-Advisor: G. B. Mills

Approved for public release; distribution unlimited

T197846

UNCLASSIFIED

SECURITY CLASSIFICATION OF THIS PAGE (Whan Data Entatad)

IOMTEBEY. CAuf ,&«,

REPORT DOCUMENTATION PAGE

READ INSTRUCTIONS
BEFORE COMPLETING FORM

1. REPORT NUMBER

2. GOVT ACCESSION NO

3. RECIPIENT'S CATALOG NUMBER

4. TITLE (and Subtitle)

Use of Historical Salinity Data in the
Computation of Depth Corrections Due
to Variations in Sound Speed

5. TYPE OF REPORT a PERIOD COVERED

Master's Thesis;
September 1980

6. PERFORMING ORG. REPORT NUMBER

7. AUTHORS

Luis Maria Cabral Leal de Faria

8. CONTRACT OR GRANT NUMBERC*;

9. PERFORMING ORGANIZATION NAME AND ADDRESS

Naval Postgraduate School
Monterey, California 93940

10. PROGRAM ELEMENT, PROJECT, TASK
AREA & WORK UNIT NUMBERS

11. CONTROLLING OFFICE NAME AND AOORESS

Naval Postgraduate School
Monterey, California 93940

12. REPORT DATE

September 1980

13. NUMBER OF PAGES

89

14. MONITORING AGENCY NAME a ADORESSf!/ dltterani from Controlling Otllca)

IS. SECURITY CLASS, (ol thla ripert)

Unclassified

Mm. OECLASSi-iCATION/ DOWNGRADING
SCHEDULE

16. DISTRIBUTION STATEMENT (o( :hi* Raport)

Approved for public release; distribution unlimited

17. DISTRIBUTION STATEMENT (ol tha abatract antarad In Block 20, II dltlarant from Raport)

18. SUPPLEMENTARY NOTES

19. KEY WORDS (Contlnua on tavaraa alda It nacaaaary and Idantlty by block number)

Echo-sounding

Corrections to echo-sounding
Vertical sound speed profiling
Velocity corrections

Hydrography
Coastal sound speed

20. ABSTRACT (Contlnua on ravaraa alda It nacaaaary and Idantlty by block numbar)

Six different methods for averaging historical salinity
values were studied for their applicability in the deter-
mination of depth corrections due to sound speed variations.
The averaging techniques range in complexity from the use of
a single value of salinity throughout the entire water column
to the use of a historical salinity average profile corrected

DO ,:°NRM73 1473
(Page 1)

EDITION OF 1 NOV 68 IS OBSOLETE
S/N 0102-014-6601 | ^

UNCLASSIFIED

SECURITY CLASSIFICATION OF THIS PAGe (Whan Data Bniarad)

UNCLASSIFIED

f u C u "1 TV CL ASSI^'C A TIQM Q> TwiS >>Qtf*»»n r>«r» t»l«f»j

for the surface salinity. Results show that historically
determined salinity can be used in the computation of the
depth corrections without exceeding the accuracy limits.
Two single values of salinity were found to be sufficient to
cover the West Coast of the United States: 31 %o , applicable
north of 40°N, and 35 %o , for use south of that latitude.

1473

DD Form
1 Jan 73
S/N 0102-014-6601

UNCLASSIFIED

SECURITY CLAMiriCATlOX Of THIS P *Gt.f**>»» Dmlm Cn„,.d>

Approved for public release; distribution unlimited.

Use of Historical Salinity Data in the Computation
of Depth Corrections Due to Variations in Sound Speed

by

Luis Maria Cabral Leal de Faria
Lieutenant Junior Grade, Portuguese Navy
Portuguese Naval Academy, 1976

Submitted in partial fulfillment of the
requirements for the degree of

MASTER OF SCIENCE IN OCEANOGRAPHY (HYDROGRAPHY)

from the

NAVAL POSTGRADUATE SCHOOL
September 1980

ABSTRACT

Six different methods for averaging historical salinity-
values were studied for their applicability in the determin-
ation of depth corrections due to sound speed variations.
The averaging techniques range in complexity from the use of
a single value of salinity throughout the entire water column
to the use of a historical salinity average profile corrected
for the surface salinity. Results show that historically
determined salinity can be used in the computation of the
depth corrections without exceeding the accuracy limits. Two
single values of salinity were found to be sufficient to cover
the West Coast of the United States: 31 %o , applicable north
of 40°N, and 33 %o » for use south of that latitude.

TABLE OF CONTENTS

I. INTRODUCTION 11

A. THE ECHOSOUNDER 11

B. ACCURACY REQUIREMENTS 13

C. SPEED OF SOUND IN THE SEA 14

D. TRADITIONAL METHODS FOR DETERMINING

DEPTH CORRECTIONS I5

1. Direct Comparison Systems 15

a. Leadline Comparison (Vertical Cast) 16

b. Bar Check I6

2. Direct Measurement of Sound Speed 17

3. Measurements of Salinity, Temperature

and Depth V

4. Historical Tables or Atlases 18

E. THE LAYER METHOD 18

1. Options 18

a. Curve-fit Method I9

b. No-curve-fit Method 19

2. Computation of the Depth Corrections 19

a. Layer Sound Speed 19

b. Layer Factor 19

c. Layer Correction 21

d. Bottom-of- layer Correction 21

e. Fathometer Depth Correction Table 21

f. Correction Curve 22

5

F. VARIATION OF THE PARAMETERS 22

G. USE OF HISTORICAL INFORMATION 24

H. OBJECTIVES 25

II. PROCEDURE AND METHODOLOGY 27

A. STUDY AREA 2 7

B. DATA SOURCES AND TREATMENT OF DATA 2 9

1. Historical Data 29

2. Observed Data 30

C. DATA ANALYSIS 3 5

1. Salinity 35

a. Gross Historical Salinity Average
(Method 1) 35

b. Fine Historical Salinity Average

(Method 2) 35

c. Historical Salinity Curve (Method 3) 3

d. Adjusted Historical Salinity Curve
(Method 4) 37

e. Inferred Value (Method 5) 37

f. Inferred Value with Surface Salinity
(Method 6) 39

2. Corrections 39

III. DISCUSSION OF THE RESULTS 45

A. USE OF HISTORICAL SALINITY DATA DERIVED

FROM ICAPS (FIRST FOUR METHODS) 45

1. Northern Area 45

2. Southern Area 48

3. Strait of Juan De Fuca and Puget Sound 49

'&

B. USE OF INFERRED SALINITY VALUE-

(FIFTH AND SIXTH METHODS) 49

6

1. Northern Area 49

2. Southern Area 51

3. Strait of Juan De Fuca and Puget Sound 51

IV. CONCLUSIONS 52

APPENDIX A: Wilson's Equation 55

APPENDIX B: Station Locations for Area and Season 59

APPENDIX C: Computer Program 67

LIST OF REFERENCES 84

INITIAL DISTRIBUTION LIST 87

LIST OF TABLES

I. Historical Salinity for the West Coast

of the United States, from ICAPS 32

II. Distribution of Stations by Area and Season 34

III. Historical Salinity Averages for the

West Coast of the United States 36

IV. Errors Resultant from the Use of Historical
Salinity Data Derived from ICAPS 46

V. Errors Resultant from the Use of a Single
Value of Salinity Throughout the Entire

Water Column 50

VI. Errors Resultant from the Use of a Single
Value of Salinity Corrected for the

Observed Surface Salinity 50

LIST OF FIGURES

1. Representative temperature, salinity and sound

speed profiles 12

2. Layer distribution for the no-curve-fit option

of the layer method 20

3. Depth correction diagram 23

4. Study area, showing the inshore limits of the

Alaskan and Californian water masses-- 28

5. (a) North Pacific Ocean locator chart, (b) Pacific
area A, and (c) Pacific area A water masses, as
defined in the ICAPS history file 31

6. Historical salinity derived from ICAPS 38

7. Correction curves for the first four methods 40

8. Depth corrections as a function of "true depth" (a)
and "true depth minus correction" (b) for true
salinity and method 1. Figure 8(c) illustrates

the difference between the two methods 42

9. Program output for a particular station 45

B-l. Location of stations for the northern area

and Strait of Juan De Fuca/Puget Sound, in winter-- 59

B-2. Location of stations for the northern area

and Strait of Juan De Fuca/Puget Sound, in spring-- 60

B-3. Location of stations for the northern area

and Strait of Juan De Fuca/Puget Sound area, in

summer 61

B-4. Location of stations for northern area and Strait

of Juan De Fuca/Puget Sound area, in fall 62

B-5. Location of stations for southern area, winter 63

B-6. Location of stations for southern area, spring 64

B-7. Location of stations for southern area, summer 65

B-8. Location of stations for southern area, fall 66

9

ACKNOWLEDGEMENTS

I would like to express my sincere appreciation to
Professor R. Bourke, thesis advisor, for his encouragement
and guidance throughout this project; to CDR. D. Nortrup,
Co-advisor before he left the Naval Postgraduate School for
a new assignment, and LCDR. J. Mills, co-advisor, for their
guidance, not only throughout the project but also throughout
all the hydrographic curriculum.

I am especially indebted to Mr. P. Stevens, of the Fleet
Numerical Oceanography Center, for his invaluable effort in
providing most of the data used; and to Mr. D. Mar, for his
assistance in computer processing. I also would like to thank
Mr. J. Green, NOAA, who provided part of the data used in the
project; and CDR. J. Compton, Royal Australian Navy, for his
cooperation in providing requested information.

Finally, I would like to express my appreciation to the
thesis typist, Ms. A. Schow, for her help and dedication.

10

1

I. INTRODUCTION

A. THE ECHOSOUNDER

Despite the development of laser- and photobathymetry ,
the echosounder, or fathometer, is still the most commonly
used method for the determination of depth in hydrography.
The transducer emits a sound pulse which is reflected by
the bottom and received back at the transducer. The time
of travel of this sound pulse is divided by two and this
value multiplied by the assumed speed of sound in sea-water,
thus giving the depth according to the expression z = v t
(distance = speed x time) . This transformation is made
either mechanically or electronically within the fathometer
itself, and the result displayed is the nominal or fatho-
meter depth beneath the transducer. This depth is not equal
to the true depth, since the assumed sound speed generally
does not equal the true speed of sound in sea-water, which
varies throughout the water column. One example of an
observed sound speed profile is presented in Figure 1.
For determining the true depth, one must know the sound speed
profile throughout the water column and apply it according to
the following expression [Greenberg and Sweers, 1972]:

Although sound speed is the correct term, most hydro-
graphic literature uses sound velocity instead.

11

off-o

QO'QS QO'OOl

QQ-QSt

00*002 00*0S2

(WJ Hld3Q

OO'OOS

OQ'OS£

oo'oon

OQ'OS OO'OOl 00*051

00*002 00*OS2
(W) HldHQ

00*006 00*056

00*00*1

oo -ds

00*

oJ\

00

^

oo-oda 00*0^2

(W) Hld3Q

oo* ode otros'e

00'

<u
l—l

■H

m

o

?H

o.

T3
0)

<u

a,
7i

G
O

-a

03

>s
■M
•H

c

•H
i— I

c/i

fH

■M
03

fH

<D

e
+->

>

■H
4->
CO
■M

c

0)

w

0)

fH

a
<u
as

5h
3
GO
•H

IX,-

12

1 r2At , . ,.
z = _ /q v (zt) dt

In this equation, v(z ) represents the sound speed at the
depth z where the signal passes at the time t.

Some echosounders have the possibility of being set to
different sound speeds, in order to represent as closely as
possible the average speed of sound in the water column. In
most cases, however, they are constructed with a calibrated
speed of sound assumed in the device, usually 1463.04 or
1500 m/s (800 or 820 fathoms per second) . The echosounders
used by the National Ocean Survey (NOS) are calibrated for
an assumed speed of sound of 1463.04 m/s which, according to
Umbach [1976] is a value reasonably close to the sound speed
in most waters surveyed. Corrections are then applied to
the depths obtained by the fathometer in order to compensate
for the use of a sound speed other than the one that actually
affects the travel of the sound pulse.

B. ACCURACY REQUIREMENTS

The International Hydrographic Bureau (IHB) established
in 1968 the following requirements for accuracy of depth
determination [International Hydrographic Bureau, 1968]:

2

Although these are depth corrections, the term usually

applied in hydrographic literature is sound velocity correc-
tions .

13

Depth Range Accuracy Requirement

0-11 fathoms (0- 20 m) 1.0 ft

11-55 fathoms (20 - 100 m) 3.0 ft

deeper than 55 fathoms \% of the depth

The above standards represent the maximum allowable error in
the measurement of depth due to all sources, including the
variations of sound speed within the water column. When con-
sidering errors due to sound speed alone, no errors can be
equal to or exceed 0.25% of the depth, which means that the
mean sound speed must be known to within ±4 m/s [Umbach, 1976],
The standards mentioned above apply only to conventional echo-
sounding systems, in which the sound pulse beam is vertical.
Other systems, applying oblique acoustic paths, require higher
accuracy standards.

C. SPEED OF SOUND IN THE SEA

Speed of sound in a fluid depends on the density (p) , and
compressibility (k) of the fluid and the ratio between the
specific heats of the fluid at constant pressure and at con-
stant volume (y) , according to the following expression
[Sverdrup, Johnson and Fleming, 1942; Bowditch, 1972]:

-V

v —

Pk

As these parameters depend on temperature, salinity and
pressure (which depends on depth) , it is more convenient

14

to measure these properties instead, and compute the sound
speed from them.

Different expressions have been developed empirically
by Kuwahara [1939], Del Grosso [1950], Wilson [1960a, b],
Leroy [1969] and, more recently, Clay and Medwin [1977] .
Mackenzie [1960] compared several of these formulas, which
showed differences between them of less than 3 m/s, or 0.2%
at zero depth [Urick, 1975] . Speeds used for depth correc-
tions in NOS are calculated using Wilson's equation [1960b],
which is described in Appendix A. This equation is the most
commonly accepted formula due to its large range of applica-
bility and its relative accuracy (0.30 m/s standard deviation)
[Wilson, 1960b, 1962].

D. TRADITIONAL METHODS FOR DETERMINING DEPTH CORRECTIONS
Several different procedures can be used to determine
the depth corrections, or correctors, to be applied to echo-
soundings. These corrections can either be found directly
or by knowing the sound speed profile. The sound speed can
either be measured directly (direct comparison systems) or
computed after the determination of some characteristics of
the water column, as stated above (oceanographic and limno-
logic determination) . Some of the different procedures are
described below:

1 . Direct Comparison Systems

The depth recorded by the echosounder is compared
to the actual depth indicated by the leadline or the depth

15

at which the bar is lowered. Because these are comparative
methods, not only are sound speed variations included in the
results but also instrument errors [Umbach, 1976] . Both
methods are limited to shallow depths and good weather con-
ditions .

a. Leadline Comparison (Vertical Cast)

The leadline is lowered to the bottom at several
depths over a range which should be at least as great as that
likely to be encountered in the survey area [The Hydrographer
of the Navy, 1969] . Several readings are taken from the
fathometer at each depth, and all are plotted individually
in order to eliminate random errors. In areas where launch
hydrography joins or overlaps ship hydrography, it is of
particular importance to use this method on both ships and
launches in order to resolve possible discrepancies in the
soundings [Umbach, 1976]. This method is limited to calm
seas, shallow depths, and areas of hard bottom for accurate
readings .

b. Bar Check

The depth of a bar indicated by the echosounder
is compared to the known depth at which the bar is lowered
below the transducer. The procedure is used at several depths
and the readings are taken both when the bar is lowered and
when it is raised. Since the bar is handled over the side,
this method is usually limited to small vessels and shallow
depths, although it is sometimes performed mechanically on

16

large vessels. The depth limitation still holds, however,
due to difficulties in maintaining vertical lowering lines
and in keeping the bar within the beam of the transducer.
This procedure should only be used in calm seas with light
winds and no strong current.

