THE INTEGRATION OF BASIN,
STREAMFLOW AND CHANNEL
CHARACTERISTICS FOR
CHANNEL CONDITION
ANALYSES
John F. Orsborn
Alan W. Johnson
Mack T. Orsborn
Prepared for:
EPA - Region 10
Surface Water Branch
Seattle, WA
Stephen C. Ralph
EPA Contract Officer
Grant No. X990717
April 2001
-------
ERRATA
6/13/01
For the EPA Report:
THE INTEGRATION OF BASIN, STREAMFLOW AND CHANNEL
CHARACTERISTICS FOR CHANNEL CONDITION ANALYSES
Orsbom, Johnson & Orsborn. 2001.
Page „. ^ .; Line As Is
Figure No.
Change To
2-18
2-18
2-18
3-1
3-15
3-36
3-36
3-47
4-2
4-6
4-8
4-19
4-22
4-34
5-8
3
4
5
2
Figure 3-9
Figure 3-19
Figure 3-19
Figure 3-25
6
Under
Summary
1st para.,
7th bullet
9th bullet,
2nd entry
Figure 4- 10
3rd
Reference
4
2
10
1
4
5
8
1
3
geometry
at-a-station;
the graphs
a series of example
models were developed
X-scale:
Ab(in/yr • mi2)
Title:
for P*A
X-scale:
P»A
Title:
versus PBE
bank full width
with the South fork.
Energy (A) (H)05
its plan view
region model
Title:
Orsborn 1999
M.T. Orsborn 2000.
geometries
at-a-station.
these graphs
a series was developed
of example models
X-scale:
Ah(mi2)
Title:
P'Ab
X-scale:
P»Ah
Title:
P-Ah
bankfull width
with the South Fork.
Energy (A) (H)aso -
their plan views
regional model
Title:
Orsborn & Orsborn 1999
M.T. Orsborn. 2000.
-------
EPA Channel Condition Project
Acknowledgments
This pilot study was funded through a grant from the Environmental Protection
Agency, Region 10, to Washington Trout, which administered the grant. In-kind,
professional participation was provided by King County, through Jeanne M.
Stypula, Senior Engineer in the Rivers Section of the Department of Natural
Resources. Jeanne provided literature research, key references, editorial advice,
meeting facilities and a practical perspective.
A Note to the Readers of this Report
If you are a person who is interested in the basic concepts and the results of this
study, but not the development of equations, we suggest that you bypass Part 3-
Methods of Analysis from pages 3-1 through 3-70. The Summary of Part 3 is on
pages 3-71 through 3-74, and should be read to fit with the rest of the report.
-------
EPA Channel Condition Project
TABLE OF CONTENTS
Acknowledgments i
A Note to the Readers of this Report i
LIST OF FIGURES iv
LIST OF TABLES ....vi
1. INTRODUCTION 1-1
Problem Definition 1-1
Channel Condition Studies 1-2
Fundamentals 1-2
Definitions 1-2
Methods of Analysis 1-4
Applications-Case Studies 1-5
Summary and Conclusions 1-5
References for Part 1 1-7
2. FUNDAMENTALS 2-1
State of Our Current Knowledge about Width Adjustment 2-1
Procedure for Approaching Width-Adjustment 2-2
Overview of Basin, Streamflow and Channel Characteristic Models 2-3
Dimensional Analysis of the Basin, Flow and Channel Characteristics 2-7
Analysis of the Basin 2-7
Analysis of the Channel 2-11
Channel Hydraulic Geometry 2-17
A Severity Factor Analysis to Determine the Influence of Flow Reduction on Channel
Characteristics 2-20
Regional Relationships between Basin Characteristics (BC) and Channel
Characteristics (CC) Using Flow Characteristics (QC) 2-23
Accounting for Changes in Channel Geometry 2-27
An Example for Evaluating Effects of Land Use Change on Channel Geometry 2-29
References for Part 2 2-33
3. METHODS OF ANALYSIS 3-1
Introduction 3-1
General Analytical Methods 3-1
OLYMPIC PENINSULA REGION 3-2
Width, Depth and Channel Area at Q1F2 3-11
Combined Relationships of Channel and Basin Characteristics for the Olympic
Peninsula 3-16
FLOOD FLOWS.. 3-16
AVERAGE ANNUAL FLOWS 3-21
7-DAY AVERAGE LOW FLOWS 3-24
Discussion of Olympic Peninsula Empirical and Combined Relationships of
Channel, Flow and Basin Characteristics 3-28
PUGET LOWLAND REGION 3-31
Database and Empirical Relationships 3-31
Regional Hydraulic Geometry 3-39
NORTHEASTERN WASHINGTON REGIONAL STREAMS .......3-50
11
-------
EPA Channel Condition Project
