TSAMD84141
TS-AMD-8414I/8453]
December 1985
TRIBUTARY INPUT OF SEDIMENTS INTO THE GREAT LAKES
Lake Michigan. Lake Huron, and Lake Erie
by
M. V. Olsen
Environmental Programs
Lockheed Engineering and Management Services Company. Inc
Las Vegas, Nevada 89114
Contract No. 68-03-3245
Project Officer
T. W. Foresman
Advanced Monitoring Systems Division
Environmental Monitoring Systems Laboratory
Las Vegas. Nevada 89114
ENVIRONMENTAL MONITORING SYSTEMS LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
LAS VEGAS. NEVADA 89114
-------
NOTICE
This document has not been peer and administratively reviewed within EPA and
for internal Agency use and distribution only.
11
-------
ABSTRACT
Water quality monitoring of Lake Michigan. Lake Huron, and Lake Erie was
performed using visual interpretation of Landsat multispectral scanner (MSS) band
data. Transparencies were acquired on the basis of proximity to 1984 Great Lakes
ship survey dates. A total of thirty-five scenes were found which corresponded to
spring, summer, and winter survey dates.
The images were visually interpreted to detect sediment input from tributarie
into the lakes. This was performed to help monitor changes in water quality and t
provide ancillary information to be used in the design of lake sample locations.
The results of this analysis are shown on overlays which attach to each print.
Several tributaries were identified as sediment sources and were annotated by name
whenever possible.
The U.S. Environmental Protection Agency's Environmental Monitoring Systems
Laboratory in Las Vegas. Nevada, prepared this report for the Agency's Environment
Services Division in Region 5 and Great Lakes Program Office.
111
-------
CONTENTS
I
Abstract
Introduction
Methodology . .
Photo Analysis
Spring Survey
Summer Survey
Winter Survey
References
FIGURES
Number
1 Great Lakes study area location
2 Approximate locations of Landsat images and dates analyzed for the
spring survey
3-14 Image analysis 6
15 Approximate locations of Landsat images and dates analyzed for the
summer survey
16-34 Image analysis 21
35 Approximate locations of Landsat images and dates analyzed for the
winter survey
36-39 Image analysis 42
v
-------
/ J'/LAWRENCEVItLE - VINCENNES
re 1. Great Lakes study area location. Approximate scale 1:5,000,000,
VI
-------
INTRODUCTION
The Great Lakes National Program Office conducted ship surveys on Lake Michig
Lake Huron, and Lake Erie (Figure 1) between April 1984 and January 1985. These w
performed to help monitor the progress towards the objectives of the 1978 Canada-
United States Great Lakes Water Quality Agreement. Water sample data were collect
for use in lake eutrophication models and to monitor changes in water quality. In
support of these objectives, this study was performed to identify sources of
sediment input from tributaries into the lakes.
Landsat multispectral scanner (MSS) band 2 transparencies covering Lake
Michigan. Lake Huron, and Lake Erie were used for the analysis. Multiple image
dates, which corresponded to the spring, summer, and winter surveys, were acquired
from the EROS Data Center in Sioux Falls, South Dakota, whenever available.
The U.S. Environmental Protection Agency's Environmental Monitoring Systems
Laboratory in Las Vegas. Nevada, prepared this report for the Agency's Environment
Services Division in Region 5 and Great Lakes National Program Office.
-------
ce 2. Approximate locations of Landsat images and dates analyzed for the spring
survey. Approximate scale 1:3,600.000.
-------
LEGE
BLACK 0\
CITY
ICE
CLOUDS
HAZE
BLUE 0V
LAKE NAME
US - UNKNOWN
UT- UNKNOWN
RIVER NAME
e 6. Image analysis of Great Lakes study area. May 15, 1984.
1:1.000.000.
Approximate scale
-------
LEGE
BLACK O
CITY
ICE
CLOUDS
HAZE
BLUE O\
LAKE NAME
US -UNKNOWI
UT-UNKNOWI
RIVER NAME
e 7. Image analysis of Great Lakes study area. March 29. 1984.
scale 1:1.000.000.
Approximate
10
-------
PHOTO ANALYSIS
SPRING SURVEY
A total of 12 Landsat band 2 images were acquired which correspond with the
spring lake survey. These images provide coverage for most of Lake Michigan, Lake
Huron, and Lake Erie. The approximate locations of these images and date of
acquisition are shown in Figure 2. The interpretation of each image is presented
Figures 3 through 14.