2 . Direct Measurement of Sound Speed

Sound speed can be measured directly by lowering a
sound velocimeter to various depths and recording the speeds
indicated. Although accuracies obtained by this method
can be of the order of a few tenths of a meter per second
[Ulonska, 1972], sound velocimeters are not generally used
because they are very expensive instruments [Calder, 1975].

3. Measurements of Salinity, Temperature and Depth
These parameters can be determined by direct meas-
urement either with Nansen casts (temperature and salinity
at specific depths) or with continuous sampling instruments
which measure salinity (or conductivity) , temperature and
depth (STD or CTD) . The latter instruments can also be of
an expendable type. Sound speed at various depths is then
computed using an equation relating salinity, temperature
and depth (pressure) to sound speed, such as Wilson's equa-
tion [1960b], or they can be extracted from tables, wherein
sound speed is tabulated for different values of temperature,
pressure and salinity [Heck and Service, 1924; Matthews,
1927, 1939; Kuwahara, 1939].

17

4. Historical Tables or Atlases

Sound speed or salinity and temperature data have
been tabulated for various areas and seasons. Matthews'
tables [1939] provide corrections to be applied to the depths
obtained with fathometers calibrated to a sound speed of 1463
or 1500 m/s. These depth corrections are given as a function
of nominal depth and are tabulated for 52 different oceanic
areas identified by Matthews [Greenberg and Sweers, 1972].
This method is not used by NOS but it is used in some coun-
tries for depths exceeding 200 m. For example, this is the
only method which Australia uses for determining corrections
in water depths of more than 200 m [Compton, 1980] .

E. THE LAYER METHOD
1 . Options

In addition to the above techniques to establish the
depth correction, yet another more sophisticated method is in
widespread use--the "summation of layers" method [Umbach,
1976] . This method is applicable when the sound speed is
determined either by direct measurement or by computation from
the depth, temperature and salinity parameters of the water
column. This method can be applied in two ways, which are
described in the "Hydroplot/Hydrolog Systems Manual" [Wallace,
1971] :

3

The Hydroplot/Hydrolog system is an automated data aqui-
sition and processing system in use by NOS, respectively with
and without a plotting capability. A description of this
system is given in the Hydrographic Manual, Fourth Ed. [Umbach,
1976].

18

a. Curve- fit Method

The water column is partitioned into layers of
varying thickness which are predetermined by the operator.
Layer mid-depths are then computed and the sound speed at
those depths interpolated either graphically or mathemati-
cally from a curve passed through the points at which the
sound speed has been computed.

b. No-curve-fit Method

The water column is partitioned as above but the
depths at which the observations have been made are consid-
ered as the mid-depths, the boundaries between layers being
defined as half-way between two successive observations of
the cast. One example illustrating the no-curve-fit method
is shown in Figure 2.

2 . Computation of the Depth Corrections

a. Layer Sound Speed

For each layer a sound speed is determined for
the corresponding mid-depth by any of the methods described
above. This value of sound speed is assumed to apply through-
out the entire layer.

b. Layer Factor

A factor is computed for each layer, based on the

sound speed for which the instrument is calibrated, using the

following expression [Umbach, 1976] :

A. - C
factor. = — =

19

§1478.00

1480.00

SOUND SPEED (M/S)
1482.00 1484.00 1486.00

1488.00

00

Figure 2. Layer distribution for the no-curve-fit
option of the layer method.

20

where A. is the actual sound speed for the i-th layer which
has been determined for the corresponding mid-depth (see
preceding paragraph) , and C is the assumed sound speed of
the instrument.

c. Layer Correction

The above factor is multiplied by the layer
thickness, Az. (the difference between the depth at the
bottom and at the top of the layer) in order to obtain the
correction to the depth (LC-) due to the variation in sound
speed of that layer:

LC • = factor . x Az .
1 11

The water column above the transducer is eliminated when
determining the corrections for the upper layer.

d. Bottom-of- layer Correction

The layer corrections are added from the surface
downward in order to obtain the total correction at the bottom

of each layer:

i
C = I LC

1 3=1 J

where C- is the correction at the bottom of the i-th layer.

e. Fathometer Depth Correction Table

The above corrections correspond to the true
depth of the bottom of each layer. To obtain the fathometer
depth to which these corrections will apply, one must sub-
tract the correction (C . ) from the true depth of the bottom

21

of the corresponding layer (z ,) :

CFD. " zi " Ci

1

where Ccr. is the fathometer depth corresponding to the i-th

1

layer .

f. Correction Curve

The values of the corrections are plotted against
the fathometer depth. A curve is drawn to fit these depth-
correction pairs. From this curve the depth correction appli-
cable to any reading of the fathometer (within the range for
which the curve has been computed) can be obtained. An example
of such a curve, for which a transducer at zero depth has been
considered, is shown in Figure 3.

F. VARIATION OF THE PARAMETERS

The variation of sound speed with depth, temperature
and salinity depends on the initial conditions of these par-
ameters. For a temperature of 14°C, salinity of 34 %o and at
zero depth (sound speed under these conditions is 1502.94 m/s)
an increase of one degree in temperature will increase sound
speed by about 4.5 m/s; an increase of 1 %o in salinity will
increase the sound speed by about 1.3 m/s; an increase in
depth of 100 m (about 10 atmospheres in pressure) will in-
crease sound speed by about 1.7 m/s [Urick, 1975]. Latitude
also has some influence, since pressure increases from the
equator to the poles because of the increase of the gravita-
tional attraction. This influence, however, is very small

22

§D. 00

CORRECTION
2.00 3.00

(M

Figure 3. Depth correction diagram

23

and can be ignored. An example of the variation of sound
speed with temperature, salinity and depth is shown in
Figure 1.

Temperature is the parameter that has the largest influ-
ence on sound speed. Also, it shows more variability with
time. Umbach [1976] states that, for the sound speed to be
accurate to within ±4 m/s, temperature and salinity must be
determined with an accuracy of +1°C and 1 %o , respectively.
However, recent studies have been performed by the Testing
Division, Office of Marine Technology (OMT), NOAA, regarding
the influence of the different parameters on sound speed.
These studies show that if temperatures can be determined
with an accuracy of ±.1°C, the salinity accuracy need only
be within ±3%o to still meet the requisite accuracy standards
[Bivins, 1976].

G. USE OF HISTORICAL INFORMATION

Historical information has been used for some time in the
computation of depth corrections for echosoundings , but usually
only for depths in excess of 200 m [Sherwood, 1974; Urick,
1975]. Yeager [1979] studied whether depth corrections of
sufficient accuracy could be determined from historical salin-
ity and temperature only. He chose an area which is repre-
sentative of the near shore region along the East Coast of
the United States. The historical sound speed data there
indicated that the temporal and spatial variability of tem-
perature exceeded the acceptable limits for sounding

24

corrections, thus precluding the use of historical data alone
for correction determination. In his study he also concluded
that, in that area (a little more than one degree square with
depths less than 200 m) salinity values were sufficiently
stable to allow acceptable depth corrections to be computed
from historically derived salinity values combined with in
situ temperature measurements.

In Australia data have been collected over the contin-
ental shelf since 1965 which show salinity to vary generally
between 54.5 and 36 %o , changing only slightly with depth
down to 300 m [Calder, 1975]. At present, the Australian
Hydrographic Service exclusively uses a value of 35 %o for
the calculation of sound speed in depths less than 200 m
[Compton, 1980].

H. OBJECTIVES

The purpose of this thesis is to analyze the possibility
of using historical salinity data averages for a large area
with in situ temperature measurements to compute depth cor-
rections and still remain within the required accuracy limits
The historical salinity data were treated using several dif-
ferent techniques which increased in accuracy as more complex
averaging schemes were used.

Temperature-salinity observations generally have to be
made at least once a month and be of sufficient number to
cover the entire area sounded [Umbach, 1976] . The use of

25

historical salinity data and XBT (expendable bathythermo-
graph) temperature information could save a significant
amount of time and money if results remain within the
accuracy standards. Also, since little time is needed for
each XBT observation, more frequent temperature measurements
could be made (both in time and space) , thus increasing the
accuracy of the corrections used throughout the survey.

26

II. PROCEDURE AND METHODOLOGY

Some of the terms to be used in the following chapters

are defined below:

Historical data: data resultant from some averaging
scheme applied to previous observations in a large
area.

Inferred data: data obtained from both historical and
observed salinities, used to represent the data for a
large region. It has not been measured, and consists
of a single value.

Station: group of measurements taken at a particular
time in a particular location. Each station consists
of several observations taken over a given depth
interval .

Observation: measurement taken at a particular depth
in one station.

True correction: depth correction resultant from the
application of the observed data.

Estimated correction: depth correction for which the
computation of some historical data were used.

Error: difference between the estimated and true
corrections .

Percentage of error: division of the error by the
depth over which the correction applies.

A. STUDY AREA

The area selected for this study was the continental shelf
region of the West Coast of the United States, extending off-
shore generally to depths of about 1500 m. The study area
can be seen in Figure 4. This area was chosen on the basis
of several factors:

27

49<

40°N

32'

Alaskan
water
mass

Study
area —
limits

Californian
water mass

Puget Sound

Strait of Juan De Fuca

127°W

(

ICAPS
'mass limits

122*

Figure 4. Study area, showing the inshore limits of the
Alaskan and Californian water masses.

28

1. It represents a fairly large area wherein a consider-
able amount of data were available. For the purpose of this
study, a large area was desired to generalize the applica-
bility of the results.

2. Most hydrographic surveys are conducted over the
continental shelf region.

3. The area was well suited for acquisition of histor-
ical data, as will be seen later in this chapter.

4. The extension to 1500 m depths guaranteed that all
shelf features, including submarine canyons, would be con-
sidered.

The study area was subdivided into three sub-areas:
Northern, Southern and Strait of Juan De Fuca and Puget Sound.
This latter area was considered separately because of its
particular salinity characteristics. The North-South division
of the Pacific Ocean waters along latitude 40°N corresponds
to the separation of the two different water masses consid-
ered in the historical salinity file.

B. DATA SOURCES AND TREATMENT OF DATA
1 . Historical Data

Historical data for this study were obtained through
the PROFIL subroutine of the Integrated Command ASW Predic-
tion System (ICAPS) . The ICAPS water mass history file was
developed by the U.S. Naval Oceanographic Office (NAVOCEANO)
and contains historical information on the water masses of

29

the Northern Hemisphere and Indian Ocean [Department of the
Navy, 1978] . Each ocean is divided into several areas and
subareas . In each of these subareas up to five different
water masses may be included. Figure 5 shows the North
Pacific Ocean, Pacific Area A divisions and the water masses
present in Area A. The water mass files are also grouped
seasonally: winter (January through March) , spring (April
through June) , summer (July through September) , fall (October
through December) .

Within the study area the ICAPS water mass file con-
sists of only two water masses: Alaskan water (north of 40°N
and Californian water (south of 40°N) . These are shown in
Figure 5(b). The inshore limits of these water masses are
shown in Figure 4. The Alaskan water mass was used for the
near shore area off Washington, although this area is not
included in either of the two water masses.

For ICAPS data retrieval and processing, several dif-
ferent program modules are available [Chace, 1979]. In this
study only the PROFIL subroutine was used. One of the out-
puts of PROFIL is historical salinity. It was used to obtain
information for each area during all four seasons. The
salinity profiles from these eight runs were punched onto
cards and are shown in Table I.
2 . Observed Data

The main source of observed data was the Fleet Numer-
ical Oceanography Center (FNOC) file entitled MOODS (Master

30

'"'-V. k-

J

<

.c

f>-^

f ^y c

*o .

gO*E IO»'E lio'f. IT*'E 190* E 163* E 180* E 183' W 130* W 139* W l»' W I0»' W 90" * 75* W

Figure 5(a)

SON

40N

33N -

JON

Figure 5(b)

Region

Water Mj»» jUgj

T200
IUn

<*C)

Max

BT (*C)
Hln K»* Position

*«<1. K)

At

TRANS 1 Tint)

KHROSIMO

7

n

1)

1»

35

55

A2

PAST TRANSITION

10

16

100

AJ

NOBPAr

FAST TRANSITION

7
12

12
16

45

57

A*

CAI.IPORNIAN
FAST TRANSITION

5
11

11
1»

60
40

A5

CAUFMRMAN

5

11

inn

M

cin r nr California

10

1(

inn

a;

ALFUTIAN
NIlRFAf

0

7

7
12

8

92

AA

ALASKAN

NORPAC

0
7

7
12

46
54

A3

AI.ASKAN

0

t

inn

AJO AJFIFTIAN

Figure 5(c)

0

A

100

Figure 5(a) North Pacific Ocean locator chart, (b) Pacific
area A, and (c) Pacific area A water masses, as defined
in the ICAPS history file.

31

o

xf

A

X

H

►—

CO

a:

<

O

O 00

2

U 0-.

<

H U

CO HH

w

3= s

o

w cs;

X tu

H

HH

oi oo

o w

w

tU E-

-3

<

CQ

>H H

<

H 00

H

i— i

z q

«*

i— i w

7

-J E-

<M

a

00 2

vf

X

V

J

< W

OJ X

r

hh E-

h-

oi

3

O P-.

'.J

H O

■>o

00

u.

