ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF ENFORCEMENT
REMOTE SENSING STUDY
GREEN WATER IN LAKE SUPERIOR
OCTOBER 1972
NATIONAL FIELD INVESTIGATIONS CENTER-DENVER
AND
NATIONAL WATER QUALITY LABORATORY
DULUTH. MINNESOTA
JANUARY 1973
tXEAl
-------
ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF ENFORCEMENT
REMOTE SENSING STUDY
GREEN WATER IN LAKE SUPERIOR
OCTOBER 1972
National Field Investigations Center
and
EPA National Water Quality Laboratory
Duluth, Minnesota
January 1973
-------
THIS REPORT WAS PREPARED BY:
1) Arthur W. Dybdahl
Physicist
Remote Sensing Programs
Process Control Branch
National Field Investigations Center-Denver
Environmental Protection Agency
2) Jim V. Rouse
Geologist
Process Control Branch
National Field Investigations Center-Denver
Environmental Protection Agency
-------
TABLE OF CONTENTS
Chapter Title Page
I INTRODUCTION 1
II MISSION PURPOSE 1
III BACKGROUND 2
IV CHRONOLOGICAL DATA 3
V AIRCRAFT SENSOR DATA 3
VI CONTROLS ON AERIAL RECONNAISSANCE DATA 9
VII FLIGHT PARAMETER DATA 11
VIII WEATHER INFORMATION 11
IX MECHANICS OF AERIAL RECONNAISSANCE DATA
INTERPRETATIONS 13
X RESULTS OF THE AERIAL RECONNAISSANCE DATA
ANALYSIS AND INTERPRETATIONS 16
XI SUMMARY AND CONCLUSIONS 25
REFERENCES 26
APPENDIX A 27
ii
-------
REMOTE SENSING STUDY
GREEN WATER IN LAKE SUPERIOR
October 1972
I. INTRODUCTION
An aerial reconnaissance study of the green water phenomenon
in Lake Superior was conducted on October 19, 1972. This effort
was requested by the Director of the National Water Quality
Laboratory, EPA, Duluth, Minnesota. The sections of Lake Superior
covered during this study are shown in Figure 1.
II. MISSION PURPOSE
The aerial reconnaissance study of the northern shore reaches
of Lake Superior was designed to fulfill the following objectives:
(a) Document the presence of the green-water phenomenon.
(b) Obtain the precise location and lake surface area of the
green water recorded.
(c) Compare, to the extent practicable, the color characteristics
of the green water mass to those recorded in the immediate
vicinity of the Reserve Mining Company's taconite tailing
effluent.
(d) Document the presence of any lake water up-welling along
the Minnesota shore.
-------
PAGE NOT
AVAILABLE
DIGITALLY
-------
-2-
III. BACKGROUND
EPA1 has carried out extensive investigations regarding the
cause or source of the green-water effect. A long term detailed
field sampling program and subsequent laboratory analysis were
initiated in September, 1968. The results, pertinent to this report,
of the study are as follows:
(a) The major cause of the "green-water phenomenon" along
Lake Superior's northern shore (refer to Figure 1) was
taconite tailings suspended in the water.
(b) Taconite tailings were characterized by a mixture of
cummingtonite, grunerite, and quartz.
(c) The taconite tailings, comprising the suspended solids
in the green-water, originated from a launder discharge
within the Reserve Mining Company facility located at
Silver Bay, Minnesota.
(d) Throughout the period of study, the green water was
not observed northeast of the Reserve Mining Company
effluent regardless of the prevailing wind direction.
(e) Water clarity in the green water, caused by the taconite
tailings, was 4 to 10 times less than the clarity in
the clear or background water in Lake Superior.
Patterns2 for the surface currents in Lake Superior have been
documented through detailed long term field studies initiated in 1966.
These patterns indicate that the characteristic surface currents,
along the U. S. section of the northern shore of Lake Superior, are
in a counter-clockwise direction.
-------
-3-
It was concluded, from a drift-bottle study^ performed by
Northern Michigan University, that the test bottles released in
Lake Superior along the Michigan shore would not (except under unusual
circumstances) gather or strand along the Minnesota's northern shore-
line, because of prevailing off-shore winds causing an upwelling
and off-shore drift.
IV. CHRONOLOGICAL DATA
The aerial reconnaissance mission was flown on October 19,
1972 between the hours of 0917 and 1108 CDT.
