Special Report Number LM9
                   April 1963

              Public Health Service
  Division of Water Supply and Pollution Control
    Great Lakes-Illinois River Basins Project


                          TABLE OF CONTENTS


INTRODUCTION                                                  1

     Description of Study                                     1
     Purpose                                                  1
     Previous Studies                                         2

STUDY PROCEDURES                                              k

RESULTS OF STUDY                                              6

     Reliability                                              6
     Flow Characteristics                                     8
     Significance of Findings                                10
     Relation to Previous Work                               11

SUMMARY                                                      12

REFERENCES                                                   14




 I    Readings in Each Quadrant


 1    Station Location
 2    Woods Hole Meter
 3    Telemetry Meter
 k-    Integrated Current Meter
 5    Portion of Typical Current Meter Speed Record
 6    Great Lakes Current Study Typical Current Station
 7    95$ Confidence Limits for Speed and Direction
 8    Savonius Rotor Calibration
 9    Comparison of Speeds at 60' and 90'
10    Comparison of Directions - 60 feet and 90 feet
11    Comparison of Directions - 90 feet and 120 feet
12    Histogram of Speed at Three Levels
13    Polar Diagram of Magnitude and Direction at 60 feet
Ik-    Polar Diagram of Magnitude and Direction at 90 feet
15    Polar Diagram of Magnitude and Direction at 120 feet


       Many of the theoretical concepts of water movements in the
oceans or the Great Lakes have defied testing because of the lack
of adequate instrumentation.  With the advent of the Savonius rotor
and the development of the Woods Hole meter less than three years
ago,, a new tool was made available to the scientist.  An important
feature of this meter, the automatic readout of recorded data, was
perfected in July 1962.  This automatic recording current meter
with the ability to record up to six months unattended, is the
basis of this study.

Description of Study

       The study of cm-rents in Lake Michigan by using current meters
began October 21, 1961 with the setting of an experimental telemetry
station.  The station was first set about 25 miles northeast of
Chicago, see Figure 1, but was removed after three days because of
a defect.  It was reset on May 15, 19&2 about fifteen miles northeast
of Chicago (Figure l).  From May 15 through November 1, 19&2, several
different types of stations were set for the purpose of evaluating
meter performance, mooring systems, and other features of the
instrumentation, preliminary to the full scale study.


       This report will describe the results of the observations
with respect to the following factors:

       1.  The reliability and sensitivity of instruments under
           field conditions.

       2,  The operating characteristics of different types of
           current mster installations.

       3.  The relationship of water velocity with time and with
           depth, for use in programming full-scale operations
           with all stations in the lake.

       U.  To secure early information concerning the existence, or
           absence of prevailing currents in the vicinity of Chicago.

       5.  The relationships between winds-on-shore and lake currents.


Previous Studies

       Other important factors pertinent to this study were also
explored including:  a literature search on past studiesj where
(geographically) to locate meter stations; the amount of time
required; the amount and kind of data to be collected; and a review
of the logistics of conducting the study.

       A detailed literature search disclosed three important studies
pertinent to the measurement of currents in Lake Michigan.  In the
years from 1892 through 189^, Mark Harrington of the U.S. Weather
Bureau released over 1500 drift bottles in the lake, of which only
about 203 were recovered during the three-year period (l).  These
returns formed the basis of the first map of Lake Michigan surface
currents.  In 195^-55, James Johnson of the U.S. Bureau of Fisheries
completed an intensive study of surface currents using drift bottles
and drift envelopes (2).  Out of 6260 releases, some 2870 were
recovered during this study.  The results indicated that during
no stable pattern of surface currents existed.  In general, Johnson
found the drift was from west to east and the north-south movements
were about equal.  A similar pattern existed in 1955 except that the
drift in the eastern areas was mainly to the north.  In 1958, Ayers
and others published the results of a series of synoptic surveys (3),
in which they applied a modification of the dynamic height method
and published the first map of Lake Michigan showing subsurface
water movement.  The method used, however, is open to question,
because of the choice of boundary conditions.

       Townsend (4) took sharp issue with Harrington on the method
of presentation of data.  He stated that in such a confined area
as Lake Michigan Harrington could not assume curved line motion but
should have used straight line motion.  Townsend's ideas were partially
verified by studies on Lake Erie (5).