'/I

l/J

£

*0>C T -t^-J" ^tti corner- (viot^-o 'TMTKVig- ^lt
"*" r^> -n »t jj lT -j -f >o ^O er 00<-<mr>tirin>0 o

r\j r j r\i r\ j rNj o.i ro ro r«~ ro ro r«-> <r xt vt >»■>*■ <T <T - 1 ^J" ^t

j-OCT 'Njrr. ^o cm — ■■ «f f*- (^~> vC <"° o Of Oct lti — *rr PO tr
—.fNir^ir OX — <»t o'1 ^T-O — irH-^r vfLPiLf>-0vC

rVJ(xj'VJrg(Vj(VirO'nrororOr^ 4" >T -3" -J- >}■ -J- ^vj-st^

.j-X—'T T (VlCTO— *CT vf *— r'**— «'0--'OU~\(VjaNCr!in
r^ rO m rO f"< rn -oi^r^^^'n(^rnrr*',0(-^"nrnrnrnrn

■XT -"*,"> O^vj— "Vjf\iC7'vfr-"riO'X3 0 T*ir\oj0>^m
■j- ,*■ .n m >n -o 'vj '.r\ r— oo cr ct- o — «'-*'"'*ic'*>>j'i.rMrv-o>o

(\iivj(\if\i(Vjrsj'fic0rOrOfOfi^'>^'4* ^"■^■>J"^n^«^'>^^"

>^^rO'^'^'*^rn'^r^^rOf^rOro(*^ror'"'^ror**",^ro

<r -r ^- ^n rn ^- x5 r^ X) o, — <— ' <vi r\j "^ -4- vj"X m *£ o -o

o c •ONNSOm>r-lo,<roinoavvO-,ii1-*>tO
t«i^nr,srONcooQ-,(vJM'ririj"ifMnii>cxi

f r»^ ^<^" ^- ^- vCr—TjOO-^fVi'Nir^.-n >ri/>ir. >c c o
rn,*-in(T>r^.;OrO-nf<"ivrJ">3'vt<f>3"J'^"<r-t-3"vt-r

— i - j n xn r- d ji^o -f* o ) ' J 3 nj j^ o -•

— <_.— <^j-\j n fin o oo \j-f"">.i"\ "5

32

Oceanographic Observation Data Set) . Some additional data
were obtained from the Pacific Marine Center (PMC) , NOAA,
Seattle, Washington. The MOODS file contains data from sea
water observations all over the world. Different types of
observations are included: STD , CTD, Nansen cast, BT or XBT,
and others. Each station contains the geographical location,
date, time and a code for the ship/agency that made the ob-
servation. Temperature, salinity, sound speed, oxygen con-
centration or other properties are also provided depending
on the type of observation. The depth interval between
observations for different stations is not standardized.

The time interval chosen for the study was the ten
year period from 1964 to 1973. A more recent period was not
chosen because a very large percentage of the recent data is
from XBT observations, which does not provide salinity infor-
mation.

A direct extract from the MOODS file was not readily
usable on the Naval Postgraduate School computer since the
MOODS data are stored with a variable block size. A nine
track tape, 1600 bits per inch, constant block size, was then
prepared at FNOC which contained temperature and salinity for
depths standardized to 0, 30, 60, 90, 120, 150, 175, 200, 250
and 300 m. Values in excess of 300 m were not given, even if
they had existed in the original data. In the meantime, some
data were punched onto cards, extracted from a printout ob-
tained from the original data. Since this data set had a

33

better profile definition, especially useful for the inter-
mediate values between the surface and 30 m, it was decided
to use this punched data in substitution for the correspond
ing stations on the tape.

A total of 3459 stations were analyzed, with the
distribution shown in Table II. For each of these stations
a depth correction was calculated based on the procedures
outlined in the previous chapter. Such corrections are con
sidered true depth corrections and are the values against
which the estimated depth corrections (based on some mean
historical salinity value) must be compared.

TABLE II
DISTRIBUTION OF THE STATIONS BY AREA AND SEASON

Total

Winter Spring Summer Fall Total

North

248

679

643

195

1765

South

450

350

229

193

1222

Puget Sound/

Strait of

127

149

100

96

472

Juan De Fuca

3459

Appendix B shows all the station locations for each
area and season. Multiple occupations of a station may have
been made which are not indicated by these figures.

34

C. DATA ANALYSIS
1 . Salinity

From the ICAPS data file, mean salinity values were
computed from the surface to a given depth, for each area
and season. This mean was determined not only for the depth
of each given value of historical salinity (Table I) but also
for the mid-depths between two successive values. Table III
contains the values obtained. These data were then used to
establish a variety of historical salinity averages to be
described below.

Estimated depth corrections were then determined for
each station using the observed temperature and an historical
salinity value derived from one of four different methods:

a. Gross Historical Salinity Average (Method 1)
This method uses a single value of salinity

throughout the entire water column. This salinity value
was determined by averaging the historical salinity data
from the surface to 200, 1000 or 3000 m. The choice of the
depth depended on whether the deepest depth for the station
was less than 200, between 200 m and 3000 m, or greater than
3000 m.

b. Fine Historical Salinity Average (Method 2)
This method was similar to the previous one

except that, instead of three possible values for salinity,
ten possible values were considered: averages from the sur-
face to 50, 100, 150, 200, 300, 500, 750, 1000, 1500 and

35

TABLE III

HISTORICAL SALINITY AVERAGES FOR THE
WEST COAST OF THE UNITED STATES

SOUTH ( <4Q N)
S S

NGKTH D'.J N)

0.0
5.0
10. 0
15.0
2u.O
25.0
30. 0
4C.0
50.0
62. 5
/5.0
d7. 5

100. 0

112.5

125.

1 37,

150,

i7e.

2oO,

225.0

250.0

2/5.0

JOG. 0

150. 0

400.0

450. 0

500.0

550.0

60C. 0

750.0

800. 0

900. 0

1000. 0

1 luO. 0

12OU.0

U50.0

1500. 0

1 750.0

2030.0

2250.0

2500.0

2750.0

3000.0

33. 250
33. 35l

33. 250
3^ .350
33. 250
33 . 350
3J .352
3 3 . 3 i 4
3 3.3-jS
3 3.363
3 3.37i
33 .3o2
3 3. 4»j3
3 j . 4 2 0
33 .44 7
3 3 . 4 t S
3 3.49a
33.544
33.401
33.645
J3.tSC
33.726

33 . 700
33. t 13
3 3.061

33. as9

23.S24
33.963
3 3 . S S 1

34 .053
3 4 . C 7 4

34. I 10
34 .145
34. 1 y4
34.201
31.234
3i -2cfa
34.306
J4.343
34.372
34.398

33. 36 0
33. 36 0
33. 360
33 .360
33. j60
33. 360
33.362
33. 364
3 3.373
3 3 . 3 3 J

33. 39 7
33.409
33.433
33.451
33.479
33.502
33.531
33.576
33.632
33.675
33.716
33.750
33.7 63
33. 034
33.000
33. SI 5
3 3.95 0
3 3.9 7 a-
34.005
34.06 4
34.06 4

34. lid
34.152
34. I fll)
34.20 7
34.239
34.271
34. J 10
34.347
3^.375
34.40 1

3 3.400
3 3.4 00
3 3.395
33.393
3 3.390
33.388
3 3.385
3 3. ? ^ l
33.379
3 3.377

33. ^9 3
3 3 .404
3 3.428
3 3.447
33.476
33.499
33.527
33.572
33.627
3 3 .669
33.711
33.746
3 3.77 0
3 i. 6 10
33 .8 7b
33.912
3 3.946

33 .974
34.001

34 .0 6 1
3 4.0 8 1

34. 1 16
34.150
34.178
34 .?C6
34 .240
34.271
34.311
34.343
34.377
34.403
34.425

34.4^0 34.423 34.425
24.440 34.442 34.444

3 3 . 40 0
3 3.40 0
3 1. 400
3 3.400
33.400
5 3.400
33.390
3 3 . 3 3 6
3 3.339
3 3.383
33.3«6
33.405
3 3.42 9
33. 44 8
33.47d
3 3.50 1
33.531
33.57b
3 3.632
3 3.674
33. 71 7

33. 752
3 3.78 4
3 3.835
33.082
33.918
3 3.95 3
2 3. 9 12
34.009
34.070
34.090

34. 125
34.159
34.187
34. 214
34.24 7
34.277
34.316
34.352
34. 379
34.404
34.425
34.444

12.

480

32.

, 24 0

32.

140

32.

3 6 0

32.

400

32.

,24 J

32.

, 140

22.

,3 60.

32.

185

12.

263

■ 32.

170

32.

,:>60

3 2.

4b7

52.

, 2o 7

32.

180

32.

3 60

32.

4S2

32.

,302

12.

232

32.

,367

3 2.

4 96

32.

324

32.

2 54

32.

,372

■3 ■> <

502

32.

,352

2.2.

,307

32.

,333

3 2i!

5GS

32 ,

,3 86

32.

, 3 60

32.

,397

32.

52 3

32.

,427

32.

4 24

32.

448

32.

,534

32.

, 4o 0

32.

,475

32.

,4 63

32.

5»12

32.

5<lQ

3?.

5 3 9

12.

564

32.

616

12.

, 5(1.1

32,

5 35

32.

617

32.

690

12.

,641

32.

,6 5 2

32.

,694

32.

74 8

1 2.

70 2

32.

7 04

32.

, 754

32.

825

12.

,7o2

32.

,775

32.

,826

3?.

tltiO

3 2 .

,jW

32.

,5 32

32.

,8 36

32.

S5H

32.

,910

32.

^JCH)

32.

,9 52

33.

,06b

33.

,0 3 2

33.

,005

33.

,0 56

33.

16 9

33.

139

33.

1 13

33.

,159

33.

, 2 4S

11.

, 21 3

3 3.

137

33.

239

3 3.

3io

33.

,29 4

33.

,271

33 .

.308

3 3.

3 75

_7 «• 4

253

33,

3 31

33,

,365

33.

,125

3 3.

,>U4

11.

,383

33,

,4 14

3 3.

502

3 3 ,

,4o5

33,

,465

33.

,492

33.

,568

ll.

.553

J i.

5 34

33.

5 5 f J

33.

,62u

33.

,60 6

33.

,591

33.

,609

33.

068

3 3.

657

3 \.

,b'*2

33.

,657

3 3.

, 7u7

3 2 .

,6y8

3 3,

,633

33.

,697

3 3,

, 74 t>

33.

.738

3 <,

.725

33 ,

.736

3 3.

M33

3 3.

, 8^:6

2 3.

, 8 16

33.

,82 3

33.

, Jf 2

33.

.85 7

33.

,846

33.

,853

3 3.

,911

33.

■ SO 7

33.

.39 7

33,

,902

3 3.

. 1>6'V

33.

.956

33.

,9 46

33.

,951

2 3 .

,990

33.

.99 7

33,

.9 86

33.

.99 1

34.

036

34.

,03 4

34.

,025

34.

,029

34.

,0b2

34,

.08 1

34.

.0 72

34,

.076

3 4.

, 126

34.

.12^

34,

.1 16

34,

.120

34.

182

34,

, 161

34.

, 172

34.

,178

34.

.233

34,

.212

34,

.224

34.

.229

34,

,2 73

34,

.272

34.

,265

34.

,269

34.

, 30b

34.

. 30 8

34.

, 301

34.

, 305

34,

.337

34,

.33 7

34,

.331

34.

.335

34.

, 363

34,

, 363

34,

,358

34,

.361

36

3000 m depth. As before, the depth greater than or equal to
the deepest depth observed was chosen. For example, if the
depth were 250 m, the salinity average would be the one cor-
responding to the 300 m water column.

c. Historical Salinity Curve (Method 3)

The value from the historical salinity curve,
which corresponded to a given depth, was obtained by interpo-
lation from the existing values extracted from the ICAPS file.

d. Adjusted Historical Salinity Curve (Method 4)
Since the salinity value observed at the sea sur-
face can sometimes deviate considerably from the historical
surface value, an alternative to the historical salinity curve
was considered. A linear adjustment was made to this curve

so that the surface value would equal the observed salinity
value, and the value at 30 m would equal the historical data
(ICAPS) at that depth. Below 30 m the historical curve was
used.

Salinity values illustrating these first four methods
as applied to a particular station are shown in Figure 6. In
light of the results obtained using these four methods, two
new approaches were taken:

e. Inferred Value (Method 5)

A single value of salinity was used for each area,
independent of season and depth. This value, which was inferred
from both the historical data and the actual observations, was
judged to represent average conditions for the area.

37

§3

SRL1NITT (PPT)
0.00 31.00 32.00 33.00 34.00 35.00

j I L

3

o

C^^^^ L -Method 3

^V\— Method 4

Method 1

-Method 2

o
o

•

1/5

Observed

O

o

o

o
o

o

\ 1

DEPTH

,00 200

1

11

o

o
o

,

o

o_

m

\,

o
O

■

o
m

o

o

■

o

o_

l
i

i
i

36.00

j

Figure 6. Historical salinity derived from ICAPS

38

f. Inferred Value With Observed Surface Salinity
(Method 6)

This method is similar to the previous method,
except that the salinity which is applied to the first layer
is the observed surface salinity. This method is also simi-
lar to method 4, except it only uses one value of salinity
throughout the column for depths other than the surface.
2 . Corrections

Each of the six methods was applied as in a normal
survey. Estimated depth corrections were computed for each
layer using the no-curve-fit method, i.e., the depth of the
observed temperature was used as the mid-depth for the layer.
The procedures indicated in section I.E. were followed using
steps a to d:

(a) Estimated sound speed for each layer.

(b) Computation of the corresponding factor.

(c) Layer correction.

(d) Bottom of layer correction

The correction curves for the observed temperature and salin-
ity values and those based upon each of the first four methods
are shown in Figure 7. The estimated correction at the bottom
of each layer was compared to the "true" correction which was
obtained using the observed salinity.

Error. = estimated correction. - true correction.
1 i l

The differences were divided by the depth, in order to obtain

39

§3.00

CORRECTION
2.00 3.00

J L

(M

y.oo

5.00

6.00

j

Method 3

Method 4
Observed

Figure 7. Correction curves for the first four methods.

40

the percentage of error for that particular depth:

Percentage of error. =

Error

zi

where z. is the depth of the bottom of the layer. If this
value were greater than or equal to 0.251, the method was
considered unsuitable at that depth, since the accuracy re-
quirements could not be met. The values compared were the
corrections at the true depth at the bottom of each layer.
In reality, the values at the fathometer depth (true depth
minus correction) should have been compared. This was not
done since the fathometer depths would probably be different
for each error. However, the resultant error from this approx-
imation is insignificant- -of the order of a few millimeters
(see Figure 8) .

All the computations for the first four methods were
executed sequentially in the same program, shown in Appendix
C. This program was also used for the fifth and sixth methods.
An output of this program for a particular station is shown in
Figure 9. The meaning of the different headings is as follows:

SHI
SH2

Z

ID
T
S

gross historical salinity average (method 1)

fine historical salinity average (method 2)

depth of the observation

depth at the bottom of each layer

observed temperature

observed salinity

41

00*0

oo-ooi oo*osi oo'oo^ oo'osa oo-aoe
(wi NQU33yyoa shniw Hid3a 3nyi

oo*ose oo'oofi

u

CO

<d

P.

3
OO

•H

U3

■a
o

— o

-p

<3J

3-

12- ^^-*

o

C-JO

CD

>
p
cd

00

o

^

.Q

C-J

O

o

o

1 1 1 l 1

i i i

00*02

00*001 00-OSI 00*002 00*09,2 00'QOe 00-QSE QO'QOti

(W) NQU33UyQ3 SHNIW Hld3Q 3fiyi

CO

<D
!h

3
ao

•H

o^o

QO'OS

QQ-QQT

oo'qsi oo•oo^ oo'osa
(W) HldBQ 3Hbl

00*006

oo-ose

rt

CO

QQ*0QTl ^

P

GO

•H

*3

.