V. AIRCRAFT SENSOR DATA
Two high performance aircraft were used to carry out this
remote sensing mission. The sensors, carried on-board each of these
aircraft, were three cameras and an Infrared Line Scanner.
Two of the three cameras were KS-87B Aerial Framing Cameras
with 6-inch (152mm) focal length lens assemblies. They were mounted
in the aircraft in their respective vertical positions. The
framing cameras were uploaded with different film and optical
filter combinations as follows:
(a) Kodak 2403 with a Wratten HF3/HF5 gelatin optical filter
combination which effectively eliminates the sunlight
scattering by the lower atmosphere. The resultant photo-
graphic data were 4.5"x4.5" black and white negatives.
This sensor was used for depth penetration below the surface
of the lake water.
-------
-4-
(b) Kodak Aerochrome Infrared Film 2443 with a Wratten 16
gelatin optical filter resulting in a 4.5"x4.5" color
transparency. This filter transmits a portion of the
visible optical spectrum, i.e., deep green, yellow,
orange, red, along with the near infrared energy from
0.7 to 1.0 microns. This film presents a modified color
or false color rendition in the processed transparency
unlike the more familiar true-color films. It has an
emulsion layer that is sensitive to the near infrared in
addition to the red and green layers, whereas the true-
color ektachrome films have red, green, and blue sensi-
tive layers. (Every color in the visible optical spectrum
is formed in the true color film by various combinations
of red, green and blue dyes similar to the red, green and
blue dots on the front of a color television picture tube.)
The modified or false color rendition comes into play
when the exposed image on the infrared film is processed.
In the finished transparency, the scene objects (trees,
plants) producing infrared exposure, appear red in color,
while red and green objects produce green and blue images,
respectively. The most important asset of this film is
its capability of recording the presence of various levels
of chlorophyll in plant growth. The leaves on a healthy
tree will record as a bright red image rather than the
usual green. The orange filter is used to keep all blue
light from reaching the film which would cause an unbalance
in the normal red, green, blue color balance.
-------
-5-
The viewing angle of the framing cameras was 41° centered about
the aircraft's nadir as shown below:
AIRCRAFT
ALTITUDE
I
GROUND LEVEL
Viewing Angle of a Framing Camera Configured with a 6-inch Focal Length.
A diagram of a typical framing camera is provided below:
Focal Plane
Shutter
Lens
Film Advances Frame by Frame
Framing Camera
-------
-6-
The remaining camera of the three mentioned above, was a
high altitude panoramic camera, the KA-55. A typical panoramic
camera is shown in the diagram below:
Scanning
Stovepipe"
pivoted at rear
nodal point
Film Advances Frame by Frame
Scanning Lens Panoramic
This camera used a 32" lens assembly, provided twice the image
magnification over that provided by the framing cameras. The lens
assembly scans in a direction perpendicular to the aircraft's line
of flight through an angle of 90° centered about the nadir of the
aircraft as shown below:
AIRCRAFT
ALTITUDE
GROUND LEVEL
The viewing angle of the lens assembly in the direction
parallel to the aircraft's line of flight was 21.1°.
-------
-7-
The panoramic camera carried a Kodak Aerographic Ektachrome S0-397
film producing a true color transparency 4.5"xl8.8" in size. No
special optical filters were used with this film in order to preserve
the true color or real world rendition of the area flown.
This camera was used to photograph as much of the green water as
possible in the aircraft's lateral direction while including the
shoreline within each frame. This was required so that the precise
location and surface area of the green water could be established.
An infrared line scanner (IRLS), which records a thermal map
of an imaged area, completed the array of airborne sensors used
on this mission. The IRLS uses an infrared detector in an electro-
optic system to record on film the amount of infrared energy
detected in the imaged area. The effective focal length of the IRLS
is 1.15 inches and the field of view is 120° perpendicular to the
line of flight.
The three basic units in an infrared reconnaissance set are
scanner optics, a detector, and a recording unit. The scanner
collects the infrared emissions from the ground and reflects them
to a parabolic mirror. The parabolic mirror focuses the infrared
emissions onto the detector. The detector converts the infrared
energy collected by the scanner into an electrical signal. In
the recording unit the electrical signal is converted to visible
light through a cathode ray tube which is then recorded on ordinary
black and white film. The diagrams below depict the optical
-------
-8-
collection system and the lateral field of view of the IRLS,
respectively.