       It is difficult to translate drift card or bottle movements
over many days into those of actual currents.  The only positive
statement that can be made about drift methods is in reference to the
points of release and retrieval.  Such movements at best can only
demonstrate a type of pattern between the times of release and
recovery.  The seemingly random changes between Ayers' and Johnson's
work graphically illustrate this point.  The more recent studies
using bottles and cards did not claim to show anything other than
the conditions that existed at that time and place.  Harrington,
however, tried to pin down specific current patterns from far less
data over a much greater time span.   However, the meters which exist
today were unknown five years ago and earlier scientists used the
tools then available.


       The dynamic height method of obtaining subsurface current
data has not been substantiated by means of current meters.  Criticism
of the method is based on two assumptions.  One, the method assumes
a zero velocity depth level which may or may not be correct; two,
the theory assumes constancy in the isothermal lines over a short
period of time, whereas it is known that internal waves may produce
very striking thermal changes in lakes and also in the ocean (6).

       Since none of the past studies shows net circulation, maximum
current speeds, effects of storms, or detailed inshore circulation,
further work is necessary to give the answers needed today.


                          STUDY PROCEDURES
       Of the various methods for the investigation of currents, a
number of fixed stations with automatic recording meters was selected
as most applicable to the Public Health Service comprehensive study
of lake water quality.

       Three types of measuring systems were considered seriously
by the U.S. Public Health Service:  The Woods Hole Oceanographic
Institution meter (7) developed by Dr. William S. Richardson (Figure
2); a telemetry system devised by the U.S. Corps of Engineers (8)
for the Public Health Service (Figure 3); and a modular system  (9)
devised by Marine Advisers, Inc. (Figure 4).  The first is self-
recording and has a mechanical recording system.  The other two are
also self-recording but have electronic recording systems.  All have
many desirable features not found in other types of current meters,
such as:

       1.  The ability to measure low current speeds, down to about
           0.03 feet per second (fps) by utilizing a Savonius rotor (8),

       2.  The ability to operate untended for up to four months
           while recording observations hourly.

       3.  The ability to measure direction to + 7.

       4.  The ability to record observations at shorter intervals
           than one hour.

       The meters have individual characteristics which make them
desirable for use in different locations.  After the trial period,
the Woods Hole meter was selected as the most suitable of the three
systems tested.

       After the method of study was selected and testing of meters
was underway, plans were devised for station locations, number of
stations required and data processing.  The selection of stations
and the number of meters per station were subjects of a conference
held in Chicago with scientists from the Great Lakes States and
Canada.  It was decided that the standard procedure of equal area
and depth coverage was the most logical method.  In order to study
all of Lake Michigan it was agreed that between 40 and 60 stations
should be set.  The number of meters to be set posed a problem
because of the variability of the thermocline.  It was readily
established that the meters should be set at fixed depths, and at


closer spacing in the upper layers.  The depths selected were:
30, 50, 75, 100 feet and each succeeding 100-foot level.  No meter
was set less than 30 feet below the surface because of problems
resulting from surface waves.

       The preliminary recording of data was different for the
various types of stations.  The telemetering station recorded
hourly on its internal recording system and transmitted other
data every four hours by radio to a receiving station.  The Woods
Hole meter at different times was used on a continuous recording
mode and then on a mode recording 50 seconds continuously once
every twenty minutes.  The telemetry system reports total revolutions
of the Savonius rotor in each four-hour interval together with one
instantaneous direction reading made at the time of reporting.  The
Woods Hole meter records continuous speed and direction values
throughout each fifty second period on a photographic film within
the instrument.  The recording for the full-scale study is set at
a continuous fifty-second recording period once each thirty minutes
for the winter months, but intervals will be reduced to twenty
minutes for the rest of the year.  The longer period was adopted
for winter operation to provide a margin of safety against exhaustion
of the power supply batteries inside each unit.  During severe
winter weather it will be impractical to visit the stations for
retrieving data, replacing batteries, and other servicing.


                           RESULTS OF STUDY
       The results are shown on a series of graphs  (Figures 7 to 15).
These graphs present results of statistical tests on the data obtained
from the current meters.  Speed readings are given  in feet per second.
Direction readings are given in terms of degrees clockwise from magnetic
north.  The readings for direction given in this report represent the
direction from which the current is coming.