Database and Empirical Relationships 3-50
Width and Channel Area at QAA 3-50
Combined Relationships of Channel and Basin Characteristics in Northeastern
Washington 3-54
Discussion of NE Washington Results 3-66
Summary Comparisons of Regional Analyses 3-71
References for Part 3 3-75
4. APPLICATIONS 4-1
Introduction 4-1
Reconnaissance 4-1
Restoration 4-2
Reconstruction 4-3
CASE STUDIES 4-5
CASE STUDY 1. HABITAT IMPROVEMENT PROJECTS IN LOWER LEBAR
CREEK BASIN 4-5
CASE STUDY 2. PLANNING AND DESIGN FOR THE RECONSTRUCTION
OF A GOLD-DREDGED STREAM, Crooked River (Idaho) Habitat Improvement
Project 4-17
CASE STUDY 3. EVALUATION OF LAND USE IMPACTS ON BIG BEEF
CREEK 4-26
CASE STUDY 4. LOWER ELWHA RIVER LOW FLOW RECONNAISSANCE
STUDY 4-32
Comparative Notes on the Four Example Projects 4-40
5. SUMMARY AND CONCLUSIONS 5-1
Natural Stream Reaches 5-4
Altered Stream Reaches 5-4
References for Part 5 5-8
6. NOTATION 6-1
7. INDEX OF TERMS AND AUTHORS 7-1
111
-------
EPA Channel Condition Project
LIST OF FIGURES
Tab-Page
Figure 2-2a. Siuslaw National Forest Basin Energy Models (Orsbom 1981) 2-9
Figure 2-2b. Detail of Low Flows for Siuslaw National Forest Basin Energy Models (Orsborn 1981) 2-10
Figure 2-3. Shear-Shape relationships for Natural and Rectangular Stream Channels 2-13
Figure 2-4. General Shear-Shape Relationships for Rectangular and Non-Rectangular (Natural)
Channels (Orsborn and Orsbom 1999a) 2-16
Figure 2-5a. San Foil River in Northeastern Washington (SPR047). Number of Points Test for 2-19
Figure 2-5b. San Foil River in Northeastern Washington (SPR047). Number of Points Test for 2-19
Figure 2-6. Multiple Severity Factor (XSF5) for Flows Less than Bankfull for Triangular,
Trapezoidal and Rectangular Idealized channels. Data In Table 2-2. (Orsborn and Deane
1976) 2-22
Figure 2-8. Cross-Sections for CCT Priority Low Flow Study Sites, Wilmont Creek Site WIL028
(Orsborn and Orsborn 1999c) 2-25
Figure 2-9. Sketch of Example Basin with 5% (1 sq. mi.) clear-cut, (not to scale) 2-29
Figure 3-1. Regional Hydraulic Geometry: Width, Velocity and Depth Versus the Two-Year,
One-Day Average Flood Flows for Olympic Peninsula Streams 3-6
Figure 3-2. Regional Hydraulic Geometry: Width, Velocity and Depth Versus Average Annual
Flows for Olympic Peninsula Streams 3-7
Figure 3-3. Regional Hydraulic Geometry: Width, Velocity and Depth Versus the Two-Year,
Seven-Day Average Low Flows for Olympic Peninsula Streams 3-8
Figure 3-4. Regional Hydraulic Geometry: Cross-Sectional Area Versus Q7L2, QAA and Q1F2
for Olympic Peninsula Streams 3-9
Figure 3-5. USGS Stream Gaging Stations on the Olympic Peninsula 3-10
Figure 3-6. Channel Characteristics versus PAb for Olympic Peninsula Streams at Q1F2 3-12
Figure 3-7. Channel Characteristics versus PAb for Olympic Peninsula Streams at QAA 3-13
Figure 3-8. Channel Characteristics versus PAb for Olympic Peninsula Streams at Q7L2 3-14
Figure 3-9. Channel Width versus Ab for Olympic Peninsula Streams at Q7L2 3-15
Figure 3-10. Q1F2 versus Ab for Olympic Peninsula Streams 3-17
Figure 3-11. Q1F2 versus PAb for Olympic Peninsula Streams 3-18
Figure 3-12. Q1F2 Predictions versus USGS Values for Width, Depth and Channel Area for
Olympic Peninsula Streams 3-20
Figure 3-13. QAA Predictions versus USGS Values for Width, Depth and Channel Area for
Olympic Peninsula Streams 3-23
Figure 3-14. Q7L2 versus PBE for Olympic Peninsula Streams 3-25
Figure 3-15. Q7L2 Predictions versus USGS Hydraulic Geometry Values for Width, Depth and
Channel Area for Olympic Peninsula Streams 3-27
Figure 3-16. North Creek Regional Analysis: Channel Area at Q1F2 vs. Basin Area (All Points) 3-33
Figure 3-17. North Creek Regional Analysis: Channel Area at Q1F2 vs. Basin Area (Selected
Points) 3-34
Figure 3-18. North Creek Regional Analysis: Channel Area at QAA vs. Basin Area (Selected
