20040722 Ver 1_Mitigation Information_20100830

Michael Baker Engineering, Inc.
8000 Regency Parkway, Suite 200
Cary, NC 27518
[phone]
August 30, 2010
Restoration Systems
Attn: Mr. Worth Creech - Project Manager
1101 Haynes Street, Suite 211
Raleigh NC 27604
Subject: Further Information Regarding Channel Stability for the
Elk Shoals Full Delivery Project.
Dear Mr. Creech,
0_0?a7_
This letter is intended to provide additional information regarding channel stability assessments
for the Elk Shoals Stream Restoration Full Delivery Project. Mr. Will Harman (Stream
Mechanics) provided an initial assessment of channel stability in his letter to you dated June 21,
2010, based on monitoring data that had been collected and his personal observations of the site.
In a meeting between the NC Ecosystem Enhancement Program (EEP), Restoration Systems
(RS), Will Harman, and Kevin Tweedy (Baker Engineering) held at the EEP offices on July 30,
2010, EEP expressed a desire to receive additional data regarding the vertical stability and
floodplain access of the channel in the area near permanent cross-section #3. The monitoring
data collected to date indicated that the channel at cross-section #3 had incised during 2009
(monitoring year 5) by approximately one foot. EEP was hesitant to support the proposed
recommendations in Will Harman's letter without additional data to demonstrate that the incision
seen at cross-section #3 was not a systemic problem that would continue to get worse.
On August 4, 2010, three staff members from Baker Engineering visited the project site and re-
surveyed permanent cross-section #3, as well as one additional riffle cross-section upstream of
#3, and one additional riffle cross-section downstream of #3. The additional cross-section
surveyed upstream of #3 (labeled here as #11) was located approximately 16 feet upstream and
was the closest riffle location to cross-section #3. The additional cross-section surveyed
downstream of #3 (labeled here as #12) was located approximately 122 feet downstream of #3 on
the next available riffle downstream, which was also the location of a constructed riffle installed
during construction of the site. The intent of these additional surveys was to determine if the
incision at cross-section #3 had increased or decreased since the surveys last fall, if the incision
was evident upstream or downstream of cross-section #3, and to what degree the remnant channel
section has access its adjacent floodplain.
Bed Elevation Assessments
The re-survey of cross-section #3 indicates that the bed elevation has risen slightly (0.2 feet);
therefore, the incision noted in the Year 5 data is not getting any deeper and has actually filled
back slightly (see Graph 1). The surveyed cross-sections upstream (#11) and downstream (#12)
of cross-section #3 are 0.7 and 1.0 feet, respectively, higher in elevation than cross-section #3
(see Graphs 2, 3, and 4). This indicates that the scour noted at cross-section 93 is a localized
phenomenon and is not characteristic of reach-wide incision or instability (see Graph 4).
Page 1 of 8
Michael Baker Engineering, Inc.
8000 Regency Parkway, Suite 200
• Cary, NC 27518
[phone]
Max
Stream BKF BKF BKF BKF BH BKF TOB
Feature Type Area Width Depth Depth W/D Ratio ER Elev Elev
Riffle E 60.1 21.01 2.86 5.97 7.35 1 > 2.4 926.5 926.74
933
931
929
927
0
925
m
W 923
921
919
Elk Shoals Creek
Cross-section #3
-----------------------------------------------------------------------------------------------------------------------
-------------------------------------------------
t"uilt --------------------

Yaer 1
Year 2
?• Yaer 3
-\
Yeer 4 Year 5
-•- Year 6 -
--o---- F rone ---e---- Banktull
100 110 120 130 140 150
Station (ft)
Graph 1. Cross-section #3 data collected on August 4, 2010 (Year 6), compared to the first five
years of monitoring data for the location.
Max
Stream BKF BKF BKF BKF BH TOB
Feature Type Area Width Depth Depth W/D Ratio ER BKF Elev Elev
Riffle E 50.2 17.02 2.95 4.49 5.76 1 > 1.8 925.96 925.97
933
931
e 929
0
927
d
U' 925
923
921
919
Elk Shoals Creek
Cross-section #11
-----------------o
0 5 10 15 20 25 30 35
Station
- -a--- Bankfull ---a--- Floodprone
Graph 2. Cross-section #11, surveyed approximately 16 ft upstream of permanent cross-section
#3 on August 4, 2010.
Page 2 of 8
Michael Baker Engineering, Inc.
8000 Regency Parkway, Suite 200
Cary, NC 27518
[phone]
Max
Stream BKF BKF BKF BKF BH BKF TOB
Feature Type Area Width De th De th W/D Ratio ER Elev Elev
Riffle E 51.5 24.67 2.09 4.21 11.81 1 ? 1.4 926 925.95
Ekl Shoals Creek
Cross-section #12
933 _ --