       Figures 10 and 11 were examined for correlation and equality
in direction.  If the absolute value of the median  difference of the
readings between two meters is significantly less than 90 a positive
correlation exists.


       The reliability and sensitivity of the meters was determined by
an examination of results, i.e., speed and direction data, rather than
by an analysis of the physical characteristics of the meters.  A study
of the data alone will not disclose systematic errors in the measure-
ments; however, such errors are believed to be small, based on tank
tests in which the sensors are towed through still  water (8).

       The data used for reliability evaluation were from two separate
stations - the telemetry station with three meters, set at 60, 90, and
120 feet, and a station with one meter of the Woods Hole type set at
120 feet.  The two stations were within 300 feet of each other.  The
recording interval for the telemetry station was once every four hours
with some observations at ten minute intervals.  The total time of
recording was 63 days.  The Woods Hole meter recorded continuously
and the data were printed out at 1.25-minute intervals.  The record
length for this meter was 6 days, July 5 to 10, 1962.

       Figure 5 shows a series of current meter readings taken
during a period of increasing flow.  The data show  two types of
changes in the speed; a long-period systematic rise and a short-
term random fluctuation.  The long-period rise shows an increasing
current speed.  The short-term random fluctuation could be due to
changes in the current speed or to instrumental error.  At this
time it is not possible to distinguish between the two changes
and they are grouped together and treated as physical measurement
limitations.   The amount of the random fluctuations is used to
estimate the precision (repeatability or consistency) of the


       The problem of variation measurement is to measure the short-
term changes without including the long-term systematic variation.
The simplest solution to this problem is to assume that the long-term
systematic variation is approximately zero over a short period of
time (Figure 5).  Thus, over a short period of time, the total
variation can be used to approximate the random fluctuation.

       Under steady-state conditions, the larger the quantity of
data collected, the more accurately it will represent the population
of which it is a sample.  However, the larger the quantity of data
used, the greater the time span.  Thus, the problem is to use a
large quantity of data without spanning a long period of time.  The
solution is to have as short a period of time between observations
as possible.

       The data in Figure 5 from the Woods Hole meter were plotted
from the readings made at five minute intervals.  The figure is a
plot of magnitude (or speed) against time.  The plot represents
a portion of a 40-hour period and indicates that the meter is
capable of detecting long-term systematic changes in speed.  However,
initial tests showed a large random variation in direction at higher
speeds, due to the type of mooring system used.

       The original mooring system used by Dr. Richardson of Woods
Hole in the Atlantic Ocean was modified for use in Lake Michigan.
The Atlantic installation used a synthetic line anchored to the
bottom and held at the surface by a large float.  Ice and severe
winter conditions prevented the use of a slack wire mooring system
in Lake Michigan.  The Project used a taut wire system with a
subsurface float which has a positive buoyancy of 550 Ibs.  A
surface float, for use in ice-free periods, was attached by a slack
line to the subsurface float.  An examination of the original test
data showed unusual variations in the direction vane.  Observation
disclosed that during windy periods the large surface float was
capable of jerking the lower taut line in any direction.  To
prevent this action and the consequent swinging of direction vanes
on the metersj the mooring system was changed to a two-line system
(Figure 6).   This present system leaves the meter line almost
completely free from external stress.

       The mooring system used at the telemetry station did not
have a large surface float relative to the subsurface float.
Therefore, the pull of the surface float did not apparently affect
the direction vane on the sensors.  Thus, the data from the telemetry
station were used in the analysis of direction reliability.  The


ten-minute time intervals were used for the analysis.  Figure 7
shows confidence limits using ten-minute intervals  (10, p. 29*0.
The results in Figure 7 show acceptable precision in both magnitude
and direction at each of the levels (60', 90'} and  120').

       The Savonius rotor has been subjected  to tests in an effort
to determine its degree of reliability, response in turbulent motion
and dependability over periods of time.  The  rotor, as presently
used, is sensitive to about 0.03 fps as shown by tow tank tests
conducted by the Corps of Engineers (8).  Tests by  Gaul indicate
that the rotor is very reliable between 0.08  fps and 70 fps (ll).

