STUDY PERFORMANCE REPORT
State: Michigan
Project No.:
Study No.: 230702
Title: Effects of sediment traps on Michigan
river channels
Period Covered:
F-80-R-10
October 1, 2008 to September 30, 2009
Study Objectives: To quantify the effect of sediment removal efforts on the channel morphology of
select Michigan streams. More specifically, to 1) identify the rate and spatial extent of change in
riverbed elevation and substrate conditions, and 2) relate these data to hydrologic, gradient, and
valley characteristics of each stream.
Summary: I re-surveyed previously-established transects in all eight rivers included in this study.
The data continue to show that excavation of sediment traps generally had only small effects on
mean channel depth and substrate, with changes occurring both upstream and downstream of the
trap. The lateral position of the channel remained constant, indicating little side cutting had
occurred. Changes in channel area remain variable and appear as likely to occur at transects
proximal to the sediment trap as at transects located further upstream or downstream. Little
difference in mean channel depth, area, or frequency of coarse substrates was evident upstream
or downstream of the study trap on the Sturgeon River, which was operated for over 20 years
before being abandoned in 2005. These results suggest that sediment trap maintenance has not
achieved the desired goals of increased downcutting and exposure of coarse substrates
downstream of the sediment traps studied.
Findings: Jobs 2, 3, and 4 were scheduled for 2008-09, and progress is reported below.
Job 2. Title: Survey bed elevations and substrate conditions.–I re-surveyed bed elevations and
recorded substrate composition from visual observations and pebble counts at previouslyestablished transects in each of eight rivers included in this study.
Job 3. Title: Analyze data.−When compared to data collected in previous years (tables 1 and 2),
data from 2009 continue to show variable changes in depth in stream reaches where sediment
traps were constructed. Trap excavation created a headcut upstream of sediment traps in the Au
Sable and Baldwin rivers as well as Silver Lead and Twomile creeks. The extent of the
headcutting varied by location and ranged from 0.1 ft in the Au Sable River and Twomile Creek
to 0.9 ft in the Baldwin River. Channel depth downstream of the sediment trap increased by 0.1–
0.3 ft in the Au Sable and Baldwin rivers and Silver Lead Creek (Table 1). No change in depth
was observed below the sediment trap in Twomile Creek, while depth below the sediment trap in
the Boardman River and both above and below the sediment traps in the East Branch Au Sable
and Little Manistee Rivers increased, indicating aggradation of the channel at these locations.
Changes in channel area were also variable and are as likely to occur at transects that are close to
the sediment trap as at transects located further upstream or downstream. The lateral position of
the channel remained the same throughout all study reaches.
Visual observation data indicated that sand substrate decreased by 13% downstream of the trap
in the Little Manistee River, with a commensurate increase in gravel substrate. Even though sand
F-80-R-10, Study 230702
substrate decreased downstream of the trap at four other locations, an accompanying decrease
upstream of the traps in these rivers indicates a drop in fine substrate throughout each reach due
to headcutting or some other larger-scale factor other than the sediment trap itself. The observed
decrease in sand substrate was paired with an increase in gravel substrate both upstream and
downstream of the trap at three of the same locations. An increase in silt and detritus occurred
with a decrease in sand throughout the study reach of the Baldwin River and an increase in sand
occurred with a decrease in silt and detritus throughout Twomile Creek. These observations
suggest that observers may be classifying sand and silt/detritus differently from year-to-year
(Table 2). Pebble count data, which are less subjective to observer bias, show that fine substrates
(D50 <1 mm) are common throughout all study reaches and have not decreased more than 6%
downstream of any sediment trap except Silver Lead Creek, where a more substantial decrease
occurred upstream of the trap (Figure 1).
