United States
            Environmental Protection
            Agency
            Air And Radiation
            (ANR-461)
EPA520/1-91-009
June 1991
f/EPA
Ocean Current Measurements
At The Farallon Islands
Low-Level Radioactive Waste
Disposal Site 1977 -1978

-------
                                         EPA 520/1-91-009
       Ocean Current Measurements
                      at the
                Farallon Islands
Low-Level Radioactive Waste Disposal Site
                   1977 - 1978
                   Prepared under
                 Contract 68-01-0796
                     Project Officer
                     Robert S. Dyer
                Office of Radiation Programs
              U.S. Environmental Protection Agency
                    Washington, D.C.
                                 Printed on Recycled Paper

-------
                                   FOREWORD
             Pursuant to Public Law 92-532 (the Marine Protection, Research and
Sanctuaries Act of 1972), the U.S. Environmental Protection Agency (EPA) has
developed criteria and regulations to govern ocean disposal  of all forms of waste,
including low-level radioactive waste (LLW) materials.

             In 1974, the EPA Office of Radiation Programs (ORP) initiated feasibility
studies to determine whether existing technologies could be  applied toward assessing the
fate of radioactive wastes that had previously been disposed in the oceans.  The ORP
developed a program for site characterization studies to determine the biological,
chemical, geological  and physical characteristics of the  marine environment in and near
sites that had been designated by the former Atomic Energy Commission (AEC) for
ocean disposal of LLW. These studies also included investigating the presence and
distribution of radionuclides within these sites.

             A primary mechanism for physically dispersing and redistributing both
soluble and particulate radioactive materials from a disposal site is the action of ocean
bottom currents.  Of particular interest is the magnitude and direction of these currents.
This report discusses the results of ocean bottom current measurements obtained from
the Farallon Islands LLW disposal site off the California coast, near San Francisco.  The
report includes a discussion of the velocity of the currents over the time  period and area
measured relative to large-scale currents off the California coast, and the possibility for
shoreward transport of LLW materials  from the Farallon Islands site.

             The Agency invites all readers of this report to send any  comments or
suggestions to Mr. Martin P. Halper, Director, Analysis and Support Division, Office of
Radiation Programs (ANR-461), U.S. Environmental Protection Agency, Washington, DC
20460.
                                                          T. Oge, ActmgyDirector
                                                          of RadiatiojPrograms
                                        111

-------
                           CONTENTS
Section                                                     Pagt

1.         INTRODUCTION                                     1

2.         SUMMARY                                          3
2.1        DESCRIPTION OF STUDY                              3
2.2        RESULTS AND CONCLUSIONS                           3

3.         MEASUREMENT PROGRAM                             7
3.1        OPERATION AND OBJECTIVES                          7
3.2        DESCRIPTION OF HARDWARE                          8
3.3        CURRENT METER ARRAY LOCATION AND OPERATION       9
3.4        BATHYMETRY                                      15
3.5        SEDIMENT INFORMATION                            15

4.         DATA REDUCTION                                  18
4.1        CURRENT METER DATA                              18
4.2        GENERATION OF BATHYMETRIC CHART                 18
4.3        GRAIN SIZE FROM SEDIMENT SAMPLES                 19

5.         CURRENT METER DATA PROCESSING                   21
5.1        DATA PRODUCTS                                    21
5.2        VACM AND AANDERAA COMPARISON                   35
5.3        MEASUREMENT ACCURACY AND TIMING ERRORS         41
5.4        CONDUCTIVITY AND SALINITY                        45

6.         DATA INTERPRETATION                              46
6.1        ANALYSIS OF CURRENTS                             46
6.2        TRANSPORT POTENTIAL                             62

7.         RECOMMENDATIONS FOR FURTHER WORK              68

8.         REFERENCES                                      71

     APPENDICES [Contained In A Separate Report - EPA 520/1-91-009/A]
          TIME HISTORY - SPEED AND DIRECTION               A-l
          TIME HISTORY PLOTS - NORTH, EAST CURRENTS         B-l
          HISTOGRAMS                                     C-l
          SCATTERGRAMS                                   D-l
          SPECTRA - NORTH, EAST COMPONENTS                E-l
          PROGRESSIVE VECTOR DIAGRAMS                    F-l
          STICK PLOTS                                     G-l
          SUMMARY STATISTICS                              H-l

-------
                               LIST OF FIGURES
Number                                                             Page

1-1   Locator Chart                                                  2

2-1   Current Meter Arrays for Radioactive Waste Disposal Site           4

3-1   Current Meter Locations                                       11
3-2   Data Coverage Plot 1977-1978                                   12
3-3   Profile Schematic                                              13
3-4   Current Meter Array Plan Schematic                             14
3-5   Locations of Core Sample  Sites                                  16
3-6   Depth Profile of Core Sample Sites                              17

4-1   Bathymetry Chart                                              20

5-1   Time History Plot, Current Speed, and Direction (Meter 2920)      22
5-2   Time History Plot, North Current, East Current (Meter 2920)       24
5-3   Histogram Current Speed  (Meter 2920)                          26
5-4   Speed Versus Direction Scattergram (Meter 2920)                 28
5-5   Power Spectrum (Meter 2920; February 1978)                     29
5-6   Progressive Vector Diagram (Meter 2919)                         31
5-7   Stick Plot (Meter 2918)                                         33
5-8   Summary Statistics (Meter 2919)                                34
5-9   VACM Reconstruction 1                                        37
5-10  VACM Reconstruction 2                                        38
5-11  VACM Reconstruction 3                                        39
5-12  VACM Reconstruction 4                                        40
5-13  Speed Gain Transfer Function for Meter 2830 Versus VACM        42
5-14  Speed Phase Transfer Function for Meter 2830 Versus VACM       43

6-1   Power Spectrum for Meter 2918                                 50
6-2   Power Spectrum for Meter 2919                                 51
6-3   Power Spectrum for Meter 2830                                 52
6-4   Power Spectrum for VACM                                     53
6-5   Power Spectrum for Meter 2920                                 54
6-6   Tidal Ellipse Calculation                                       55
6-7   Tidal Ellipse for Meter 2918                                     56
6-8   Tidal Ellipse for Meter 2919                                     57
6-9   Tidal Ellipse for Meter 2830                                     58
6-10  Tidal Ellipse for VACM                                         59
6-11  Tidal Ellipse for Meter 2920                                     60
6-12  Exceedence Curves for Total Measurement Periods                 66
6-13  Exceedence Curves for October through December 1977            67

                                       vii

-------
                               LIST OF TABLES
Number                                                           Page


3-1   Farallon Islands Current Meter Locations                         9

3-2   Data Record                                                  10

3-3   Box Cores Collected for Sediment Studies During
       R/V VELERO IV Cruise (August-September 1977)                15

5-1   Estimated Clock Drift                                         44

6-1   Major Current Energy Sources for Entire Operating
       Period  of Each Meter                                         49

6-2   Vector-averaged Current Velocities                              61

6-3   Sediment Surface Grain Size Data from Farallon
       Islands LLW Site Survey Cores                                63

6-4   Percent of Total Time Measurement Speeds Exceed 20 cm/s        65
                                       IX

-------
1.  INTRODUCTION
      This report describes the work performed by Interstate Electronics Corporation
(IEC) under contract  to the U.S. Environmental Protection Agency's (EPA) Office  of
Radiation Programs (ORP). It presents data from an ocean current measurement study,
conducted during 1977 and 1978, in the area of the U.S. low-level radioactive waste (LLW)
disposal  site near  the  Farallon Islands (Figure 1-1) off  the coast of San Francisco,
California. It also contains other oceanographic and environmental data acquired from the
same area.

      The purpose of this  study was to measure near-bottom and bottom currents in the
area, and utilize available historical data, to determine the potential for transport of LLW
from the disposal site  toward  populated areas  in the vicinity of San Francisco.  The
remainder of this report is sub-divided into sections, as follows:
             Section 2 describes the study and summarizes results and
             conclusions;

             Section 3 discusses the measurement program that generated
             the data analyzed in this report;

             Section 4 describes the steps taken to improve the quality of
             data from two of the current meters, and gives an assessment
             of the overall quality and temporal  coverage of the current
             meter data;

             Section 5 discusses data processing and verification;

             Section 6 contains an interpretation of the current meter data
             combined with other available data taken at the site during
             previous studies conducted in 1974  and 1975,  including an
             analysis of the current, sediment, and bathymetric data and of
             transport mechanisms related to the data;

             Section 7 contains recommendations for future sampling;

             Section 8 lists the references utilized during the study; and,

             the Appendices contain computer generated graphical displays
             of the current meters' output data.

-------
D DUMP SITES

O STATIONS SAMPLED FOR WATER, SEDIMENT, OR  BIOTA

O CURRENT METER ARRAYS
                           Figure 1-1.   Locator Chart.

            Dashed line indicates area defined [11 as the Waste Disposal Area; squares
      A, B, and C are disposal sites.  Circled numbers 1 through 10 are the locations of
      January 1977 sampling stations [2].  Hexagons B, C, and D indicate the locations
      of current meter arrays during this (1977-1978) study.

-------
2. SUMMARY

      This section  briefly summarizes the major events of this current  measurement
program, the subsequent analysis of data, and the conclusions obtained.  The analysis of
data included the following five tasks: (1) assessing data quality, (2) comparing data from
the different types of current meters used, (3) determining relative clock drift and timing
errors, (4) evaluating the measured current components, and (5) estimating potential for
sediment transport of LLW from the disposal site. The conclusions derived from each task
are given in Sub-Section 2.2., below.  Examples to support these conclusions are discussed
in the appropriate portions of Sections 3 through 7 of this report.
2.1 DESCRIPTION OF STUDY

      In October 1977 four mooring line arrays containing seven current meters were
deployed in the vicinity of the LLW disposal site near the Farallon Islands  (Figure 2-1).
The instruments on the arrays included six Aanderaa model 5 Recording Current Meters
(RCM), and one Sea-Link Vector-Averaging Current Meter (VACM). All were pre-set for
an operational period of approximately 90 days, but actual deployment lasted for about a
year.

      During recovery, in October 1978, it was observed that many of the meters were
capable of continuous data recording for a longer period of time than was expected (Figure
3-2).  This  is especially true for  the VACM, which continued to record data for the  entire
year.

      It was also observed during recovery that the transmitter on the  acoustic release
mechanism for Array A did not transmit when interrogated. When the release mechanism
was activated the array could not be located. Thus, two of the RCM 5 meters (numbers
2916 and 2917) were lost.
2.2 RESULTS AND CONCLUSIONS

2.2.1  Data Quality

      Data was recovered from five current meters-four Aanderaas and one VACM. The
VACM recorded data for 1 year, while the Aanderaas recorded for a period of 2 to 4Vi
months each.  After recovery of the  meters  and subsequent initial inspection, it was
observed that  the VACM was still  recording data, while two of the Aanderaa meters
(numbers 2919 and 2920) had filled their output tapes.  The other two Aanderaa meters
(numbers 2830 and 2918) had experienced battery failures during their operation. Details
regarding the operation and performance of the hardware can be found in Sections 3.1 and
3.3.

-------
                                  TO SURFACE
ARRAY B
         ARRAY D
                      ARRAY C
                                                              ARRAY A
       -  SUBSURFACE BUOYS
    A   ANCHOR/ACOUSTIC RELEASE
       AANDERM  CURRENT METER
       VECTOR-AVERAGING CURRENT METER (VACM)
    (NOT DRAWN TO SCALE)
Figure 2-1.   Current Meter Arrays  for Radioactive Waste Disposal  Site
                                      4

-------
      All of the data recovered from the VACM and two of the Aanderaa meters (numbers
2918 and 2919) was of a quality suitable for analysis and interpretation.  The data from the
other two Aanderaa meters (numbers 2830 and 2920) had to be recovered by a special
technique, as described in Section 4.1, but after this additional processing it was also found
to be of good quality. There were a few isolated data points in some of the records for each
of the meters which had obvious nonphysical values due to either electronic noise or
mechanical problems, but these were replaced by interpolated values yielding a continuous
record of high quality.  The  data was then translated to metric engineering units for
analyses and generation of data  products. See Section 4.1 for further discussion.

      The statistical  properties of the recovered data were generally in accordance with
expected physical processes, although it became  evident that the VACM data record
contained a number of transpositions in data sequence. In addition, Aanderaa # 2830
apparently contained  gaps  in its recording. Approximately 9 hours of data were missing
from the beginning of the record commencing October 26,1977. Further discussion appears
in Section 5.2.

2.2.2  Comparison of the VACM and Aanderaa Meters

      After taking into account the observed inconsistencies in the VACM data sequence,
the corresponding VACM and Aanderaa meter # 2830 records were compared by examining
the time series and obtaining the speed transfer function.  The data records were found to
be a relatively close match.  The speed transfer function was relatively flat and fell within
a range of 0.8 to 1.05, with the VACM giving a generally larger magnitude than observed
on  meter  # 2830. The phase portion of the transfer function was linear, indicating no
substantial timing rate shifts between the two meters (see Section 5.2).