       The rotor is omni-directional, which means that it will
sense motion from any direction in a horizontal plane when set with
its axis of rotation vertical.  The orbital motion  of waves produces
a surge forward and then backward as it passes.  This effect on a
Savonius rotor would show as an increased speed.  Wave action,
during severe weather conditions, can be detected as deep as 50
feet.  For this reason, the uppermost current meter is at the 30
foot level and will only be affected during the more adverse
conditions.  Figure 8 shows a typical rating  curve  for the rotor
made by the Corps of Engineers.

       The direction data could not be used for the Woods Hole
meter because of the problems inherent in the old system of mooring;
however, the data from the telemetry station  showed precision in
both magnitude and direction.

Flow Characteristics

       A preliminary examination was made of  the flow data.  The
following were investigated:  relationship between  wind and water,
relationship between readings at different stations, relationship
between readings at different levels, range of speeds, and possible
predominant directions at different speeds.

       It should be remembered that much of the report depends on
statistical testing for relationships or correlations between two
variables.  A positive result (correlation) implies ninety five per cent
confidence that a relationship or correlation does  exist.  A
negative result (no correlation) implies one  of the following two
possibilities:   either no relationship exists between the variables
in question or a relationship does exist but  the nature of the
statistical test or the data was such that the relationship could
not be detected.


       Two limitations  on the applicability of the findings in
this report should be mentioned.   It  should be remembered that the
current meter data represents samples from a changing physical
system.  Thus, physical relationships which may occur for one
period of time may not  necessarily occur for any other period of
time.  For example, there is a definite possibility of systematic,
seasonal differences that are not  reflected in the tests reported
herein.  Also, the relations which are here established represent
average effects.  The only thing necessary to produce a non-random
effect during the whole period of  time is a large enough non-random
effect some of the time to affect  the overall result.  The conclusion
that can be obtained from a positive  result is that one can expect
(with 95$ confidence) that the result held more than fifty per cent
of the time, for the period of study.  (10, p. 28o).

       An attempt was made to correlate wind data and the current
data.  It should be noted that wind data was estimated (from shore
based stations) for the 30 foot level (above the lake surface) and
the current data was taken at 60 feet below the surface.  Various
relationships were tried, using a  variety of ways of weighing past
wind daily averages and correlating with present currents.  Wo
positive correlation could be found.  This does not necessarily
mean that wind does not affect water  currents.  There is every
reason to assume that such a relationship exists, but detection
may be difficult at the greater depths.

       A test was made  to correlate currents at different depths
(60*, 90', and 120') at the same station.  Both magnitude and
direction showed positive correlation at 95$ confidence (10).
Figure 9 shows the magnitude (speed) relationship at sixty and
ninety feet.  The positive correlation can easily be seen (lO, p. 29?) 
It can also be seen that the speeds at 60 feet are greater than
speeds at 90 feet on the average (a difference in medians at 95$
confidence is shown).   Speeds at the 120 foot depth were generally
lower than the minimum  detectable  speeds on the current meter.
Figures 10 and 11 show  a comparison of directions at different
depths.  The current at 90' was normally to the right of the
current at 60'.  The current at 120' appeared to the left of the
current at 90' on the average.   This fact was shown at 95$
confidence.   The theory of the Ekman Spiral, states in part, that
currents tend to rotate cum sole,  that is, clockwise in the
northern hemisphere and decrease in speed with increasing depth (l5)
The clockwise rotation  is due to rotation of the earth.  It is an
open question if the Ekman spiral  theory applies to Lake Michigan
because of its relatively shallow  depth.


       Figure 12 shows a histogram which is designed to  show the
range of speeds at different depths.  The abscissa is a  group of
speed ranges and the ordinate is the per cent of time that the
readings fall into the range specified in the abscissa.  It can
be readily seen that there are differences in speeds at  different
depths.  The greater speeds are the bars which are farther to the

       Figures 13, 14, and 15 show polar coordinate diagrams of
speed and direction at the three depths observed.  Table 1 shows
the number and per cent of observations in each quadrant, the
average angle and speed for each quadrant and the maximum number
of times the current was found in a specific quadrant.   The number
of times the readings were in the northwest quadrant was unusually
high, indicating a 95$ probability non-random effect (12, p. 35)
One meter showed northwest currents over four days in succession
(Table l).