Six of these images provide coverage for nearly all of Lake Michigan (Figures
through 8). Visual interpretation of these images resulted in the detection of fi
apparent sediment outfalls. Three of these were identified by name and are
annotated in Figures 7 and 8. The two remaining tributaries (Figures 5 and 8) cou
not be identified by name and were marked as unknown tributaries. Additional
turbidity was observed in Figures 5. 7. and 8 which could not be attributed to any
point source. Consequently, these patterns were marked as originating from an
unknown source. The source of turbidity observed in Figure 8 may be obscured by i
which was detected along that shoreline.
Figures 9, 10. and 11 provide coverage of most of Lake Huron during the
spring. The only location where sediments were observed was in Saginaw Bay (Figur
11). The Saginaw River. Quanicassee River, and Wiscoggin Drain all seem to be
contributing to the turbidity of that bay. The turbidity patterns indicate that t
flow is north out of the bay and under the ice (Figure 10), flowing around the poi
and proceeding south along the west shore of the lake. The ice, which is still
remaining along much of the shoreline, may be obscuring additional sediment inputs
The final area of analysis for the spring survey was Lake Erie (Figures 12. 1
and 14). Ten tributaries were observed which appeared to be discharging turbid
water into the lake. All of these were identified by name. Most are located alon
the southern shoreline near Sandusky. Ohio (Figure 12). Several additional sedime
plumes were observed but could not be traced to their source. The most prominent
these is in the southwestern section of the lake (Figure 12). This plume appears
originate to the west of the image, somewhere in the vicinity of Toledo. Ohio.
-------
LEGE
BLACK O>
CITY
ICE
CLOUDS
HAZE
BLUE OV
LAKE NAME
US - UNKNOWN
UT•UNKNOWN
RIVER NAME
e 3. Image analysis of Great Lakes study area. April 29, 1984.
scale 1:1.000.000.
Approximate
-------
BLACK OV
CITY
ICE
CLOUDS
HAZE
BLUE OVI
LAKE NAME
US - UNKNOWN
UT-UNKNOWN
RIVER NAME
re 4. Image analysis of Great Lakes study area. April 21. 1984
scale 1:1.000.000.
Approximate
-------
LEGE
BLACK 0
CITY
ICE
CLOUDS
HAZE
BLUE O\
LAKE NAME
US -UNKNOWf
UT-UNKNOWr
RIVER NAME
e 5. Image analysis of Great Lakes study area, May 7. 1984.
1:1.000.000.
Approximate scale
8
-------
LEGE
BLACK O\
CITY
ICE
CLOUDS
HAZE
BLUEOV
LAKE NAME
US - UNKNOWN
UT - UNKNOWfy
RIVER NAME
ce 9. Image analysis of Great Lakes study area. April 8, 1984.
scale 1:1.000.000.
Approximate
12
-------
LEGE
BLACK 0\
CITY
ICE
CLOUDS
HAZE
BLUE OV
LAKE NAME
US - UNKNOWN
UT- UNKNOWN
RIVER NAME
e 8. Image analysis of Great Lakes study area. March 29, 1984.
scale 1:1,000,000.
Approximate
11
-------
LEGEN
BLACK OVI
CITY
ICE
CLOUDS
HAZE
BLUE OVE
LAKE NAME
US - UNKNOWN:
UT- UNKNOWN
RIVER NAME
ce 12. Image analysis of Great Lakes study area. April 25, 1984
scale 1:1.000.000.
Approximate
15
-------
LEG
BLACK C
CITY
ICE
CLOUDS
HAZE
BLUEQT
LAKE NAME
US -UNKNOWI
UT-UNKNOWI
RIVER NAME
e 13. Image analysis of Great Lakes study area. April 10, 1984
scale 1:1.000.000.
Approximate
16
-------
LEGEI
BLACK OV
CITY
ICE
CLOUDS
HAZE
BLUE OV
LAKE NAME
US - UNKNOWN
UT- UNKNOWN
RIVER NAME
e 14. Image analysis of Great Lakes study area. April 2. 1984.
scale 1:1.000.000.
Approximate
17
-------
e 15
Approximate locations of Landsat images and dates analyzed for the summer
survey. Approximate scale 1:3.600,000.
18
-------
PHOTO ANALYSIS
SUMMER SURVEY
Nineteen Landsat band 2 images were selected which corresponded with the sumro
lake survey. These scenes provide complete coverage of Lake Michigan, Lake Huron,
and Lake Erie. Figure 15 shows the approximate locations of these scenes as well
the date of acquisition. The interpretation of each image is presented in Figures
16 through 34.