2.2.3  Relative Clock Drift

      The semidiurnal tidal occurrences were examined to determine the degree of relative
clock drift for the 5 meters. Analyses in both the time and frequency domains indicated
that although clock drift appeared to be of a larger magnitude than advertised by the meter
manufacturers, relative timing errors were within a range suitable for cross-analysis of all
five current meter data records.  Section 5.3 addresses the details of this analysis.

2.2.4  Evaluation of Measured Current Components

      For each meter the short-term periodic and long-term average current components
were examined (see Section 6.1)  with the following  results:

      a.     In the deep western part of the dump site, most of the current energy is from
             12-hour tides in the east-west direction;

-------
      b.    In the deep eastern part of the dump site, most of the current energy is from
            12-hour tides with a  slightly greater component of energy in the north-south
            direction;

      c.    The total current energy is significantly greater in  the deep western portion
            of the site than in the deep eastern portion;

      d.    The VACM and Aanderaa meter # 2830 in the open ocean (midwater) show
            a relatively equal distribution of tidal energy and also have a significant low
            frequency component in the north-south direction;  and,

      e.    Long-term currents  average less than 2 cm/s  and move in the northeast
            direction for the bottom measurements (Aanderaa meters  2918, 2919, and
            2920) and the  southeast direction  for  the deep midwater  (VACM  and
            Aanderaa # 2830) measurements.
2.2.5  Estimation of Sediment Transport Potential

      Analysis of sediment distributions has indicated that fine-grain sediments composed
the majority of sediment volume ^. Since these sediments can occasionally be suspended
when water velocities exceed about 20 cm/s, an estimate of transport potential was obtained
by determining the extent to which these velocities occur.  Average currents were then used
to determine the probable distance and direction of transport (see Section 6.2).  The general
conclusions are as follows:

      a.     In the deep western part of the site, current measurements exceed 20 cm/s no
             more than 3 percent of the time. This may be sufficient to suspend fine-grain
             sediments (silt and clay), providing a potential for transport in the water
             column.

      b.     Long-term average near-bottom currents move north and eastward throughout
             the site.  Average current speeds diminish from 1.7 cm/s at the deep western
             end of the site to 0.17 cm/s at the eastern end, thus it appears that this vector
             decreases with proximity to the shore.

      c.      Results, though inconclusive, indicate that there may be reason to  further
             investigate the  extent and  scale of transport as it relates to contaminated
             sediment.

-------
3.  MEASUREMENT PROGRAM
      On October 25, 1977 four arrays containing six Aanderaa current meters and one
VACM were installed in an area of the radioactive waste disposal site near the Farallon
Islands  (Figure 1-1). This section discusses the procedures and operations used in the
deployment of the meters and the subsequent acquisition of data.
3.1 OPERATION AND OBJECTIVES

      A total of seven current meters were to be emplaced using four arrays, supported
by subsurface buoys (Figure 2-1). One array (A) was lost, consequently no data were
recovered from Aanderaa meters 2916 and 2917. The remaining four Aanderaa meters
and the VACM functioned and recorded current velocity, temperature, and conductivity
measurements for between 2 months and 1 year for each meter, giving a cumulative
volume of data totaling 800 days.

      The selection of the current measurement hardware was based on a number of
factors including  (1) cost, (2) anticipated performance at the depths and durations
required for this program, and (3) known handling and reliability characteristics. The
Aanderaa RCM, model 5, was deemed most suitable in terms of these considerations.
The VACM was deployed immediately above one of the Aanderaa meters in order to
study the operational characteristics of the two meter types relative to each other.

      The study was designed to provide preliminary information about the speed and
direction of water moving through the region of the disposal  site.  This information,
along with the results of previous physical and chemical studies of the region, was
intended to aid in the development of an understanding of the natural physical processes
and any subsequent impact related to potential or existing transport of radioactive
materials from the  disposal area.

      Deployment  of the arrays  (Figure 2-1) was planned to establish two approximately
triangular planar surfaces of measurement;  one parallel to and within a few meters of
the bottom, (arrays A, B, and D,  and  the other approximately parallel to the 3,000-foot
depth level (arrays A, B, and C).   In addition, the positioning the VACM and Aanderaa
meter # 2830 close together on the same string would allow for a direct comparison of
these two meter types in a midwater location, where the assumption could be made that
both meters were  experiencing the same water mass dynamics.

-------
3.2 DESCRIPTION OF HARDWARE

       The Sea-Link VACM records current speed and direction in cartesian coordinates,
average water temperatures, and timing information sampled at an adjustable interval
for a period as  long as 1 year. Current speed is obtained from a savonius rotor, and
direction is measured relative to magnetic north by means of a neutrally buoyant
magnetic-coupled vane mounted above the rotor.

       The VACM records eight current speed and direction measurements with each
turn of the rotor.  The readings are divided into components, combined, and referenced
to time and temperature on a cassette tape recorder at preselected intervals. This
averaging is believed to filter out high frequencies caused by mooring motion.  A 15-
minute interval was selected for this measurement program.  The meter is powered by a
self-contained alkaline battery pack with a life expectancy of 9 months at an average
current speed of 150 cm/s.  The operational ranges and accuracies for measured
parameters are:
             Current speed:             2.5 to 300 cm/s

             Current direction:         0* to 360' 2.8*

             Temperature:             -2*C to 35*C O.10'C

             Timing accuracy:          6 minutes/year

       The Aanderaa RCM-5  current meter also uses a savonius rotor to record current
speed. The meter is mounted in front of a large vane so that the rotor is always pivoted
into the oncoming current. In addition to current speed and direction and water
temperature, the instrument may contain optional devices for measuring conductivity
and water depth.  One of the  meters used in this study, # 2920, recorded conductivity by
means of an inductively coupled toroidal coil. The measurement interval was preselected
at 20 minutes to allow for approximately 4 months of recording.  Battery life is 3
months.  Operational ranges and accuracies for measured parameters are:

             Current speed:            1.5 to 250 cm/s

             Current direction':        0' to 360* 0.1*

             Conductivity:             0.0 to 60 millimoho/cm

             Timing accuracy:          12 minutes/year

      'Aanderaa RCM's have demonstrated compass errors due to magnetization of
nickel plating in pressure cases.  The meters deployed for this survey used new epoxy-
coated pressure cases with nickel only in the upper 0-ring region.

                                        8

-------
3.3  CURRENT METER ARRAY LOCATION AND OPERATION

      Table 3-1 and Figure 3-1 give the locations, site depths, and meter depths of each
current meter in the water column. Figure 3-1 also shows the local bathymetry in
relation to the current meters.  The meters were deployed along a gully running roughly
east-west.  The VACM and Aanderaa meter # 2830 were attached on the same array
anchored at the bottom in a depression, and the other Aanderaa meters were anchored
on a ridge just north of the gully with meters 2918 and 2919 attached on the same array.
Bottom topography increased from east to west
                Table 3-1.  Farallon Islands Current Meter Locations
Meter
(Array)
Aanderaa
2920 (B)
Aanderaa
2830 (C)
VACM (C)
Aanderaa
2918 (D)
Aanderaa
2919 (D)
Start
10/25/77
10/25/77
10/25/77
10/25/77
10/25/77
End
3/15/78
2/06/78
10/24/78
12/21/77
3/10/78
North
Latitude:
3736'36"
3736'52"
3736'52"
3736'51"
3736'51"
Wesl
Longitude
12307'32"
12314'46"
12314'46"
12317'27"
12317'27"
Site
Bep*
914
1372
1372
1829
1829
Meter
Depth
911
912
911
1800
1826
      Table 3-2 and Figure 3-2 show the beginning and end of the data records, and the
number of days of recorded data for each current meter. The four Aanderaa meters
recorded from 57 to 141 days of data at 20-minute intervals while the VACM recorded
one complete year of data at 15-minute intervals.

      Figures 3-3 and 3-4 show the meter array schematically in profile and plan views,
respectively.  The loss of the two meters on Array A (Figure 2-1) resulted in a loss of a
third reference point, which was necessary for the full definition of the two planes
discussed in Section 3-1; however, a sufficient amount of information was recorded by
the remaining meters from which to draw preliminary conclusions.

-------
                               Table 3-2. Data Record
Meter
Aanderaa 2920
Aanderaa 2830
VACM
Aanderaa 2918
Aanderaa 2919
Start Date
10/25/77
10/25/77
10/25/77
10/25/77
10/25/77
End Date
03/15/78
02/06/78
10/24/78
12/21/77
03/10/78
Number of
Days
141
104
365
57
136
Number of
JJata Records
10154
7547
35018
4016
9800
       Although the array was designed to acquire water flow information for 3 months,
considerably more data were acquired, most notably from the VACM. The following
observations were made during the recovery operation conducted on October 24,1978,1
year after deployment:
             Meter # 2830 collected data on approximately 75 percent of the output
             tape before battery failure occurred.  Timing problems with the tape drive
             mechanism necessitated a special data recovery technique described in
             Section 4.1.
             Meter # 2918 collected data on about 50 percent of the tape before
             experiencing a battery failure.
             Meters 2919 and 2920 had both completely filled their respective output
             tapes. Meter # 2920 had experienced tape drive problems similar to the
             problems experienced with meter # 2830 (see Section 4.1).
            The VACM meter was still functioning at the time of its recovery.
                                       10

-------
3736'60"
3736'55
37 36'50'
3736'45"
3736'40"
3736'35"
          ARRAY
                                              LONGITUDE  (W)
                  12317'
12315'
12313'
                    1239'
                                o
                                o
                                     \
                                     \

                                     +ARRAY C
    'ft

o
og
1237'
1235'
                                                                                ARRAY B
                                   (VERTICAL EXAGGERATION = 19:1)
                               Figure 3-1.  Current Meter Locations

-------
AANDERAA   .57 DAYS
NO. 2918

AANDERAA   .
NO. 2919
                  136 DAYS
AANDERAA   ,_
NO. 2920
                   141  DAYS
AANDERAA   h
NO. 2830
               104 DAYS
    VACM   I-
                                  365 DAYS
OCT
NOV|DEC
1977
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
1978
              Figure  3-2.   Data Coverage Plot 1977-1978
                                 12

-------
                                             LONGITUDE (W)
     500
SI   1000
UJ
Q
    1500
   2000
12317'      12315'
  ~i	r
                                           12313'
                    S/N  2918
                    S/N  2919
                                 o;
 123U09'
	1	
                                               12307'
                 12305'
                                  "
     VACM
     S/N  2830
                                                                                S/N 2920
                0 km
3.9 km

  HORIZONTAL DISPLACEMENT FROM ARRAY D
14.4 km
                                     Figure 3-3.   Profile Schematic

-------
   N
   >,
ARRAY D
                         ARRAY  C
                                                      ARRAY D - AANDERAA 2918,  2919

                                                      ARRAY C - AANDERAA 2830,  VACM

                                                      ARRAY B - AANDERAA 2920
                                       TOP VIEW
                                  CURRENT METER ARRAY


      (HORIZONTAL SCALE =  10 TIMES VERTICAL SCALE)
                          Figure 3-4.   Current Meter Array Plan Schematic

-------
3.4 BATHYMETRY

      Bottom depth profiles were acquired during surveys of the Farallon Islands LLW
disposal site in 1974,1975, and 1977. The equipment used during the 1977 survey
consisted of two precision depth recorders, a GIFFT recorder serving as the primary
sounding device, and an EDO-Western recorder for backup.  Sounding profiles were
made with both devices.

      Navigation equipment for the 1977 survey included a Motorola Mini-Ranger III,
which received reference signals from Point Reyes and Montara Point This allowed
position readings with an estimated accuracy of approximately 200 meters.

3.5 SEDIMENT INFORMATION

      During August and September 1977, six box cores were collected for sediment
analysis ^.  The locations of those coring sites are shown in Figures 3-5 and 3-6 and
Table 3-3. The sample at Station 41A was not fully recovered because the jaws of the
box core did  not fully close during retrieval.
       Table 3-3.   Box cores collected for sedimentological studies during R/V
                         tO IV cruise (August-September 1977)
Station
<
13A
2A
47
39
41A
48
Coring Bate
8/31/77
8/3 1/77
9/1/77
9/1/77
9/1/77
9/1/77
Latitude
37*38.1'N
3739.8'N
37-383'N
37*38.0'N
37-38.0'N
37-36.6'N
Longitude
123*08.0'W
123-07.1^
123-14.0W
123-17.0W
123-20.7^
123-12.7W
Water
Depth (in)
1042
878
1335
1469
2350
1216
Total
Recovery
Cm)
33
60
34
37
surface
grab
45
       Sediment subcore samples were obtained from each of the box cores (except
Station 41A) and were investigated for sediment properties, x-radiographic properties,
and radiochemical properties.  See ^ for the procedures followed to obtain and analyze
the samples.
                                        15

-------
       I2330'
123 00
37 50' -
3730  -
     D  NUCLEAR WASTE DUMP SUB5ITES
     *   BOX CORING STATIONS


  AB  IS AN  EAST-WEST TRANSECT AT NORTH LATITUDE 37 38'

     Figure 3-5.   Locations of Core Sample Sites ^.
                           16

-------
       r SEA LEVEL
oc.
LU
UJ
I
I-
o.
    500 -
   1000-
   1500-
   2000 -
   2500
  3000-
                                               NUCLEAR WASTE  DUMP SUBSITES

                                            I   BOX  CORING  STATIONS
                                                  10 Km
                                    VERTICAL  EXAGERATION  8:1
                12320.7' 12317' 12314'
                                            12308'
                                     -OFFSHORE
                          AB IS EAST-WEST TRANSECT
              Figure 3-6.   Depth Profile of Core Sample Sites [31.