       Valuable information is presently obtainable from the few
meters used.  Two current meter stations placed in the same area
showed correlation in direction, on the average.  Some information
is now available concerning variation of water speed with depth.
As expected, relatively high speeds at one depth will on the
average lead to relatively high speeds at another depth.  Also,
the speed of the currents showed a decrease with depth.  The
direction of currents correlated positively at different depths.
Differences in direction at different depths were shown  to be
detectable.  On the average the current at 90' was shown to be to
the right of currents at 60'.  The currents at 120' appeared to the
left of those at 90'.  As mentioned earlier, the current speeds at
120' were frequently below the minimum threshold capability of the
rotor.  It is probable that the vane response to movement is also
close to its minimum range of detection.  More observations at
higher speeds are needed to determine if the currents normally
shift in a clockwise movement with increasing depth.  The speed
ranges are shown on the histogram plotted.  The polar coordinate
diagrams and Table 1 indicate a predominance of currents from
the northwest during the period sampled.

Significance of Findings

       The data show that an effluent discharged into the lake, at
any depth, can have a prevailing direction of movement.  During the
period of May to July 19^1, this movement at the test station would
normally have been from the northwest at the three levels observed.


The maximum consecutive readings from the northwest varied from
three to more than four days with an average drift of 3 or more
miles per day for the 60 and 90 foot levels.  The rate of flow
is expected to be higher in the upper layers.

       Although mixing occurs during severe weather conditions the
exact rate or amount is unknown.  Earlier studies when the Chicago
River was diverted into the lake in September I960, shows that the
discharged waters maintained a detectable coliform count for five
days (13).  Therefore, it may be possible that even during periods
of high velocities for eight or more hours the effluent will not
be mixed into the surrounding water.

       The worst condition appears to be when an effluent would
remain in the general vicinity of its discharge point because of
extremely low velocities.  After a period of build-up, the
concentrated effluent might then be moved, en masse, by the
current.  Relatively slow currents could move the effluent for
four or more days in one direction, as shown in Table 1, or a
fast moving current could produce the same result within five
to eight hours.

Relation to Previous Work

       Little or no relationships can be made to previous work.
The present study, although made from test data, far exceeds the
current data collected from all other studies on Lake Michigan,

       Ayers, et al. demonstrated the variations in the subsurface
layers for specific time periods (3).  Surface current studies
using drift bottles and cards, which average over a period of time,
may be better explained by the test data.

       Frequently, the drift bottle or card flow will show quite
variable conditions when obtained after being adrift for one or two
days.  Long term drift periods, weeks or more, many times show a type
of "uniform" movement.  The polar diagrams showed that the currents
at a given station can be in any direction; however, there may be a
prevailing dominance from a particular quadrant over a long period
of time.

       Present work indicates a far greater variability of flow than
has been previously demonstrated.  Johnson's work on Lake Michigan
gave some indication of the variability from his transects (2).  The
current speed ranges from near zero to more than 1.3 fps.  The
ranges observed are certainly not the maximum values that can be


       The data and performance tests from the first meters set in
Lake Michigan have been examined and evaluated.  The tests were made
to evaluate the type of meters to be used and reliability of the
instruments' internal components.  The test data were examined for
persistency of movement and the variations of speed with depth.

       A literature search disclosed three intensive studies.  These
investigations, using drift bottles or drift cards and dynamic height
methods, were limited in scope and disclosed the random variations
found in the surface layers.  Only Harrington in the 1895 studies
attempted to show that a prevailing pattern existed.

       Three types of meter systems were evaluated and the Woods Hole
meter was selected because of the established performance record of
its internal components, type of data collected and lower cost.

       The test data were examined to determine the reliability and
sensitivity of meters for measuring both speed and direction.  The
data from the instruments using one type of mooring system indicated
that the large surface float had an effect on the vane direction.
The mooring system was altered to eliminate this possible source of

       The study did not show a positive correlation between the
estimated wind and the currents at the 60 foot level.  However,
detection may be difficult using shore based wind data and currents
at the 60 foot depth.