I observed little difference in bed elevations and substrate upstream and downstream of a trap
that was recently abandoned in the Sturgeon River. Although the trap was maintained from 1984
to 2005, mean depth upstream and downstream of the trap is very similar (Table 3). Contrary to
expectation, both visual observation and pebble count data indicate coarser substrate upstream of
the trap compared to downstream (Table 3, Figure 1). However, median particle size was again
fine at both locations (D50 <0.15 mm), indicating the trap did not achieve the objective of
exposing coarse substrate.
Job 4. Title: Write annual performance reports.−This progress report was prepared.
Prepared by: Todd C. Wills
Date: September 30, 2009
2
F-80-R-10, Study 230702
Table 1.–Summary of change in mean channel depth following sediment trap excavation
for seven study reaches over an eight-year period. Post-excavation data are in bold. Preexcavation data were not collected in the Boardman or Little Manistee rivers. U = upstream, d
= downstream of sediment trap.
River
Au Sable
Baldwin
Boardman
E Br Au Sable
Little Manistee
Silver Lead Cr
Twomile Cr
1
Location
02
03
u
d
u
d
u
d
u
d
u
d
u
d
u
d
1.9
1.9
–
–
–
–
1.8
2.1
3.4
3.7
–
–
–
–
2.0
2.1
–
–
2.6
2.2
1.9
2.1
3.3
3.7
1.6
1.4
–
–
Mean channel depth (ft)
04 05 06 07 08
1.9
2.1
–
–
2.6
2.1
1.9
2.0
–
–
1.7
1.5
2.4
2.3
–
–
–
–
–
–
–
–
3.3
3.6
2.0
1.6
2.6
2.5
Total number of depth measurements collected.
3
–
–
4.0
4.2
–
–
–
–
–
–
1.9
1.7
2.7
2.5
–
–
4.9
4.6
–
–
–
–
3.2
3.2
–
–
–
–
2.1
2.0
–
–
–
–
–
–
–
–
–
–
–
–
09
N1
505
2.0
2.0 1,033
334
4.9
493
4.5
274
2.6
608
1.8
312
1.7
632
1.9
474
3.0
3.2 1,038
417
2.0
891
1.7
307
2.5
724
2.3
Net
change (ft)
0.1
0.1
0.9
0.3
0.0
-0.4
-0.1
-0.2
-0.4
-0.5
0.4
0.2
0.1
0.0
F-80-R-10, Study 230702
Table 2.–Summary of net change in visually classified substrate data following sediment trap
excavation for seven study river reaches. Pre-excavation data were not collected in the Boardman
or Little Manistee rivers. Substrate categories (and dimensions in mm) were silt (0.004–0.063),
sand (0.064–2), gravel (3–64), small cobble (65–128), large cobble (129–256) and other (clay,
boulder, or wood). U = upstream, d = downstream of sediment trap.
River
(Time period)
Silt or
Location detritus
Net change (%)
Small
Gravel cobble
Sand
Large
cobble
Other
N1
Au Sable
(2002-09)
u
d
-5
17
-19
-31
15
7
0
0
0
0
0
8
362
738
Baldwin
(2006-09)
u
d
19
16
-21
-12
1
0
0
0
0
0
1
-4
166
293
Boardman
(2003-09)
u
d
-1
-3
3
6
-8
-5
-5
-5
0
0
10
8
161
373
E Br Au Sable
(2002-09)
u
d
0
4
-25
-16
29
7
0
0
0
0
-4
5
218
412
Little Manistee
(2002-09)
u
d
0
2
0
-13
0
13
0
-1
0
0
0
0
340
735
Silver Lead Cr
(2003-09)
u
d
2
8
-49
-24
41
19
0
0
3
0
3
-3
304
637
Twomile Cr
(2004-09)
u
d
-30
-24
47
37
0
0
0
0
0
0
-17
-14
191
464
1
Total number of substrate observations.
4
F-80-R-10, Study 230702
Table 3.–Summary of mean channel depth and visually classified substrate in the Sturgeon River.