C -3

t/1

03

O TJ

X

O

v -M

<->

03

CD

P

' S

CD
g

•

-3

-C

£

O

P

03

2

Q

P

CD

1^

T3

P

CD

•H

r-

CD

c

P

3

•H

P

r— 1

c

+->

03

CD

-

CO

CD
2

<-H

CD

P

o

3

CD

p -a

rj

P

o

CD

•H

P

U

P

o

C

U

*-H

CD

fi

P

3

, — .,

CD

<4-t

X> *-H

* — '

"-H

03

•H

z

•3

t/1

c

03

o

CD

•H

X

00

P

P

C

o

o

CD

00

•H

P

CD

+->

P

p

u

o

03

0)

u

p

p

p

f-H

en

CO

o

3

3

u

S

i—i

H

i— (

,£■;

s

H

P

a,-c

— \

CD

p

u

Q

CV

1 '

CD co

■

■d

•

CD

00

CD

P

3

3

CD

P

GO

P

4-> •

H

3 :

: ti.

CO

tu

42

m

XI

n

,,

X

0*

—

0

-o

•

IT *

£

m

<3>.-0

3r*-

"5

—4

.-«> J"»

r>^

2

•

J->f

0 0

U">

•J-'N

. •

S.

—

rr,.-y

\z

■r^i*

— •

• •

• • ■•

h—

JJ

l3

*• • •

~I

>—

f— ~*

— IN

*-

<

< n

TT

/»

'3

- -J

VJi/1

u omDono^ocoooo o • 000

Q. • • • O « «0

ouooo'joooooooo 00 »

I I o

lt — xu* u- ^n o; -j- — «cc -tim— < — ^ru*

vj u-ou'jgu3uouuou o • — On

a. .».»• o • »<j

OOOOOOOOOOOOOO 00 •

o

(NifMf 3- 00 — o >*•—'- n"ici->0 i\ir>.ir\

rvi f>.i\a u" jifl — o-^ jjNf "! — — O <NiO —

O r\ir-, — — — — — CCoOOOO O • <N— Is-

u. t ♦ . • • . « •• u t ,(_>

ajr- .r^-inrji\i>r — .— >& -J>* OOOO OOC OOOUOO'J OO •

<7" ■c-O'nj v c*r»r» iQ^-r-NO O
<—j**t ■Jvf~-ujL/,.y<\j.jr— ir^ —

• • » • Xx3"u",r^'--<xr— Oco^cru-vr- x^«N

T'C^T-r-.rMO^J^irr^iv^crx — < -* >r ^ c" -c >■•"> c*, «- • O x ^ J^ -j- "*"i o .ji-n —

XX n T x a- Xx r X -ocrr~-r— O (N^^i — — . — — — — >oO OOO q , nj— f-»

%f^Tvrvr«r>r^f^-vrsrsr>T>r^ a ....,,... q , ,0

— — — — — — — — — —— — — Oo — OCO0OO0OOOO OO •

o

r*-\Xr"ir'-u~><viN>^- — •* a ^TfO-C
"^i^r-OO-rr-

^- j.(M-rr-«r j-\ O f- u-\ -J- — mo u

«"> 00~r~r— '"'^0<M.ru~'*'M"\i';rj.x ai r^OOO — « — — OOOOO— 'l— I

> C" j'-nx ncc xxor xx>xr»-f"- Q ooCivCCqOooOOCO +->

>r.r%r-r>i'>t«J,vr«J,-r>J-v}'-j-sr o trt

— —— — — —————— — a occooooooooooa , ,

I I +-•

XPvJrvjOx — C^-J- — <Oir\rOtn'\) (/)

-J- O J-— •— <0^-C— •"1.J- (MO O 2 O— — •— < •\j:NI-«VNI'Nl<->j — •— «— <0 .

.• O OOO— OOOOOOOOOO **

<\j.<Nir\iOC7>Nj'Nir"\.}'J*ro<\j'.-JCP'"» i_) • >>«•. ««...i.> C3

> (J'O^/ai-OiOCJX^DDOir^r. lj UOJQ OOOOOOOOOO r— I

i,Wri^irifr>n^i^i,K,'ni''i

•j o -» jj u» — .i^-tn %jf»> x^O— «

,-iM^«irt«\i»iiTi.^^ i-vm*"'

3

(M — -n -u-j- r— "isr fVj* -j/>r q-.c>— U

x (Nj u"\u"> X1 <N O" -r — • •%! x o -r m at -<n^os-ir\iN-«rtUco*)vf -H

i\OhIO-<(?H(Ui<1J-<- 3 (530 OO — _-<—.— .-.OOO ij

ioox-ioir. -»>aso>^f^'jo -i-r o o ••••>••••••••!

• ••t.... O 00-500000000000 "

— f\J ^, O O- -r <\j ."■-\ ,TM/"> rO <\i O :> CO n3

>a,^J>T3TXX'oXXX;X(--f»- Q,

a oooo— — — — — — — — — w

&j -o — in ^r ^ m o Q cr — "Ni ii^ ^- O •»•••«••••••••

C" r<" •* X f^ >i vr >»■ O rr — »0 .-ri ^j- Q OOOOCOOOOOOOOO j^

— >r "Odin JCNvf -o;I3 CcnfM CT

> CT7- UTgO.'^vT^TM 3 7> » <+-|

xx" "ox a-ooD xxxxr-f^

— -<— —<—«-•»<«-.— <-h-«——«—i >r xf-^r t cra>rr>Of- — f^J— <>J"tr\

ai O'Vjvr ■ou-ivj^oOmr'-iMx-G-^ 3

a d,

O OOCJOO — — f\4fS|!Vipn(*) sf^o

ec(M«-uoor->oc7s-ro>'"-oxoo 'J~

• x— ^c?0 oi»-xo — a; f*i-o«M Z3

i/> o^o^l", o— vrf-x^JC^c^o .rir^M>ro^o^;r-— r^ocj'io O

s

c ooooi~ oc^vry-crx x j_,

v cc — pJ'OsQo- mjmOiN hA

i/i •M-jvr.r-f1*'- itn'S j.'-u'o — <m — x x*-r, j- -j- ^- j^rNj -n— ^j. °<J

X ••••■••••#«•<• rsi oo?j\i»-f>r-i"o minjij— ><j O

oif>nic*j.'ni«t.tii»»('if»i«i"Mi'n*i j ••••••••• £V

Q OOO J— •— — V4<\jtMr»li'1 ■r*f>

cn

OO "DO— — — ■NjfM'Ni-ri -r\ ^lA J. 0

3 .. >H

3

•H

o "5 ">o oo 3000 oo'J ~3 "^ Oirv-r '-o 3<~>— iTi— o-r ■cn -r >-o Q

ifN"^ — u> 0O-< 'J*0 •* ^ — —• O X. B>-vj- J< r. x M J- ■* — — -< vT-0 — ^ OJ OO

• •••• •• o • • — u ••

— — 0'^> jNB-ajo^r> ^3-0^\ OOO 3o- <— •-••M'm ,«i i -fS\ h- iu 2

— -t~* •• *x vi j

SI > CS —

x ut a ►-

►- LU <T

O "3 OOOiA ^LfioU^OUM^O SaOOOiH'fllftO^O^iflS i ^. UJ —

3 o .lu n 3>

-< tP^-^OMO — -J— '■f^N'OO — ♦ if\4*iiTi3MO-4 0-'L',iM OO O O " -J -U

— M^ O x— -^ o o^vjr- -1*^ -i^j* ca-">o,o-j">-f3' 'JJ < Q

— — — — ■NJ.Ni'^.-^l — — — — r%jN— VT1 i. J.O >

-3 0< O

h- I- Cjtf

O 30 "3 OOO 300 DO"DO Oj^TOC 11OJ30Q3 "t 4: 2 ^1 <

■ ^ .•>.. ......... ~j .«•«.• -»- j- jj a.' •• O

— ->j"-\ jai^.O'm j-'*-o> -i* j^o* -i^j -o .1 ?>■ j m •<• n 7 ~t j- ^ x ^t "e<t<

— _— • — is^-jm — • — — — • ini (Ni "n "Z ~* UJ <lu^-

43

HS : historical salinity value (method 3)

HS' : adjusted historical salinity curve (method 4)

V : sound speed computed from Z, T and S

Vl-4: sound speed computed using methods 1-4

COR : correction applied at depth ID computed
from V

C0R1-4: correction obtained using methods 1-4

DC0R1-4: error of methods 1-4 (difference between
C0R1-4 and COR)

PC1-4: percentage of error of methods 1-4
(DC0R1-4 divided by ID)

44

III. DISCUSSION OF THE RESULTS

Based on the foregoing analysis, four different possi-
bilities have been considered for each of the six salinity
averaging methods tested:

(1) No error exceeds the limits at any given depth.

(2) The error limit is exceeded only in the surface
layer.

(3) The error limit is exceeded between the surface
layer and the layer which contains the 30 m depth.

(4) Excessive errors have been found for layers deeper
than 30 m.

A. USE OF HISTORICAL SALINITY DATA DERIVED FROM ICAPS
(FIRST FOUR METHODS)

The results obtained from using historical salinity data
derived from the ICAPS water mass file are shown in Table IV
A discussion of these results, categorized by area, follows.

1 . Northern Area

The large variation of surface salinity in this
region caused the depth correction for the first layer to
exceed the acceptable limits for the first three salinity
averaging methods considered. Excessive errors were found
for 31.6%, 28.0% and 24.4% of the stations examined, respec-
tively. A slight improvement is noted between methods 1 and
3, indicating that some benefit is derived from the use of
an historical mean salinity profile. However, when the

45

TABLE IV

ERRORS RESULTANT FROM THE USE OF HISTORICAL SALINITY

DERIVED FROM ICAPS (METHODS 1 TO 4) .

Results are shown for the surface layer and for water

depths shallower or deeper than 30 m. Numbers indicate

the number of stations which exceeded the accuracy

standards. Percentages are indicated in parenthesis.

.ria Tnrai Method 1 Method 2 Method 3 Method 4

"rv-a l0Cdl S. L. £30 >30 S.L. <30 >30 S.I.. <30 >30 S.L. <30 >30

Winter 248 93 38 J 78 24 2 73 21 2 0 0 0

(38.3) (1S.3) (1.2) (31.5) (9.7) (U.8) (29.4) (8.5) (0.8) (0.0) (0.0) (0.0)

idling 07y 301 100 44 278 151 31 235 135 25 0 0 0

(44.3) (24.4) (6.5) 1,40.9) (22.2) (4.0) (34.0) (ly.9) (3.7) (0.0) (0.0) (0.0)

North iu....er 043 1S1 So 1 13U 27 0 115 21 0 0 0 0

(23.5) (8.7) (0.2) (20.2) (4.2) (0.0) (17. V) (3.3) (0.0) (0.0) (0.0) (0.0)

Fall I'.'S 10 3 0 8 3 0 7 3 0 0 0 0

(S.l) 11.5, (0.0) (4.1) (1.5) (0.0) (3.o) (1.5) (0.0) (0.0) (.0.0) (0.0)

lota! 176S 557 2<>3 48 494 205 33 430 180 27 0 0 0

(31. b) (14. i>) (2.7) (.28.0) (11. o) (1.9) (24.4) (10.2) (1.5) (0.0) (0.0) (0.0)

Winter 4 50 1 0 0 0 0 0 u 0 0 0 0 0

(0.2) (0.0) (0.0) (0.0) (0.0) (0.0) (0.0) (0.0) (0.0) (0.0) (0.0) (0.0)

Spring 350 000 000 0 00 000

(0.0) (0.0) (0.0) (0.0) (0.0) (0.0) (0.0) (0.0) (0.0) (0.0) (0.0) (0.0)

South Suumer 229 3 0 6 10 0 U o 0 0 0 0

(1.3) (0.0) (0.0) (0.4) (0.0) (0.0) (0.0) (0.0) (0.0) 10.0) (0.0) (0.0)

Fall 193 001 000 000 000

(0.0) (0.0) (0.5) (0.0) (0.0) (0.0) {0.0) (O.J) ;0.0) (0.0) (0.0) (0.0)

1222 4 0 1 10 0 ii 0 0 0 0 0

(0.3) 10.O) (0.1) (0.1) (0.0) (0.0) (il. 0) iil.O) in ". (0.0) (0.0) (0.0)

Winter 127 110 103 99 102 90 92 83 79 7o 0 1 IS

(BO.l,) (81.1) (78.0) (80.3) (75.0) (72.4) (oS.l) (o2.2) (59 8) (0.0) (0.3) (11.8)

Spru.g 149 131 129 125 12o 125 121 125 123 119 o i 3

(87.9) (80.0) (83.9) (84.0) (83.9) (8L.2) t83.9) (82 :.» (79. y) (0.0) (0.7) (2.0)

«S!itS?rnd/ Suirar 100 71 69 64 70 68 c4 70 t,3 0 2 8

Juan Dc Fuca (71-"J (°y-°J (b4-u) (7ll-°J (t>8-u) (biA)) '70-tM ibi'0> r,'A)i ClMM (2-0) t8-0)

Full 90 65 64 58 58 So i4 47 3o 34 0 4 4

(07.7) (0O.7) (00.4) (O0.4) (58.3) (So. 3) (49.0) { 37 Sj (35.4) (0.0) (4.2) (4.2.)

Total 472 377 3Ts— 346 356 345 331 32 r. 301 286 ~0~ T 3fi"

(79.9) (77.3) (73.3) (75.4) (73.1) 70.1) (68.9) 163.8) (00.0) (O.O) (1.7) (0.4)

46

historical salinity profile was adjusted to match the surface
value (method 4) , there were no stations which exceeded the
acceptable limits.

The depth error of the first layer was often so large
that sometimes the offset created induced an unacceptably
large error in the second layer. This occurred for 14.9%,
11.61 and 10.21 of the stations examined, respectively. In
very few cases (2.7%, 1.9% and 1.5%) the excessive difference
in surface salinities created an excessive error which extended
to the third layer.

There are two main reasons why this error is so large:
1) the observed salinity extends to very low values, and 2)
most of the data only are defined at the surface and 30 m
isopleths which creates a very wide first layer (15 m) over
which that value is applied. The errors which occur due to
low surface salinities generally do not cause the accuracy
limits to be exceeded except close to the surface, for depths
rarely exceeding 30 m. The salinity may be very low at the
surface but it increases quite rapidly to nominal oceanic
values in a few meters.

The error due to low surface salinities is of par-
ticular importance during the spring season (April-June) due
to the large outflow of the rivers of Northern California and
Oregon, especially the Columbia River. It is also noticeable
in the winter season, but is considerably less important in
summer. An example where this situation frequently occurred

47

was in the area of influence of the Columbia River, where
on one occasion a salinity as low as 8.02 %> was observed
during the spring season.
2 . Southern Area

Considering the least sophisticated of the salinity
averaging schemes (method 1) , only five stations out of 1222
(0.41) showed any kind of deviation which exceeded the accept-
able depth limits. Of these five, only one station, located
adjacent to the Golden Gate Bridge in San Francisco, exceeded
those limits in a layer other than the surface. For this
station, unacceptable depth errors were found at depths
between 70 and 80 m due to an inversion in the salinity pro-
file. Of the other four stations, only one exceeded the
error limit when the second method was used. All these sta-
tions were located in water more than 300 m deep, except one
located about 8 miles south of Point Reyes, about half-way
to the Farallon Islands. None of these observations exceeded
the limits for the third and fourth methods. The general
pattern resulted in a slight improvement in accuracy from the
first to the third methods, and a significant improvement
when the fourth method was applied. However, there were cases
where, because of the large range in salinity, method 1 was
more accurate than method 3. The pattern of the accuracy of
the four methods was similar to that experienced in the
Northern area.