931
--------------------------------------------------------------------------------------------------------------- --- -
929
2
.
927
W
925 ---------------------------------------------------------------------------------------------
923
921

919
0 5 10 15 20 25 30 35 40
Station
---o-- Bankfull ---o-- Floodprone
Graph 3. Cross-section #12, surveyed approximately 122 ft downstream of permanent cross-
section #3 on August 4, 2010.
Comparison of TW Elevations Along Channel Near Permanent Cross-section #3
925
924
923
922
0
.q
d 921
a
920
919
918
¦ Nov. 2009
XS 1 1 _? Aug. 2010
XS #3
¦? ---------------"
¦
XS #12

2200 2250 2300 2350 2400 2450
Longitudinal Station (ft)
Graph 4. Comparison of thalweg elevations for cross-sections (XS) #3, #11, and #12. Data show
that the scour at cross-section #3 is a localized phenomenon.
Page 3 of 8
Michael Baker Engineering, Inc.
8000 Regency Parkway, Suite 200
Cary, NC 27518
[phone]
Observations during the August 4 field visit provided insight into the cause of the scour at cross-
section #3. After construction of the site, a relatively uniform riffle existed in the location of
cross-section #3, and was the reason that a permanent cross-section was selected for that location.
The riffle discharged into a pool around the downstream meander bend. During the past two
years, a small bar feature has developed upstream of cross-section #3, reducing the width of the
base-flow channel just upstream and causing a scour feature to develop at cross-section #3 due to
the flow convergence caused by the bar feature. The location of cross-section #3 is now
functioning more as a run feature that extends down into the downstream meander pool, while the
upstream head of riffle has steepened due to the development of the bar feature. This evolution
sequence is illustrated in Figure 1 below. The bar feature can be seen in Graph 2 and Photo 2,
taken during the field surveys on August 4, 2010.
Stream channels are dynamic systems, and the changes in the bed profile described above are
considered to be part of the natural evolution and seasonal changes in bed characteristics that are
common for all streams. During the field surveys on August 4, no areas of instability were
observed in the vicinity around cross-sections #3 and #11, and vegetation was quite dense on the
channel banks (see Photos 1 and 2).
Floodplain Access
The Final Restoration Plan document for the project, dated October 2004 (amended), summarized
assessments that were performed during the project design phase to evaluate the cross-sectional
area of the remnant channel reach. Pages from the Restoration Plan that discuss the results of
those assessments are provided in Attachment 1. The assessments included an evaluation of
cross-sections along the remnant channel to evaluate its cross-sectional area. Cross-sections were
cut every 200 feet through the topographic survey data collected for the remnant channel, and are
plotted with associated bankfull cross-sectional areas and estimated bank height ratios (BHR) in
Attachment #2. The data show that at the upstream end of the remnant channel (approximately
20+00 to 24+00), cross-sectional areas were less than needed to carry the bankfull flow, due to
partial filling of this section during past disturbance (pre-restoration). During restoration, this
section of the remnant channel was opened back up to a cross-sectional area of approximately 50
to 60 square feet. From approximate station 26+00 to 26+00 to 30+00, the cross-sectional area of
the remnant channel ranged from approximately 50 to 60 square feet in its existing condition, and
matched well with bankfull estimates from regional curve data. From station 32+00 to 34+00, the
remnant channel was slightly incised, with an estimated BHR of approximately 1.5. A BHR of
1.5 is considered moderately incised, but still allows floodplain access at large flows. Because
this reach had a significant amount of mature woody vegetation on the channel banks, the level of
incision was not severe, the reach was rather short, and low BHRs above and below the reach
afforded appropriate floodplain access, the reach was considered to be sufficiently stable to
receive restored flows. Since construction, no stability issues have been identified along this
reach.
In summary, the assessments conducted during the restoration plan development evaluated the
cross-sectional area of the remnant channel and found that the cross-sectional area at top-of-bank
for the majority of the channel matched well with the Piedmont Regional Curve information (see
page 4-1 and 4-2 of Attachment 1); therefore, it was concluded that restoring flows to the remnant
channel section would provide appropriate reconnection with the adjacent floodplain.
Page 4 of 8
Michael Baker Engineering, Inc.
8000 Regency Parkway, Suite 200
Cary, NC 27518
[phone]
To further investigate the floodplain connectivity of the remnant channel with more current data,
the top-of-bank cross-sectional areas of the cross-sections surveyed on August 4, 2010 were
compared to the design bankfull cross-sectional area of Elk Shoals Creek. Page 6-1 in
Attachment 1 provides the geomorphology design table for Elk Shoals Creek, and gives the
design bankfull cross-sectional area as 53 square feet. For cross-sections 3, 11, and 12, cross-