-------
4. DATA REDUCTION

      In this section the data reduction for the current meters is described through the
initial stages of processing, and the quality of the resulting data is assessed.  Steps taken
to improve the quality of the data are also described.
4.1 CURRENT METER DATA

      The initial processing of the current data was accomplished in five stages: (1)
data tapes were translated into a computer-readable digital format, (2) data tapes were
checked for format errors such as missing bits in a word, missing words in a record, or
missing records (these errors were then flagged for later processing), (3) digital counts
from the translated data were converted to metric  engineering units,  (4) data in
engineering units were checked for physically unreasonable values, and (5) both format
errors and physically unreasonable data were removed and replaced by linearly
interpolated values.

      When the resulting data were examined it was found that Aanderaa meters 2830
and 2920  contained a  large amount of unexplained scatter and were deemed unsuitable
for analysis.  In an attempt to recover the data the raw binary data tapes from these
meters were examined with an oscilloscope.  It was found that the pulse widths for the
binary ones and zeros varied beyond manufacturer specifications during the record.
However,  the ratio of the pulse widths to the time  between the trailing edges of
consecutive pulses remained constant. The tapes for meters 2830 and 2920 were then
retranslated using this ratio to distinguish between ones and zeros in the record. In this
way, it was possible to recover the data for the entire period of record for both current
meters.

      Verification of the current meter data in terms of relative calibration and timing
accuracy is discussed in Section 5.2.
4.2 GENERATION OF BATHYMETRIC CHART

      The bathymetry obtained was annotated to the ship's position at the time of
deployment, and the resulting position and depth coordinates were compared with those
found on an oceanographic data chart, "LBL Publication 92," produced by the Lawrence
Berkeley Laboratory ^, and with bathymetric data from previous  surveys in 1974 and
1975.  The bathymetric measurements taken during the deployment of the meters
                                        18

-------
generally corroborate the larger features depicted on published bathymetry. However,
smaller features were observed in the local bottom terrain which are not shown in the
standard publications.  This situation was also observed with regard to the soundings
taken during the 1975 survey.  The soundings performed in 1977 indicate the presence of
an east-west running depression as shown in Figure 2-1. A  section of the LBL chart has
been reproduced and is shown in Figure 4-1.
43 GRAIN SIZE FROM SEDIMENT SAMPLES

       The tube cores (subcores) described in Section 3.5 were separated by 2-millimeter
and 63-micrometer sieves to determine the percentages of sand, silt, and clay.  Sand is
defined as being composed of particles between 2 millimeters and 63 micrometers in
diameter, while silt and clay are less than 63 micrometers in size.

       Cumulative probability distributions of grain size were obtained for each of the
sampling stations (see ^).  Statistical parameters, including mean, median, standard
deviation, skewness, and kurtosis were obtained for each distribution as a function of
depth beneath the bottom surface in order to determine the cross-sectional profile of the
sediments. This information is used in Section 6.2 for determination of transport
potential from the site.
                                        19

-------
38N
                                                                          Kilometers
                                                                           Nautical
                                                                            Miles
                                            WASTE
                                            DISPOSAL
                                            AREA
                                                                 BATHYMETRY PLOT

                                                                 100-METER INTERVALS

                                                                 500-METER INTERVALS
   30'
20'
10'      123W     50'
10'      122W
                           Figure 4-1.   Bathymetry Chart

-------
5.  CURRENT METER DATA PROCESSING
      This section describes the processing and verification of current meter data that
were performed prior to interpretation.  The data from each meter were separated into
data sets after preparation (e.g., translation, conversion, editing, and interpolation)
described in the previous section. The processing consisted of applying various
statistical algorithms to the data in both the time and frequency domain. The output
consisted  of time history plots, histograms, scattergrams, progressive vector diagrams,
stick plots, power spectral density plots, and listings of summary statistics.
5.1 DATAPRODUCTS

       Various statistical functions were performed on the data to examine different
aspects of current conditions at the site.  Time history plots chart continuous
measurements of current flow and direction; histograms and scattergrams provide
individual and combined distributions of the measured quantities; power spectra show
periodic components in the data; and progressive vector diagrams and stick plots
illustrate the long-term average components in the data.

       Definitions and samples of the data products prepared from the reduced data are
given below.  A complete set of plots and tables are contained in  the Appendices.
5.1.1  Time History Plots of Current Speed and Direction

       Monthly plots are included for each current meter on a horizontal scale of 1 day
per division.  Speed is plotted on a scale of 5 cm/s per division from 0 to 40 cm/s, and
current direction is plotted on a scale of 45 degrees per division. To aid in the
interpretation of current directions, the pen was lifted whenever two consecutive current
directions differed by more than 180 degrees, with the assumption that the current
vector went through 0 degrees (True) between these readings (e.g., for three successive
readings of 359', 0*, and 1', the pen would be up between the first and second point, but
down between the second and third). A reading of 0 degrees (True) for current direction
implies that the current is moving toward true north, while 45* corresponds to northeast,
etc.  A sample plot for February 1978 is given for Aanderaa meter # 2920 in Figure 5-1.
The average velocity for this sample is approximately 4 cm/s, with a mildly predominate
northeast direction. The complete set is provided in Appendix A.
                                        21

-------
                           RRNDERRR  2920
            TIME  HISTORY  PLOT    CURRENT  SPEED
0
(_>
LU
1 I 2 13
         r4
yij-
f 6 I 7 Is" I 9 "1 "*i 0 "rtr"T 12 13.
                           li> ~l 16 In I" 18 T19 I 20 ~1 21 122 1 23 1 24 I 2'j> 26 ' 27 ' 28
                        FEDRUHHY  1978
   M1
' \  ,\
o
LLJ
a
                           RRNUtRRR  2920
            TIME  HISTORY  PEOT    CURRENT  DIRECTION
      '3 ' i  I :,  I 6  T 7  Ta 1 j I 10 r n ( 12 I 13 f u I it. 1 i I iv 1 is I i 1 20 I 21 1 i.1" I"22i I 24


                        FEBRUHRl  1978

            Figure 5-1.  Time History Plot, Current Speed, and Direction
                          (Meter No. 2920)

-------
      The time history plots provide a quick and complete visual summary of the
activity of the recorded data over the extent of the plot.  Both long-term and short-term
trends are visible in the display and implications can be made concerning the
periodicity, stationarity, ergodicity, and overall quality of the data.  Periodicity may be
defined as the degree to which the time history is composed of periodic events.
Stationarity and ergodocity are closely allied concepts.  A data set may be considered
stationary if its statistical properties (i.e., mean, variance, etc.) do not exhibit significant
trends with time, and it is ergodic if the statistical properties of its data subsets do not
significantly differ for different subsets.  Stationarity and ergodicity are necessary
preconditions for spectral analysis.  Visual inspection can frequently be used to pick out
questionable data records (such as excessive values represented by spikes) and locate
places where there was a sensor failure (flat spots in the data or unexplained changes in
the character of the time series).

       The plot of current speed in Figure 5-1 clearly shows the tidal periods in the data
and also shows that the data are reasonably stationary and ergodic.  The data is
evidently of good quality,  as there are no unexpected changes in the overall appearance
of the plot.  Peak events are visible on the 9th, 13th, 23rd, 24th and 27th of the month.
The direction time  series shows evidence of the tidal rotation in the corkscrew
appearance of the data (i.e., the way it appears to increase through a 360* range with the
passage of time).
5.1.2  Time History Plots of U and V Current Components

       The current speeds and directions are also represented in U (eastward-flowing)
and V (northward-flowing) components.  The plots are provided on a horizontal scale of
1 day per division and a vertical scale of 10 cm/s per division.  The U and V components
are derived as follows:

                    U = S  sin(a)
                    V = S  cos(a)
             where
                    U = East component (cm/s)
                    V = North component (cm/s)
                    S = Current speed (cm/s)
                    a = Current direction (degrees true)
      A sample plot for February 1978 is given for meter # 2920 in Figure 5-2. The
complete set is contained in Appendix B.

                                         23

-------
                            flRNDERflfl  2920
            TIME HISTORY PLOT    NORTH  CURRENT
S 2-
2 ~T~3  f 4 ~TS 1 6 I 7 ~ T8 Tg I
                      jo \\ 12
                                     n  a! 9
                  T2o rVi~T22~1 23
                                                       I 26 '27 I za '
                        FEBRUflRY  1978
                                       2920
            TIME  HISTORY PLOT    ERST  CURRENT
n I |2 I |3 I u I |'j I IS 1|7 I 18 ' 19

J;LBRUl1RY  1 978
                                               "22 I 23 I 24 I 2b "I 26 TJ7 "1 28~>
               Figure 5-2. Time History Plot, Horth Current, East Current
                             (Meter No. 2920)

-------
      These plots corroborate the speed and direction plots by showing the tidal periods
as before, and by revealing no abrupt changes in statistical character and no
nonstationary processes. The currents appear to be more or less symmetrical with
respect to direction, showing that the predominant factors affecting current are periodic
in nature.  Tidal influences seem to prevail in these plots.  These observations are
supported by inspection of the associated power spectral density plots (see Section 5.1.5).
5.13  Histograms

       Histograms for each meter are included to provide the distributions of the
measured quantities over the entire period of record.  The following histograms were
calculated:
             Current speed using 2.O cm/s intervals

             Eastward  (U) current component using 4.O cm/s intervals

             Northward (V) current component using 4.0 cm/s intervals

             Conductivity (for meter # 2920 only) using O.02 millimoho intervals

       A sample histogram of current speed for meter # 2920 is given in Figure 5-3.
The complete set is contained in Appendix C.

       The frequency histogram displays the frequency of occurrence within specified
class  intervals for the parameter  shown in the headings.  The class intervals are along
the horizontal axis, with the frequency of occurrence within each interval shown on the
vertical axis. The lower bound for each interval is greater than or equal to the value for
the class, and less than the value of the next higher class (i.e., the histogram for speed is
in 2 cm/s intervals and  interval class "10" represents all the occurrences of speed greater
than or equal to 10.0 cm/s and less than 12.0). Note that this representation will make a
distribution appear to shift to the left by half a class interval.

       Actual frequency for each interval is listed vertically in the heading above the
interval. The total number of events (in this case  10154) is found to the right of the
heading. The resolution of the frequency bars is variable depending on  the data being
plotted.  The frequency  scale is adjusted to give maximum resolution  on each plot.  Data
flagged as erroneous are not counted in any interval or in the total number of events.

       The histogram of Figure 5-3 assumes the general form expected for a non-
negative parameter such as speed. The histogram is skewed towards the origin, or zero
interval class (where 1500 events  out of 10154 are  observed), and displays a gradual
taper  towards the higher amplitudes to the right. There are no isolated events in the
higher class intervals, suggesting  that there are probably no unreasonable spikes in the
corresponding time series plot.
                                         25

-------
INTERSTRTC CLBCTTONICS CORP.
OCFJRNIC ENGINEERING DIV.
gWHEIM. CR  "IW 772-2811                                                               SITE DEPTH -  gu METERS

TI^TQI i SrTM nSrF1^J?9^5 ,  l"10^ S/N ~ 292    METER REF - 7H9   RNCHORED BUOT B   METER DEPTH -  911 METERS
ipSl^LHTION DRTE 10/24/77 IQW HRS   RNCHOR DN   35 HRS   RNCHOR RELEflSE  620HRS   RECOVERT  S/Hi/78 620 HRS
POSITION  37 36 56. ON 125  7  32.1W MflG. VRRIRTION - -17 DECS  TIME ZONE -  POT CGMT  H>8 HRS)    SRMP INTV - 20 MIN

                                               FREQUENCY HISTOGRflM
                                                       SPEED
     1 2
     5961
     032311
     22^082400000000000000000000000000000000000000000000
     31
FREQ. 1 1 3
     12051
     88297520000000000000000000000000000000000000000000
3118  	
                                                                                                             TOTflL
                                                                                                             N3. OF
                                                                                                             EVENTS

                                                                                                              10154


                                                                                                             ERRORS

                                                                                                                  0



1870






1247






0623



       \


                 * WHW*VBBlHHMMM_KM_HH>BMa>a*WW^**HW^W**^MMM(<|llB*  JBW.H^_aw^^W^MH  ^
1    2
0    0
                    3
                    0
4
0
5
0
6
0
7
0

SPEED
8
0
1
0
0
1
1
0
1    1
3    4
0    0
1    1
5    6
0    0
1    1
7    8
0    0
                                                       2.00 CM/SEC  INTERVftLS
                             Figure 5-3.  Histogram Current Speed  (Meter No.  2920)

-------
5.1.4  Scattergrams

       Scattergrams for each meter are included to provide joint distributions of
measured quantities taken over the entire period of record. The following Scattergrams
were calculated:

             Current speed versus current direction  with speed binned at 2 cm/s
             intervals and direction binned at 18 degree intervals.