Figures 16 through 23 provide coverage of Lake Michigan. Analysis of these
images identified turbidity patterns which appear to originate from the outfall of
several tributaries. A total of 12 tributaries were identified as apparent source
of sediment input. Six of these were identified by name while the remaining six
were marked as unknown tributaries. All of these were located along the eastern o
southern shoreline of the lake. Turbidity patterns were also observed within Gree
Bay (Figure 17) but could not be traced to a source point.
The analysis of Lake Huron was performed using Figures 20, 24-27, 29 and 30.
Interpretation of these images resulted in the identification of five sediment
outfalls. Three of these were identified by name as the Saint Mary's River (Figur
24), Saginaw River, and Quanicassee River (Figure 27). The remaining two were
marked as unknown tributaries. An additional sediment plume was observed along tin
southern shore of the lake (Figure 27) but could not be traced to its origin. It
appeared to originate to the east, beyond the edge of the image. That area is shoi
in Figure 30 but because the satellite coverage was not of the same date, no
conclusions about the sediment source could be made. Additional light tonal
patterns were observed in the middle of the lake in Figure 26 but appear to be the
result of fog.
19
-------
The final area to be analyzed for the summer survey was Lake Erie. Coverage
this lake is provided by Figures 28 and 30 through 34. Analysis of these images
detected several tributaries which appear to be discharging turbid water into the
lake. Six of these tributaries are located at the west end of the lake (Figures 2
and 31) and are identified by name. The Toussaint and Sandusky Rivers were
annotated on each of these two image dates because of overlap between adjacent
Landsat images. A light haze was also observed in this area for one of the dates
(Figure 28) and should not be misinterpreted as turbidity. Two additional point
sources were identified in Figures 32 and 34 but could not be named. Finally,
turbid waters were observed along the north shore of the lake (Figure 30) but no
source could be delineated. Clouds are obscuring part of this plume and may also
covering the source point.
20
-------
LEGE
BLACK O\
CITY
ICE
CLOUDS
HAZE
BLUE OV
LAKE NAME
US - UNKNOWN
UT- UNKNOWN
RIVER NAME
e 16. Image analysis of Great Lakes study area. August 19. 1984.
scale 1:1.000.000.
Approximate
21
-------
LEGE
BLACK C
CITY
ICE
CLOUDS
HAZE
BLUE OV
LAKE NAME
US -UNKNOWf
UT-UNKNOWr
RIVER NAME
ce 17. Image analysis of Great Lakes study area. August 19. 1984
scale 1:1,000.000.
Approximate
22
-------
LEGEF
BLACK OV
CITY
ICE
CLOUDS
HAZE
BLUE OVI
LAKE NAME
US - UNKNOWN
UT- UNKNOWN
RIVER NAME
:e 18. Image analysis of Great Lakes study area. July 18. 1984.
scale 1:1.000.000.
Approximate
23
-------
LEGE
BLACK O
CITY
ICE
CLOUDS
HAZE
BLUE OV
LAKE NAME
US - UNKNOWN
UT-UNKNOWr
RIVER NAME
e 19. Image analysis of Great Lakes study area. July 18. 1984
scale 1:1.000.000.
Approximate
24
-------
LEGEI
BLACK O\
CITY
ICE
CLOUDS
HAZE
BLUE OV
LAKE NAME
US - UNKNOWN
UT - UNKNOWNS
RIVER NAME
re 20. Image analysis of Great Lakes study area. August 12. 1984.
scale 1:1.000.000.
Approximate
25
-------
LEGE
BLACK O
CITY
ICE
CLOUDS
HAZE
BLUEOV
LAKE NAME
US - UNKNOWN
UT - UNKNOWIV
RIVER NAME
e 21. Image analysis of Great Lakes study area. August 12, 1984.
scale 1:1,000,000.
Approximate
26
-------
LEGEI
BLACK 0V
CITY
ICE
CLOUDS
HAZE
BLUE OVE
LAKE NAME
US - UNKNOWN 5
UT-UNKNOWN:
RIVER NAME
-------
LEGE
oi
study
StU
1984
caie
28
BLACK O
CITY
ICE
CLOUDS
HAZE
BLUE O'
LAKE NAME
US -UNKNOW
UT-UNKNOW
RIVER NAME
te
-------
LEGE
BLACK
CITY
ICE
CLOUDS
HAZE
BLUEOV
LAKE NAME
US - UNKNOWN
UT- UNKNOWN
RIVER NAME
e 24. Image analysis of Great Lakes study area. July 20. 1984.
scale 1:1.000.000.