sectional areas at top-of-bank were 60, 50, and 52 square feet, respectively, as shown in Graphs 1,
2, and 3. This provides further confidence that the remnant channel is of appropriate size to carry
the bankfull flow with floodplain access available for flows larger than bankfull.
It is our assessment that the changes in bed elevation observed at cross-section #3 do not
represent channel instability, but rather natural changes in bed features that occur on all streams
over time. We also conclude that the remnant channel section of the project is adequately sized to
provide appropriate floodplain access during storm events. If you have any further questions or
would like additional information, please contact me at [phone].
Sincerely,
Kevin Tweedy, PE
Michael Baker Engineering, Inc.
Page 5 of 8
FLOW
0
xs 94
Bar Fea
Michael Baker Engineering, Inc.
8000 Regency Parkway, Suite 200
Cary, NC 27518
[phone]
FLOW
Figure 1. Schematic of the changes in bed profile and character in the vicinity of cross-section
#3, based on observations and data collected during field surveys on August 4, 2010.
Page 6 of 8
Year 1
I Year 5
Michael Baker Engineering, Inc.
8000 Regency Parkway, Suite 200
Cary, NC 27518
[phone]
Photo 1. Looking at right bank of cross-section #11. Note stability of the bank and vegetation.
Page 7 of 8
Michael Baker Engineering, Inc.
8000 Regency Parkway, Suite 200
Cary, NC 27518
[phone]
vegetation that is growing on the bar.
Page 8 of 8
Photo 2. Looking at left bank of cross-section #11. The bar feature is evident from herbaceous
Attachment 1
Pages from the Sink Property - Elk Shoals Creek Restoration Plan that reference assessments
done on the remnant channel section of Elk Shoals Creek.
eventually leads to increased heights and slopes of stream banks, and when critical bank
heights are exceeded, the banks begin to fail and mass wasting of soil and rock leads to
channel widening. Incision and widening continue migrating upstream, a process
commonly referred to as a head-cut. Eventually the mass wasting slows and the stream
begins to aggrade with a new low-flow channel forming in the sediment deposits. By the
end of the evolutionary process, a stable stream with dimension, pattern, and profile
similar to those of undisturbed channels forms in the deposited alluvium but with a much
narrower floodplain. The new channel is at a lower elevation than its original form with
a new floodplain constructed of alluvial material. The old floodplain remains a dry
terrace (FISRWG, 1998). The time required to reach a state of quasi-equilibrium is
highly variable, but generally is on the order of decades.
2.2 Elk Shoals Creek Existing Conditions
Buck Engineering calculated the project watershed size at the point where Elk Shoals
Creek crosses Old Concord Church Road as 4.6 square miles. Two unnamed tributaries
flow into Elk Shoals Creek within the project limits. The watershed areas for UT1 and
UT2 are 0.38 square miles and 0.5 square miles, respectively.
Within the project site, all of the reaches have been straightened and impacted by
agricultural practices. The straightening of Elk Shoals Creek has led to incision, which
has resulted in instability and degradation. The two unnamed tributaries are also unstable
and experiencing high amounts of bank erosion because they have downcut to the
entrenched bed elevation of Elk Shoals Creek. This downcutting has resulted in high
bank height ratios and overly steep slopes. Elk Shoals Creek and its tributaries are in the
beginning stages of pattern development; however, based on the existing conditions, the
reaches will continue eroding and depositing sediment for some time before stability is
achieved. A stable stream pattern and floodplain should be restored in order to minimize
the amount of sediment being deposited into the receiving waters.
An abandoned historic reach of Elk Shoals Creek was located to the east of the existing
channel near the confluence with UT I. The historic reach was abandoned when the creek
was channelized and diverted to its existing channel, which flows down the western edge
of the floodplain. The historic section of channel is approximately 1,200 feet in length
and appears to represent a stable cross section and pattern of Elk Shoals Creek.
Existing condition parameters in Table 2-3 reflect conditions in Elk Shoals Creek and
UT1 within the project site; the parameters shown for UT2 were measured immediately
upstream of the project reach.
Sink Property - Elk Shoals Creek Mitigation Plan 2-3 Buck Engineering
Table 4-1 NC Rural Piedmont Curve Equations.
North Carolina Piedmont Rural Regional Curve Equations
Harman et al., 1999
Qbkf = 89.039 A,,, W=0.95
Abkf = 21.43 A,, R =0.91
Wbkf = 11.89 AW 11.41 R2=0.92
Dbkf = 1.5 AW R2=0.88
4.3 Bankfull Verification in the Project Watershed
The preferred method of verifying hydraulic geometry relationships within a project
watershed is to survey a nearby gage site and compare the results to the appropriate
regional curve. Buck Engineering consulted the United States Geological Survey
(USGS) website to locate gages within the Catawba River Basin (HUC 03050101). The
gage at Norwood Creek near Troutman, within this HUC, was used in the development of
the North Carolina rural Piedmont regional curve (Harman et al., 1999). Buck
Engineering visited the site to assess its continuing stability. The dimension, pattern, and