Detector
i m m
1
AIRCRAFT
ALTITUDE
i
Basic Two-Sided Coaxial Rotating
Mirror Optical System
GROUND LEVEL
Field-of-View of the IRLS
The Appendix contains information pertinent to aerial sensors
in respect to:
- Focal length
- Angle of view
- Effects of focal length and altitude on scale
and ground coverage.
-------
-9-
VI. CONTROLS ON AERIAL RECONNAISSANCE DATA
This mission was flown with two high performance reconnaissance
aircraft. The exposure, processing and subsequent interpretation
of the photographic films were under the control of the National
Field Investigations Center - Denver, EPA.
The precise flight lines, shown in Figure 1, the respective
altitudes of each aircraft and the approximate time of flight
were specified to the flight crews. The film and optical filters
were provided by NFIC-Denver. The respective exposure levels for
the film were specified to the personnel installing the film in the
aerial cameras. They were as follows:
(a) Camera Station 1 - Infrared film 2443 has an aerial
exposure index (AEI) of 10 with a Wratten 12 yellow
optical filter. Camera was set on AEI of 12 with a
Wratten 16 orange filter (1/3 stop underexposed).
(b) Camera Station 2 - Tri-X black and white film 2403 has an
AEI of 250. Camera was set on AEI of 150 with HF-3/HF-5
haze cutting optical filters (2/3 stop overexposed).
(c) Camera Station 3 - S0-397 true color film has an AEI of 12.
Camera was set on AEI of 12 with no external optical
filters.
The film was processed in processors manufactured by Eastman
Kodak Company. The infrared and true-color Ektachrome films were
-------
-10-
processed in the Ektachrome RT Processor, Model 1811, Type M,
Federal Stock Number 6740-109-2987PK, Part Number 460250. This machine
uses Kodak EA-5 chemicals. The temperature of the respective
chemicals in the processor and the film process rate, in feet per
minute, are the important parameters. Their values were specified
as follows:
1) Prehardner 115°F
2) Neutralizer 115°F
3) First Developer 115°F
4) First Stop Bath 115°F
5) Color Developer 120°F
6) Second Stop Bath 120°F
7) Bleach 125°F
8) Fixer 120°F
9) Stabilizer 120°F
The film process rate was 9 feet per minute. The nine chemical
baths, mentioned above, comprise the EA-5 process used for the
color films. The temperature and pressure of the fresh water supplied
to the processor was 120°F and 45 pounds per square inch minimum
respectively. The fresh water is used to wash the film immediately
before entering the dryers.
The black and white film 2403 was processed in a Kodak Versamat
Model 11-CM processor using Kodak 641 chemicals. This process contains
only two chemical baths which are the developer and fixer. During
processing, these were maintained at 85°F with a film process rate
of 12 feet per minute. Fresh water temperature was maintained at
85°F with a pressure greater than 45 pounds per square inch.
-------
-11-
With complete control over the patterns of flight, film type
and exposure, film processing and photographic interpretation, the
true color film S0-397 is a true and exact representation of the
actual scene recorded by the reconnaissance aircraft on October 19, 1972,
VII. FLIGHT PARAMETER DATA
The flight parameter data consists of the following entities:
(a) direction-of-flight of the aircraft (line-of-flight)
(b) air speed of aircraft
(c) aircraft altitude above ground level (AGL)
(d) time of flight.
The values of these parameters are as follows:
Flight Line 1 Flight Line 2 Flight Lines 3-7
Air Speed 360 Knots 360 Knots 360 Knots
Altitude 16,000 feet AGL 7,500 feet AGL 7,500 feet AGL
The above mentioned flight lines are depicted in Figure 1.
The time of flight for this study was 19 October 1972 at 0917
to 1108 hours CDT.
VIII. WEATHER INFORMATION*
The direction and speed of the wind at Duluth, Minnesota on
October 17 - 19, 1972 were the following: (The wind direction
is measured at a particular angle clockwise from true north.)