Points) 3-35
Figure 3-19. North Creek Regional Analysis, Channel Area at QAA for P« A 3-36
Figure 3-20. North Creek Regional Analysis of Depth at Q1F2 3-37
Figure 3-21. North Creek Regional Analysis of Channel Width at Bankfull Flow (Q1F2) 3-38
Figure 3-22. Q1F2 versus PAb for Eight Puget Lowland USGS Gages for their Periods of Record
(Data from Table 3-7) 3-40
IV
-------
EPA Channel Condition Project
Figure 3-23. Channel Width, Depth and Area Estimations versus Hydraulic Geometry Values at
Q1F2 for Puget Lowland Streams •. 3-43
Figure 3-24. Channel Width, Depth and Area Estimations versus Hydraulic Geometry Values at
QAA for Puget Lowland Streams 3-46
Figure 3-25. Q7L2 versus PBE for Eight Puget Lowland USGS Stations for Their Periods of
Record (Data from Table 3-7) 3-47
Figure 3-26. Channel Width, Depth and Area Estimations versus Hydraulic Geometry Values at
Q7L2 for Puget Lowland Streams 3-49
Figure 3-27. W and Ac versus Ab at QAA in NE Washington 3-52
Figure 3-28. W and Ac versus Basin Energy at QAA in NE Washington 3-53
Figure 3-29. Regional Models of Width, Depth, Velocity and Channel Area Related to Q1F2 at
USGS Stations in NE Washington.... 3-57
Figure 3-30. Regional Models of Width, Depth, Velocity and Channel Area Related to QAA at
USGS Stations in NE Washington..... 3-58
Figure 3-31. Regional Models of Width, Depth, Velocity and Channel Area Related to Q7L2 at
USGS Stations in NE Washington 3-59
Figure 3-32. Q1F2 as a function of Basin Energy for Selected USGS Stations in NE Washington 3-61
Figure 3-33. Channel Width, Depth and Area Estimations versus Hydraulic Geometry Values at
Q1F2 for NE Washington 3-63
Figure 3-34. Channel Width, Depth and Area Estimations versus Hydraulic Geometry Values at
QAA for NE Washington 3-65
Figure 3-35. Q7L2 versus Annual Precipitation times Basin Energy (PBE) for Four Northeast
Washington USGS Gages 3-68
Figure 3-36. Channel Width, Depth and Area Estimations versus Hydraulic Geometry Values at
Q7L2 for NE Washington 3-70
Figure 4-1. Location Map for the LeBar Creek Project (Not to Scale) 4-5
Figure 4-2. Project Basin and Stream Map for LeBar Creek 4-9
Figure 4-3. LeBar Creek Habitat Project Map and Baselines , 4-15
Figure 4-4. LeBar Creek Habitat Project Details of HIU 4 Modifications 4-16
Figure 4-5. Crooked River Study Site Plan and Regional Location 4-17
Figure 4-6. Staff Gage Locations 4-23
Figure 4-7. Bankfull Flow Area (AB), Top Width (WB), and Mean Depth (DB), Related to BankfuU
Flow (QB) in the Crooked River Study Region 4-24
Figure 4-8. Location of USGS surface-water stations in the Hood Canal Watershed (USGS 1995) 4-26
Figure 4-9. Location Map for the Lower Elwha Project (from DNAI1994) 4-33
Figure 4-10. Schematic Representation of Measurement Sites and Facilities, Lower Elwha Low
Flow Study 4-34
Figure 4-11. August to November, 1938 Hydrograph, Recession Flows (Elwha Gage #12045500 &
NF Skokomish Gage #12056500) 4-36
Figure 4-12. August to November, 1962 Hydrograph, Recession Flows (Elwha Gage #12045500 &
NF Skokomish Gage #12056500) 4-36
-------
EPA Channel Condition Project
LIST OF TABLES
Tab-Page
Table 2-2. Measured and Modeled Values of Average Annual Flow, Width, Depth and Velocity
for Deer, Fall and Flynn Creeks in the Oregon Mid-coast Region 2-15
Table 2-3. Components of Severity Factor Analysis of Dimensionless Ratios for Triangular,
Trapezoidal and Rectangular Channels (Orsborn and Deane 1976) 2-21
Table 2-4. Basin Characteristics and Peak Flows for Figure 2-9 under Pre-Logging Conditions 2-30
Table 2-5. Comparison of Pre- and Post-Logging Peak Flows at the Check Points 2-30
Table 2-6. Natural and Post-Logging Channel Widths for Estimated Average Daily Flood
Conditions 2-31
Table 3-1. Basin characteristics for Gaging Stations on the Olympic Peninsula ..3-3