48

3 . Strait of Juan De Fuca and Puget Sound

Although method 4 demonstrated a significant increase
in accuracy, none of the four methods which used historical
salinity derived from the ICAPS file was deemed accurate
enough. This file does not define any water mass applicable
in the Strait of Juan De Fuca and Puget Sound, and the extrap
olation of the ICAPS data to fit this region proved to be
insufficient in accuracy.

B. USE OF INFERRED SALINITY VALUE
(FIFTH AND SIXTH METHODS)

Salinity values of 31, 33 and 29 °/oo were selected as
representative of the Northern, Southern and Strait of Juan
De Fuca/Puget Sound areas, respectively. These values were
applied both throughout the entire water column (method 5)
or else were replaced by the observed surface salinity value
for the first layer only (method 6) . The number of stations
that exceeded the acceptable limits for each of these methods
is shown in Tables V and VI. A further analysis of these
results follows.

1 . Northern Area

Of the 1765 stations examined, 523 stations (18.3%)
had a surface layer error that exceeded the acceptable limits
when a value of 31 %o was used throughout the entire water
column Oethod 5). Only 117 (6.6%) exceeded the limits in
deeper layers, and only one in a layer deeper than 30 m.

49

TABLE V

ERRORS RESULTANT FROM THE USE OF A SINGLE VALUE OF SALINITY
THROUGHOUT THE ENTIRE WATER COLUMN (METHOD 5)*

Season: Winter Spring Summer Fall

V.ca Salinity Tola! S.L. i3u -JO Total S.L. i30 >30 Total S.L. <30 >J0 Toral S L. i30 >30

North 31 '/« 24b 34 y 0 679 197 luU 1 043 33 8 f: 193 4 0 0

(13.7) (3. 0J CO-") C2S.UJ (14.7) (0.1) (13.7) (1.2) (0.U) (2.1) (0.0) (0.0)

South 33 °/» 4S0 0 0 0 3S0 0 u 0 229 0 0 0 193 0 0 0

(0.0) (0.0) (0.0) (0.0) (0.0) (0.0) (0.0) (0.0) (0.0) (0.0) (0.0) (0.0)

Pug.-t S01.W 29'/„ 127 22 16 a 149 3S 16 IS 100 28 22 18 96 25 13 12

Strait or (17.3) (12.6) (6.3) (2S.S) (10.7) (10.1) (28.0) (22.0) (18.0) (26,0) (13.3) (12.5)
JulH De i:uca

*Table descriptions as in Table IV.

TABLE VI

ERRORS RESULTANT FROM THE USE OF A SINGLE VALUE OF SALINITY
THROUGHOUT THE ENTIRE WATER COLUMN (METHOD 6)*

Season: Winter Spring Summer FjH_

Area

Salinit> Toral S.L. ^30 >30 Total S.L. <30 >30 Total , L ,H >3Q Total S.u. >20 >M

«• !" AAA " AAA M> A

(i 0 0 3S0 0 0 0

(0.0) (0.0) (0.0) (0.0) (0.0) (0.0)

KtoTd/ "^ lV (0°0) (0°0) (4°7) ^ (0°0) (0°0) (IS,

»*» ^ t^,Xu^ iS° roVoVo°0) A m C0°0) C0°0) (0Uo)

4 96 C 0 11

Juan De Fuca

*Table descriptions as in Table IV

50

When the observed surface salinity value was used in the
first layer, no stations exhibited an unacceptable error.

2 . Southern Area

No excessive errors were found when either method 5
or 6 was applied, for which a constant value of 33 %o was
assumed for this area.

3. Strait of Juan De Fuca and Puget Sound

A salinity value of 29 %o was assumed for this area.
Of the 472 stations examined, 113 stations (23.91) exceeded
the limits when this value was used by itself, but only 23
(4.91) when it was substituted in the first layer by the
observed surface salinity value.

It has been seen that for the three areas, and for any
of the methods considered, most of the errors exceeding the
accuracy limits occurred in the surface layer only. This does
not affect the accuracy standards when the depth is deeper
than about 30 m, which represents most of the area over which
hydrographic surveys are conducted.

51

IV. CONCLUSIONS

The results shown in the previous chapter indicate that
historical data can indeed be used in the computation of
depth corrections due to variations in sound speed for the
West Coast of the United States. Data derived from the ICAPS
salinity file are completely acceptable only in the Southern
area, although good approximations can be made in the Northern
area. The situation in the Northern area, however, can be an
artifact created by the observed data format. If the surface
value as observed in the sea is applied to adjust the histor-
ical salinity curve from ICAPS, then the results are completely
correct for all depths and seasons, both in the Northern and
Southern areas. This by itself would indicate sufficient
confidence to apply this method to actual hydrographic sur-
veys in the areas considered. It would only require the use
of a table, such as Table I, and the measurement of the sur-
face salinity, the temperature being measured with an XBT.
The advantages of this method over what is presently being
done are significant, both in time and cost.

ICAPS water masses are representative of very large areas
of deep water. For application to coastal hydrography it is
better to create a file of the shelf area exclusively. A
ten year period may be representative, but a longer period,
including as many observations as possible should be better.

52

The use of a single salinity value derived from a large
number of observations is the easiest method to be applied
in the computation of depth corrections. One further step
is to make a surface salinity observation and use that value
for the computation of the first layer correction. The im-
provement of this method over the method currently in use,
from a practical point of view, is remarkable.

This conclusion is applicable to the entire West Coast
of the United States, excluding the Strait of Juan De Fuca
and Puget Sound where the results were not favorable, and
to San Francisco Bay area where there were insufficient data

In summary, the applicability of the use of historical
data in the computation of depth corrections is as follows:

1 . Northern Area

A single value of salinity of 31 %0 is applicable
throughout the entire water column except for the surface,
where the salinity must be measured. In these situations,
the more complex method 4, wherein the historical salinity
profile derived from the ICAPS' data file is adjusted to the
surface value, may also be used.

2 . Southern Area

The use of a single salinity value of 33 %0 , appli-
cable throughout the entire water column, is sufficiently
accurate for this region. Alternatively, the historical
salinity curve derived from ICAPS can also be used with no
adjustment for the surface layer.

53

3. Strait of Juan De Fuca and Puget Sound

The use of historical salinity data derived from any
of the methods above is not recommended in this area.

54

APPENDIX A
WILSON'S EQUATION

Wilson first presented his empirical formula for the
computation of the speed of sound in sea-water in the June
1960 issue of the Journal of the Acoustical Society of Amer-
ica [Wilson, 1960a]. In the October issue of the same year,
a slightly modified version of this formula was introduced,
with an extended range of applicability. This version is
the formula commonly used and accepted as best representing
the sound speed dependence on pressure, temperature and

salinity. It fits data within the ranges of -4°C to 30°C

2 ' 2
in temperature, 1 kg/cm to 1000 kg/cm in pressure and 0 %o

to 37 %o in salinity, with a standard deviation of about

0.30 m/s [Wilson, 1960b]4.

A value of 1449.14 is given for the sound speed at T = 0°C.
P= 0.0 kg/cm and S = 35 %o • Corrections are added to this
reference speed in order to compensate for the variations in
pressure, temperature and salinity, either alone or in a
combined effect. The revised formula is as follows:

An analysis of this formula for values of pressure,
temperature and salinity outside the range for which the
equation was derived showed that the error can go up to
2.13 m/s for extreme conditions of temperature and salinity
expected to occur in the sea [Wilson, 1962] .

55

V = 1449.14 + VT + Vp + Vs + VSTp

where VT = 4.5721T- 4.4532 x 10_2T2 -2.6045 x 10 ~4 T3
+ 7.9851 x 10"6 T4

Vp = 1.60272 x 10"1 P + 1.0268 x 10"5 P2

+ 3.5216 x 10"9 P3 - 3.3603 x 10"12 P4

Vs = 1.39799(S-35) + 1.69202 x 10_3(S-3 5)2
VSTp= (S - 35) (-1.1244 x 10"2 T + 7.7711 x 10"7 T2
+ 7.7016 x 10"5 P - 1.2943 x 10"7 P2
+ 3.1580 x 10"8 PT + 1.5790 x 10"9 PT2)
+ P(-1.8607 x 10"4 T + 7.4812 x 10"6 T2
+ 4.5283 x 10"8 T3) + P2(-2.5294 x 10"7 T
+ 1.8563 x 10"9 T2) + P3(-1.9646 x 10"10T)

The units of temperature, pressure, salinity and sound
speed are, respectively, degrees Celsius, kilograms per square
centimeter, parts per thousand and meters per second. While
temperature and salinity are usually directly measured,
pressure is computed from depth. The dependence of pressure
on depth is not linear and several expressions may be applied
for this computation. An iterative process is used by NOS
to determine pressure [Wallace, 1971]. This method was
extracted from the User's Guide to NODC ' s Data Services
[NOAA, Department of Commerce, 1974], where it is described
as follows:

56

a = -9.3445863 x 10"2 + 8.14876577 x lO^S
o

- 4.8249614 x 10"4 S2 + 6.7678614 x 10"6 S3

g = 0.980616 - 2.5928 x 1O~3(cos20) + 6.9 x 1O~6(cos20)2

P(surface) = 10.1325 decibars

g. = gQ + 1.101 x 10'7(di + di_1)

Pi - (1 + 10'3 aT)/R

R = 1 - [4.886 x 10"6P/(1 + 1.83 x 10"5P)]

+ P[-2.2072 x 10"7 + 3.673 x 10_8T - 6.63 x 10_10T2

+ 4.0 x 10"12T^ + a (1.725xl0~8 - 3.28xlO_10T

o

+ 4.0 x 10"12T2) + a2(-4.5 x 10"11 + 10_12T)]

+ P2[-6.68 x 10" 14 - 1.24064 x 10" 12T + 2.14 x 10"14T2

+ a (-4.248 x 10"13 + 1.206 x 10" 14 T - 2 . 0 x 10"16T2)
o

+ aQ2 (1.8 x 10"15 - 6.0 x 10"17T)] + P-5 (1 . 5 x 10 " 17T)

1. A first approximation gives PV and p ! :

P'.' = P. . + p. ,g. (d. - d. ,)
l i-l Mi-lsi^ l 1-1/

with P!1 computed this way, p.' is calculated.

2. A second approximation using p'. gives P'. and p.

P'. = P. . + l/2(p. . + p.' )g.(d. - d. ..)
l i-l ^Mi-1 yi •/6iv i i-l7

with P1. p. is calculated.

l i

57

3. Finally, with p.

P. = P. . + 1/2 Cp- i + P-)g- Cd- - d. ,)
i l-l ^ l-l i/6iv i i-I'

then, P = 0.10197 P.

In the above expressions, the subscript indicates the depth
that is being computed, and the number of primes is inversely
proportional to the accuracy of the approximation. The mean-
ing of the different parameters is the following:

d = depth in meters

T = temperature in degrees Celsius

S = salinity in parts per thousand

p = density

P = pressure in kilograms per square centimeter

0 = latitude in degrees.

58

49°N

X

h- (

Q

w
a*

a,

<

2

o

CO

<
w

CO

Q

<
W

<

OS

o

CO

<
u

O

o

l-H

<
Eh
CO

40°N

127°W

122°W

Figure B-l. Location of stations for the northern area
and Strait of Juan De Fuca/Puget Sound, in winter.

59

49°N

40°N

[ .» * • 't

127°W

122°W

Figure B-2. Location of stations for the northern area
and Strait of Juan De Fuca/Puget Sound, in spring.

60

49°N

40°N

127°W

122°W

Figure B-3. Location of stations for the northern area
and Strait of Juan De Fuca/Puget Sound area, in summer

61

49°N

40°N

127°W

122°W

Figure B-4. Location of stations for northern area and
Strait of Juan De Fuca/Puget Sound area, in fall.

62

0"V
» »

125°W

117°W

Figure B-5. Location of stations for southern area, winter.

63

40°M

125°W

117°W

Figure B-6. Location of stations for southern area, spring

64

125°W

117°W

Figure B- 7. Location of stations for southern area, summer

65

125°W

117°W

Figure B-8. Location of stations for southern area, fall.

66

APPENDIX C

COMPUTER PROGRAM

C PROGRAM SVC

C

C PROGRAM SVC USES HISTORICALY DERIVED SALINITY IN THE

C COMPUTATION OF DEPTH CORRECTIONS DUE TO VARIATIONS

C IN SOUND SPEED, IT USES FOUR DIFFERENT METHODS

C FOR DERIVING THE HISTORICAL VALUES. THE DIFFERENT

C DEPTH CORRECTIONS ARE COMPARED TO THE CORRECTION

C OBTAINED USING THE OBSERVED SALINITY.