             Northward (V) versus eastward (U) current components with both
             components binned at 4 cm/s intervals.

       A sample plot of current speed versus direction for meter # 2920 is given in
Figure 5-4.  The complete set is contained in Appendix D.

       The scatter diagram is a two-dimensional histogram that tabulates the number of
occurrences of specified combinations of two variables. When the number of events in
each box is  divided by the total number of observations, the result may be interpreted as
the joint probability density. The class intervals are  indicated on the ordinate and the
abscissa.  The size of the matrix is limited to 20 by 20.  The range for each interval is
limited to integer values.  Data flagged as erroneous  in either of the variables causes the
occurrence to be dropped from the calculation.  The total number of correct occurrences
represented in the diagram is printed at  the top of each diagram.

       Scattergrams can be  used to assess the degree of statistical dependence of two
measured parameters. The  scatter diagram of Figure 5-4 shows the relationship of
current speed to current direction for meter # 2920.  It is apparent that the greatest
current speeds occurred when the current direction was north, and no currents larger
than 14 cm/s were observed  with a southerly direction.  The scattergram is smoothly
distributed over the 360 degree range, and there are no speed/direction events isolated by
more than one class interval from the distribution. There also are no conspicuous holes.
5.1.5  Spectral Density Plots of U and V Components

      These plots were derived by the method of averaged periodograms from the U and
V components of the current time series.  The scale of the monthly spectral plot was
determined by the peak spectral density in order to aid in the resolution of spectral
peaks. The horizontal scale is frequency in cycles/hour and the vertical  scale is spectral
density in units of velocity squared times hours, where velocity is in cm/s.  A sample plot
for February 1978 is given for meter # 2920 in Figure 5-5.  The complete set is
contained in Appendix E.
                                         27

-------
INTEftSTWE ELECTRONICS CORP.
GCERNIC ENGINEERING DIV.
flNFlHEIM.  Cfl C7H4)  772-2811                                                                SITE DEPTH -  91H  METERS
^Irn, ,Tnr?nM S2W5fl^-5     METER S/N - 2920     METER  REF - 7M   RNCHORED BUOT B   METER DEPTH -  911  METERS
inlTrHiS11^ S^US^iS7 19W HRS   nNCM3R DN   35 HRS   nNCHOR REVERSE  620HRS   RECOVERY  3/U/78 620 MRS
POSITION   37 36 36. ON 123  7 32. IN MflG. VRRIRTION -  -17  DECS  TIME ZONE -  PDT CGMT  +08 HRS)   SRMP INTV - 20 MIN
       SPEELJ
      CM/SEC
    ua.oo
    36.c
    32. C
    26.

    2U.

    22.

    20.

    18.

    16.

    14.

    12.

    1Q.

     8.C
                                                  SCRTTER DIflGRRM
                                         NUMBER  OF  EVENTS PER  MflTRIX UNIT
                                              SPEED     VS         01R
TOTRL  NUMBER OF EVENTSi  1015H
   FLRGGED DRTfl POINTS!      0
.GO
.00
.00
.00
.00
.00
.00
.00
.00
.00
.ac
.00
.00
.00
.aa
.00
.00
.00
.00
nn









l

l
2
1
8
18
56
158
155
93










1
1
3
6
19
16
109
180
183
8











1
2
5
1U
21
78
1145
165
71








1


2
2
9
17
ue
67
178
176
79









1

1
1
7
21
45
87
167
181
95












6
5
21
35
81
186
169
90











1
2
U
6
28
98
175
173
103











1

2
7
29
89
199
164
115












1
2
1
20
61
136
169
103













1
8
10
11
112
113
86













3
3
18
46
122
162
76












1
6
17
25
63
116
155
66











1
10
15
10
59
92
110
136
53










1
1
6
16
37
61
80
155
164
52










1

U
11
17
38
8>1
ma
161
57










1
1
2
11
20
32
U7
137
me
6









1
1

1
7
10
22
51
117
121
49










3

1
2
8*
21
61
115
121
60







2
2
5
2
3
4
7
9
22
67
128
129
59






2
2
2
4
7
4
5
8
13
25
50
160
140
63
        0.00       36.00      72.00      100.00     144.00     160.00     216.00      52.00      269.00     321.00     360.00
             16.00      54.00      90.00      126.00     162.00     193.00     234.00      270.00     306.00     342.00

                                                         DIR
                                                       DEC

                        Figure  5-4.  Speed Versus Direction Scattergram  (Meter No. 2920)

-------
                            NORTH CURRENT
    320

    280-

    240

 I 200
CM
    160
 00
 -v.
 5 120


     80


     40
      0
      0.
    00      0.25      0.50      0.75      1.00      1.25      1.50
                            CPH
                            EAST CURRENT
C\J

 00
 o
320

280

240

200

160

120

 80

 40
      0
      0.
    00      0.25      0.50      0.75      1.00      1.25      1.50
                            CPH
     Figure 5-5.  Power  Spectrum (Meter No. 2920; February  1978)

                                 29

-------
       The spectral density function of a parameter displays the amount of variance
 around the mean caused by components that occur with a regular frequency.  Thus, it is
 a method of describing the mean square value or relative energy of a time series as a
 function of frequency.  Periodic components of the data, such as diurnal and
 semidiurnal tidal currents, are represented by peaks in the frequency spectrum.  The
 relative size of these peaks is an indication of the relative importance of the
 corresponding periodic process.  Higher peaks demonstrate that more energy is
 contained in the measured parameter at the respective frequency.

       The spectral density can be thought of as the Fourier transform of the
 autocorrelation function, and is calculated using standard FFT algorithms. Figure 5-5
 clearly shows the spectral peak for the semidiurnal tides in both the east and north
 currents during February 1978.  Some energy at higher frequencies is also visible,
 especially for the east current, but nothing significant exists at frequencies higher than
 0.5 cycles per hour.
 5.1.6 Progressive Vector Diagrams

       Progressive Vector Diagrams (PVDs) are provided for each of the 5 meters
 showing the cumulative vector sums of the east and north current components over the
 life of the meter. The PVDs were calculated from time series of low pass filtered data in
 order to eliminate tidal variations from the plot, and give a smooth curve depicting long-
 term trends. Each of the east and north component time series were filtered with a
 cosine-tapered symmetric filter with a cutoff frequency of 0.020 cycles per hour
 (corresponding to a 50-hour period). The PVDs were then plotted to a uniform scale of
 40 kilometers per inch. A sample plot is given in Figure 5-6 and the remaining plots are
 included in Appendix F.

       PVDs provide a visual representation of the sequential current observations
 recorded by a current meter. Current speeds and directions are broken into their U and
 V components and  are summed from consecutive data and plotted in vector form-tip to
 tail.

       An important fact to remember when using PVDs, is that the instrument is fixed
 in space as it records current data. The vector sum of the components is not the path of
 a water particle through space, but rather the sum of instantaneous current
 measurements at one fixed point.  The current velocity field must be determined from a
 sufficiently defined  array of meters before water particle trajectories can be derived.
 However, a PVD can be used to support observations about current behavior in the
 immediate vicinity of the meter, or to compare measurements from separate locations.
 The PVD in Figure 5-6 for meter # 2919 shows the prevailing direction to be toward the
 north with a  slight tendency to the east.  The current vector is relatively straight,
 suggesting that a fairly steady residual (after removal of tides) current exists at this
point.
                                         30

-------
  FflRflLLON ISlflNO  CURRENT  METER DflTR
       flflNOERflfl METER  2919
       CCT 25. 1977  - M3R  10,  1973
STSHT.iO/25/77   3 0
EMO  . 3/10/73   7 0
S3WUHG INTEftVBLi   ;.00 W5
  PROGRESSIVE VECTOR OlflC-RflM
U        CM/SEC          V

               -^.  220.
K. KN
                                         _^ 202
                                         __ MO. ,
                                         __ IQQJ
                                         _.  30.
                                         __ SO.
                                         _- 40.,
                                                            CM/SEC
-120.   -100.    -80.    -SO,    -40.     -20.     1.     2C.     40.     SO.     80,     100.
                                         _L  -20.
           Figure 5-6.   Progressive Vector  Diagram (Meter  No. 2919)
                                         31

-------
 5.1.7 Stick Plots

       Three-hour averages of speed and direction are plotted on a polar coordinate
 system that moves through time.  Stick plots showing current speed and direction are
 included in Appendix G.  The length of the vector represents speed, and the angle of the
 vector represents direction in the true geographic coordinate system (degrees of direction
 increase clockwise from north, which is vertically oriented on the diagram).  Time is
 indicated along the bottom of the diagram. Plotted beneath each stick plot is a time
 series plot of speed for quick cross-reference.  Peaks in the time series plot can be
 matched with their respective vectors on the stick plot.

       The stick plot contains the same information as the PVD, except that the vectors
 are plotted on  a horizontal scale instead of end to end. Figure 5-7 shows a stick plot for
 meter # 2918. Inspection reveals a wide range of scatter in both current magnitude and
 direction, but there appears to be a strong tendency for the current to travel east-west,
 as evidenced by the relatively shallow angle between the stick plot vectors and the
 horizontal axis.
 5.1.8 Summary Statistics

       Daily, monthly, and whole study summary statistics are compiled for each current
 meter.  The statistics consist of mean, minimum, and maximum values, and also
 standard deviation (SIG), skewness (SKW), and kurtosis (KUR).  These statistics are
 calculated for current speed and north (V) and east (U) current components.  A sample
 listing of daily statistics for meter # 2919 is given in Figure 5-8.  Note that the time
 given for each entry in the table is the time at which summary statistics were generated
 for all preceding data since the last entry. The remaining tabulations are included in
 Appendix H.

       The summary statistics show the first four moments of the data distribution and
 can be used to assess data quality and stationarity.  For stationary data the statistical
 properties remain constant over time.  Consequently, if the statistical moments (mean,
 variance, skewness, and kurtosis) remain within reasonable ranges, the time series  can
 be considered stationary.

       The variance or the standard deviation provide measures of the dispersion of the
 data about its mean. An excessive degree of dispersion can indicate discontinuities in
 the time series, Or bimodality in the distribution, while a very small dispersion  (i.e.,
 standard deviation approaching zero) may indicate the  absence of a meaningful signal.
This may occur in the event of a sensor failure.

      The skewness represents the degree of asymmetry, or departure from symmetry,
of a statistical distribution. If the distribution has a longer tail to the right than to the

                                         32

-------
   .
no :u-2o/7i   i?
HfEKOCIHC IfclfR.IL. 110. OH 1IH
                                                            AANDERAA METER  #2918
                                       SPEED
 STICK PLOTS
CM/SEC         01R      OEG
                                                            20  Jt ;2 23  ^* M 1$  21  2t 23 30  .  2  3
                                              Figure  5-7.   Stick Plot (Meter No.  2918)

-------
INTERSTATE ELECTRONICS CORP.
OCEANIC ENGINEERING QEV
ANAHEIM.CA (71*) 772-2811
SUMMARY STATS FOR U
START TIME: 10/25/77 05:
METER TYPE - AANDERAA RCH-5
1NTH
10
10
10
10
 10
10
11
11
11
11
11
11
11
11
-: 11
11
11
11
1 1
11
11
11
: 11
11
11
11
11
11
11
11
11
11
 11
11
11
11
12
12
12
1 2
12
12
1?
12
=. 12
j 12
12
ii 12
12
OAY
26
27
28
29
30
31
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
23
29
30
1
2
3
4
5
6
7
8
9
10
11
12
13
TR
77
77
77
77
77
77
77
77
77
77
77
77
77
77
77
77
77
77
77
77
77
77
77
77
77
77
77
77
77
77
77
77
77
77
77
77
77
77
77
77
77
77
77
77
77
77
77
77
TIME
8
3
3
3
3
a
8
3
,3
3
3
3
3
3
3
8
3
3
3
3
3
3
3
8
8
3
3
3
3
3
3
3
3
3
3
3
3
3
3
;)
MEAN
.92
.23
1.00
.14
1.70
1.27
2.30
-.63
.47
-.01
-.32
.70
-.04
-1.09
.31
-.52
-.57
.50
1.00
-.36
1.59
2.15
.76
1.17
1.12
1.21
1.15
2.43
-.30
.71
.61
.97
-.82
-1.07
.71
.59
1.28
.30
1.33
.59
.97
-.15
-.54
-.48
-1.64
.57
-.98
-.93
.00
MIN
-7
-10
-3
-11
-7
-7
-6
-11
-7
-1C
-10
-a
-11
-22
-20
-20
-20
-5
-10
-11
-7
-10
-9
-12
-14
-15
-9
-22
-8
-7
-14
-23
-22
-14
-16
-9
-10
-15
-ft
-8
-18
-2ft
-22
-26
-25
-25
55 - END TIH:
METER  - 2919
MAX
10
11
12
13
12
7
14
7
7
17
12
12
18
12
14
12
3
9
a
9
11
11
14
19
14
14
13
9
3
11
13
14
23
8
19
11
13
12
11
10
12
16
14
14
14
15
SIG
5.06
3.76
3.55
5.60
4.39
3.66
5.12
4.56
3.96
5.86
4.92
5.01
6.53
8.68
3.12
7.94
7.29
4.24
4.41
4.40
3.97
4.25
4.72
5.19
7.29
7.47
4.89
5.51
3.06
4.03
3.96
5.90
3.76
a. 33
8.55
6.31
5.43
5.30
7.33
6.52
5.19
4.42
7.25
8.57
J.77
7.23
3.57
9.45
9.68
: 3/13/78 07:48
SAMP INTV - 20 HIM
SKU
.13
-.03
.16
.29
-.15
-.50
.55
-.34
-.20
.96
.27
.13
.24
-.08
-.56
-.12
-.76
-.28
-.09
.01
-.22
-.82
-.27
.04
-.29
.16
.13
-.31
-.93
-.04
-.06
-.63
-.34
-.57
.35
-1.25
.73
-.07
-.64
-.69
.07
.43
-.oa
-.01
.26
-.73
-.90
-.71
-.93
KUR
1.49
3.37
1.91
2.52
2.51
2.17
2.30
2.13
1.90
3.26
2.22
2.03
1.59
2.54
2.29
2.63
3.25
1.97
2.20
2.20
2.26
2.80
2.09
2.94
2.19
2.51
2.39
2.19
3.36
2.32
2.19
2.57
3.00
2.93
1.87
3.66
3. SB
1.75
2.ao
2.91
1.74
2.45
2.08
2.09
3.52
3.35
3.39
2.90
3.48
          Figure 5-8.   Summary  Statistics (Meter No. 2919)
                                   34

-------
left of its mean, it has a positive skewness. Likewise, it has a negative skewness if the
distribution is slanted to the left. Quantities which are expected to follow a Gaussian
distribution, such as the east and north current components, tend to exhibit little
skewness. Current speed, on the other hand, should tend towards a positive skewness
since the distribution is bounded on the left by zero.