Table 3-2. Calculated Values of At-a-Station Hydraulic Geometry for Three characteristic Flows
at USGS Gaging Stations on the Olympic Peninsula 3-4
Table 3-3. Channel Properties at Q1F2 (Average Flood), QAA (Average Annual Flow) and Q7L2
(Average Low Flow) for Olympic Peninsula USGS Gaging Stations, including W/D Values 3-5
Table 3-4. Channel Width, Depth and Area Comparison at Q1F2 for Olympic Peninsula Streams...3-19
Table 3-5. Channel Width, Depth and Area Comparison at QAA for Olympic Peninsula Streams ...3-22
Table 3-6. Channel Width, Depth and Area Comparison at Q7L2 for Olympic Peninsula Streams...3-26
Table 3-7. Basic Streamflow, Channel and Basin Data for the Puget Lowland Region 3-32
Table 3-8. Channel Width, Depth, and Area Comparison at Q1F2 for Puget Lowlands 3-42
Table 3-9. Channel Width, Depth, and Area Comparison at QAA for Puget Lowlands 3-45
Table 3-10. Channel Width, Depth, and Area Comparison at Q7L2 for Puget Lowlands 3-48
Table 3-11. USGS Stations, Basin Area, Basin Energy, and Channel Width (W) and Area (Ac) for
Developing CCBC Preliminary Models at Average Annual Flow (QAA) in NE Washington ..3-51
Table 3-12. At-a-Station Channel Geometry Summary: USGS Regional Stations, Northeastern
Washington 3-55
Table 3-13. Characteristic Flows for NE Washington USGS Gages 3-56
Table 3-14. Data from Table 3-9, BASIN CHARACTERISTICS (Orsborn & Orsborn, 1997) 3-61
Table 3-15. Channel Width, Depth and Area Comparison at Q1F2 for NE Washington 3-62
Table 3-16. Channel Width, Depth and Area Comparison at QAA for NE Washington 3-64
Table 3-17. Summary of characteristic Seasonal Q7L2 Low Flows (Orsborn & Orsborn, 1999) 3-67
Table 3-18. Channel Width, Depth and Area Comparison at Q7L2 for NE Washington 3-69
Table 3-19. Comparison of Three Regional Sets of HYDRAULIC GEOMETRY Equations for
Three Characteristic Flows 3-71
Table 3-20. Ranges of Flows and Average Annual Basin Precipitation (P) in the Three Regions of
Washington Used in Regional Models of Hydraulic Geometry 3-72
Table 3-21. Comparison of Maximum and Minimum Unit Values in cfs per square mile for Q1F2,
QAA and Q7L2 in the Three Regions for the Ranges of Flow in Table 3-20 3-73
Table 3-22. Comparison of Regional Equations for Estimating CHARACTERISTIC FLOWS in the
Three Regions (Streamflow Equations) 3-73
Table 4-1. Geomorphic Characteristics of LeBar Creek Basin 4-8
Table 4-2. Percent of LeBar Creek Basin Logged, and Estimated Annual Miles of Road
Constructed (Based on % cut) 4-10
Table 4-3. LeBar Creek Basin Threshold Rating and Their Severity Compared to the Standard
Thresholds Listed Above 4-12
Table 4-4. Conditions and Explanation of Terms (see text for details), Crooked River habitat
Improvements— Alternatives Matrix 4-20
Table 4-5. Sample Analysis— Use with Table 1— Explanation of Terms 4-21
Table 4-6. Predicted and Actual Dimensions of the Big Beef Creek Channel at Station 50+50 4-29
VI
-------
EPA Channel Condition Project
Table 4-7, Modeled Hydraulic Geometry Parameters for Various Flows at Site Al and Different
Rates of Diversion , 4-39
Table 4-8. Comparative Emphasis on Tasks and Objectives for Four Example Projects (0-10 High
in Relative Amount of Activity, or Importance to Each Project) ,.,4-40
vn
-------
EPA Channel Condition Project
1. INTRODUCTION
Problem Definition
With the recent listings of chinook salmon, bull trout, and other sahnonids under
the Endangered Species Act, the quality of streams and fish habitat has become a
primary concern in the Pacific Northwest. In an effort to increase the survival of
these listed salmonids, resource managers have accelerated extensive (and often
expensive) programs to restore aquatic habitat degraded by various land use
activities. Often these efforts take place without benefit of a template of stream
channel conditions to target conditions the restoration plans attempt to emulate.