*This information was provided by EPA National Water Quality Laboratory,
6201 Congdon Boulevard, Duluth, Minnesota 55804.
-------
-12-
Day
Oct. 17
Oct. 18
Oct. 19
Time
1100
1200 (Noon)
1300
1400
1500
1600
1700
1800
1900
2000
2100
2200
2300
2400 (Mdnt)
0100
0200
0300
0400
0500
0600
0700
0800
0900
1000
1100
1200 (Noon)
1300
1400
1500
1600
1700
1800
1900
2000
2100
2200
2300
2400 (Mdnt)
0100
0200
0300
0400
0500
0600
0700
0800
0900
1000
1100
1200 (Noon)
TABLE VII-1
Wind Direction
290°
290
280
320
280
280
280
320
300
310
320
320
320
300
300°
310
270
280
290
310
300
310
310
330
330
340
340
310
300
280
330
320
300
270
280
290
270
240
250°
250
260
260
260
260
250
220
230
260
210
260
Wind Speed (Knots)
14
14
15
12
11
14
13
11
9
10
13
9
8
6
5
6
6
8
5
6
8
9
11
10
11
11
13
12
11
10 (gusts to 19)
14 (gusts to 21)
11 (gusts to 17)
9
6
7
8
8
6
6
6
5
7
7
8
7
4
6
7
10
13
-------
-13-
The daily amounts of rainfall in the Duluth vicinity, recorded
for the 20 days preceding the flight, are provided below:
TABLE VII-2
Rainfall (inches)
October 1972
Day Amount Day Amount
Oct. 1 trace Oct. 11 trace
2 trace 12 0.0
3 0.04 13 0.0
4 0.0 14 trace
5 0.04 15 0.0
6 0.0 16 trace
9 0.0 17 0.0
8 0.01 18 trace
9 trace 19 0.0
10 0.29
The last measurable precipitation occurred on October 10, 1972.
There was virtually no rain for eight days preceding the mission,
thus precluding the possibility of land run-off in the Silver Bay
vicinity at the time of flight.
The wind direction as observed by the direction of travel of a
smoke plume, from within the Reserve Mining Company facility, was
measured from the photographic film to be 263°. This area was
photographed at approximately 1000 hours. The wind direction at
Duluth, at this particular time, was 260° as given in the table above.
IX. MECHANICS OF AERIAL RECONNAISSANCE DATA INTERPRETATIONS
In order to document the magnitude of the surface area of Lake
Superior covered by the large green-water mass, the precise location
of the latter was plotted on a series of.USGS 15 Minute (Scale 1:62,500)
maps. Thus, the surface area affected by the green water mass was
-------
-14-
measured and calculated from these maps. The location of the green
water is shown in Figure 2.
There was no ground truth, in the form of water samples,
obtained on October 19, 1972 to correlate the constituency of the
green-water mass in Lake Superior with that of the Reserve Mining
Company's taconite effluent. Consequentially, a detailed color
analysis was conducted on the color films to show that the green-
water mass and the taconite effluent have identical color character-
istics. The mechanics of this analysis is outlined in the following
paragraphs.
The original photographic transparencies were subjected to
optical tests whereby the light transmittance* through the film
was precisely measured. Each preselected physical point on a
given transparency (in most cases five points per transparency)
was measured for optical transmittance with the use of a Macbeth
Corporation TD-203AM transmission densitometer. This system measures
film transmittance on a scale from 100% to 0.01%. It also provides
transmittance in terms of film density on a scale from 0.00 to 4.00,
where- 100% transmittance is equivalent to 0 density units, 10%
to 1.0 density units, 1% to 2.0 density units, 0.1% to 3.0 density
units, and 0.01% to 4.0 density units. The transmittance measure-
ments were made through red, green, blue and yellow (visual) optical
*Light transmittance is defined as
Amount of light transmitted through an object
Light Transmittance = Amount of light incident upon the object.
-------
PAGE NOT
AVAILABLE
DIGITALLY
-------
-15-
fliters* contained within the densitometer. These filters provide
the optical transmittance or density values only in their respective
colors of the optical spectrum. For example, the blue filter
transmits light colored from violet through the common blue.
It is opaque or does not transmit the deep greens, yellow, orange,
and red. This type and degree of color isolation is required to
analyze the respective color densities in the photographic films.
The term "color analysis" refers to the measurement of the various
film densities, through a particular color filter in the densitometer,
and the subsequent technical interpretation of this data.
The color analysis was, for the most part, performed on the
false color infrared imagery rather than the true color film for
the following reasons:
(a) The green and background waters are differentiated only
by various shades of green in the true color film.
Through the previously discussed modified color rendition,
the false color infrared film shows the green water as being
bright blue in color and the background waters as a very
dark grayish-green. The latter provides a significantly
wider color separation between the two types of water for
a precise color analysis than does the true color data.
(b) The affects of high altitude atmospheric haze is not
present in the false color infrared film where as there is a
definite influence upon rendition of the true color film.
The former film eliminates this interference factor in the
color analysis.
*These filters were manufactured and calibrated to established specifications
by Eastman Kodak Company, Rochester, New York 14650.