      The kurtosis is a measure of the degree of peakedness of a distribution. The
normal  distribution has a kurtosis of 3. A distribution with a  steeper peak has a higher
kurtosis and is called leptokurtic. A low kurtosis, corresponding to a platykurtic
condition, is indicative of a relatively flat-topped distribution.  A failed sensor can
frequently be identified by the presence of a leptokurtic condition in the  measured data
since, in this case, the measured signal will frequently assume  a constant or near
constant value.  Platykurtic  data may indicate bimodality in the distribution.

      The minima and maxima are useful in isolating extreme events for closer study,
and may also be used to corroborate the amount of dispersion  indicated  by the standard
deviation.

      Figure 5-8 lists daily  summary statistics for the east current from meter # 2919.
Based on the physical nature of the data, the east current would be expected to follow a
more or less Gaussian distribution, and a  glance at the tabulated values  of kurtosis,
which range between  1.49 and 3.68,  suggest a mild degree of platykurtosis. There is also
a degree of negative skewness, indicating the presence of a slant towards the right.  This
observation is verified by the frequency histogram (Appendix C).  Consequently the PVD
for meter # 2919 might be expected to occupy the first or fourth quadrants, and it does
in fact does occupy the first quadrant.
5.2 VACM AND AANDERAA COMPARISON

      The VACM and Aanderra meter # 2830 were located less than 1 meter apart on
the same string and were almost 0.5 kilometer above the bottom.  These two meters were
compared by looking at the gain and phase transfer functions between their respective
speed measurements.  Since the meters had different sample rates, the VACM output
was interpolated to give equivalent sample rates for comparison. In the ideal case when
the meters measure exactly the same phenomena, the transfer function magnitude would
be equal to one at all frequencies.  The phase angle would show any lead or lag between
the meters for different frequencies, and would be linear.

      Prior to calculating  the transfer function between the two meters, the lower order
statistical functions were reviewed to determine how closely the  statistical parameters
were in  agreement. It was  discovered that the PVDs were not in agreement,  and that the

                                         35

-------
VACM data was displaced in time relative to the Aanderaa meter by more than 2
months.  Close visual inspection of the two sets of time series indicated that the two
meters were measuring the same events over the following intervals:

      Aanderaa # 2830                      VACM

       10/25/77  4:22                   1/09/78 16:41
        2/06/78  23:42                   4/24/78 21:06

      It is possible to visually match these two data records on a peak-by-peak (and
indeed almost a point-by-point) basis.

      Thus, it is apparent that at least some of the VACM data was transposed in time.
Evidently, at some point during the initial handling and transcription of the raw VACM
data tape, portions of the data record were processed out of sequence, and the altered
data sequence was retained during all subsequent processing until its discovery during
the analysis described in this report.  Further scrutiny of the VACM's PVD revealed the
presence of many unaccounted for cusps in the progressive vector path, suggesting that
these might be break points where transpositions occurred.  Efforts were made to
reconstruct the correct sequence of data by examining  the data around the cusps and
matching trends with data beginning or ending at other cusps. Each of the
reconstructions began with the portion that matches the PVD of meter # 2830. In
Figure 5-9, a new PVD was constructed by removing the known out-of-sequence data
segment at the beginning of the record to the end of the record while leaving the
remainder  of the sequence intact. In Figure 5-10, an effort was made to match the
directions of the low-pass filtered data on either  side of the cusp. Figure 5-11 was
constructed in the same manner, but using unfiltered PVD segments so that short
duration direction trends could be matched.  Figure 5-12 was made by observing the
visual record of spring and neap tides in the unfiltered PVD  sequence, and attempting to
construct a logical sequence from these observations. No  one single reconstruction
presented itself as being indisputably correct, and several  possible alternatives are shown
in Figures 5-9 through 5-12. Note that all of the reconstructions, as well as  the original
PVD, result in a markedly eastward vector average.

      In accordance with the above findings, the comparison between meter # 2830 and
the VACM used the 104-day stretch of VACM data starting on January 9, 1978. It was
possible to visually identify corresponding events in the two sets of series, so a timing
comparison was made between the two meters by matching peaks visually and
comparing  elapsed times since the beginning of the record.  This leads to the observation
that, although the relative drift was approximately 17 minutes for the last 80 percent of
both records, the VACM appeared to lag behind  meter #  2830 by around 9  hours for the
first 20 percent of the compared intervals. Visual inspection of the speed time series for
meter # 2830 suggests that there are gaps in the beginning of its data.
                                        36

-------
                            VACM RECONSTRUCTION  1

                                 A JAN
                                 B FEB
                                 C MAR
                                 D APR
                                   JUN
                                   MAY
                                 6 JUL
                                 N SEP
                                   OCT
                                   AUG
Figure 5-9.  VACM Reconstruction 1

                37

-------
                              VACM RECONSTRUCTION 2
                                      NOV
                                      AUG
                                    K OCT
                                    L DEC
Figure 5-10.  VACM Reconstruction 2

                 38

-------
                              VACM RECONSTRUCTION 3

                                  A JAN
                                  B FEB
                                    MAR
                                    JUN
                                    APR
                                    MAY
                                    JUL
                                  H SEP
                                  I OCT
                                  J AUG
                                  K NOV
                                  L DEC
Figure 5-11.  VACM Reconstruction 3

                39

-------
                                 VACM RECONSTRUCTION 4
                                        JAN
                                        FEB
                                        MAR
                                      D APR
                                      E
                                      F
                                      G
MAY
JUL
JUN
                                      H SEP
Figure 5-12.  VACM Reconstruction 4
                                        OCT
                                        AUG
                                        NOV
                                        DEC
                 40

-------
      The speed transfer function was obtained between meter # 2830 and the VACM
by # 2830/VACM.  The unfiltered time series were used so that high-frequency events
would be retained in the comparison. The gain and phase functions are shown in
Figures 5-13 and 5-14, respectively. It is seen that the speed transfer function is fairly
flat and within a range of 0.8 to 1.05, with the VACM generally giving a larger
magnitude than # 2830. The phase portion of the transfer function is linear, indicating
no substantial timing rate shifts between the two meters. This is the type of gain and
phase relationship that should  exist between similar sensors measuring the same
parameter under equivalent conditions.
5.3 MEASUREMENT ACCURACY AND TIMING ERRORS

      The main known phenomenon in the current measurement field is the occurrence
of tides.  All of the current measurements contained a significant amount of tidal energy
at approximately 12-hour intervals.  This  information was used to determine timing
errors. Once it had been determined with an initial  inspection of the time series that
there were no gross timing errors other than the one described for the VACM in the
previous  section, meter timing was evaluated by locating tidal peaks and comparing
elapsed time since meter startup with the time obtained by multiplying the tidal period
by the number of elapsed tidal periods.
      Semidiurnal tides contain two major frequency components, one due to the
gravitational influence of the moon, the other due to the pull of the sun. The lunar
component has a period of approximately 12 hours, 25 minutes and is the major
component. The solar contribution has a 12-hour period and accounts for around 46.6
percent as much influence as that of the lunar tidal contribution. In accordance with
these facts, the tidal peaks were assumed to be 12 hours 25 minutes apart and heavily
contaminated by other frequencies, mainly the solar 12-hour tide.

      Several tidal peaks were selected from the time series for each meter to establish
clock drift as a function of time. Peaks were selected for incorporation into the timing
calculation based on the following criteria:

             only peaks which had the greatest  speed magnitude in a given
             24-hour  period were selected; and,

             these speed peaks were used only if they occurred
             simultaneously with the daily maximum for the east-west
             current component.
                                        41

-------
    1.60
    1.401
    1.20
    1.00
    0.80
5  0.60
5

    0.40
    0.20-
    0.00
       0.00    0.02    0.04    0.06    0.08    0.10    0.12    0.14    0.16    0.18    0.20     0.22    0.24
                                                FREQUENCY CPH
                  Figure  5-13.   Speed Gain Transfer Function for Meter No. 2830 Versus VACM

-------
 200
-200.00
      0.00    0.02    0.04    0.06    0.08    0.10    0.12    0.14    0.16     0.18     0.20     0.22    0.24
                                                  FREQ CPH
               Figure  5-14.   Speed  Phase Transfer Function for Meter No. 2830 Versus VACM

-------
      A reference peak was chosen from the beginning of the time series, and the
elapsed number of lunar semidiurnal periods (12 hours, 25 minutes) was calculated for
each candidate peak with reference to the reference peak.

      The calculated elapsed tidal intervals consisted of an integral number of periods
plus some residual fraction of a period.  Since the relative clock drift for each meter was
found to be within one tidal period for each meter (see bottom of page 41) it was
expected that the calculated residuals would be close to each other so long as the
candidate peaks were all coincident with lunar semidiurnal tides.

      In order to eliminate peaks that were not coincident with the lunar tides, the
residuals were histogrammed to tenths of a tidal period to determine which bins
contained the greatest numbers of peaks. In this manner those peaks belonging in the
most frequented bins were readily identified as lunar  semidiurnal tidal peaks, while the
remainder were discarded.

      A least-squares regression analysis was performed on the final set of residuals to
obtain a linear relationship between the residuals and the elapsed number of periods.
The slope from this relationship is the estimated rate of clock drift for the meter under
consideration.  Table 5-1 contains the estimated clock drift for each of the five meters.
                          Table 5-1.  Estimated clock drift
Meter
2830
2918
2919
2920
VACM
Days of Operation
104
57
136
141
104*
Estimated % JMft
Rate
0.024
0.049
-0.035
-0.004
0.011
Estimated
Maximum Time
Shift (minutes)
35
40
-68
-8
17
             * Only 104 days were used since this was all that could be
             matched with Aanderaa No. 2830.
      These results show the meters to be within 108 minutes of each other during the
4 month measurement period. This is less than the advertised accuracies (6 minutes per
year for the Aanderaa meters and 12 minutes per year for the VACM), but sufficient to
perform the type of analyses done for this study.
                                        44

-------
5.4  CONDUCTIVITY AND SALINITY

      The conductivity measurements taken by meter # 2920 were reduced to salinity
for the measurement period October 25, 1977 to March 15, 1978. Salinity was calculated
by normalizing the conductivity measurements to 15*C and one atmosphere pressure,
and then applying the following empirical formula (which is given in the operating
manual for the Aanderaa current meters):

      S = -0.08996 + 28.2972R + 12.8O832R2 - 1O.67869R3

                   + 5.98624R4 - 132311R5

      where   S = Salinity in parts per thousand (ppt)

              R = The ratio of conductivity of a sample (with a temperature of 15*C
                   and pressure equal to atmosphere) to the conductivity of sample
                   (with a temperature of 15*C, pressure equal to 1 atmosphere, and
                   having a salinity of 35 ppt)
      The calculated mean value of salinity for the measurement period was found to be
34.49 ppt. This value is slightly below the historical range (34.6 to 35 ppt) for deep
bottom water ^. The calculated monthly means and ranges are listed below (units  =
ppt).
             Month            Minimum          Mean       Maximum

             Octl977           34.49              34.59         34.71
             Novl977           34.23              34.57         34.71
             Dec 1977           33.79              34.52         34.64
             Jan 1978           33.61              34.49         34.64
             Febl978           34.29              34.42         34.55
             Mar 1978           34.25              3433         34.53
             Full Period         33.61              34.49         34.71
      Note: Appendix C contains a histogram of the conductivity measurements
            from which these salinity statistics were derived.