profile appeared stable and the stream had access to its floodplain during bankfull events.
The proximity of this site provides confidence that the application of the regional curve to
the project reaches is appropriate.
The bankfull stage of the main channel of Elk Shoals Creek, as well as the tributary
channels, was identified in the field; the indicators were a break in slope on flat
depositional features and the back of point bars. The relationship of bankfull cross
sectional area versus drainage area was then compared to regional curve to determine
whether the relationships are similar. Bankfull data for the project reaches are compared
with the North Carolina rural Piedmont regional curve in Figure 4-1. The cross-sectional
areas consistently plot within the 95% confidence interval. This provides evidence that
bankfull was correctly identified in the field.
The abandoned historic reach of Elk Shoals Creek provided additional confidence of the
application of the regional curve. The historic section appears to represent a stable cross
section of Elk Shoals Creek and plots within the acceptable limits of the regional curve
(Figure 4-1).
Two additional reference reaches located in the same physiographic region as the project
site were used to further corroborate the results: West Branch to Tibbs Run and an
unnamed tributary to Cane Creek (Figure 4-2). These sites were selected based on the
confidence with which bankfull features were identified, the apparent cross-section
stability, the natural state of the stream, and the stability of stream pattern. Both of these
sites had clear bankfull indicators located at, or near, the top of banks. One
representative riffle cross section was surveyed at the each site. The drainage areas were
Sink Property - Elk Shoals Creek Mitigation Plan 4-2 Buck Engineering
determined based on watershed delineation from USGS topographic quadrangles. These
points were plotted on the Piedmont regional curve, along with the data from Elk Shoals
Creek (Figure 4-1). The data plotted within acceptable limits and thus verify that the
relationships in this basin are similar to those of the Piedmont region. It is concluded that
because the surveyed cross-sectional areas of the stream reaches correlate well with
regional curve data, bankfull was correctly identified on the stream reaches.
NC Rural Piedmont Regional Curve
1000 Remnant
Channel
Section
i 100 ----Loµer95%
d - - Upper 95%
? Regional a Data
? Elk Shoals - Historic
yC ?, ¦ UT #1
a UT #2
1 r • Trib. to Cane Creek
O
cC
? ? West Branch of Tibbs Rim
? Elk Shoals - Existing
Power (Regional Curve Data)
1
0.1 1 10 100 1000
Y = 21.43x068 Waters hed Are a (Sq. Mi.)
R2 = 0.95
Figure 4-1 NC Rural Piedmont Regional Curve with Surveyed Bankfull Cross-
Sectional Areas for Elk Shoals Reaches and Reference Reaches. (Project data
points were not used in determining the regression line.)
Reference reach data for all project reference reaches is included in Appendix 3.
Sink Property - Elk Shoals Creek Mitigation Plan 4-3 Buck Engineering
of approaches may be used to optimize the design. The final selection of design criteria
is discussed in the design chapter.
5.5 Design Criteria Selection for Sink Property - Elk Shoals Creek
Buck Engineering evaluated the historic abandoned channel of Elk Shoals Creek as a
potential reference reach. The reach appears to exhibit a stable cross-section and pattern,
and plots within the 95% confidence interval of the North Carolina Piedmont rural
regional curve (see Section 4.3). Therefore, it is included as part of the design criteria
(Table 5-1). Due to its proximity to the project reach, the historic channel was selected as
the primary source of reference reach information.
Buck Engineering included data from two additional reference reaches in the North
Carolina Department of Transportation Reference Reach Database. Data from West
Branch to Tibbs Run and an unnamed tributary to Cane Creek appear in Table 5-1.
These reaches were determined to be appropriate references for this project because their
valley type, slope, and bed material compare to the project reaches and they are located in
the same physiographic region.
The Hydraulic Design Manual for Stream Restoration Projects by the US Army Corps of
Engineers (Copeland et al., 2001) was also consulted as a reference. Recommended
values for meander length and radius of curvature ratios are presented in Table 5-1. Buck
Engineering typically consults the USACE manual for pattern criteria. This is due to the
fact that most reference reach surveys in the eastern United States show radius of
curvature divided by bankfull width ratios much less than 1.5. However, the USACE
manual recommends a ratio greater than 2.0 to maintain stability in free-forming systems.
Since most stream restoration projects are constructed on floodplains denude of woody
vegetation, Buck Engineering often uses the corps recommended value rather than
reference reach data.
Sink Property -Elk Shoals Creek Mitigation Plan 5-6 Buck Engineering
Table 6-1 Natural Channel Design Parameters for Elk Shoals Creek and
Unnamed Tributaries
Sink Pronertv - Desien Parameters
Elk Shoals Creek Unnamed Trib. #1 Unnamed Trib. #2
Design Stream
Values
- Design Stream
Values Design Stream
Values Rationale