Natural resource management agencies and regulators need some reliable means
to evaluate the status and trends in the physical condition of stream channels and
associated aquatic habitats. Given the dynamic nature of channel form, it can be
difficult to distinguish natural variability in watershed processes from those
changes associated with human activities. The purpose of this project was to
evaluate the concept of "regional indices of channel morphology" for typical
stream types found in Washington, and to determine if they can provide a useful
diagnostic and predictive tool to help evaluate existing and potential channel
characteristics.
The negative influence of various land use activities on the hydrology and
geomorphology of streams have been extensively investigated and documented
(Hammer, 1972; Leopold, 1973; Arnold et al, 1982; Booth, 1990). The transition of
a watershed from a natural to an altered state includes removing vegetation,
compacting soils, creating impervious surfaces, and altering natural drainage
networks. These actions change fundamental watershed processes that control
the rates and distribution of surface water runoff and sediment budgets. An
early Northwest example of these detrimental conditions was the study
completed on Big Beef Creek by Made] (1978). She was able to show how
logging, impoundment and development changed the channel geometry,
increased the sediment load, and contributed to the decline of the coho
population of Big Beef Creek. This thesis is summarized as one of the case
studies in Part 4.
When conducting stream studies of habitat assessment and other water resource
investigations, it is desirable to determine the condition of the stream and the
watershed factors controlling the characteristics of the stream. For example:
• Is the stream in a natural, stable condition (i.e. an appropriate
geomorphic state, as best as we can define that state);
• If the stream is in an "unnatural" condition, how far removed is it from
its natural condition; and
1-1
-------
EPA Channel Condition Project
• What is the potential for returning the stream to its "natural"
condition?
But what is natural? How can we "fix" (determine the essential elements of) a
stream channel and fish habitat if we don't know what is broken? If a stream is
"broken" (disrupted by change), what should it look like?
The objectives of this report are to:
1. discuss methods for measuring and assessing the condition (i.e.
naturalness) of a stream reach;
2. summarize a systematic method for characterizing the existing state
of a stream reach; and
3. provide examples of procedures and models for determining stream
condition in terms of basin, channel characteristics and flow.
Channel Condition Studies
Channel condition studies, when coupled with stream hydrology, lead to the
following applications: the design of bridges and culverts; channel capacities;
flood plain inundation; instream flow analysis and usability of habitat; habitat
modification; upstream fish passage during migration seasons; temperature
effects; availability of rearing habitat in pools and side channels; diversions;
flow reservations; water availability studies, habitat productivity; and water
supply analysis .
Fundamentals
Part 2 of this report covers the following topics and forms the analytical basis of
the study: the state of our current knowledge about stream width adjustment;
an overview of basin, streamflow and channel characteristics; dimensional
analyses of the basins and channels; channel hydraulic geometry; the influences
of flow reductions on channel characteristics; regional relationships between
basin and channel characteristics; and accounting for changes in channel
geometry. We conclude Part 2 with recommended steps for utilizing channel
indices as tools for protection and recovery of stream habitats.
Definitions
The term "state" is synonymous with "condition" and deals with existing
conditions. The existing physical condition of a stream reach is compared to a
baseline or reference set of conditions in other natural stream reaches. The
1-2
-------
EPA Channel Condition Project
compared conditions would include: flow regime and reach slope, which dictate
channel pattern and hydraulic geometry. Taken together these measures define
the site conditions in three dimensions, and show the results of the dynamic
forces involved.