            Appendix H contains summary statistics of the conductivity measurements.
                                       45

-------
6. DATA INTERPRETATION

      Three major pieces of information concerning the Farallon Islands LLW Disposal
Site are combined in this section to evaluate the type of currents observed at the site and
to derive an estimate for transport potential of waste material from the site.  The
information used for this evaluation includes the current measurements processed as
described in Section 5, the bathymetry describing the depths and slopes in the
measurement area, and (see ^) an evaluation of sediment grain size distributions.

6.1 ANALYSIS OF CURRENTS

      Most of the energy stored in deep ocean currents can be characterized by five
components listed below.

      Current Component            Period                   Frequency

      Large scale                     > 6 months               > 0.0002 cph
      Meteorological                 3 to 7 days               0.006 to  0.014 cph
      Tidal (major)                  12 and  24 hrs            0.083 and 0.042 cph
      Inertial                        12 hr/sin (lat)            Sin (lat)/12 hr
      Internal waves                 10 min to 8 hr           0.125 to  6.0 cph

      Large-scale currents balance pressure gradients resulting from density variations
and the Coriolis Force. Since the total measurement time for most current meters is
less than 4 months, this current component will  appear as part of the mean,  or average
of the current  measurement time series. Meteorological currents result from stress at
the sea surface caused by the average wind speed and strength (i.e., length of blow and
fetch area) over the ocean. This component will appear as a low frequency part of the
current measurement spectrum.  Tidal currents result from sea surface displacements
caused by astronomical forces. There are approximately 400 different harmonic
components of tides that can be classified into three principal components; semidiurnal
(12 hours), diurnal (24 hours), and long period (0.5 month, 1 month, 6 months).  The
semidiurnal tides appear to make up a large part (close to 50 percent) of the total
current energy at the Farallon site. Inertial currents may be excited by other currents
and propagate because of the coriolis force until they are damped by friction. The
period of this component is approximately 20 hours at the latitude of the measurement
field.  Internal waves and their associated currents are also excited by other currents and
propagate along density gradients due to stratification of the ocean. Some energy at
internal wave frequencies in the deepwater measurements was observed in the data from
meter # 2919 for the months of December and January, as can be seen from
examination of the spectra in Appendix E.
                                        46

-------
      The interpretation of the current data from the present study and from previous
studies ( '*! and ^ ) is described below.  The data are resolved into periodic
components, such as tidal currents, and long-term drift currents.  The periodic
components are assessed through an examination of the spectral frequency components
and the drift currents are assessed by taking long-term vector averages of the current
time series.  These data products are described in  Section 5.1. The relationship between
currents measured at the site and large-scale current patterns is also discussed.
6.1.1  Periodic Current Components

      The periodic components in the current measurements can be evaluated using the
spectrum generated from the measurement time series.  The monthly spectra for each
meter are given in Appendix E. It can be seen that the measurements are dominated by
12-hour periods corresponding to  semidiurnal tides.  In general, there is also significant
energy in the 20 to 24-hour region, as well as very low frequencies. Occasionally there is
also a peak in the spectrum with a 6-hour period in the north-south direction.  This
situation was most prominent during December 1977 and January 1978 for Aanderaa
current meter # 2919.  It appears to be the first harmonic of the dominant 12-hour tidal
current, and its exact cause is unknown. One possibility is that this  peak is due to an
internal wave generated by strong tidal currents, which reach their maxima during this
period as they flow into and interact with the sloped bathymetry in the area of the
trench south of the deployment site.

      To examine the major current energy components in more  detail, the data was
decimated (i.e., resampled with larger sampling intervals) using a finite impulse
response digital filter with the  following characteristics:

                   Cutoff frequency: 0.1 cph
                   Symmetrical, low pass
                   Number of weights: 121
                   Cosine taper
                   Decimation interval:  3 hours

      The spectra for the decimated time series are shown in Figures 6-1 through 6-5.
Each spectrum corresponds to the entire measurement record for the respective current
meter. The total  relative energy, or statistical variance, in the current measurement
record can be calculated from:

             N                    N
             Z   S (fi) A fi  =  1/N Z  (Xi2 - X 2)
           ,=1                    ,=1
                                        47

-------
      Where       Xi  = measurement time series

                    X   = mean of series

                    N   = total number of samples; i = 1, 2,...N

                   S(fi) = spectral value at frequency fi

                   Afi   = frequency resolution (reciprocal of signal length)

      The relative energy near various frequencies of interest can be calculated from:

                       j+k
                   E  =  Z S (fi) Afi
                       >=j-k

where 'k' is some number of frequency bins about the frequency bin 'j' of interest; the
results are given in Table 6-1. This table shows relative energy (variance) values  for low
frequencies (less than or equal to 0.01 cycles per hour), 24-hour tides with 20-hour
inertial currents,  12-hour tides, and the total mean square energy in the record. It is
seen that for the deep near-bottom current meters (# 2918 and # 2919), most of the
energy is related to 12-hour tides and is along the east-west axis. However, for the
bottom meter # 2920 in the east most section of the measurement area, the current
energy (including semidiurnal tides) is slightly greater in the north-south direction.
Aanderaa meter # 2830 and the VACM have a more equal energy distribution with a
significant low-frequency component dominated by the north-south direction. In
conclusion, the periodic tidal energy is dominant in the east-west direction for the deep-
site location, and is more multidirectional at the other measurement points.  The
difference between the total energy and the sum of the current components in Table 6-1
is general noise that is relatively flat through the remainder of the spectrum.

      To give an indication of the orientation of the rotary tidal current in the area,
tidal ellipses were calculated for  the 12-hour energy components listed in Table 6-1.  The
calculation was performed as described in Figure 6-6, using the cross-spectrum between
the north-south and east-west current measurements to compute the phase angle.  The
results are shown in Figures 6-7  through 6-11.  As would be expected, the near-bottom
ellipses have a larger ratio  of major to minor axes than those for meters in the midwater
column (Aanderra # 2830 and the VACM).

      The measured current speeds generally range between 0 and 20 cm/s, as can be
seen from the time series plots, histograms, and scattergrams given in Appendices A, C,
and D, respectively. The speeds exceed 20 cm/s less than 3 percent of the time, and this
is discussed in the next section as being relevant to sediment transport.
                                        48

-------
Table 6-1. Major Current Energy Sources for Entire Operating Period of Each Meter
Mster
2918 N-S
2918 E-W
2919 N-S
2919 E-W
2830 N-S
2830 E-W
VACM N-S
VACM E-W
2920 N-S
2920 E-W
Lew Frequency
to 0.01 cph
0.4
0.2
6.6
2.5
9.4
3.1
-
Current Components
Tide (24 hour)
Inertial $0 hour)
4.4
0.5
2.9
2.7
2.1
2.4
2.8
0.2
0.4
Tide
(12 how)
0.083 cph
0.5
14.6
0.7
26.1
43
5.2
6.0
6.9
1.4
0.8
Total
Energy
em/s*
2.2
24.2
4.6
35.0
17.6
13.0
22.1
16.8
43
3.8
                                      49

-------
                            NORTH CURRENT
 o
 3C

CVJ
 o
                                         SEMIDIURNAL TIDES
     0.000     0.028     0.056     0.083

                               CPH
0.111
0.139    0.167
                             EAST CURRENT
                                           SEMIDIURNAL  TIDES
     0.000     0.028     0.056     0.083

                               CPH
0.111
0.139
0.167
              Figure 6-1.  Power Spectrum for Meter No.  2918


                                  50

-------
                            NORTH CURRENT
      0
     0.000
                            DIURNAL TIDES
                                    SEMIDIURNAL TIDES
           0.028    0.056     0.083     0.111     0.139
                             CPH
                                                     0.167
                             EAST CURRENT
 o
 o
 X
 CXL
 O
 ^
CVI
 oo
 \
 O
160
140
120
100
 80
 60
 40 j
 20

 0.000
DIURNAL TIDES
                                        SEMIDIURNAL TIDES
               0.028
               0.056
 0.083
CPH
0.111
0.139
0.167
            Figure 6-2.   Power Spectrum for Meter No. 2919
                                  51

-------
                          NORTH CURRENT
    0.000
                                     SEMIDIURNAL TIDES
           20-HOUR PERIOD
          0.028
0.056
 0.083
CPH
0.111
0.139
0.167
320

280

240

200

160
oo
    80

    40
                           EAST CURRENT
          20-HOUR PERIOD
                                         SEMIDIURNAL TIDES
                                   \
 0.000    0.028    0.056
                                 0.083     0.111
                                CPH
                            0.139     0.167
           Figure 6-3.   Power Spectrum for Meter No. 2830

                                 52

-------
            NORTH CURRENT
800
700

2 600
X

LUNAR
SEMIDIURNAL
1 TIDES

/
g 500 I!
1 P
z 400
CM
\ 300
~ 200

100
0



DIURNAL 20-HOUR
TIDES, PERIOD
k - \ji.__ J
r



SOLAR
SEMIDIURNAL
TIDES
/
s
\

0.000 0.028 0.056 0.083 0.111 0.139 0.167
CPH
EAST CURRENT
640

560
480
o
*" 400
X
= 320
o
in
CM_ 240
 16
80
LUNAR









ll rmiDMAi 20-HOUR
| D URNAL pERIOD
M TIDESv /
L x ./
Q|" W A^WA^^s^y
^_^ SEMIDIURNAL
TIDES




SOLAR
SEMIDIURNAL
TIDES

V-..I
0.000 0.028 0.056 0.083 0.111 0.139 0.167
              CPH
Figure 6-4.  Power Spectrum for VACM



                 53

-------
                 NORTH CURRENT
160 1
140
o
- 120
X
5 100
o
xT 80
oo
s 60
o
40
20
0;





^/SEMIDIURNAL






-"V^yVV^N-^-N/A_/Ay'~~v^~ ^^^^\J
^ TIDES





\^ 	 ^__
!i iii
0.000 0.028 0.056 0.083 0.111 0.139 0.167
CPH
EAST CURRENT
320
280
240

1 200
DT
^ 160
o 120
*" **
80
40
n
u
0.








DIURNAL
TIDES^
^^ 1
1
y\ /V A XI  A HA^J, ^ A
/ \J Vw^WvUvJ V^^V^X/^
^^SEMIDIURNAL
TIDES





V
v^ 	 _
000 0.028 0.056 0.083 0.111 0.139 0.167
CPH
Figure 6-5.  Power Spectrum for Meter No. 2920



                       54

-------
     ROTARY TIDAL CURRENT:
     WHERE
           Cx =  Ax sin (wt +  0X)  (east/west)
           Cy =  Ay sin (wt +  0y)  (north/south)


THE VECTOR (T(t)  WILL PARAMETERIZE AN  ELLIPSE
BOUNDED BY THE RECTANGLE
xl fi Ax and ICyl  Ay
C

-*x

/
/^

y
Ay
/
/
-Ay
AX

   Figure 6-6.  Tidal Ellipse Calculation
                     55

-------
 8.00
 6.00-
 4.00-
 2.00-
-2.00
-4.00
-6.QO--
-8.00
  -12.
00  -10.00   -8.00   -6.00    -4.00   -2.00    0.00     2.00    4.00     6.00     8.00   10.00   12.00
                                            2918 E/W

                         Figure 6-7.  Tidal Ellipse for Meter No. 2918

-------
  8.00
  6.00
  4.00
  2.00--
CT.0.00
i-H

CM
 -2.00-
 -4.00
 -6.00-
 -8.00
   -12.00  -10.00   -8.00
-6.00   -4.00   -2.00    0.00    2.00

                       2919  E/W
4.00    6.00    8.00    10.00   12.00
                                 Figure 6-8.   Tidal  Ellipse for Meter No. 2919

-------
      8,00
      6.00-  -  -  
     4.00
      2.00
   c/o
   o 0.00
   co
OJ 00
OO CM
    -2.00
    -4.00-
     -6.00-

    -8.00

       -12.
00  -10.00   -8.00   -6.00    -4.00    -2.00    0.00    2100    4.00    6.00    8.00    10.00   12.00

                                             2830  E/W
                                 Figure 6-9.  Tidal Ellipse  for Meter No. 2830

-------
  S.OOr
  6.00
  4.00-
  2.00	
  0.00
o

-------
  8.00
  6.00--
   4.00-
   2.00-

CO

  o.oo-
CM


 -2.00-
  -4.00-
  -6.00-
  -8.00
    -12
00  -10.00   -8.00   -6.00    -4.00
-2.00    0.00    2.00
       2920 E/W
4.00    6.00     8.00    10.00   12.00
                              Figure 6-11.   Tidal Ellipse for Meter No.  2920

-------
6.1.2  Average Current Components

      Long-term average currents were calculated by taking vector averages of the
current velocities for each meter.  The results are shown in Table 6-2 and depicted
graphically by the PVDs given in Appendix F. The average currents  are much smaller
than the periodic currents (less than 2 cm/s) and move in the northeast quadrant for the
bottom measurements and the southeast quadrant for the midwater column
measurements.