Parameter
MIN
-
MAX
F
MIN
MAX
MIN
MAX
F
Drainage Area, DA (sq mi) 4.6 0.38 0.5
Stream Type (Rosgen) C4 C4 C4 Note 1
Bankfull Discharge, Qbkf (cfs) 270 44 54 Note 2
Bankfull Riffle XSEC Area, Abkf (sq ft) 53.0 11.0 13.0 Note 2
Bankfull Mean Velocity, Vbkf (ft/s) 5.1 4.0 4.2 V=Q/A
Bankfull Riffle Width, Wbkf (ft) 23.0 25.2 10.5 11.5 11.4 12.5 Note 3
Bankfull Riffle Mean Depth, Dbkf (ft) 2.1 2.1 1.0 1.0 1.0 1.1 d=A/W
Width to Depth Ratio, W/D (ft/ft) 10.0 12.0 10.0 12.0 10.0 12.0 Note 4
Width Floodprone Area, Wfpa (ft) 230 275 230 275 230 275
Entrenchment Ratio, Wfpa/Wbkf (ft/ft) 9.1 10.9 21.9 23.9 20.2 22.0 Note 5
Riffle Max Depth @ bkf, Dmax (ft) 2.5 2.9 1.1 1.5 1.2 1.6
Riffle Max Depth Ratio, Dmax/Dbkf 1.2 1.4 1.2 1.4 1.2 1.4 Note 6
Max Depth @ tob, Dmaxtob (ft) 2.5 2.9 1.1 1.5 1.2 1.6
Bank Height Ratio, Dtob/Dmax (ft/ft) 1.0 1.0 1.0 1.0 1.0 1.0 Note 7
Meander Length, Lm (ft) 161 277 73 126 80 137
Meander Length Ratio, Lm/Wbkf * 7.0 11.0 7.0 11.0 7.0 11.0 Note 8
Radius of Curvature, Rc (ft) 46 88 21 40 23 44
Rc Ratio, Rc/Wbkf * 2.0 3.5 2.0 3.5 2.0 3.5 Note 8
Belt Width, Wblt (ft) 81 126 31 57 34 62
Meander Width Ratio, Wblt/Wbkf * 3.5 5.0 3.0 5.0 3.0 5.0 Note 8
Sinuosity, K 1.23 1.23 1.50 1.50 1.40 1.40 TW len.Nal. len.
Valley Slope, Sval (ft/ft) 0.0033 0.0033 0.0055 0.0055 0.0070 0.0070
Channel Slope, Schan (ft/ft) 0.0027 0.0027 0.0037 0.0037 0.0050 0.0050 Sval / K
Slope Riffle, Srif (ft/ft) 0.0035 0.0043 0.0048 0.0059 0.0065 0.0080
Riffle Slope Ratio, Srif/Schan 1.3 1.6 1.3 1.6 1.3 1.6 Note 9
Slope Run, Snm (ft/ft) 0.0013 0.0021 0.0018 0.0029 0.0025 0.0040
Run Slope Ratio, Srun/Schan 0.5 0.8 0.5 0.8 0.5 0.8 Note 9
Slope Glide, Sglide, (ft/ft) 0.0008 0.0013 0.0011 0.0018 0.0015 0.0025
Glide Slope Ratio, Sglide/Schan 0.3 0.5 0.3 0.5 0.3 0.5 Note 9
Slope Pool, Spool (ft/ft) 0.0005 0.0005 0.0007 0.0007 0.0010 0.0010
Pool Slope Ratio, Spool/Schan 0.20 0.20 0.20 0.20 0.20 0.20 Note 9
Pool Max Depth, Dmaxpool (ft) 4.2 6.3 1.9 3.1 2.1 3.4
Pool Max Depth Ratio, Dmaxpool/Dbkf 2.0 3.0 2.0 3.0 2.0 3.0 Note 8
Pool Width, Wpool (ft) 29.9 42.9 13.6 19.5 14.8 21.2
Pool Width Ratio, Wpool/Wbkf 1.3 1.7 1.3 1.7 1.3 1.7 Note 10
Pool-Pool Spacing, Lps (ft) 115.1 176.5 52.4 80.4 57.0 87.4
Pool-Pool Spacing Ratio, Lps/Wbkf 5.0 7.0 5.0 7.0 5.0 7.0 Note 11
d16 (mm) 0.06 0.06
d35 (mm) 0.27 0.27
d50 (mm) 2.6 2.6
d84 (mm) 27 27
d95i" c; 40 40
Sink Property - Elk Shoals Creek Mitigation Plan 6-4 Buck Engineering
Attachment 2
Cross-sectional Area Data for the Remnant Channel Section of the
Elk Shoals Full Delivery Project.
/
s fill w
a
?4 00-00,
d° CD 'A 1 9
Y ------ ------------
e
o l g9?,,es
?Z<- -
2'? ? Qua
S 8? ? t CI?~ U
Lu8z
13
ZJZ
G ?O !
2
J
Z Z W LL]
O Z
Of z
w e
NJe/fOW $m
z `ao5 03? g ?W
0 LL
?O 10 6E am
F- w I-- °noNUza m$
D W a wmz
Uo $
Z
y'
Z Q
8
O
Q
U >
W W
O
cn Q $I
H ?
z
U 1? 0
U
LL
O = y?
y
O w z ?IU
rOO to
O O w ! YLU
wm
P!; N?"nN?,l?w 5
------------------
Cl) c 3 Ifp? ? 'si
zz
?yy ?.•
F`
wgw??? ,3 ? ? 1 1 ,1.
f o?
1
\, Ngg}??j aa
_ R1?1 I !1'i!? ?¢LL N?i?
w boo
00
O+ ?? YW
co wo
LE.'9?.:'2 u6F's0-8V-Wa-St LLIB?cw(E\'.6?s?????a
I
_
i
I I -