C

C

INTEGER ON,DD,YY,TNO,TNSO

REAL lOlfLLtLflDtLC

DIMENSION AVE (43 ,8 ) , AVEK 43 ) , COR ( 30 ) ,C0R1 ( 30) ,C0R2(30)

1 ,C0R3(30) ,C0R4(30) . DC0R1 ( 30 ) ,DC0R2( 30 ) ,DC0R3(30),

2 DC0R4(30),DC0RR(30) ,DCR1 1(2000) ,DCR22( 2000) ,

3 DCR33 ( 2000 ),DCR44( 2 000), 10(59) ,ID1(43),L(59),L1(43)

4 , PC (30 I , PC 1(30) ,PCL 1(2000) , PC2 (30 ) , PC2 2 ( 20 00 ) ,

5 PC3(30),PC33(200 0),PC4(30) ,PC44(2000) ,S(30),

6 SH(22f8) ,SHU(30)tSH22(30) . SH3 ( 30 ) , SH4( 30 ) ,SHH(22),

7 SHHH( 22) ,T(30) ,V(30) ,V1(30) ,V2(30) ,V3(30) ,V4(30) ,

8 Z(22),Z1(30)

100 FORMAT <F7.1,8F7.3)

101 FORMAT <3I2,I5,A1,I6,A1,1X,A4,I3)

102 FORMAT (10F7.3)

200 FORMAT (• 1 • ,34X, ■ SOUTH (<40 N )•, 1 8X, » NORTH 040 N)«//
1 24X,«Z,,2(6X,»W« ,2(6Xt "S1 ) ,6X, •FIV3X)/I

201 FORMAT ( 20X , F6.0, 1 X,4F7 .2 ,3X, 4F7. 2 )

202 FORMAT (• 1« , 37X, • SOUTH (<40 N )», 18X ,• NORTH 040 N) • //
1 25X, "Z«,2( 7X,«W»,2(6X,«S«),6X,«F«,2X)/)

203 FORMAT ( 20X ,F 7. I , 2X,4F7.3 ,3X, 4F7 .3 )

204 FORMAT (• 1» f ' STATION NUMBER :', IX, A4//» DATE: »,I2,
1 2(«.ltI2)J

205 FORMAT (/» LAT : •,I4,A1/' LONG: «,I5,A1)

206 FORMAT (/• SH1:»,F7.3/' SH2:»,F7.3/)

207 FORMAT ( / 5X , » Z« ♦ 5X , • ID • ,7X, • T * , 7X , ■ S • ,6X, • HS • , 5X,

1 •HS* •• ,7X, •V*,7X,IV1«,7X,' V2» ,7X,' V3» , 7X,« V4'/)

208 FORMAT ( 2 F7. 1 ,F9.3 , 2X, 3F7.3 ,2 X, 5F 9. 3)

209 FORMAT ( //5X, «Z» , 5X , • I D • ,6X , • COR1 ,3X, ■ COR 1 ' »3X, «C0R2 • ,

1 3X,«C0R3,,3X, •C0R4« , 5X,»QC0R1» , 2X . • DC0R2 • , 2X ,

2 'DOORS' ,2X, 'DC0R4',5X,'PC1' ,4X, ^PCa* ,4X, • PC 3' ,4X ,

3 'PC4'/)

210 FORMAT ( 2F7 . 1 ,2X, 5F 7.3 , 2X, 4F7 .3 , 2 X, 4F7. 3 )

211 FORMAT </• NUMBER OF DEPTHS:', 15)

212 FORMAT (//• NUM8ER OF OBSERVATIONS OFF : • , 53X , 41 7/
1 • PERCENTAGE:1 ,71X,4F7. 2)

213 FORMAT (/• LARGEST VALUE OBSERVED: • f59X, 4F7. 3/

1 ■ MEAN:' ,77X,4F7.3/' STANDARD DEVIATION:1,

2 63X,4F7.4)

214 FORMAT (//HOX.'X X'/U1X,'X X' /l 12X , • X • / 1 1 IX,
1 'X X'/llOXt'X X1)

215 FORMAT ( • 1' ,45X, 27 ( • ♦ » ) /46X, • ♦ • , 25X , ' + • /46X,

1 • + STATISTICS +'/46X,'+» ,25X,'*',/

2 46X,27('+' )//)

216 FORMAT ( / /49X , « PC 1 ■ ,4X , • PC2' , 4X, • PC3 • ,4X, • PC4 • )

217 FORMAT (/• TOTAL NUMBER OF OSERVATIONS: • , 1 1 1/ • TOTAL'

1 ,• NUMBER OF OBSERVATIONS OFF : • , 11X, 41 7/

2 • PERCENTAGE:1 , 34X, 4F7.2// • NUMBER OF STATIONS:',

3 120/' NUMBER 3F STATIONS OFF : • ,2 IX, 41 If

4 • PERCENTAGE:1 ,34X,4F7. 2/) m _, „ ,

218 FORMAT (/• LARGEST SINGLE VALUE OBSERVED :•, 15X, 4F7. 3/

1 • LARGEST MEAN VALUE OBSERVED :•, 17X, 4F7. 3/

2 ■ LARGEST STANDARD DEVIATION OBSERVED: '. 9X.4F7 .3 )

219 FORMAT (//• MEAM OF THE MEANS : • , 27X ,4F7. 3/ ■ STANDARD1
1 ,' DEVIATION OF THE MEANS: ■ , 13X , 4F7.4)

67

FAC(VI)={ VI-VO/VC
PCTG(M,N) =FLGAT(M)/FL0AT(N)*100.
C

C VC IS THE SOUND SPEED ASSUMED IN THE FATHOMETER.

VC=1463.04

C

c

C 1. COMPUTE HISTORICAL SALIMITY AVERAGES

c

C

N=2 2

DO 10 1=1 ,N

READ(5tI00) Z(I ),(SH(I ,K) ,K=1,8)

10 CONTINUE
WRITE (6,200)

WRITE (6,201) (Z( I) ,(SH(I,K) ,K=1,8) ,I = 1,N)

J=0

M=2*N-1

DO 13 K=l,8

DO 11 1 = 1, N

SHH(I) = SH( I,K)

11 CONTIMUE

CALL MEANK J ,N , M , Z , SHH, AVE 1 , ID1 ,Li )
00 12 1 = 1, M

AVE(I,K)=AVE1( I)

12 CONTINUE
J=l

13 CONTINUE
WRITE (6,202)
WRITE (6,203) ( ID1 ( I ) , ( AVE( I , K) , K=l , 8 ) , I =1 , M )

C

C

0N=0

TNO=0

N0FF1=0

N0FF2=0

N0FF3=0

N0FF4=0

DC0R1L=0.

DC0R2L=0.

DC0R3L=0.

0COR4L=0.

PC1L=0.

PC2L=0.

PC3L=0.

PC4L=0.

DCR1ML=0.

DCR2ML=0.

DCR3ML=0.

DCR4ML=0.

PC1ML=0.

PC2ML=0.

PC3ML=0.

PC4ML=0.

DCR1SL=0.

DCR2SL=0.

DCR3SL=0.

DCR4SL=0.

PC1SL=0.

PC2SL=0.

PC3SL=0.

PC4SL=0.

NS0FF1=0

NSOFF2=0

NS0FF3=0

NSOFF4-=0

68

c

c

C 2. READ DATA

c

C

20 READ (5,1011 YY, MM , DD, LAT,H1, L0NGrH2,NS, KONT
IF (DD.EQ.O) GO TO 80

WRITE (6,204) NS,MM,DD,YY

WRITE (6,205) UT , HI, LONG, H2

N0FF11=0

N0FF22=0

N0FF33=0

N0FF44=0

DCR1TL=0.

DCR2TL=0.

DCR3TL=0.

DCR4TL=0.

PC1TL=0.

PC2TL=0.

PC3TL=0.

PC4TL=0.

READ (5,102) (Zl( I ), 1 = 1,10)

IF (KONT.EQ.O) SO TO 23

DO 22 1=2,10

IF (Zl( D.NE.O.) GO TO 21

N=I-1

GO TO 26

21 IF (I.EQ.10) N=I
11 CONTINUE

GO TO 26

23 READ (5,102) (Zl ( I ) , 1 = 1 1 , 20 )
DO 25 1=11,20

IF (Zl(I).NE.O.) GO TO 24

N=I-1

GO TO 26

24 IF (I.EQ.20) M=I

25 CONTINUE

26 READ (5,102) (T(I),I=1,N)
READ (5,102) (S( I) , 1 = 1, N)
0N=0N+1

IF (LAT.GT.4000) GO TO 27
K=l

IF (MM.GE.4) K=2
IF (MM.GE.7) <=3
IF (MM.GE.iO) K=4
GO TO 28

27 CONTINUE

K=5

IF (MM.GE.4) K=6
IF (MM.GE.7) K = 7
IF (MM.GE.IO) K=8

28 CONTINUE
C

C 3. MAKE COMPUTATIONS WITH TRUE VALUES

c

C

CALL LAYER (N,Z1,L, ID)

CALL C0RREC(LAT,VC,N,Z1 ,T,S , L , V, COR )
C

C

C 4. MAKE COMPUTATIONS WITH 200, 1000,3000 M AVERAGES

C

SH1=AVE(19,K)

IF (ZKN) .GT.200.) SH1=AVE( 33 ,K )

IF (ZKN) .GT. 1000. ) SH1 = AVE (43 ,K)

69

DO 40 1=1 .N

SH1KI )=SH1

40 CONTINUE

CALL C0RREC(LAT,VC,N,Z1,T,SHU,L,V1,C0R1)
CALL DCRR(N,C0R1,C0R,ID,DC0R1 ,PC1)
DO 41 1=1, N

AP1=ABS(PC1(I) )

IF ( AP1.GT.PC1TL) PC1TL=AP1

IF (AP1.GT..25) NOFFll=NOFFli+l

41 CONTINUE

IF (PC1TL.GT.PC1L) PC1L=PCITL

CALL STAT ( N, PC 1, PC 1M, PC1S)

IF (ABS(PCIM) .GT.PC1ML) PC1ML=A8S ( PC1M )

IF (ABS(PCIS) .GT.PC1SL) PC1SL=ABS ( PC1S)

PC1K0N) = PC1M

PRC11=PCTG(N0FF11,N)

IF (N0FF11.GT.0) MS0FF1=NS0FF 1+1

C

c

C 5. MAKE COMPUTATIONS WITH 50, 100, 150, 200, 300, 500,

C 750, 1000, 1500, 3000 M AVERAGES

c

C

SH2=AVE(9,K)

IF (ZKN) .GT.50.) SH2=A VE ( 1 3, K)
IF (ZKNJ .GT.100. ) SH2 = AVE( 17, K)
IF (ZKN) .GT.150. ) SH2=AVE(19,K)
IF (ZKN) .GT.200. ) SH2 = AVE( 23 ,K )
IF (ZKN) .GT.300. ) SH2=AVE( 27 ,K )
IF (ZKN) .GT. 500.) SH2=AVE( 32 ,K)
IF (ZKN) .GT. 750.) SH2=AVE( 33 ,K)
IF (ZKN) .GT. 1000. ) SH2=AVE ( 3 7,K )
IF (ZKN) .GT. 1500. ) SH2=AVE ( 43 ,K )
DO 50 1=1, N

SH22(I )=SH2

50 CONTINUE

CALL C0RREC(LAT,VC,N,Z1,T,SH2 2,L,V2,C0R2)

WRITE (6,206) SH1,SH2

CALL DCRR(N,C0R2,C0R,ID,DC0R2,PC2)

DO 51 1=1, N

AP2=ABS(PC2(I) )

IF ( AP2.GT.PC2TL) PC2TL=AP2

IF (AP2.GT..25) N0FF22=N0FF22+ 1

51 CONTINUE

IF (PC2TL.GT.PC2L) PC2L=PC2TL
C

CALL STAT ( N , PC2, PC2M, PC2S)

IF (ABS(PC2M) .GT.PC2ML) PC2ML=A3S ( PC2M )

IF (ABS(PC2S) .GT.PC2SL) PC2SL=ABS ( PC2S I

PC22(0N)=PC2M

PRC22=PCTG(NOFF22,N)

IF (N0FF22.GT.0) NS0FF2=NS0FF2+i
C

C 6. MAKE COMPUTATIONS FOR EACH CORRESPONDING HISTORICAL

C SALINITY VALJE

c

C

DO 60 1=1 ,22

SHHH( I)=SH(I,K)

60 CONTINUE ,....-,.

CALL INTRPL ( 22, Z , SHHH ,N, Zl ,SH3 )
CALL CORREC(LAT,VC,N,Zl ,T , SH3 ,L , V3,C0R3)
CALL DCRR(N,COR3,COR,ID,DCOR3,PC3 )
DO 61 1=1 ,N

AP3=ABS(PC3(I) )

IF (AP3.GT.PC3TL) PC3TL=AP3

70

IF (AP3.GT..25) N0FF33=N0FF33+1
61 CONTINUE

IF (PC3TL.GT.PC3L) PC3L=PC3TL

CALL STAT (N , PC3, PC3M, PC3S )

IF (ABS(PC3M) .GT.PC3ML) PC3ML = ABS ( PC3M )

IF (ABSCPC3S) .GT.PC3SL) PC3SL=A3S ( PC3S)

PC33(ON)=PC3M

PRC33=PCTG(N0FF33,N)

IF (N0FF33.GT.0) NS0FF3=NS0FF 3+1

C

c

C 7. MAKE COMPUTATIONS WITH ADJUSTED HISTORICAL

C SALINITY CURVE

c

C

ADJ=S(1)-SH3( 1)
DF=30.0-Z1(1)

DO 71 1=1, N

IF (ZKI J.GE.30.0) GO TO 70

SH4( I)=SH3(I)+ADJ*(3 0.0-Z1( I ) )/DF

GO TO 71

70 SH4( I)=SH3(I)

71 CONTINUE

CALL C0RREC(LAT,VC,N,Z1,T,SH4,L,V4,C0R^)
CALL DCRR(N,COR^,COR,ID,DC0R4,PC4)
DO 72 1=1, N

AP4=A3S(PC4( I) )

IF ( AP4.GT.PC4TL) PC4TL=AP4

IF (AP4..GT..25) N0FF44=N0FF44+1

72 CONTINUE

IF (PC4TL,GT.PC4L) PC4L=PC4TL
CALL STAT ( N , PC4, PC4M, PC4S)
IF (ABS(PC4M) .GT.PC4ML) PC4ML=ABS( PC4M)
IF (ABS(PC4S) .GT.PC4SL) PC4SL=ABS ( PC4S)
PC44(0N)=PC4M
PRC44=PCTG(N0FF<>4,N)
IF (NOFF<f4.GT.O) NS0FF4=NSQFF4+1
WRITE (6,207)

WRITE (6,208) ( Zl ( I ) , I D( II , T( I ) , S ( I ) , SH3( I ) , SH4( I ) ,
1 V(I ),Vi(I),V2(I) ,V3(I), V4( I If 1=1, N)

WRITE (6,209)
WRITE (6,210) (Zl( I ),ID(I),COR( I) , C0R1 ( I ) ,C0R2( I),

1 C0R3(I ),C0R4(I) ,DCORK I) , DC0R2 ( I ) , DC0R3 ( I) ,DC0R4(I )

2 ,PC1(I ),PC2(I),PC3( I),PC4( I) ,I=1,N)
WRITE (6,211) N

WRITE (6,212) N0FFllfN0FF22»N0FF33fN0FF44f PRC11,
1 PRC22,PRC33,PRC44

WRITE (6,213) PC1TL,PC2TL,PC3TL,PC4TL,PC1M,PC2M,
1 PC3M,PC4M,PC1S,PC2S,PC3S,PC4S

N0FF=N0FF11+N0FF22+N0FF33+N0FF44

IF (NOFF.NE.O) WRITE (6,214)

TNO=TNO+M

NQFF1=N0FF1+N0FF11

N0FF2=NQFF2+N0FF22

NOFF3=NOFF3+NOFF33

N0FF4=N0FF4+N0FF44

GO TO 20

C

c

C 8. COMPUTE STATISTICS

c

C

80 CONTINUE

PRC1=PCTG(N0FF1,TN0)
PRC2=PCTG(N0FF2,TN0)
PRC3=PCTG(N0FF3,TN0)
PRC4=PCTG(N0FF4,TN0)

71

PRCT1=PCTG(NSQFF1,0N)

PRCT2=PCTG(NS0FF2,0N)

PRCT3=PCTG(NSQFF3,0N)

PRCT4=PCTG(NS0FF^,0N)

CALL STAT (ON t PCI 1 t PC11M, PC11S)

CALL STAT ( ON ,PC22 , PC22M, PC22 S)

CALL STAT ( ON ,PC33 , PC33M, PC33S )

CALL STAT ( ON ,PC^4, PC44M, PC44S)

WRITE (6,215)

WRITE (6,216)

WRITE (6,217) TN0,N0FF1,N0FF2,N0FF3,N0FF4,PRC1,PRC2,

1 PRC3,PRC4,0N,NS0FF1,NS0FF2,NS0FF3,NS0FF4,PRCT1,

2 PRCT2, PRCT 3»PRC T4

WRITE (6,218) PC1L,PC2L,PC3L,PC4L,PC1ML,PC2ML,
1 PC3ML,PC4ML,PC1SL,PC2SL,PC3SL,PC4SL

WRITE (6,2L9) PCI 1M,PC22M,PC33M, PC44M,PC 1 IS, PC22S ,
1 PC33S,PC44S

STOP

END

72

c

c

C SUBROUTINE MEAN1

C

SUBROUTINE MEAN1 ( J , N,K ,Z , P , AVE , I D ,L )

C

C SUBROUTINE MEAN1 MAKES A WEIGHTED AVERAGE OF PARAMETER

C P, USING SUBROUTINE LAYER1 FOR COMPUTATION OF

C THE WEIGHTS.