Substrate categories (and dimensions in mm) were silt (0.004–0.063), sand (0.064–2), gravel (3–64),
small cobble (65–128), large cobble (129–256) and other (clay, boulder, or wood). U = upstream, d =
downstream of sediment trap.
Substrate composition (%)
Year
2008
2009
Net
change
1
2
Location
u
d
u
d
u
d
Mean
depth
(ft)
2.9
3.0
2.9
3.0
0.0
0.0
N1
108
155
216
311
Silt or
detritus
20
20
13
17
-7
-3
Sand
41
56
63
75
22
19
Total number of depth measurements collected.
Total number of substrate observations.
5
Gravel
12
2
14
1
2
-1
Small
cobble
9
0
0
0
-9
0
Large
cobble
0
0
0
0
0
0
Other
18
23
10
8
-9
-15
N2
92
131
92
132
F-80-R-10, Study 230702
Au Sable River
Baldwin River
100%
Percent finer than
Percent finer than
100%
80%
60%
40%
20%
0%
0.01
0.1
1
10
100
1000
80%
60%
40%
20%
0%
0.01
10000
0.1
1
Particle size (mm)
US 1 yr pre-dig (2006)
US 3 yrs post-dig (2009)
DS 7 yrs post dig (2009)
DS 1 yr pre-dig (2006)
DS 3 yrs post-dig (2009)
East Branch Au Sable River
Percent finer than
Percent finer than
100%
60%
40%
20%
0.1
1
10
100
1000
80%
60%
40%
20%
0%
0.01
10000
0.1
1
Particle size (mm)
100
1000
10000
US 2 yrs post-dig (2004)
US 7 yrs post-dig (2009)
US 2 yrs post-dig (2004)
US 7 yrs post-dig (2009)
DS 2 yrs post-dig (2004)
DS 7 yrs post-dig (2009)
DS 2 yrs post-dig (2004)
DS 7 yrs post-dig (2009)
Silver Lead Creek
100%
100%
Percent finer than
Percent finer than
10
Particle size (mm)
Little Manistee River
80%
60%
40%
20%
0.1
1
10
100
1000
80%
60%
40%
20%
0%
0.01
10000
0.1
1
Particle size (mm)
10
100
1000
10000
Particle size (mm)
US 4 yrs post-dig (2005)
US 8 yrs post-dig (2009)
US 1 yr pre-dig (2004)
US 5 yrs post-dig (2009)
DS 4 yrs post-dig (2005)
DS 8 yrs post-dig (2009)
DS 1 yr pre-dig (2004)
DS 5 yrs post-dig (2009)
Sturgeon River
Twomile Creek
100%
Percent finer than
100%
Percent finer than
10000
US 7 yrs post-dig (2009)
80%
80%
60%
40%
20%
0%
0.01
1000
DS 2 yrs post-dig (2004)
100%
0%
0.01
100
US 2 yrs post-dig (2004)
Boardman River
0%
0.01
10
Particle size (mm)
0.1
1
10
100
1000
80%
60%
40%
20%
0%
0.01
10000
Particle size (mm)
0.1
1
10
100
1000
10000
Particle size (mm)
US 3 yrs post-abandon (2008)
US 4 yrs post-abandon (2009)
US 1 yr pre-dig (2004)
US 5 yrs post-dig (2009)
DS 3 yrs post-abandon (2008)
DS 4 yrs post-abandon (2009)
DS 1 yr pre-dig (2004)
DS 5 yrs post-dig (2009)
Figure 1.–Cumulative frequency of substrate particle size from pebble counts in eight study
rivers. Substrate particle size categories (and dimensions in mm) were organic and clay (0–0.04mm),
silt (0.05–0.062mm), sand (0.063–2mm), very fine gravel (2–4mm), fine gravel (5–8mm), medium
gravel (9–16mm), coarse gravel (17–32mm), very coarse gravel (33–64mm), small cobble (65–
128mm), large cobble (129–256mm), small boulder (257–512mm), and medium boulder (>512mm).
US = upstream, DS = downstream.
6