                    Table 6-2. Vector-averaged Current Velocities
Meter
2918
2919
2830
VACM
2920
Speed
(cm/s)
1.40
1.70
0.67
0.90
0.17
Direction
(degrees true)
69
2
137
152
57
      In an attempt to determine characteristics of near-bottom circulation patterns
within the area, an earlier measurement program (^) deployed two meters above the
seafloor near the 1700-meter site in 1975.  The current meters, emplaced 21 August 1975
with recordings over a 27-day period, were located 16 kilometers apart and occupied a
square centered at 37*38'N and 123*18'W. After removal of the tidal components, bottom
currents were found to be clearly north with a mean direction of 004*  azimuth and mean
velocity of 133 cm/s in general agreement with the present (1977-1978) study results.

      Although the detailed current structure of the ocean is complex and is not totally
understood, there are certain large-scale patterns caused by global forces such as the
prevailing winds, tides, and rotation of the earth. These forces produce large-scale
patterns in both the surface and deepwater currents that can be seen in  the data.

      The  major surface current off California is the California Current. It moves
southward and is part of the north subtropical gyre in the Pacific Ocean, caused by the
prevailing winds in the Northern Hemisphere, the continental boundary  of North
America, and the Coriolis Force. Like all wind-driven currents, it does not penetrate
beyond a depth of approximately 100 meters.  There is a large-scale deepwater current
off California which flows northward, and tends to offset the  southward  California
Current, but  it is much slower.
                                        61

-------
      The Coriolis Force (caused by the earth's rotation) tends to give the southward
surface current an offshore (westward) component, and the northward deepwater current
an onshore (eastward) component. In the coastal region the colder bottom water tends
to rise to take the place of the surface water that has been transported offshore.

      The general trend of the survey data agrees with this global phenomena, but
shows significant variation due to local conditions such as bathymetry. This can be seen
from examining the periodic current component energy in Table 6-1 and the long-term
average currents in Table 6-2.
6.2 TRANSPORT POTENTIAL

      In this section the current data, bathymetry, and sediment grain size are
evaluated to determine an estimate of the potential for sediment transport from the
waste disposal site. It should be noted that this inquiry makes no claim to present any
quantitative estimates of sediment transport paths. However, the data do justify a
qualitative assessment of the likelihood that sediments will become suspended, and
should this occur, in which general directions transport is most likely. The method
followed includes a review of the determined grain size distribution (see * ')$ a
determination of the potential for suspension of these grain sizes relative to observed
current speeds; and, an estimation of transport speed and direction from long-term
average currents at each measurement location.

      Transport may take place in essentially two modes:  (1)  transport within the
water mass which can be inferred from measurements of two-dimensional observations
such as those presented in this report, and (2) bedload transport.  To realize transport
within the water mass, material must be suspended in that water mass. One important
assumption for transport analysis is that suspended material follows water  mass motion
as a particle of water and without divergence.

      Transport at the interface of the water and sediment is called  bedload transport.
This is a mobilization of the deposited material which moves it horizontally across the
bottom, possibly resuspending it for short intervals in a siltation process.

      This section considers only the transport of fine-grained sediments (silts and
clays) in the water mass for the following reasons:

            Grain size analysis (see ^) shows these relatively fine
            sediments are present throughout the site region and
            adjacent areas. In combination they comprise more  than
            half, and usually nearer to 65 percent of the sediment
            volume (Table 6-3).
                                        62

-------
             The fine-grain sediments offer large surface area-to-volume
             ratios that enhances the ability to act as a carrier for
             adsorbed radioactive material.

             The site contains packaged waste material. Only when the
             implosion of the container causes exposure of the surrounding
             sediments and water to radioactivity is transport of concern.

             Coarse material is restricted to transport  by major, impulsive
             events such as slumps or turbidity flows which move short
             distances with long time periods of no activity in between.

             Fine sediment, coupled with currents that exceed 20 cm/s,
             offer the major possibility for transport over long distances,
             and at high enough speeds to be seriously considered for impact
             on regions outside site perimeters.  This speed range is
             approximately the threshold above which sediment resuspension
                       r^i
             can occur (l J).
             Table 6-3.   Sediment Surface Grain Size Data
                         from Farallon Islands LLW Site Survey Cores
Station
Humber
13A
48
47
39
Sample Depth
0 to 1 cm
0 to 1 cm
0 to 1 cm
0 to 1 cm
Weight Percent
Sand
28.8
13.8
15.9
38.0
Silt
44.8
65.7
61.0
36.5
Clay
26.8
20.5
23.1
25.5
Proportional
Sediments
(% Fines)
71.6%
86.2%
84.1%
62.0%
      The data available for this study consist of the observations of currents described
in Section 6.1, and grain size distributions (see I3J) from four locations in the area of
the dump site extending from water depths of 878 meters to 1350 meters.  See [3] for
complete descriptions of sampling locations and the analytical techniques applied. It is
important to note that a wet-sieve technique was used to determine the distributions at
the sample locations.
                                        63

-------
      The arrangement of current meter arrays for the 1977-1978 study was established
to describe movement of water at positions near bottom (within 1.5 meters) in an effort
to infer where particulate matter would travel while moving near the waste material
lying on the bottom. Other meters were suspended at locations in the water column as
follows:

             A current meter about 30 meters above the bottom at the
             deepest location to estimate the depth of the boundary
             layer dynamics and the shift of speed and direction due
             to boundary layer influences.

             Current meters at 911 meters below the surface some 10
             kilometers seaward of the near-bottom meter location in
             914 meters of water. This approximates a plane permitting
             examination of the horizontal large-scale motions onshore
             or offshore from  the shallower site.
       Because of the nature of the study the conservative assumption is that fine
sediments in the form of silts and clays can become suspended in the water mass at
current velocities exceeding approximately 20 cm/s.  The percent of time that the
measured speeds exceeded various velocities was determined for each current meter.
The results for 20 cm/s are listed in Table 6-4 and two exceedence diagrams are given in
Figures 6-12 and 6-13.  Figure 6-12 shows the percent exceedences for all five meters
with all of the measurement period considered for each meter.  Figure 6-13 shows
exceedences with the measurement periods truncated to cover the same period (October
through December 1977). The suspension threshold velocity will vary with grain size and
cohesivity, but the effect of choosing a different threshold than 20 cm/s can be estimated
by referencing these figures.

       The following discussion uses a 20 cm/s threshold as its basis.  At each near-
bottom measurement location, the speeds do become great enough to cause some
suspension of the fine-grain sediments. Of the two bottom meters, # 2919 had the
greatest percentage of large current speeds, approximately ten times that of meter #
2920.  Both meters operated for more than 4.5 months.  Since the grain size distribution
is similar at both meter locations, it is concluded that a greater potential exists for fine
sediment suspension at the western (down slope) end of the site than at the eastern end.
The fine-grain sediment will drop out of suspension only at very low current speeds. The
speed histograms in Appendix C imply that there will be current velocities sufficient to
keep the sediment in suspension for periods of days to weeks once it becomes suspended.
                                        64

-------
        Table 6-4.  Percent of Total Time Measurement Speeds Exceed 20 cm/s
Aitay


B
C
C
D
D
Meter


2920
2830
VACM
2918
2919
Percent
of Time

0.23
0.01
0.08
0.77
2.74
Itotal
Measurement
(Days)
141
104
365
57
136
      For the purposes of this inquiry, transport of fine-grain sediments in suspension
can be estimated by examining the long-term vector-averaged velocities given in Table
6-2.  At the deep western part of the site, meter # 2919 measured an average speed of
1.7 cm/s (1.46 tan/day) due north.  Meter # 2918 is 100 meters above # 2919 and gives
similar average speeds, predominately in the eastward direction. For sediment in the
water column, the general direction of transport is to the northeast (up slope).  This can
be seen in greater detail from the PVDs in Appendix F.  However the average speed
measured by # 2920 at the eastern (up slope) end of the site is 0.17 cm/day to the east,
which is significantly less than that measured at the western end.  Also, # 2830 and the
VACM show a southeast movement of 0.67 and 0.90 cm/s, respectively. Since these
meters effectively define a horizontal plane with meter # 2920, the onshore transport
diminishes significantly at the east end of the site. It therefore appears that there is
little chance for any  significant transport to continue toward shore; however, there are
no explicit measurements to verify this.

      In  conclusion, current measurements  suggest that fine-grain sediment may
become suspended with the greatest likelihood of suspension being in deep waters at the
western end of the site.  Long-term average currents move eastward within the site, but
diminish at the shoaling eastern part. The potential for sediment suspension and
transport  exists, but  is small.  If necessary, it can be further defined with more detailed
analysis, as suggested in  Section 7.
                                        65

-------
  32.00
  28.00
  24.00
  20.00
16.00
212.00
   8.00
   4.00
   0.
       ,00   12.00
14.00   16,00   18.00   20.00   22.00   24.00    26.00    28.00    30.00    32.00   34,
                             SPEED (CM/S)
00
                        Figure 6-12.  Exceedence Curves for Total Measurement Periods

-------
32.00
28.00
 0.00-
   10.00    12.00    14.00   16.00   18.00   20.00   22.00   24.00   26.00   28.00   30.00   32.00   34.00
                                                 SPEED  (CM/S)
                   Figure  6-13.   Exceedence Curves for October through December 1977

-------
7.  RECOMMENDATIONS FOR FURTHER STUDY

      A major task for this study was to prepare recommendations for future sampling
requirements necessary to develop reliable predictive models for the potential transport
of soluble and particulate radioactive materials from formerly used U.S. Pacific Ocean
LLW disposal sites.  Section 6 described the mechanisms for potential transport based
on data accumulated to date. Detailed knowledge of current speeds and sediment
composition at the disposal site will yield qualitative information as to whether
sediments that may contain radioactive particles from the waste materials could be
transported from the site. Transport by ocean currents is believed to be an important
mechanism for the possible redistribution of the radioactive materials.  The measured
speeds from the 1977 survey were great enough so that at least some transport could
occur.  However, without a more detailed picture of the physical mechanisms involved,
the amount and distribution (i.e., direction)  of the radioactive sediments cannot be
determined.

       To develop predictive models which will indicate the  probable quantity and
direction of the transport of radioactive waste from a disposal site, it is necessary to first
develop a conceptual model of the important physical processes involved and their
relationships. Presently, there are  some gaps in  the knowledge of the physics of
sediment transport and dispersion  of dissolved and suspended particles. However, this
should not be a deterrent to the model formulation for various reasons.  First, the
monitoring and assessment of the potential environmental effects from waste disposal
sites is of great concern, and a good conceptual model can be of great use in the design
and planning of future measurement programs.  The results of the surveys can be used
to enhance the model, which will then result in a more effective monitoring program.
Second, it may be sufficient to use the model to derive an upper limit for the transport
of waste materials.  The upper limit can take the extreme case of additive adverse
assumptions, and if the resulting exposure is still within acceptable levels, further
research of the physical processes becomes less important.  If, on the other hand, the
results are not within acceptable limits, the model may be further elaborated to see if a
more realistic (but more complicated) picture of the physical situation will produce
more acceptable results.  Thus, in the case of the Farallon Islands study, one may
consider the curve for the threshold of sediment  transport and  suspension as a function
of sediment grain size and current speed, to be a very simple conceptual model. If the
measured current velocities were not great enough to transport sediment, there would be
a strong indication that this is not a problem at  the dump site unless the measurements
were taken during an unusually calm period. Since this was not the case during this
study, it is recommended that further development of a conceptual model be initiated for
the distribution and dispersal of radioactive materials from waste disposal sites. This
model may take various forms, but is generally a specification of the important  physical
quantities and boundary conditions with a set of mathematical equations expressing
their relationships.

                                        68

-------
      One possible approach would be to consider the current speed and direction data
as inputs of a stochastic process, which would generate a set of transport probability
distributions based on a well-defined sediment transport formulation such as the
Hjulstrom curves (m). This formulation would assign probabilities of initial and
sustained suspended transport in terms of current speed and mean grain diameter.
Implementing this approach would involve the use of standard Monte Carlo
computational procedures. The probabilistic approach could take into account
unmeasured or inherently stochastic aspects of the processes being evaluated (i.e., the
variability of suspension threshold velocities with the cohesivity of fine grained
sediments).  It would also offer some direct insight into the degree of uncertainty
involved in the predicted transport modes, and  statistical confidence units could be
ascribed.

      To be of the greatest possible utility, the conceptual model (verified and modified
by measured data) should be supported by a numerical model.  The numerical model is
a realization of the conceptual model, which will provide quantitative estimates of the
process under investigation for conditions relevant to the area under study.  The
numerical model can be as simple as a mathematical formula or statistical relationship
between the conditions in  the area and the resulting  effects, or as complex as an
evaluation of the coupled equations of motion of all of the important quantities.  It is
important to also estimate the accuracy of the model to assure that the quality of its
predictions are within acceptable limits.  It is also important not to make the model
more detailed than is justified by the knowledge of the conditions in the area and the
physical processes involved. The model can be  realized in modular form so that as more
information about the area and mechanisms are gathered, it can be easily enhanced to
give more accurate predictions.  It can then be  run under a variety of conditions and
assumptions and easily applied to other areas that contain related processes.