!

I

-
1 ?
sea
I o
I I I 1 I:. I
II

M
\ l m
m
n
O - m? I O -_ Q / O
o % Nm
?c < ' N
N fi
J N
n =
<m
I

a L.
1

? .J I I I _ ! I pr

H

I g
3 ? I ?_ ? I. I

L
I
R
?

IF,
I
I

m m m - m m m m -y.- _ __.. rn m m

$ $
1_ -;- - -
h h . - I I

_ _
i

?. :JT
- h

,

..

s
I g
.. - ;

t
I 7 7:

- I _

1 ? _ T I
i -7

I 11
I
I'
o 11 O
N 11 = .. o
J

o O
O O
O co
N II
I o
o 0
+
I tr
I +
o
- tw
am I +

n
m
M r
J co
r
I I
- I

O I
_
"
-
I
I I
I
?
_
o

I
?
I
-
I
,
n - o

I ?
?
I -44
I
N
- - - - ; I r - - -- I :I
? I 1 ,. I ? I n
? I
n
s I I I ' I . I
I i- I

7 ?
T s
i I

r ?
S I
9

n
t 1 I ?
.

1

a I
i I

N
I
? I I

I I I

_ F _ , -.

I I
-

I I I

I I

- I
N ' N
- ? _
L . I
?
I , I
-
? _
j ?
.
I
I
-
r
I I -
I
I
O - O
? I
n I _ _
I :
I I 0
- I _
I N
o
I
T o

?
I ? I I , i ? ?

O - a
pOD

I
n ? / n

r - I I
I`
o

I I j _ I
I
N
1 I I i- ? ? 1:.

1 . I
N I 1 ? I I
_. I ? I Z .I .
r + I
3
3 r ? I i I ?
- t --' - _ - -

I I , I

LH L.'