-------
-16-
X. RESULTS OF THE AERIAL RECONNAISSANCE DATA ANALYSIS AND INTERPRETATIONS
The surface area of Lake Superior affected by the green water is
shown in Figure 2. Two generations or densities of the green water
have been plotted. The first generation is the heaviest concentra-
tion appearing heavy green and the second is comprised of lesser
concentrations appearing lighter green in color. The magnitude of
the total surface area of green water was calculated to be approximately
66 square miles.
No trace of the green water was recorded from a point immediately
north of the Reserve Mining Company launder effluent along shore to
a point near Tofte, Minnesota.
The true-color photographic data, recorded by a high altitude
panoramic camera, is presented in this report as Figure 3 through
Figure 46. The green water is clearly shown in the prints in
a light milky-green rendition, while the background waters appear
in a darker greenish-blue color. The above mentioned figures are
located in a packet near the back of this report. They can be
trimmed and, beginning with Figure 3, put together in sequence to
form a mosaic of the area from a point near Two Harbors, Minnesota
to that near Silver Bay.
It is seen from the above mentioned prints that the green water
did not form a continuum from the Reserve Mining Company effluent
to the location of the mass near Split Rock, which is approximately
8.9 statute miles southwest of Silver Bay along shore. For this
reason, the detailed color analysis, discussed in Section IX, was
performed on the film for following areas:
-------
-17-
(a) Characteristic color of the blue on background water
northeast of Silver Bay where there was no detectable
levels of green water.
(b) Characteristic color of the green-water along the effluent
delta at Silver Bay. This is depicted by the bright,
prominent blue areas of Figures 47, 48 and 49.
(c) Characteristic color of the green-water in the large
mass from Beaver Bay to Encampment Island, a distance of
approximately 16.8 miles along shore. This area is shown
in the false color infrared prints labeled Figures 50
through 87.
(d) Characteristic color of the blue or background water in
the areas where sharp or distinctive boundaries exist
between the green and background waters. One such loca-
tion was in the vicinity of Split Rock.
A total of 83 different frames or transparencies were tested
for optical film densities with the densitometer. Each of these
values was graphically plotted to form characteristic curves for
the green water at the Reserve Mining Company effluent delta, the
background or blue water northeast of Silver Bay, the green water
in the 66 square mile mass, the background/green water near Split
Rock, and the typical outflowing river waters. The river water is
called commonly a "tea" due to the high natural organic waste
content giving rise to a grayish brown characteristic color. The
typical graphic plots for the above are presented in Figures 88 through
93 where the optical film density is along the ordinate or vertical
axis.
-------
0.0
-------
n ' *•
.._....
......
,,,-,,„,,
lfiast:fit; Swtr;fi
0.8
o.ot
-------
So: Sauace Mile Gn en Water
-------
2.
2
0.8
0.0
-------
0.8
o.ou
-------
2.
. , : ,
'
0.8
0.0
-------
-18-
It was discussed in Section VIII that the green and blue are
the dominant colors, in the modified rendition, of the two types of
water (background and green) in the false color infrared film. The
smaller the blue optical density value, the more blue there is
present in the transparency, i.e., the blue transmittance is greater.
The graphic plots depicted in Figures 89 and 91 are derived from
the imagery recorded over the Lake Superior background on blue water.
The two respective plots have the same general shapes with the blue
optical density in each being greater than the green optical density.
This is a result of the blue transmittance in the original trans-
parencies being less than that of the green. This curve shape is
characteristic of all those generated from the background water
imagery, during the color analysis. The plots depicted in Figures 90
and 92 are obtained from the imagery recorded over two different
locations in the sixty six square mile green-water mass whose posi-
tion is shown in Figure 2. These curves also have the same general
shape. The blue optical density is seen to be significantly less,
in these two figures, than that of the green. This results from the
blue transmittance in the original transparencies being greater
than that of the green. The dominant color in this case, is
the bright blue recalling that the green-water mass records as
blue in the false color infrared film. Likewise, this curve shape is
characteristic of all those generated from the imagery obtained
over the green-water mass. This shape is unique, based on spectral
characteristics, for the taconite tailings present in water.
-------
-19-
Areas in the water of high turbidity caused by land run-off, for
example, would display a different characteristic curve than that
for the green water mass. The former would be yellow-gray or
yellow brown rather than green.