      A relatively simple procedure for estimating the potential for transport is to
implement suspension and bedload transport criteria for a distribution of grain sizes on
each of the raw current velocity/direction time series to produce output time series with
the following features:
             Whenever it is assessed that there is potential for transport
             (i.e., particle is suspended) the output velocity time series
             will be set equal to the input time series multiplied by the
             estimated probability of suspension.

             When transport is deemed unlikely  (i.e., particle is out of
             suspension) the output velocity series will be zeroed.

             The direction time series is kept identical with the input
             direction series.
                                         69

-------
      The resulting velocity/direction output series will be a representation of the
motion of sediment at the given current meter for each grain size.  This time series can
then be used to generate PVDs and scattergrams to visually represent the potential for
transport as a function of direction.  The sensitivity of these estimates (and hence the
amount of uncertainty) will be largely reflected in the amount of variation of the PVDs
as a function of grain size.

      The most direct and accurate determinations of the physical conditions in an area
of the ocean are determined from direct measurements.  Measurement programs are,
however, expensive, and therefore limited in both spatial extent and temporal coverage.
It is usually essential that any oceanographic investigation be supported by some
measured data, as was provided by the present Farallon Islands LLW disposal site study.
It is desirable that the program be designed with a conceptual model of the important
processes taken into account, and that the results be  used to test, verify, and calibrate
the model.

      It is believed that additional current measurements would yield a greater
understanding of the potential for transport of radioactive materials in the case of the
Farallon Islands LLW disposal site.  Measurements of deepwater currents are rare and
difficult to obtain, and the success of the survey analyzed in this report is encouraging.
However, only one of the meters was  capable of recording data for an entire year.  The
others recorded data for the late fall  and winter seasons only. In addition, it is not
known whether the data was taken during a typical year, or a year with unusually low or
high activity. Additional data would  help to answer this question and provide a greatly
increased statistical data base from which assessments of conditions in the area can be
made.  Statistical data for normal and extreme conditions are of importance. Once both
of these data bases have been collected and are large  enough to inspire confidence, the
area can be  characterized  in a statistical sense, and projections of current related
processes can be derived for the future.
      If more current measurements are to be obtained, it is recommended that a
careful calibration program on all instruments be carried out. The present study
provided an excellent opportunity to compare the responses of two commonly used
current meters:  the Aanderaa and VACM.  These gave the same general statistical
results, but differed somewhat in their detailed time series measurements under the
same conditions.
                                        70

-------
8.     REFERENCES


111    Joseph, A.  1957.  United States' Sea Disposal Operations, A Summary to
      December 1956.  Technical Information Service Report No. WASH-73. United
      States Atomic Energy Commission, Washington, DC.


121    Noshkin, V. E.  1978.  Radionuclides in the Marine Environment near the
      Farallon Islands.  Lawrence Livermore Laboratory, Univ. of California,
      Livermore.
[31    Dayal, R., Duedall, I.W., Fuhrmann, M. and Heaton, M.G.  1979.  Sediment and
      Water Column Properties at the Farallon Islands Radioactive Waste Dumpsites.
      Final report to the Office of Radiation Programs.  U. S. Environmental Protection
      Agency, Washington, DC.


141    Wilde, P.  1976.  Oceanographic Data Off Central California, 37* to 40* North
      Including the Delgada Deep Sea Fan. LBL Pub. 92. Lawrence Berkeley
      Laboratory, Univ. of California, Berkeley.
151    Sverdrup, H., Johnson, M.U. and Fleming, R.H.  1955. The Oceans - Their
      Physics, Chemistry and General Biology. 6th ed.  Prentice-Hall, Inc. Englewood
      Cliffs, NJ.
      U.S. Environmental Protection Agency.  1983.  Analysis of Ocean Current Meter
      Records Obtained from a 1975 Deployment off the Farallon Islands, California.
      Office of Radiation Programs, Washington, DC.  EPA 520/1-83-019.
      Dyer, R. 1976. Environmental Surveys of Two Deep Sea Radioactive Waste
      Disposal Sites Using Submersibles. Proceedings of an International Symposium
      on Management of Radioactive Wastes From the Nuclear Fuel Cycle, Vol.  I.
      International Atomic Energy Agency, Vienna, Austria. Pages 317-338.  IAEA-SM-
      207/65.
181    Hjulstrom, R. 1935. Studies of the Morphological Activity of Rivers as
      Illustrated by the River Fyris. Bull, of the Geol. Inst, Uppsala, Sweden. Vol. 25,
      Pages 221-527.

                                       71

-------
9.     BIBLIOGRAPHY
      Aanderaa Instruments. 1976.  Operating Manual for Recording Current Meter
      Model 5. Aanderaa Instruments, Bergen, Norway.

      AMF Sea-Link Systems. 1976. Vector Averaging Current Meter.  AMF Sea-Link
      Systems, Herndon, VA.

      Bendat, J.S. and Piersol, A.G.  1971. Random Data: Analysis and Measurement
      Procedures. Wiley-Interscience, NY.

      Bouma, A.H. 1962. Sedimentology of Some Flyash Deposits, Elsevier,
      Amsterdam, Netherlands.

      Carney, R.  S. 1979. A Report on the Invertebrate Megafauna Collected by
      Otter Trawl at the Farallon Islands Radioactive Waste Disposal Site during the
      August, September and October, 1977 Cruises of the R/V Velero.  Prepared for
      the Office of Radiation Programs.  U.S. Environmental Protection Agency,
      Washington, DC.

      Conomos, T.J., McCulloch, D.S, Peterson, D.H., and Carlson P.R.  1971. Drift of
      Surface and Near Bottom Waters of the  San Francisco Bay System, March 1970-
      April 1971.  Miscellaneous Field Studies Map, M.F. 333.  U.S. Geological Survey,
      Reston, Virginia.

      Dayal R., Oakley, S. and Duedall, I.W. 1976.  Sediment-Geothermal Studies of
      the 2800 m Nuclear Waste Disposal Site. Final report to the Office of Radiation
      Programs, U.S.  Environmental Protection Agency, Washington, DC.

      Defant, A.  1961.  Physical Oceanography Vols. I & II. MacMillan Co., NY.

      Friedman,  G.M. and Sanders,  J.E.   1978. Principles of Sedimentology. John
      Wiley & Sons, NY.

      Goldberg, E.D. et al., eds.  1977. The Sea: Ideas and Observations on Progress in
      the Study of the Seas. Vol. 6:  Mar. Modeling.  John Wiley and Sons, NY.

      Gotshall, D.W. and Dyer, R.S.  1987.  Deepwater Demersal Fishes Observed From
      The Submersible AVALON (DSRV-2) Off The Farallon Islands, 24 June 1985.
      Marine Resources Technical Report # 55. California Department of Fish and
      Game, Long Beach, California.
                                       72

-------
BIBLIOGRAPHY (continued)
Hill, M.N., ed.  1962.  The Seas:  Ideas and Observations on Progress in the
Study of the Seas. Vol. 1:  Phys. Oceanog. John Wiley and Sons, NY.

Keenan, Pat. 1980. Aanderaa RCM Compass Errors.  Exposure; a Newsletter for
Ocean Technologists.  Vol. 8, No. 2.

Kuenen, P.H.  1953.  Significant  Features of Graded Bedding. Bull, of American
Assoc. of Petrol. Geol. Vol. 37., Pages 1044-1066.

Reish, D.J.  1978.  Study of the Benthic Invertebrates Collected from the United
States Radioactive Waste Disposal Site off the Farallon Islands, California.
Prepared for the Office of Radiation Programs.  U. S. Environmental Protection
Agency, Washington, DC.

Schell, W.R. and Sugai, S. 1978. Radionuclides in Water, Sediment and
Biological Samples Collected in August-October, 1977 at the Radioactive Waste
Disposal Site Near the Farallon Islands. Final report to the Office of Radiation
Programs, U.S. Environmental Protection Agency, Washington, DC.

Shepard, F. P.  1973.  Submarine Geology. 3rd ed. Harper & Row Pub., NY.

Swift, D.J. et al., eds.  1973.  Shelf Sediment Transport: Process &  Pattern.
Academic Press, Inc., NY.

U.S. Environmental Protection Agency.  1975. A Survey of the Farallon Islands
500 Fathom Radioactive Waste Disposal Site.  Office of Radiation Programs,
Washington, DC. Tech Note ORP 75-1.

U.S. Environmental Protection Agency.  1982. Analysis  of Current Meter Records
at the Northwest Atlantic 2800 meter Radioactive Waste Dumpsite.  Office of
Radiation Programs, Washington, DC.  EPA 520/1-82-002.

U.S. Environmental Protection Agency.  1983. Survey of the Marine Benthic
Infauna Collected from the United States Radioactive Waste Disposal Sites off the
Farallon Islands, California.  Office of Radiation Programs, Washington, DC.
EPA 520/1-83-006.

U.S. Environmental Protection Agency.  1988. A Study of Deep-Ocean Currents
Near the 3800 m Low-Level Radioactive Waste Disposal Site, May 1984 - May
1986. Office of Radiation Programs, Washington, DC. EPA 520/1-88-007.
                                  73

-------
BIBLIOGRAPHY (continued)
U.S. Environmental Protection Agency. 1988. Waste Package Performance
Criteria for Deepsea Disposal of Low-Level Radioactive Waste.  Office of
Radiation Programs, Washington, DC.  EPA 520/1-88-009.

U.S. Environmental Protection Agency. 1988. A Monitoring Program for
Radionuclides in Marketplace Seafoods. Office of Radiation Programs,
Washington, DC.  EPA 520/1-88-010.

U.S. Environmental Protection Agency. 1988. Sediment Monitoring Parameters
and Rationale for Characterizing Deep-Ocean Low-Level Radioactive Waste
Disposal Sites.  Office of Radiation Programs, Washington, DC.  EPA 520/1-87-
011.

U.S. Environmental Protection Agency. 1989. A Review and Evaluation of
Principles Used in the Estimation of Radiation Doses Associated with  Deep-sea
Disposal of Low-Level Radioactive Waste.  Office of Radiation Programs,
Washington, DC.  EPA 520/1-89-019.

U.S. Environmental Protection Agency. 1990. Analysis and Evaluation of a
Radioactive Waste Package Retrieved from the Farallon Islands 900-meter
Disposal Site.  Office of Radiation Programs, Washington, DC.  EPA 520/1-90-
014.

U.S. Environmental Protection Agency. 1990. Recovery of Low-Level Radioactive
Waste Packages from Deep-Ocean Disposal Sites.  Office of Radiation  Programs,
Washington, DC.  EPA 520/1-90-027.
                                 74

-------
                               TECHNICAL REPORT DATA
                         (Please read Instructions on the reverse before completing)
1. REPORT NO.
 EPA  520/1-91-009
                          2.
                                                     3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
 Ocean  Current Measurements  at the Farallon
 Islands  Low-Level Radioactive. Waste Disposal
 Site,  1977-1-978
           5. REPORT DATE
              April  1991
           6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
 D. Crabbs,  R.  Crane, and  D.  Friedlander
                                                     8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
 Interstate Electronics  Corporation
 1001  East  Ball Road
 Anaheim, CA 92803
                                                     10. PROGRAM ELEMENT NO.
            11. CONTRACT/GRANT NO.

              Contract 68-01-0796
12. SPONSORING AGENCY NAME AND ADDRESS
  U.S. Environmental Protection Agency
  Office  of  Radiation Programs (ANR-461)
  401 M Street,  SW
  Washington,  DC 20460
            13. TYPE OF REPORT AND PERIOD COVERED
              Final
            14. SPONSORING AGENCY CODE

              ANR-461
15. SUPPLEMENTARY NOTES
16. ABSTRACT
           This report presents data from an  ocean current measurement
     study,  conducted during  1977 and 1978,  in  the area of the Farallon
     Islands low-level radioactive waste (LLW)  disposal site off  the
     coast of San Francisco,  California.  The purpose of this study was to
     measure near-bottom  and  bottom currents in the area, and utilize
     available historical  data,  to determine the potential for transport
     of  LLW from the disposal site toward populated areas in the  vicinity
     of  San Francisco.  Interpretation of the current meter data  combined
     with  other available  data taken at the  site during previous  studies
     in  1974 and 1975 is  also presented.  The appendices to the report
     contain computer generated  graphical displays of the output  data
     from  all the current  meters.
17.
                            KEY WORDS AND DOCUMENT ANALYSIS
                DESCRIPTORS
                                         b.lDENTIFIERS/OPEN ENDED TERMS
                        :.  COSATi Field/Group
  1.  ocean  disposal
  2.  low-level radioactive  waste
      disposal
  3.  radionuclide transport
  4.  ocean  bottom currents
18. DISTRIBUTION STATEMENT

  Release  Unlimited
19. SECURITY CLASS (Tins Report!
 Unclassified
84
20. SECURITY CLASS (This page;
 Unclassified
                        22. PRICE
EPA Form 2220-1 (Rv. 4-77)   PREVIOUS EDITION is OBSOLETE

-------