Approximate
29
-------
LEGE
BLACK O
CfTY
ICE
CLOUDS
HAZE
BLUE 0\
LAKE NAME
US - UNKNOWI
UT - UNKNOWI
RIVER NAME
re 25. Image analysis of Great Lakes study area. July 29, 1984
scale 1:1.000,000.
Approximate
30
-------
LEGENC
BLACK OVEI
CITY
ICE
CLOUDS
HAZE
BLUEOVE
LAKE NAME
US - UNKNOWN!
UT - UNKNOWN
RIVER NAME
«. 30. image analysis of Great LaKes study area. July 22. 1984
scale 1:1.000.000.
35
Approximate
-------
LEj
BLACK
CITY
ICE
CLOUDS
HAZE
BLUEO
LAKE NAME
US - UNKNOVU
UT-UNKNOW
RIVER NAME
Approximate
36
-------
LEGEr
BLACK OV
CITY
ICE
CLOUDS
HAZE
BLUE OV
LAKE NAME
US - UNKNOWIN
UT - UNKIMOWf
RIVER NAME
ure 32. lma,e analysis of Great LaRes study «... August 16. 1984.
scale 1:1.000.000.
37
Approximate
-------
LEGI
BLACK C
CITY
ICE
CLOUDS
HAZE
BLUE OV
LAKE NAME
-------
PHOTO ANALYSIS
WINTER SURVEY
Four Landsat band 2 images were used to identify sediment input from
tributaries during the winter lake survey. These images provide coverage of the
western shoreline of Lake Michigan (Figure 35). No other images were available f(
the study area because of cloud cover and data processing delays at the EROS Data
Center. The interpretation of these images is presented in Figures 36 through 39.
Visual interpretation of these images resulted in the detection of six appare
sediment outfalls. Five of these were identified by name and are marked on the
attached overlays in Figures 37-39. The remaining outfall which could not be name
is marked as an unknown tributary (Figure 39). Additional sediments were observed
to the south of these outfalls in the vicinity of Michigan City. Indiana. Analysi
of these turbidity patterns indicates that shoreline currents are flowing north.
This indicates that the source of these sediments is to the south and therefore
under the clouds. Conseguently, the sediments were annotated as coming from an
unknown source.
41
-------
LEGEts
BLACK 0V
CITY
ICE
CLOUDS
HAZE
BLUE OV
LAKE NAME
US - UNKNOWN
UT - UNKNOWIV
RIVER NAME
ice 36. image analysis of Great Lakes study area. January 3. 1985.
scale 1:1.000.000.
42
Approximate
-------
PHOTO ANALYSIS
WINTER SURVEY
Four Landsat band 2 images were used to identify sediment input from
tributaries during the winter lake survey. These images provide coverage of the
western shoreline of Lake Michigan (Figure 35). No other images were available for
the study area because of cloud cover and data processing delays at the EROS Data
Center. The interpretation of these images is presented in Figures 36 through 39.
Visual interpretation of these images resulted in the detection of six apparer
sediment outfalls. Five of these were identified by name and are marked on the
attached overlays in Figures 37-39. The remaining outfall which could not be named
is marked as an unknown tributary (Figure 39). Additional sediments were observed
to the south of these outfalls in the vicinity of Michigan City, Indiana. Analysis
of these turbidity patterns indicates that shoreline currents are flowing north.
This indicates that the source of these sediments is to the south and therefore
under the clouds. Consequently, the sediments were annotated as coming from an
unknown source.
41
-------
LEG
BLACK(
CITY
ICE
CLOUDS
HAZE
BLUEC
LAKE NAME
US -UNKNOV
UT-UNKNOV
RIVER NAME
*-*'•,'„.
are 36. Image analysis of Great Lakes study area. January 3. 1985
scale 1:1.000.000.
42
Approximate
-------
LEGEN
BLACK OVE
CITY
ICE
CLOUDS
HAZE
BLUEOVEI
LAKE NAME
US - UNKNOWN S
UT- UNKNOWN 7
RIVER NAME
39. Image analysis of Great Lakes study area, January 3, 1985
scale 1:1,000,000.
Approximate
45
-------
REFERENCES
Hoffer. R. "Biological and Physical Considerations in Applying Computer-Aided
Analysis Techniques to Remote Sensor Data," chapter 5 of Remote Sensing: The
Quantitative Approach, Swain, P., and S. Davis, eds.. McGraw-Hill, New York.
New York. 1978.
47
------- ' ,r /•* ;_ *• : -» •**• ------- |