An optical density characteristic curve was plotted from the
original transparencies recorded over the Reserve Mining Company
effluent delta at Silver Bay. It is depicted in Figure 88. Its
shape is nearly identical to those of Figures 90 and 92, characteristic
of the green-water mass. In Figure 92, the green optical density
minus the blue optical density is 0.22 and in Figure 88, it is 0.24.
The difference between these two numbers is quite small, approximately
8%. This further supports the similarity of the two curves. Also,
a characteristic curve was plotted for the river water, in this area,
flowing into Lake Superior called a freshwater tea as mentioned pre-
viously in this section. The curve is shown in Figure 93. Its
shape is completely different from those for the background water,
the green-water mass, and the green water at the Reserve Mining
Company effluent delta. It is concluded that this water source made
no contribution to the observed green water effect.
It is important to mention that there was only one small area,
where plant growth, containing chlorophyl, was detected on the rocks
at water level northeast of Silver Bay. This area was 7.7 miles
northeast of the Reserve Mining Company effluent delta. But, from a
point along shore, approximately 4,000 feet northeast of the delta
to Two Harbors, Minnesota, there were many areas of plant growth on
the rocks at water level.
-------
-20-
Optical studies were conducted on the properties of taconite
tailings to further explain the green water phenomenon recorded in
the aerial reconnaissance imagery. A wet sample of taconite tailings
was taken, by the EPA National Water Quality Laboratory in Duluth,
from the Reserve Mining Company east launder on October 2, 1972
at 2230 hours CDT. A partial of this sample was subjected to the
optical tests at NFIC-Denver. The wet sample was put through 250
mesh and 325 mesh U.S.A. Standard Testing Sieves. The sieve set
collected particle sizes of "45 to 63 microns" and "less than 45
microns" respectively (1 micron = one millionth meter) These
particular particle sizes were chosen for testing over the larger
sizes, because they will remain in suspension in water for much
longer periods of time. The instrument used for these tests was a
Beckman DK-2A spectrophotometer with an integrating sphere or
reflectance head attached. This instrument measures the amount of
light reflected from a sample based upon a known amount of incident
light, as a function of wavelength or respective color of the inci-
dent light. The reflective curves, for the above-mentioned taconite
particle sizes, are presented in Figure 94. This curve begins at
0.4 microns which is in the violet region of the optical spectrum,
passes through the visible region (deep blue through deep red to
the human eye) into the near infrared and finally into the inter-
mediate infrared region ending at 3.5 microns for Curve 1 and 2.5
microns for Curve 2. The green water effect, seen by average
human observers, is caused only by the light in the 0.45 to 0.68 micron
region in the visible spectrum. (The average human eye cannot see
-------
PAGE NOT
AVAILABLE
DIGITALLY
-------
-21-
the deep blue and deep red colors as easily as the green, yellow
and orange colors). In the region from 0.5 to 0.7 microns (deep
blue green to red respectively) the curves are reasonably flat and
nearly horizontal. In this area of the visible spectrum the taconite
is called neutral density substance, i.e. it reflects all colors
from blue green through red equally. In the area from 0.4 to 0.49 microns
(violet to deep blue green) the reflectance tends quickly toward zero.
This is referred to as a substance exhibiting minus-blue spectral
characteristics. This is the reason for the smaller taconite
fines appearing yellow-gray in color. If the reflectance in the
blue were equal to that in the yellow, green and red, the fines
would be gray rather than the observed yellow-gray.
With this in mind, the reflectance or scattering of light by
a spherical taconite particle is considered, as shown in Figure 95.
The incident sunlight strikes the particle and is reflected or
scattered off at various angles depicted by the green lines. This
physical interaction is governed by the Law of Reflection in
Geometrical Optics. A particle that is not spherical, such as
having projections and indentations, will produce a scattering effect
far more diffuse than the spherical one selected for simplicity of
explanation.
Now, place the taconite particle in water, as shown in Figure 96.
The incident sunlight strikes the water surface, is bent or
refracted and subsequently strikes the spherical particle as shown by
the green lines. The light is scattered from the particle and portions
of the light again emerge from the water. This is the light viewed
by a human observer or camera. The characteristic of clear, background
-------
INCIDENT SUNLIGHT
SCATTERED LIGHT
SCATTERED LIGHT
PARTICLE IS ASSUMED AS SPHERICAL FOR SIMPLICITY
OF EXPLANATION OF LIGHT SCATTERING.