       The median values for speed and direction were examined for
random fluctuations and the 95$ confidence limits indicate the
instruments had a sufficient degree of precision.  Currents at the
90 foot level were somewhat slower and to the right of those at
the 60 foot level.  Although the vane directions at 120' were not
to the right of those at 90', as mentioned earlier, the general
direction of flow was nearly the same for both depths.  Normally
the speeds were greater at 60' as compared to those at 90' or 120'.
However,, on occasions the speeds were highest at the 90' depth.
These variations are possibly due to changes of the thermocline
which could produce the observed departures from this rule.

       Polar diagrams for a two month period show a predominant flow
from the northwest quadrant for all depths.  The maximum consecutive


readings in each quadrant show that movement from the northwest
can persist for more than four days, whereas the maximum consecutive
readings from the other three quadrants did not exceed one and
one-half days.

       The data tabulated for May to July 1962, show that under
certain conditions an effluent could move, at low speed with
relatively little mixing, for over four days.  Movement during
other periods of the year may show that there are other quadrants
in which the current will predominate for longer periods of time.


 1.  Harrington, M. W.   Surface  Currents  of The Great Lakes,  as Deduced
     from the Movements  of Bottle Papers  During The Seasons of 1892, 1893,
     and 1894, U.  S. Department  Agric., Weather Bureau,  Bulletin B (rev.
     ed.) (1895).  p. 1-14.

 2.  Johnson, J. H.  Surface  Currents  in  Lake Michigan,  1954 and 1955
     United States Fish  and Wildlife Service Special Scientific Report,
     Fisheries No. 338  (I960).p.  1-120.

 3.  Ayers, J. C., Chandler,  D.  C., Lauff,  G.  H.,  Powers,  C.  F., and
     Henson, E. B,  Currents  and Water Masses of Lake Michigan.  Great
     Lakes Research Institute, Publication  Wo.  3 (1958).

 4.  Townsend, Col. C.   Effect Upon the Climate of the Lake States by a
     Change in the Natural Current  of  Lake  Michigan.House Doc. 762,
     63rd Congress, 2nd  Session, Appendix C (1913-14).   p.  40-71.

 5.  Verber, James L.  Surface Water Movements  in  Western  Lake Erie.
     Ohio Jour. Science,  53:  42-46  (1953).

 6.  Verber, James L.  Short  Time Variation of  Temperature in Lake
     Mendota, Wisconsin.  Ohio Jour. Science,  53:  72-76  (1953).

 7.  Richardson, W. S.   Instruction Manual  for  Recording Current Meter.
     Woods Hole Oceanographic Institution.   Ref. Wo. 62-6  (1962).   p. 10.

 8.  Measurement and Telemetry of Current Velocity and Direction.
     U. S. Eng, Waterways Exp. Sta. Proj. Bull.  62-1 (1962).   p. 8.

 9-  Frantz, D. and Putz, R.  Unsolicited Proposal for an  Untended
     Current^ Temperature and Wind  Data Acquisition System.   Marine
     Advisers and ORE, Inc. (1962).

10.  Dixon,  W. and Massey, F.  Introduction to  Statistical Analysis.
     McGraw Hill, W, I.  (1959).  P- 228.

11.  Gaul, R., Snodgrass, J.  and Cretzler,  D.   Some Dynamical Properties
     of The Savonius Rotor Current  Meter.   Marine  Sciences Instrumentation,
     Vol. 2, ISA, Plenum Press, N.  Y.  (19637!

12.  Tate, M. W.  and Clelland, R. C.   Won Parametric and Shortcut
     Statistics.   Interstate  Printers  and Publishers,  Inc., Danville,
     111. (1957).  P.  171.


13-  U. S. Public Health Service, Movements  in  Lake Michigan of Water
     Discharged Ik- September 196l from The Chicago Sanitary Canal System,
     U. S. Exhibit Wo. h, Chicago Diversion  Case, Mimeo.,  Unpublished

Ik.  Sverdrup, H, U. et al.  The Oceans.  Prentice-Hall, New York (1946).
     p. 10^9.


                                   Table I

                          Readings in Each Quadrant*

                              Meter 1 - 60' Deep

Quadrant   Maximum Consecutive     Observation              Average**
           Readings in Quadrant  Number    Per Cent    Magnitude      Direction
                                           of Time       fps            deg.