C

C

C MEANING OF THE PARAMETERS:

C INPUT:

C "J" - FLAG; AVOIO RECOMPJTAT I ON OF THE WEIGHTS IF

C THEY HAVE BEEN COMPUTED BEFORE.

C =0 IF SUBROUTINE HAS NOT YET BEEN USED

C WITH THE SAME WEIGHTS.

C =1 (OR DIFFERENT THAN ZERO) IF SAME

C WEIGHTS ARE TO APPLY.

C "N" - NUMBER OF OBSERVATIONS.

C "K" - = 2N-1 - DIMENSION OF ARRAYS AVE, ID, L

C "ZM - ARRAY FOR THE DEPTH. DIMENSION N.

C Z(l) MUST BE ZERO.

C "P" - ARRAY FOR THE QUANTITY TO BE AVERAGED.

C DIMENSION N.

C

C OUTPUT:

C "AVE"- AVERAGE ARRAY. WILL CONTAIN THE WEIGHTED

C AVERAGE FOR EACH DEPTH GIVE^ AND ALSO

C FOR EACH INTERMEDIATE DEPTH. DIMENSION K.

C "ID" - INTERMEDIATE DEPTH. DIMENSION K (SEE

C SUBROUTINE LAYEFU).

C "L" - LAYERS. DIMENSION K (SEE SUBROUTINE

C LAYER1).

C

DIMENSION Z(N),P(K) , AVE(K ) , ID (K ) , L(K),R(43)
IF (J.NE.O) GO TO 300

CALL LAYER1(N,K,Z,L,ID)

300 CONTINUE
SUM=0.D0
AVE(1)=P( 1)
DO 301 1=2, K

J=IFIX(FL0AT(I+1 )/2. )
R(I)=P(J)

SUM=SUM+L( I-1)*R(I)
AVE(I)=SUM/ID( I )

301 CONTINUE
RETURN
END

73

c

c

C SUBROUTINE LAYER1

c

C

SUBROUTINE LAYER1 (N,K,Z,L,ID)

C

C SUBROUTINE LAYER1 COMPUTES THE INTERMEDIATE DEPTHS

C OF A GIVEN WATER COLUMN AND THE LAYERS BETWEEN

C THEM. IT DIFFERS FROM SUBROUTINE LAYER BY

C CONSIDERING EACH GIVEN DEPTH AS AN INTERMEDIATE

C DEPTH.

C

C MEANING OF THE PARAMETERS:

C INPUT:

C "N" - NUMBER OF DEPTHS GIVEN.

C "K" - = 2N-1 - NUMBER OF INTERMEDIATE DEPTHS.

C DIMENSION OF ARRAYS Lt ID.

C "Z" - ARRAY FOR THE DEPTH. DIMENSION N.

C Z(ll MUST BE ZERO.

C

C OUTPUT:

C "L" - LAYERS IN BETWEEN EACH INTERMEDIATE DEPTH.

C K-l LAYERS ARE CONSIDERED.

C "ID" - INTERMEDIATE DEPTHS. BOTH EACH GIVEN

C DEPTH AND HALF-WAY IN BETWEEN TWO

C SUCCESSIVE GIVEN DEPTHS ARE CONSIDERED.

C IDU) IS THE SURFACE AND ID(K)THE LAST

C GIVEN DEPTH. DIMENSION K.

REAL ID,L

DIMENSION Z(N),ID(K),L(K)

M=N-1

DO 401 1=1, M
J=2*I

ID(J)=(Z( I )+Z( I+l))/2.
IF ( IDC J). EQ.700.) ID(J)=75Q.

401 CONTINUE

DO 402 1=1 fN
J=2*I-1
ID(J)=Z( I)

402 CONTINUE
C LAYERS

L(1)=ID(2>

J = J-1

DO 403 1=2, J

L(I) = ID( I+U-IDC I)

403 CONTINUE
RETURN
END

74

c

c

C SUBROUTINE LAYER

C
C

SUBROUTINE LAYER (NfZtLtIO)
C

C SUBROUTINE LAYER COMPUTES THE INTERMEDIATE DEPTHS

C OF A GIVEN WATER COLUMN AND THE LAYERS BETWEEN

C THEM. IT DIFFERS FROM LAYER1 BY NOT CONSIDERING

C THE GIVEN DEPTHS AS INTERMEDIATE DEPTHS.

C

C MEANING 3F THE PARAMETERS:

C INPUT:

C "N" - NUMBER OF DEPTHS GIVEN.

C "Z" - ARRAY FOR THE DEPTH. DIMENSION N.

C Z(l) MUST BE ZERO.

C

C OUTPUT:

C "L" - ARRAY FOR THE LAYERS. DIMENSION N.

C "ID" - ARRAY FOR THE INTERMEDIATE DEPTHS.

C DIMENSION N.

REAL ID,L

DIMENSION Z(N), LCN), ID(N)
C

C FIND INTERMEDIATE DEPTHS ID
C

M=N-1

DO 501 1*1, M

ID(I)=(Z( I +1H-ZC I))/2.

501 CONTINUE
ID(N)=Z(N)

C

C FIND LAYERS

L(1)=ID(1 )

DO 502 1=2, N

L(I) = ID(I)-ID( 1-1)

502 CONTINUE
RETURN
END

75

c

c

C SUBROUTINE CORREC

c

C

SUBROUTINE CORREC ( LAT , VC ,N ,Z ,T , S , L , V,COR)
C

C SUBROUTINE CORREC COMPUTES THE DEPTH CORRECTIONS

C DUE TO VARIATIONS IN SOUND SPEED. IT USES

C SUBROUTINE WILSON TO COMPUTE SOUND VELOCITY

C AT DEPTH Z, AMD APPLIES THE CORRECTIONS TO THE

C ASSUMEO SOUND SPEED VC, CORRECTIONS ARE DETERMINED

C BY LAYERS L AND SUMMED ALGEBRAICALLY TO GIVE THE

C CORRECTION AT THE BOTTOM OF EACH LAYER.
C

C MEANING OF THE PARAMETERS:

C INPUT:

C "LAT"- LATITJDE IN DEGREES.

C "VC" - SOUND SPEED ASSUMED IN THE FATHOMETER.

C "N" - NUMBER OF DEPTHS GIVEN.

C "Z" - ARRAY FOR THE DEPTH. OIMENSION N.

C Z(l) MUST BE ZERO.

C "T" - ARRAY FOR THE TEMPERATURE. DIMENSION N.

C "S" - ARRAY FOR THE SALINITY. DIMENSION N.
C

C OUTPUT:

C "V" - ARRAY FOR THE SOUND SPEED. DIMENSION N.

C "COR"- ARRAY FOR THE CORRECTIONS. OIMENSION N.

C

REAL L,LC

DIMENSION Z(N),r(N),S(N),L(N) ,COR(N)tV(N)

FAC( VI)=( VI-VO/VC

CALL WILSON ( LAT. Z { 1) , T ( 1) , S( 1) t V ( 1 ) )

LC=L(l)*FAC(V(l)>

COR(l)=LC

DO 600 I=2fN

CALL WILSON (LAT.Z(I)tT(I).S(I)tVm)

LC=L( I )*FAC(V< I) )

C0R(I)=C0R(I-1 )+LC
600 CONTINUE
RETURN
END

76

C SUBROUTINE OCRR

c

C

SUBROUTINE OCRR { N , COR1 ,COR, I D, DCOR,PC )
C

C SUBROUTINE OCRR GIVES THE DIFFERENCES BETWEEN

C THE DEPTH CORRECTIONS OBTAINED BY A PARTICULAR

C METHOD AND A REFERENCE METHOD.
C

C MEANING OF THE PARAMETERS:

C INPUT:

C "N" - NUMBER OF DEPTHS GIVEN.

C "C0R1"- ARRAY FOR THE CORRECTIONS FROM THE

C METHOD USED. DIMENSION N.

C "COR" - ARRAY FOR THE CORRECTIONS FOR THE

C REFERENCE METHOD. DIMENSION N.

C "ID" - INTERMEDIATE DETHS TO WHICH THE CORRECTIONS

C APPLY. DIMENSION N.
C

C OUTPUT:

C "DCOR"- ARRAY FOR THE DIFFERENCE IN THE

C CORRECTIONS. DIMENSION N.

C "PC" - PERCENTEGE OF THE DIFFERENCE TO

r THE DEPTH

C "PC" - ARRAY FOR THE PERCENTAGE OF THE

C DIFFERENCE TO THE DEPTH. DIMENSION N.

C

REAL ID

DIMENSION COR UN) » COR( N ) » ID< M ) , DCOR ( N) , PC i N )

DO 700 1=1, N

OCORd ) = C0R1< I )-COR( I)

PC(I)=DC0RU)/ID(I)*100.
700 CONTINUE
RETURN
END

77

c

c

C SUBROUTINE STAT
c

SUBROUTINE STAT ( M , A, ME AN, STOEV)
C

C SUBROUTINE STAT COMPUTES THE MEAN AND STANDARD
C DEVIATION OF THE ARRAY A, WHICH HAS M ELEMENTS
C
C DIMENSION MUST BE ASSIGNED IN MAIN PROGRAM FOR ARRAY A

C

DIMENSION ACM)
REAL MEAN
C

C COMPUTE MEAN

C

IF (M.GT.l) GO TD 800
MEAN=A(1)
STDEV=0.
RETURN

800 SUM=0.

DO 801 1=1, M

SUM=SUM+A( I)

801 CONTINUE
MEAN=SUM/FLOAT(M)

C

C COMPUTE STANDARD DEVIATION

C

SUM=0.

DO 802 J=1,M

TERM=( A< J)-MEAN)**2

SUM=SJM+TERM
802 CONTINUE

STDEV=SQRT( SUM/FL0AT(M-1) )

RETURN

END

78

c

c

C SUBROUTINE WILSON

C

C

SUBROUTINE WILSON ( LAT , Z, T, S, C)
C

C SUBROUTINE WILSON COMPUTES THE SOUND VELOCITY IN WATER

C USING WILSON'S EQUATION (I960). THE EQUATIONS USED

C IN THIS SUBROUTINE WERE EXTRACTED FROM "USER'S

C GUIDE TO NODC'S DATA SERVICES (NOAA, DEPARTMENT

C CF COMMERCE, 1974).

C

C THE PRESSURE USED IN THIS EQUATION IS COMPUTED IN

C AN ITERATIVE PROCESS ALONG DEPTH. THE SUBROUTINE

C SHOULD BE CALLED SEQUENTIALLY FOR THE DIFFERENT

C DEPTHS, STARTING FROM THE SURFACE (Z=3).

C

C MEANING OF THE PARAMETERS:

C INPUT:

C "LAT"- LATITJDE IN DEGREES.

C "Z" - DEPTH.

C "T" - TEMPERATURE.

C "S" - SALINITY.

C

C OUTPUT:

C "C" - SOUND SPEED.

C

T2=T**2

T3=T**3

T4=T**4

S2=S**2

S3=S**3

IF (Z.NE.O.) GO TO 900
Pl=10.1325
P = P1

DLAT=FLQ AT (L AT ) / 100 .
X=C0S(2.*DLAT*3.1415 927/13 0.)
Z1=0.

900 CONTINUE
C

C COMPUTATION OF DENSITY

C

Fl=-(T-3. 98)**2*(T+283.)/(503.57*(T+67.2S) )

F2=1.0843E-6*T3-9. 8185E-5*T2+4. 7867E-3*T

F3=1.667E-8*T3-8.164E-7*T2+1.803E-5*T

F=6.76786136E-6*S3-4.82 496140E-4*S2+8.14876577E-1*S

DCl=3.895414E-2

DC2=-0.22 5845 86
C

SGMT=F1+(F+DC1)*( 1 .-F2+F3*( F+DC2 ) )
C

C COMPUTATION OF PRESSURE

C

SGM0=-9.3 44 5863E-2+8.148765 77E-l*S-4.8 249614E-4*S2+

1 6.7678614E-6*S3
S3M02=SGM0**2

G0= 0.9806 16-2. 592 8E-3*X+6.9E-6*X**2
GI=G0+1.101E-7*{Z+Z1)
1=0
TP = P1

901 TP2=TP**2
TP3=TP**3

R=l.-(4.886E-6*TP/( l.+l .83E-5*TP ) )+TP*( -2 . 2072E-7+

1 3.673E-8*T-6.b3E-10*T2*4.E-12*T3+SGM0*( 1.7 25E-8-

2 3.28E-10*T+4.0E-12*T2)+SGM02*(-4.5E-ll+i.fc-12*T) ) +

3 TP2*(-6.68E-14-1 .240 64E- 12*T+2 . 14E- 14*T2+

79

4 SGM0*{-4.248E-13+1.206E-14*T-2.E-16*T2)+

5 SGM02*(1.8E-15-6.0E-17*T) ) +TP3*1 . 5E- 17*T
RQ=(1.+1. E-3*SGMT)/R

IF (Z.EQ.O.) GO TO 903

1 = 1 + 1
IF (I.NE. L) GO TO 902

TP=PH-R01*GI*(Z-Z1)

GO TO 901

902 TP«Pi+.5*(R01*R0)*GI*(Z-21)
IF (I.EQ.2) GO T3 901

C

903 P=.10197*TP
C

C SOUND SPEED EQUATION

C

c
c

P2=P**2
P3=P**3

VP=1.6027 2E-l*P+1.0263E-5*P2-»-3.5216E-9*P3
1 -3.3603E-12*f>4
VS=1.3979 9*(S-35, ) + 1.69202E-3*< S-35. )**2
VT=4.5721*T-4.453 2E-2*T2-2.6045E-4*T3+7.98 51E-6*T4
VSTP=(S-3 5.)*(-i. 12 44E-2*T+7. 771 1E-7*T2*7 . 7016E-5*P

1 -1.2 943E-7*P2+3.158E-8*P*T+1.579E-9*P*T2)

2 +P*(-1.860 7E-4*T+7.4812E-6*T2«-4. 5283E-8*T3)

3 +P2*(-2.52 94E-7*T+1.8563E-9*T2)+P3*(-1.9646E-10*T)

C=1449.14*VP+VS+VT+VSTP

P1 = TP
Z1 = Z

R01=R0
RETURN
END

80

c

c

C SUBROUTINE INTRPL

c

r

SUBROUTINE I NTRPL ( L , X ,Y, N, U, V)

C

C SUBROUTINE INTRPL IS PART OF THE IBM SCIENTIFIC

C SUBROUTINE PACKAGE EXISTING AT THE NAVAL

C POSTGRADUATE SCHOOL COMPUTER CENTER.