?6P'Id.-u ?..?? sa [L10?n?a5-x??6?.ep?? `•k '_
OIOZ f'b
d Mechanics
June 21, 2010
Worth Creech
Project Manager
Restoration Systems
1101 Haynes Street, Suite 211
Raleigh NC 27604
5645 Normanshire Drive
Raleigh, NC 27606
[phone]
Subject: Elk Shoals Stream Restoration Project - Stability Assessment of Remnant Channel and
Headcut Evaluation of Unnamed Tributary (UT) 1.
Dear Mr. Creech:
On May 25, 2010, members of the Inter-Agency Review Team (IRT) and the NC Ecosystem
Enhancement Program (EEP) met with staff from Restoration Systems (RS) and Michael Baker
Engineering, Inc. The meeting occurred on site and was for the purpose of closing out the
project. During the site visit, concerns were raised by the IRT about the stability of the remnant
channel due to multiple locations where bank erosion was observed. In addition, a small headcut
was observed on UT 1. Concerns were raised about long term bed stability as a result of the
headcut. Specific questions from the IRT expressed to RS include:
1. Is the bank erosion in the remnant channel caused by localized influences or is it symptomatic
of system wide instability?
2. Will the bank erosion problems in the remnant channel cause stability problems downstream
of the remnant channel?
3. What will be done to repair the headcut problem at UT 1?
Since I was involved in the assessment and design of the Elk Shoals project, RS has asked me to
provide comments / answers to these questions. Worth Creech and I visited the site on June 16,
2010 to further investigate these questions. The following is my response based on that site visit
and a review of the restoration plan and Stn year monitoring report.
Bank Erosion in Remnant Channel: Local or System Wide Instability
During the June 16`x' site visit, observations and measurements were made to determine if the
remnant channel was vertically and laterally stable at a reach / system scale. System wide
vertically instability would be caused by headcutting and subsequent bed degradation. This
concern was addressed during the design by including constructed riffles at stations 24+00 and
32+50. Both of these constructed riffles were in place and functioning during the site visit. In
addition, there were no observed headcuts anywhere in the remnant channel. Cross Section 3
does show a lower bed elevation in the Year 5 (2009) monitoring report; however, the overall
grade was held by the downstream constructed riffle and no headcuts were observed upstream of
the cross section. This degradation is likely short term and is due to recent floods. The stream
type for this reach is an E5 and the beds of these channels typically fluctuate between wet and
dry periods, more so than gravel bed streams. Based on these observations, the overall profile of
the remnant channel is stable.
To evaluate the bank erosion areas for local versus system wide stability, the pattern of erosion
and the cause of the erosion were investigated. System wide lateral instability is often indicated
by the formation of alternating point bars. This most often occurs in straight, incised channels; as
the bank on one side of the stream begins to erode, a point forms on the opposite side. As this
happens, the downstream riffle begins to form at an angle to the valley's fall line (often around
45 degrees). This redirects the velocity vectors towards the opposite bank, causing additional
bank erosion and another point bar (thus the term alternating point bars). This is how channelized
streams work to create a new belt width and to re-establish a sinuous pattern. During the June
16th field visit, we looked for signs of alternating point bars and areas where the stream would be
trying to re-establish a new belt width. In addition, we looked for riffle areas where both banks
were eroding and the bankfull width to depth ratio was increasing and mid channel bars were
forming. This would indicate that the channel was widening. Neither scenario was observed.
There were no areas in the remnant channel where alternating point bars were forming or riffles
that had high W/D ratios or mid channel bars. Based on these observations, there is no reason to
believe that the remnant channel has system wide vertical or lateral stability problems.
However, localized bank erosion was observed. The worst area was the left meander bend at
approximately station 22+00. Photos of this bank are shown below in Figure 1.
The erosion rate for this area was assessed using the Bank Erosion Hazard Index (BEHI) and
Near Bank Stress (NBS) assessment. A detailed description of this method is provided in
Watershed Assessment of River Stability and Sediment Supply (WARSSS) by Dave Rosgen and
is available on the EPA web page at http://www.ega.g_ov/warsss/. Results show that the BEHI
score was High (36.1) and the NBS was also High. The NBS score was assessed using the radius
of curvature ratio method. Using the Colorado curve, the erosion rate was 0.6 feet/year. More
important than the erosion rate, however, is the cause of the erosion and the potential for self
Page 12
Figure la. Looking Downstream
Figure lb. Looking upstream
recovery. Figure Ib shows a large debris jam that is located on the downstream "head of riffle."
This debris was deposited during a large storm and created a flow blockage. It is likely that
during the large event, a back eddy formed along the outside of the meander bend. Currently,
there is not enough woody vegetation along the streambank and the bank began to erode. Given
the flow dynamics and the tight radius of curvature, it is also likely that this bend will continue to
erode for a significant period of time before it begins to heal. Therefore, this bank should be
repaired by removing the debris jam and stabilizing the banks with root wads or toe wood and
transplants.
During the field visit, Worth pointed out another area of concern. This area was located at
approximately station 24+30 and Cross Section 4. Photos of this bank are shown in Figure 2.
Results indicate that the BEHI score was Moderate (22.7) and the NBS score was also Moderate
using the radius of curvature ratio method. The erosion rate using the CO curve was 0.25 feet per
year. Unlike the bend at station 22+00, the apex of this bank is protected by two large trees,
which provide bank protection through root depth, density and overall surface protection.
Previous erosion was observed near the upper and lower portion of the bend, around the root
wads. The upper portion was likely caused by the upstream constructed riffle. This riffle is fairly
steep and points the velocity vectors into the bank. However, the bank did not show signs of
fresh erosion even though there had been recent flooding. The lower bank showed that erosion
had occurred in prior years, probably soon after the root wad was installed. This root wad was