Figure 95. Scattering of Sunlight By A Spherical Particle
-------
INCIDENT
SUNLIGHT
SCATTERED
LIGHT
SCATTERED
LIGHT
WATER LEVEL)
SPHERICAL
PARTICLE
Figure 96. Scattering of Sunlight by Spherical
Taconite Particle in Water
-------
-22-
water in Lake Superior is a dark greenish-blue. The bluish color is
caused by the reflection of the light incident upon the lake's sur-
face from the blue sky.) With the presence of large numbers of
taconite particles in the water, this background color is lightened
significantly by the scattering process shown in Figure 96. The
dark-green color is brightened and becomes a lighter green. The
water containing taconite does not show as yellow or some other
color because taconite reflects all colors from blue green through
red equally, thus the characteristic green color of the water is
retained, but, appears much lighter.
The origin of the green water effect is explained with the
use of the Infrared Line Scanner data or the so-called "thermal
maps."
Figure 97 is a high altitude (16,000 feet AGL) thermal map
of the shore line from fro Harbors, Minnesota to a point near
Tofte, Minnesota, from left to right for the reader. The Reserve
Mining Company effluent delta is in the center of this map. Figure
98 is the thermal map, at an altitude of 7,500 feet AGL, of the
shore line from Two Harbors to the above mentioned effluent delta.
Figure 99 is the thermal map, at 7,500 feet AGL, of the shore line
from the delta to a point near Tofte, Minnesota.
In these maps, the white areas are warm white, while the dark
areas are colder. In Figure 97, notice the dark gray area along
shore, in the Lake, through the full length of the thermal map. The
temperature of the water in this area is cooler than the water temp-
erature in the white areas near the bottom of the map. These rela-
tive temperature indications are for the surface of the water only.
-------
PAGE NOT
AVAILABLE
DIGITALLY
-------
-23-
Water is opaque or does not transmit infrared energy in the thermal
band from 8 to 14 microns. The maximum penetration beneath the
water's surface is 0.01 cm. With this IRLS imagery and known
facts about Lake Superior circulation, it is possible to explain
the observed areas of green water.
At the time of flight, Lake Superior was in a near-isothermal
condition. The entire lake mass was near the temperature of maximum
density for water. Under this isothermal condition, wind-generated
currents have a significant influence on circulation of the entire lake
depth, as graphically portrayed by Hough.5
It is generally accepted (Beeton and Chandler)6 that Lake Superior
currents are quickly responsive to wind changes. As illustrated by
the wind data [Table VII-1], there had been strong offshore winds
for two days prior to the time of flight. Shear along the wind-water
interface resulted in a surface current toward the southeast. The
bearing of this current is to the right of the wind vector caused by
the Ekman effect 6»7 resulting from the Coriolis force. The surface
current moved the surface layer of water toward the Apostle Islands.
This layer was slightly warmer than the underlying water, as a
result of solar heating. This is depicted by the red arrows in Figure 100.
The development of an offshore surface current required the
counter-development of a bottom current in opposition to the surface
current. The bottom current moved to the northwest and surfaced
as an "upwelling" along the northern shore, as indicated by the cooler
water along the shore, in the thermal imagery [Figure 97, 98, 99].
The site of the upwelling was influenced by bottom topography and the
relative calm in the lee of the northern shore. The vertical
-------
PAGE NOT
AVAILABLE
DIGITALLY
-------
-24-
circulation pattern is depicted in Figure 100.
EPA divers have previously testified to the presence of billowy
"clouds" of fine taconite tailings (brown arrows on Figure 100)
being detached from the main density current. The bottom current
(blue arrows in Figure 100) would move opposite to the density current,
and would override the density layer. This is indicated by the tri-
angular shape of the gray area of surface water in the center of
Figure 97. This results from the bottom current being forced upward
over the density layer.
The bottom current upwells along the entire northern shoreline
covered by the flights. "Green water" conditions were present only
from Beaver Bay to the vicinity of Encampment Island. This is
because the net movement of suspended tailings solids was a combina-
tion of returning bottom current and the prevalent counter-clockwise
lake circulation. A predominant longshore current from Silver Bay
toward Duluth is clearly indicated by a geomorphic study of shoreline
features on the imagery and is consistent with published current
studies.3
The thermal imagery indicates a sharp boundary between the
green and blue water masses, especially along the northeastern
boundary of the green-water mass (east of Split Rock) as seen in
Figures 99 and 101. This is further evidence of a superimposition
of a counter-clockwise circulation and an upwelling bottom current.