C

C INTERPOLATION OF A SINGLE-VALUED FUNCTION

C THIS SUBROUTINE INTERPOLATES, FROM VALUES OF THE FUNCTION

C GIVEN AS ORDINATES OF INPUT DATA POINTS IN AN X-Y PLANE

C AND FOR A GIVEN SET OF X VALUES (ABSCISSAS}, THE VALUES OF

C A SINGLE-VALUED FUNCTION Y * Y( X) .

C THE INPUT PARAMETERS ARE

C L = NUMBER OF INPUT DATA POINTS

C (MUST BE 2 OR GREATER)

C X = ARRAY OF DIMENSION L STORING THE X VALUES

C (ABSCISSAS) OF INPUT DATA POINTS

C (IN ASCENDING ORDER)

C Y = ARRAY OF DIMENSION L STORING THE Y VALUES

C (ORDINATES) OF INPUT DATA POINTS

C N = NUMBER OF POINTS AT WHICH INTERPOLATION OF THE

C Y VALUE (ORDINATE) IS DESIRED

C (MUST BE 1 OR GREATER)

C U = ARRAY OF DIMENSION N STORING THE X VALUES

C (ABSCISSAS) OF DESIRED POINTS

C THE OUTPUT PARAMETER IS

C V = ARRAY OF DIMENSION N WHERE THE INTERPOLATED Y

C VALJES (ORDINATES) ARE TO BE DISPLAYED

C DECLARATION STATEMENTS

DIMENSION X(L) ,Y(L) ,U(N) ,V(N)
EQUIVALENCE (PO, X3 ) , ( QO, Y3) , (Ql , T3 )
REAL M1,M2,M3,M4,M5 m „

EQUIVALENCE (UK, DX ) , ( IMN,X2, AL , Ml ) , ( I MX , X5, A5,M5 ) ,
i (JtSWrSA) , (Y2,W2,W4,Q2),(Y5,W3,Q3)

C PRELIMINARY PROCESSING

10 LO=L
LM1=L0-1
LM2=LM1-1
LP1=L0+1
NO = N

IF(LM2.LT.O) GO TO 90
IF(NO.LE.O) GO TO 91

DO 11 I=2,L0

IF(X( I-l)-X(I) ) 11,95,96

11 CONTINUE
IPV=0

C MAIN DO-LOOP

DO 80 K=1,N0
UK=U(K)
C ROUTINE TO LOCATE THE DESIRED POINT

20 IFUM2.EQ.0) GO TO 27
IF(UK.GE.X( LO) ) GO TO 26
IF(UK.LT.X( 1)) GO TO 25
IMN=2

IMX=LO

21 I=( IMN+IMX)/2
IF(UK.GE.X( I)) GO TO 23

22 IMX=I

GO TO 24

23 IMN*I+1 „ „,

24 IF(IMX.GT.IMN) GO TO 21
I=IMX

GO TO 30

81

25 1=1

GO TO 30

26 I=L?1

GO TO 30

27 1=2
CHECK IF I = IPV

30 IFU.E3.IPV)
IPV=I
ROUTINES TO P

40

41

42
43
45

46
47

48

ICK UP
TO ESTIMATE
J = I
IF(J.EQ.l)
IF(J.EQ.LPl)
X3=X( J-l)
Y3=Y(J-1)
X4=X(J>
Y4=Y( J)
A3=X4-X3
M3=(Y4-Y3)/A3
IFCLM2.EQ.0)
IFCJ.EQ.2)
X2=X{ J-2)
Y2=Y( J-2)
A2=X3-X2
M2=< Y3-Y2)/A2
IF(J.EQ.LO)
X5=X( J>1)
Y5=YU*1)
A4=X5-X4
M4=(Y5-Y4)/A4
IFU.EQ.2)
GO TO 45
M4=M3+M3-M2
GO TO 45
M2=M3
M4=M3

IF(J.LE.3)
Al=X2-X(J-3)
Ml=(Y2-Y(J-3))/Al
GO TO 47
M1=M2+M2-M3
IF(J.GE.LMl)
A5=X< J+2)-X5
M5=(Y(J+2)-Y5)/A5
GO TO 50
M5=M4+M4-M3

GO TC 70

NECESSARY X AND Y VALUES
THEM IF NECESSARY

AND

J=2
J=L0

GO
GO

TO
TO

43
41

GO TO 42

M2=M3+M3-M4

GO TO 46

GO TO 48

C NUMERICAL DIFFERENTIATION

50 IFU.EQ.LP1 ) GO TO 52
W2=ABS( M4-M3)

W3=ABS( M2-M1)

SW=W2+W3

IF(SW.ME.O.O) GO TO 51

W2=0.5

W3=0.5

SW=1.0

51 T3 = <W2*M2+W3=M3 )/SW
IF(I.EQ.l) GO TO 54

52 W3=ABS(M5-M4)
W4=ABS(M3-M2)
SW=W3 + W4

IF(SW.NE.O.O) GO TO 53
W3=0.5

W4=0.5
SW=1.0

53 T4=( W3*M3+W4*MV)/SW
IF(I.NE.LPl) GO TO 60
T3 = T4

82

54

DET

60

COM
70
80

ERR
90

91

95

96
97
99

C FOR
2090
2091
2095
2096
2097
2099

SA=A2+A
T4=0.5*
X3=X4
Y3=Y4
A3=A2
M3=M4
GO TO
T4=T3
SA=A3+A
T3=0.5*
X3=X3-4
Y3=Y3-^
A3 = A4
M3=M2
ERMINATION
Q2=(2.0
Q3=(-M3
PUTATION 0
DX=UK-P
V(K)=QO

RETURN
OR EXIT

WRITE (6,

GO TO

WRITE

GO TO

WRITE

GO TO

WRITE

WRITE

WRITE

RETURN
MAT STATEM

FORMAT (IX

FORMAT( IX

FORMAT! IX

F3RMATQX

F0RMATC6H

F0RMATC6H
1 36

ENO

(M4+M5-A2*(A2-A3)*(M2-M3 )/(SA*SA ) )

50

99

(6,

99

(6,

97

(6,

(6,

(6,

(MH-M2-A4*(A3-A4)*(M3-M4)/(SA*SA) )

4

2*A4

OF THE COEFFICIENTS
*<M3-T3)*M3-T4)/A3
-M3 + T3«-T4)/(A3*A3)
F THE POLYNOMIAL
0
+DX*(Q1+DX*(Q2+DX*Q3) )

2090)

2091)

2095)

2096)

2097) I,X(I)

2099) L0,N0

ENTS
/22H
/22H
/27H

/33H
I

L

*** L = 1 OR LESS./)

*** N = 0 OR LESS./)

*** IDENTICAL X VALUES./)

*** X VALUES OUT OF SEQUENCE./)

= ,I7,10X,6HXU) =,E12.3)

=, I7tl0X,3HN =» 17/

H ERROR DETECTEO IN ROUTINE

INTRPL)

83

LIST OF REFERENCES

1. Bivins , L.E., Requirements for XSTD probe, Memorandum

to Associate Director, Office of Marine Technology, NOAA,
13 August 1976.

2. Bowditch, N. , American Practical Navigator, 1975 Edition,
Defense Mapping Agency Hydrographic/ Topographic Center,
1975.

3. Calder, M. , Calibration of echo-sounders for offshore
soundings using temperature and depth, International
Hydrographic Review, V. 52, No. 2, p. 13-17, July 1975.

4. Chace, A.B., Naval Postgraduate School User's Manual

for the Integrated Command ASW Prediction System (ICAPS),
Oceanography Department, Naval Postgraduate School, June
1979.

5. Clay, S.C., and Medwin, H. , Acoustical Oceanography,
Principles and Applications, Wiley - Interscience , 1977.

6. Compton, J.S., for the Hydrographer , Royal Australian
Navy (AOD) , personal communication, subject: Correction
of offshore echosounders , 13 August 1980.

7. Del Grosso, V.A., Velocity of sound in sea water at zero
depth, U.S. Naval Research Laboratory Report 4002, 1952.

8. Greenberg, D.A., and Sweers, H.E., Possible improvement
of Matthews tables in areas of Canadian data holdings,
International Hydrographic Review, V. 49, No. 1, p. 127-
151, January 1972.

9. Heck, N.H., and Service, J.H., Velocity of sound in sea
water, Special Publication 108, U.S. Department of Com-
merce, U.S. Govt. Printing Office, 1924.

10. Hydrographer of the Navy (the), Admiralty Manual of
Hydrographic Surveying, Volume two, Chapter 3, Sounding,
N.P. 134b (3) , 1969.

11. Ingham, A.E., Sea Surveying, V. 1, Wiley, 1975.

12. International Hydrographic Bureau, Accuracy standards
recommended for hydrographic surveys, Special Publication
44, Monaco, 1968.

84

Kuwahara, S. , Velocities of sound in sea water and
calculation of the velocity for use in sonic sounding,
Hydrographic Review, V. 16, p. 123, November 1939.

14. Leroy, C.C., Development of simple equations for accurate
and more realistic calculation of the speed of sound in
sea water, Journal of the Acoustical Society of America,
V. 46, No. 1, part 2, p. 216, July 1969.

15. Mackenzie, K.V., Formulas for the computation of sound
speed in sea water, Journal of the Acoustical Society
of America, V. 32, No. 1, p. 100, January 1960.

Matthews, D.J., H.D. 282, Hydrographic Department, First
Ed., Admiralty, London, 1927.

Matthews, D.J., Tables of the velocity of sound in pure
water and sea water for use in echo- sounding and sound
ranging, H.D. 282, Hydrographic Department, Admiralty,
London, 1939.

18. Sherwood, R. , The use of a computer for correcting ocean
soundings, International Hydrographic Review, V. 51,

No. 2, p. 37-50, July 1974.

19. Sverdrup, H.V. , Johnson, M.W., and Fleming, R.H., The
Oceans, Their Physics, Chemistry and General Biology,
Prentice Hall, Inc., 1942.

20. Ulonska, A. , A more accurate hydrographic survey by direct
measurement of sound velocity in the sea, International
Hydrographic Review, V. 49, No. 1, January 1972.

21. Umbach, M.J., Hydrographic Manual, Fourth Edition, U.S.
Department of Commerce, U.S. Govt . Printing Office, 1976.

22. Urick, R.J. , Principles of Underwater Sound, Second Ed.,
McGraw-Hill, Inc., 1975.

23. U.S. Department of Commerce, National Oceanic and Atmos-
pheric Administration, Key to Oceanographic Records
Documentation No. 1, User's Guide to NODC ' s Data Services,
February 1974.

24. U.S. Department of the Navy, NSTL Station, The ICAPS
Water Mass History File, Naval Oceanographic Office
Reference Publication 19, May 19 78.

25. Wallace, J.L., Hydroplot/Hydrolog Systems Manual, U.S.
Department of Commerce, NOAA, 1971.

85

26. Wilson, W.D., Speed of sound in sea water as a function
of depth, temperature and salinity, Journal of the
Acoustical Society of America, V. 32, No. 6 , p. 5T4 ,
June 1960a.

27. Wilson, W.D., Equation for the speed of sound in sea
water, Journal of the Acoustical Society of America,
V. 32, No. 10, p. 1357, October 1960b.

28. Wilson, W.D., Extrapolation of the equation for the speed
of sound in sea water, Journal of the Acoustical Society
of America, V. 34, No. 6, p. 866, June 1962.

29. Yeager, D.W. , Coastal controls in vertical sound speed
determination and correction to echo-soundings , M.S.
Thesis, Naval Postgraduate School, Monterey, 1979.

86

INITIAL DISTRIBUTION LIST

No. Copies

Defense Technical Information Center 2

Cameron Station
Alexandria, VA 22314

Library, Code 0142 2

Naval Postgraduate School
Monterey, CA 93940

Chairman, Code 68 1

Department of Oceanography
Naval Postgraduate School
Monterey, CA 93940

Chairman, Code 63 1

Department of Meteorology
Naval Postgraduate School
Monterey, CA 93940

Assoc. Professor R.H. Bourke, Code 68 Bk 1

Department of Oceanography
Naval Postgraduate School
Monterey, CA 93940

LCDR G.B. Mills, Code 68 Mi 1

Department of Oceanography
Naval Postgraduate School
Monterey, CA 93940

2? Ten. L.L. Faria 1

Instituto Hidrografico
R. das Trinas, 49
Lisboa, Portugal

Director 1

Naval Oceanography Division

Navy Observatory

34th and Massachusetts Avenue NW

Washington, D.C. 20390

Commander 1

Naval Oceanography Command

NSTL Station

Bay St. Louis, MS 39529

87

10. Commanding Officer

Naval Oceanographic Office

NSTL Station

Bay St. Louis, MS 59529

11. Commanding Officer
Naval Ocean Research and

Development Activity
NSTL Station
Bay St. Louis, MS 39529

12. Director (Code PPH)
Defense Mapping Agency

Bldg. 56, U.S. Naval Observatory
Washington, D.C. 20305

13. Director (Code HO)

Defense Mapping Agency Hydrographic

Topographic Center
6500 Brookes Lane
Washington, D.C. 20315

14. Director (Code PSD-MC)
Defense Mapping School
Fort Belvoir, VA 22060

15. Director

National Ocean Survey (C)
National Oceanic and Atmospheric

Administration
Rockville, MD 20852

16. Chief, Program Planning and Liaison (NC-2)
National Oceanic and Atmospheric

Administration
Rockville, MD 20852

17. Chief, Marine Surveys and Maps (C3)
National Oceanic and Atmospheric

Administration
Rockville, MD 20852

18. Director

Pacific Marine Center - NOAA
1801 Fairview Avenue East
Seattle, WA 98102

19. Director

Atlantic Marine Center - NOAA
439 W. York Street
Norfolk, VA 23510

88

20. Director do
Instituto Hidrograf ico
R. das Trinas, 49
Lisboa, Portugal

21. Direccao do Servico de Instrucao e Treino
Ministerio da Marinha

Praca do Comercio
Lisboa, Portugal

22. CDR. D.E. Nortrup
Commanding Officer
NOAA Ship Peirce
439, West York Street
Norfolk, VA 23510

23. LCDR. D.W. Yeager
Field Procedures Officer
439, West York Street
Norfolk, VA 23510

24. LT. A. Dreves
NOAA Ship Davidson
FPO Seattle, WA 98102

25. Ms. P. A. Eaton
SMC 2181

Naval Postgraduate School
Monterey, CA 93940

26. Cap. Ten. F.V. Abreu
Instituto Hidrografico
R. das Trinas, 49
Lisboa, Portugal

27. LCDR. D.D. Winter
SMC 1745

Naval Postgraduate School
Monterey, CA 93940

28. LT. K.W. Perrin

NOAA Ship Mt. Mitchell
439, West York Street
Norfolk, VA 23510

89

Thesis

F2253

c.1

Thesi s

F2253

c.l

19166H
Fan a

Use of historical

sal ini ty data in the

computation of depth

correction due to

variations in sound

speed.

191660

Faria

Use of historical
sal ini ty data in the
computation of depth
correction due to
variations in sound
speed.

thesF2253

Ummlm^r!Cai Salini,V da'a in the c

3 2768 002 13364 7

DUDLEY KNOX LIBRARY