Mapping Watershed-scale
Streambank Erosion Using EPABANCS for Determining TMDLs
Paul Ayers, K. Hensley, D. Wade
Biosystems Engineering & Soil Science
University of Tennessee
K. Swinson (Geosyntec), B. Connell (Trutta)
University of Tennessee Alum
Sediment in Streams
• Where does the sediment come from
– upland erosion?
– in stream erosion?
• How much sediment comes from these
sources?
• Some sediment
TMDLs ignore
streambank erosion
Research Objectives
To develop and demonstrate a rapid watershed-scale data
collection method to predict total daily sediment load (TDSL)
and total maximum daily load (TMDL) for streambank erosion
• Implement Streambank Video Mapping System (SVMS)
– Map streambank condition
• Utilize Bank Assessment for Non-point source Consequences of
Sediment (EPA BANCS) method
– Evaluate erodibility and erosivity
– Predict Bank Erosion Rate (BER)
• Utilize flow data
– Predict TDSLs and develop TMDLs
Driftwood River
Assessment
• Located in Indiana,
adjacent to Camp
Atterbury
• Funded by Department of
Defense Legacy Program
(through Army COE
CERL)
• Look at stream bank
erodibility
• 12 mile section mapped
Flow
WATERPROOF CASE
SONAR TRANSDUCER
LASERS
VIDEO CAMERAS
Streambank Video
Mapping System (SVMS)




GPS Position
Depth
Width
Video
 streambank condition
Mapping In Progress
EPA - Watershed Assessment of River Stability &
Sediment Supply (WARSSS)
• Bank Assessment for Non-point source Consequences of
Sediment (BANCS) model
• BANCS – predicts annual streambank rescission (in/yr) at a
sample location (Bank Erosion Rate)
• BER = f (BEHI, NBS)
• BEHI – Bank Erosion Hazard Index
• NBS- Near-Bank Stress = f(RC/W, etc)
• Estimate site-specific annual and total daily streambank
erosion (tons/feet of streambank)
• Determine areas of high streambank erosion rates
• TMDL determinations
SVMS
Ocular
Assessment
Rc/W
BEHI*
NBS
(erodibility)
(erosivity)
Implementing
EPA BANCS
Method
BER
USGS flow
data
TDSL
SVMS
Ocular
Assessment
Rc/W
BEHI*
NBS
(erodibility)
(erosivity)
Implementing
EPA BANCS
Method
BER
USGS flow
data
TDSL
Modified Bank Erosion Hazard Index
(BEHI*)
Modified
Bank Erosion
Hazard
Rating
Value
Low
Index
Moderate
Value
Index
High
Value
Index
Very
High
Extreme
Value
Index
BankHeight
to Bankfull
Height
(Ratio)
1.0-1.19
2.45
1.2-1.5
4.95
1.6-2.0
6.95
> 2.1
9
Riparian
Diversity
Bank
Surface
x2
Angle
Protection Index
(%)
(Degrees)
( %)
Totals
Optimal
0-60
55-100
4.9
2.45
2.45
12.2–23.2
Sub Opt
61-80
30-54
9.9
4.95
4.95
23.2–30.5
Marginal
81-90
15-29
13.9
6.95
6.95
30.5–38.0
Poor
> 91
< 14
18
9
9
38.0–45.0
Based on Connell, 2012
BEHI* Score = 38.85
Rating = Very High
Bank Angle = 6.95, BH/BF Ratio = 6.95, Surface Protection = 18, Riparian Diversity = 6.95
Bank Angle (deg) Bank Height (ft)
0 - 60 61 - 80 81 - 90
=2.45 =4.95 =6.95
14:39
> 91
=9
Surface Protection
(Avg. %)
Riparian Diversity
100-56 55-30 29-15
0-1ft 1ft-3ft 3ft - 6ft 6ft-9ft 9ft-12ft 12ft-18ft =2.45 =4.95 =6.95
14:39
< 14
=9
14:39
Optimal Sub Opt Marginal Poor
=4.9
=9.9
=13.9
=18
14:39
BEHI* Total
12.3 – 23.2
23.2 – 30.5
30.5 – 38.0
38.0 – 45.0
Low
Moderate
High
Very High
Georeferenced
Data in ArcGIS
SVMS
Ocular
Assessment
Rc/W
BEHI*
NBS
(erodibility)
(erosivity)
Implementing
EPA BANCS
Method
BER
USGS flow
data
TDSL
Near Bank Stress (NBS)
Determining NBS from
Rc/W:
• Rc = Radius of curvature
• W = River width
• NBS
– Categories very low to
extreme
– Values 1 to 6
NBS
Description
NBS
Value
Rc/W
Very Low
1
> 3.0
Low
2
2.21 - 3.0
Moderate
3
2.01 – 2.2
High
4
1.81 – 2.0
Very High
5
1.5 – 1.8
Extreme
6
< 1.5
WARSS, www.epa.gov/warsss