Stereoscopic examination of the boundary area on the film reveals
that the green-water mass underlies blue water for a short distance
northeast of this boundary.
-------
EAST
MINNESOTA
SHORE
Taconite
Tailings
Deposited
on Bottom
Zone of Downflow
of the Warmer
Surface Water
is Approximately
900 Feet
Taconite Density Layer
in Motion Toward Trench
Eddy Currents Remove Fine
Tailings from the Density
Layer and Form Suspension
in the Upwelling Water.
This is Due to the Interactions
of Opposing Currents
Figure 100. Schematic Diagram of Wind/Water Interactions Forming Green Water
-------
PAGE NOT
AVAILABLE
DIGITALLY
-------
-25-
XI. SUMMARY AND CONCLUSIONS
The presence of a large mass of green water was recorded along
the Minnesota shore of Lake Superior, between Beaver Bay and Encampment
Island, on 19 October 1972 between the hours of 0917 and 1108 hours
CDT. The mass covered approximately sixty-six square miles of lake
surface area.
The color characteristics of the green-water mass and of the
green water in the immediate area of the Reserve Mining Company
taconite effluent were essentially identical indicating that the
mass was made up of taconite tailings.
The infrared data shows that the green water effect was caused
by wind-induced upwelling along shore, bringing fine suspended
tailings to the surface. The tailings solids cause the green color
due to reflecting and scattering incident sunlight. For two days
prior to flight, the upwelling resulted from strong offshore winds,
together with near-isothermal conditions in the lake.
Numerous small areas of plant growth containing chlorophyl, were
detected on the aerial reconnaissance data from a point near the
Reserve Mining Company effluent delta to Two Harbors, Minnesota
while only one similar spot was found from a point near the delta
to Tofte, Minnesota.
The data and conclusions given in this report, apply only to
the conditions recorded on October 19, 1972.
-------
-26-
REFERENCES
1. Proceedings from the Lake Superior Enforcement Conference,
Second Session, April 29-30, 1970, Volume 1, page 223,
Effects of Taconite on Lake Superior.
2. Proceedings from Lake Superior Enforcement Conference,
Executive Session, May 13-15, 1969, September 30-October 1, 1969,
Volume 1, page 67, An Appraisal of Water Pollution in the
Lake Superior Basin.
3. Drift - Bottle Study of the Surface Currents of Lake Superior,
J. D. Hughes, J. P. Farrell and E. C. Monahan, Department of
Geography, Northern Michigan University, Marquette, Michigan.
4. Geology of the Great Lakes, Jack L. Hough, University of Illinois
Press, 1958. Figure 22 on page 51.
5. The St. Lawrence Great Lakes: Limnology in North America,
A. M. Beeton and D. C. Chandler, D. G. Frey, editor, University
of Wisconsin Press 1966, page 539.
6. General Oceanography, Gunter Dietrich, John Wiley and Sons,
Copyright 1963, page 340ff.
7. Elements of Physical Oceanography, Hugh J. McLellan, Pergamon
Press, 1965.
-------
-27-
APPENDIX
Focal Length, Angle of View, and the Effects of Focal Length and Altitude
-------
28
The focal length of the aerial sensors affects the size (or scale)
of the resulting imagery. At any given altitude, the image size
chariges in direct proportion to changes in focal length. Also for a
given focal length, the image size is inversely proportional to the
altitude.
The angle of view of a sensor is a function of the focal length
and the image format size. The importance of the angle of view is
its relationship to the amount of target area recorded in the imagery.
Refer to the following diagrams: A. Focal length of a simple lens.
B. Effect of focal length on scale and ground coverage. C. effect
of altitude on scale and ground coverage.
Point at
Infinity
Reproduction of
point at mfimty-
[— Focal Length*
I
-Parallel light rays from infinite
distance and a single point source.
Diagram A. Focal Length of a Simple Lens
Focal length is the distance from the lens (A) to the film (B)
-------
29
3-Inch Focal Length // \ \
// 20,
Ft
20,000
Ft
6-Inch Focal Length
30.000 Ft
12-Inch Focal Length
20,000
Ft
500 Ft
7,500 Ft
—/ /— 5,000 Ft
18-Inch Focal Length
DIAGRAM B Effect of Focal Length on Scale and Ground Coverage
30,000 Ft
.22,500 Ft
10,000 Ft
5,000 Ft
,500 Ft
/— 7,500 Ft
3-Inch Focal Length
DIAGRAM C Effect of Altitude on Scale and Ground Coverage
------- |