probably installed too far into the channel and it truncated the radius of curvature. This is how
most root wads were installed at this time. Subsequent storms removed the fill over the root wad
and essentially re-established the appropriate radius of curvature. This part of the bank is
currently stable and has shrubs growing along the edge. The root wad essentially serves as a
cover log now. Given that the BEHI score is at the lower end on the Moderate rating (break
between Low and Moderate is 20) and that the bank seems to be healing or trending towards
greater stability not instability, I would not recommend additional bank treatment. The toe of the
bank is stable and it is providing good habitat. Further evidence for this is shown in Cross
Section 4. Pool depth has fluctuated from year to year based on weather patterns; however, the
left bank has not migrated. This is due to the large trees along the apex of the left bank where the
cross section was taken.
Page 13
Figure 2a. Looking Downstream
Figure 2b. Lower part of bank with root wad
There were other areas of minor bank erosion and long stretches without any erosion. Examples
are shown below in Figure 3. A BEHI and NBS assessment was performed on the bank shown
in Figure 3a to represent these areas. There were probably 4 to 6 of these areas in the reach.
Figure 3a. Minor bank erosion at 21+50 Figure 3b. No bank erosion
Results from the BEHI assessment show a score of High (35.2) and a NBS score of Very Low
for a total erosion rate of 0.15 feet per year, which had to be extrapolated on the CO curve
because the data range begins at a rating of Low. These results are important because banks in
E5 channels are vertical (low W/D) and sometimes will lose vegetation. When this happens, the
BEHI score increases. However, if the NBS value remains low, very little erosion occurs and the
bank heals on its own over time.
Affect of Remnant Reach Erosion on Downstream Reach
With the exception of the bank erosion at station 22+00, the remnant channel is functioning like
most E5 reference reaches. There is no indication of vertical instability and the constructed
riffles are working as designed to prevent future degradation. The lateral instability is caused by
localized problems, like debris jams and lack of woody vegetation. Other bank erosion areas are
minor and consistent with E5 reference reaches. Therefore, there is no risk of the remnant reach
causing stability problems to downstream (or upstream) reaches.
Headcut Repair in UT 1
The headcut at approximately station 15+00 is directly upstream of a cross vane and is shown in
Figure 4. The headcut is approximately one foot deep and appears to have formed when a piping
failure occurred along the right side of the cross vane. This headcut must be repaired to prevent
further bed degradation. If left un-repaired, the best case scenario is that the headcut will advance
to the upstream pool and stop, making the pool shallower. The worst case scenario is that the
headcut advances through the entire reach, creating an incised channel. The most likely scenario
is that it would only headcut through one or two pools. However, the repair would be minor
compared to the benefits gained. There are two options for the repair.
Page 14
Option one is to repair the right arm of the cross vane to prevent the piping. If the invert of the
cross vane is at the same elevation or higher than the top of the headcut, the headcut will be
under backwater conditions and should aggrade back to the original bed elevation. If not, option
two is to fix the right arm of the cross vane and add a constructed riffle extending upstream past
the headcut.
Large Woody Debris
During the site visit, Worth mentioned that there was some concern by the IRT about a tree that
had fallen into the channel. A photo of the tree is shown below in Figure 5.
I assume that the concern is that this debris could cause bank erosion. That is true; however, the
wood is located in a fairly straight section and there are mature trees on both banks. Therefore,
future bank erosion would likely be mild to moderate. In my opinion, the bedform complexity
and habitat diversity provided by the large woody debris exceeds the risk of instability. This is a
very natural part of how E5 streams function. Nevertheless, if stability is a concern, the debris
Page 15
Figure 4. Headcut at station 15+00 on UT I
Figure 5. Large Woody Debris
could be removed by hand with a chainsaw and manual labor, which would prevent damage to
the channel and buffer.
Beaver Dam
Again during the site visit, we observed a new beaver dam on the grade control J-hook at station
34+50. The beaver dam is backing up water into the remnant channel, which is why the water
surface looks flat in figure 3b. A picture of the dam is shown below in Figure 6.
This is a relatively good location for a beaver dam. The section immediately upstream of the dam
is a cross over reach, which connects the remnant channel with the new channel alignment. The
old channel, which has oxbow ponds, joins the main channel in this reach. The bank height ratios
through this section are very good (1.0) and the connection to the oxbow provides a stream /
wetland complex. The bed will aggrade while the beaver dam is in place, but since there is grade
control underneath the dam and upstream, at the end of the remnant channel, the bed will re-
establish itself to the original grade once the beaver dam washes out during natural flood events.
Past monitoring of other projects have shown this same process. It appears that the key to
maintaining stability with beavers is to have bank height ratios near 1.0 and vegetation that is
well established. Therefore, beavers that establish in stream restoration projects in years 1-3
often cause problems, but beavers that establish later do not. However, if the IRT disagrees with
this or is uncomfortable with this approach, it would not be too destructive to the buffer to
remove the dam when the other repairs are made.
Conclusions and Recommendations
The following conclusions and recommendations are provided based on the above discussion.
1. There are no signs of system wide instability in the remnant channel, vertical or lateral.
2. There is no risk that localized bank erosion in the remnant channel will threaten downstream
or upstream project reaches.
3. Bank erosion at station 22+00 should be repaired and the woody debris should be removed.
4. The headcut in UT 1 must be repaired.
5. Removing the large woody debris from the straight section of the remnant channel is optional.
Page 16
Figure 6a. Beaver Dam at 34+50
Figure 6b. Looking upstream of beaver dam
6. Removing the beaver dam is optional.
Please let me know if you have further questions.
Sincerely,
Will Harman, PG
Principal
Page 7