  0-90           4                62       2k.3       .185            42
 90-180          3                33       12.9       -191           139
180-2700          3                58       22.8       .216           228
2TO-3600         20               102       40.0       .222           310

                              Meter 2 - 90' Deep

  0-90           4                77       25.6       .173            43
 90-1800          9                59       19-7       .144           131
180-270          5                60       20.0       .182           228
270-360         26               104       34.7       .251           316

                              Meter 3 - 120' Deep

  0-90           3                66       22.0       .018            38
 90-180          6                63       21.0       .004           138
180-270          3                73       24.3       .005           223
270-360         13                98       32.7       .011           314

*Readings are approximately four hours apart.
**Arithmetic average of magnitude and direction.



                                               Subsurface Current Station
                                                     October, 1961
Current Station
 May, 1962
                                        GREAT  LAKES  8  ILLINOIS
                                         RIVER  BASINS  PROJECT
                                     CURRENT METER  STATIONS
                                    DEPT. OF HEALTH, EDUCATION, Q WELFARE
                                           PUBLIC HEALTH SERVICE

                                    REGION  V          CHICAGO,  ILLINOIS
                                                            FIGURE I

      Surface Buoy  Containing
      Anemometer and Light
Background  - Rotor of Current Meter
Foreground  - Vane of Current Meter
                      GREAT LAKES  8  ILLINOIS
                       RIVER  BASINS PROJECT
                      WOODS  HOLE   METER

                         PUBLIC HEALTH SERVICE

                  REGION V           CHICAGO, ILLINOIS
                                          FIGURE  2

 Instrument Buoy - Approximate Height 9 Feet
        -  * _  ''"',  "   "".***",'* /-v "-* * *
Savonlus  Meter With Direction Vane and Digitizer
                            GREAT LAKES 8   ILLINOIS
                             RIVER  BASINS PROJECT
                             TELEMETRY   METER
                        DEPT. OF HEALTH, EDUCATION, 8 WELFARE
                               PUBLIC HEALTH SERVICE
                        REGION V           CHICAGO, ILLINOIS

                                                FIGURE  3

     Current and Temperature Sensor
                                     Surface Buoy Containing
                                     Anemometer and Navigation
                                        GREAT  LAKES  8  ILLINOIS
                                         RIVER  BASINS  PROJECT
                                     INTEGRATED  CURRENT METER
                                     DEPT. OF HEALTH, EDUCATION, 8 WELFARE

                                           PUBLIC HEALTH SERVICE

                                     REGION V          CHICAGO,  ILLINOIS
GPO 8278Q4
                                                            FIGURE 4




>  -a

a)  O


O t-
z o
 ' ~5
-J 0
~ QC
to 2
S <
. CD

a: <
a: P
^3 <
_J (M
o Q cn
^1 85 rf
U_ ^i
o cc ^
1 2

_j j
UJ _,
rft C_) "
- > 0
Z (T <
1 5 5
Q <
^ s

t- o
< -
1 a >
o z
._ 2
Q. a 
j UJ Uj

Q Lt





O ~r5
_O 0
^i" ^-
I  2
^  5








                                                            FIGURE 5


                                                                                   UJ   _
                                                                                                                  r  o
                                                                                                                   - o
                                                                                                                   - o
                                                                                                      FIGURE  8



     SpMtf vro94 ovtr four tourt. toston-
     torvous direction moMir4 at nd of
     ch Hriod.
     PSo4: May 26,1962-July 26,1962

     X Average velocity for quadrant (Table I)

         V           CMfcAGO,

    SpMd vro9d ovtr four hours. lnton-
    tonous direction mtoturcd at nd of
    och period.
    Prod: May 26,1962-July 26,1962.
    The  three points located oulside the graph
    limits are- 3IO,.82 f.p.s.; 288,.76f.p.s.;
    and  252,.62f.p.s.
    X Average velocity for quadrant (Table I)


    KEOfOW  V
                                                                   FIGURE  14

    Speed averaged over four hours. Instan-
    taneous direction measured at end of
    each period.
    Period: May 26,1962-July 26, 1962

    X Average velocity tor quadrant (Table I)
                                       OEPT. OF  HEALTH, EDUCATION, 8 WELFARE
                                              PUBLIC  HEALTH SERVICE

                                       REGION V           CHICAGO,  ILLINOIS
GPO 827804
                                                                FIGURE 15