Stream Erosivity (RC/W)
SVMS
Ocular
Assessment
Rc/W
BEHI*
NBS
(erodibility)
(erosivity)
Implementing
EPA BANCS
Method
BER
USGS gage
data
TDSL
North Carolina Piedmont Region
Bank Erosion Prediction Curve
http://www.bae.ncsu.edu/programs/extension/wqg/srp/
Predicting
Total Daily
Sediment
Load
(TDSL)
SVMS
Ocular
Assessment
Rc/W
BEHI*
NBS
(erodibility)
(erosivity)
BER
USGS flow
data
TDSL
USGS Gage Station
03363000
Average Daily Flow (cfs) for
Driftwood River (1942 - 1991)
4000
3500
Flow Rate (cfs)
3000
2500
2000
1500
1000
500
0
1
21
41
61
81
101 121 141 161 181 201 221 241 261 281 301 321 341 361
Julian Day
http://waterdata.usgs.gov/nwis/rt
USGS Stream Gage Data
Estimating:
• Lateral Erosion Rate = f(Q 0.6)
– Q = flow rate
– Based on a power factor (Evans et al., 2003)
• Erosion Potential
– LERi / ∑i LER
Julian
Day
Month
Number
Day
Number Begin Yr
End Yr
Mean
Flow
1
1
1
2
1
3
1
LER
Erosion
Potential
% Erosion
Potential
1969
2011
662
49.2623
0.0037
0.3693
2
1969
2011
578
45.4105
0.0034
0.3405
3
1969
2011
550
44.0776
0.0033
0.3305
http://waterdata.usgs.gov/nwis/rt
Flow Erosion Potential vs. Flow Rate
% Daily Flow Erosion Potential
0.6
0.5
0.4
0.3
0.2
0.1
0
0
500
1000
1500
2000
2500
Flow Rate (cfs)
3000
3500
4000
Predicting
Total Daily
Sediment
Load
(TDSL)
SVMS
Ocular
Assessment
Rc/W
BEHI*
NBS
(erodibility)
(erosivity)
BER
USGS flow
data
From site-specific
BER and erosion
potential based on
USGS flow data
TDSL
Predicting TDSL
Estimating Sediment Loads:
• Site-specific daily erosion rate (DER)
– BER (ft/yr) x % erosion potential
• Site-specific daily sediment load (DSL)
– DER x bank height x bank length x soil density
• Total daily sediment load (TDSL)
– ∑ Site-specific DSL
Storm Event on the Driftwood River
1800
Flow Rate (cfs)
1600
1400
1200
1000
800
600
30-May 31-May 1-Jun
2-Jun
3-Jun
4-Jun
Day
5-Jun
6-Jun
7-Jun
8-Jun
9-Jun
Flow
Storm Event on the Driftwood River
1800
20
18
16
14
1400
12
1200
10
8
1000
6
TDSL (tons/day)
Flow Rate (cfs)
1600
4
800
2
600
30-May 31-May 1-Jun
2-Jun
3-Jun
4-Jun
Day
5-Jun
6-Jun
7-Jun
8-Jun
0
9-Jun
Flow
TDSL
SVMS
Streambank
Ocular
Assessment
Rc/W
BEHI*
NBS
(erodibility)
(erosivity)
BER
Bank erosion
rate
USGS flow
data
Developing
Sediment
Total
Maximum
Daily Load
(TMDL) using
EPA BANCS
TDSL
Total daily
sediment load
TMDL
For Sediment
Developing TMDLs
Using Equation
• Long Term Average (LTA) = µ
– Mean TDSL over a desired time period
• Coefficient of variation (CV) = standard deviation
µ (mean)
• σ2 = ln(CV2+1)
• Recurring interval: 365 days, z score = 2.778
• Maximum Daily Load (MDL) = LTA * e (zσ – 0.5σ^2)
An approach for using load duration curves in developing TMDLs (EPA, 2006)
Developing TMDLs
Driftwood River
• Long Term Average = 13 tons/day
– Approximately 980 cfs
• SDev= 5.57 tons/day
• CV = 0.369
• 12mi Reach MDL
– 70 tons/day
• MDL flow rate
– Approximately 15,200cfs
– Occurs <1% of the time over a 49yr period
Driftwood River Peak Flow Rate
Benefits of Streambank Video Mapping
•
•
•
•
•
•
•
•
•
Integrate with the EPA BANCS method
Predict erodibility and erosivity
Estimate total annual or daily streambank erosion
Develop TMDL utilizing EPA protocol
Evaluate restoration impacts
Cover large areas (10 miles/day) - continuous
Non-intrusive, no access required
Permanent historical video database
The advantage of video mapping every foot of stream and bank
is that 1) the total daily streambank erosion can be
determined, and 2) the locations of high erosion can be
identified and managed
Questions?
Contact:
Paul Ayers
Biosystems Engineering
University of Tennessee
[email protected]