In order to communicate and visualize stream conditions, a form of classification
is helpful. We have chosen to use the illustrations in Rosgen (1996) as that
visualization tool. Although there are some analytical shortcomings in Rosgen
design procedures, the classification system is very useful (Miller and Ritter
1996).
Mackin (1948) introduced the concept of the "graded stream" in which there is a
long-term balance between erosion and deposition. More specifically:
"A graded stream is one in which, over a period of years, slope is delicately
adjusted to provide, with available discharge and the prevailing channel
characteristics, just the velocity required for transportation of all of the load
supplied from above".
Burkham (1981) looked at the uncertainties associated with changes in stream
channel form and quoted Blench's (1957) "in-regime" theory :
"...that average values of the quantities we appreciate as constituting
regime do not show a definite trend over some interval—usually of the
order of a score [twenty] or two of years ... [rivers in regime]
demonstrate themselves to us in the form of varying discharges,
breadths, depths, velocities, meander patterns, sediment contents, and
so forth, but their average behavior does not usually change greatly
over small periods of historic time."
Note that Blench's use of "average" behavior is very similar to Mackin's
definition of a graded stream. But, it seems that Mackin's emphasis that the
slope is "delicately adjusted" by the flow regime is the most revealing
component of both concepts, because slope represents the rate of expenditure of
potential energy.
We are going to use dimensionless ratios in our analysis. The ratios of forces are
referred to in fluid mechanics as dimensionless numbers. For example the
Froude number is the ratio of inertia to gravity forces, or the ratio of the
resistance to change to the gravity forces (change), or the ratio of the flow
velocity to the velocity of a gravity (surface) wave in a channel. It is written as
NF (Chan)= V/ (gD) a50 (1-1)
where V is the mean velocity, g is the acceleration due to gravity, and
D is the mean depth of flow. All terms are in a consistent system of units. An
important geomorphic use of the Froude number was developed for watersheds
by Strahler (1958) where the relief, H, is used in place of the channel mean depth,
D. The Froude number of the watershed is:
1-3
-------
EPA Channel Condition Project
Np (Shed)= V/( gH) 0'50 (1-2)
This dimensionless ratio of forces will be used with Eq. (1-1) to relate basin to
channel characteristics in Part 2 - Fundamentals.
In streams, and in all "open channel" flow conditions (the water surface is open
to the atmosphere), the Froude number is used to design physical hydraulic
models. For design conditions, such as bankfull flow, the Froude number in the
model is made equal to the Froude number in the prototype, or
Np Model = Np Prototype (1-3)
Streams in nature operate in the same manner as physical stream models,
wherein little streams mimic big streams. As long as we use the appropriate
dimensionless ratios, we can avoid scale effects when combining channel and
basin characteristics, and in relating one stream to another.
One dimensionless ratio commonly used in habitat work is W/D, or the water
surface width of the channel (W) divided by the mean hydraulic depth (D). The
depth (D) is calculated by D = AC/W, where Ac is the channel cross-sectional area
at that flow. But, this in essence reduces all channels to equivalent rectangular
shapes. It might be more descriptive to write W/D as
W/D = W/[D(Dmax/D)], or W/Dmax (1-4)
Using Dmax incorporates the shape of the channel. For example: for rectangular
channels, D = Dmax; and for triangular channels, Dmax = 2D or D = 0.5 Dmax
. Also, triangular channel cross sections (such as those in bends), usually have a
constant W/D over a range of flows. If W/Dmax is used, a triangular section
with the same flow area as the rectangular section, will have a W/D max that is
half of the W/D for the rectangular section. Therefore the triangular cross-
section provides greater depth habitat at reduced flows.
Methods of Analysis
To extend the fundamental relationships developed in Part 2, three regional
stream channel databases have been selected in Washington State for comparison
in Part 3: (1) the Olympic Peninsula; (2) some of the lowland streams north and
east of Seattle; and (3) the mixed mountainous and agricultural region of
Northeast Washington located east of the Columbia River, and north of Grand
Coulee dam.
1-4
-------
EPA Channel Condition Project
The Olympic Peninsula stream gage locations include basins having a diverse
mixture of geology, with streams flowing through valleys ranging from bedrock,
through boulders to sandy gravels. The lowland streams east of Puget Sound are
less diverse in character than the Olympic streams and are experiencing
urbanization to varying degrees. The northeast Washington streams included in
the analysis have experienced diversions for irrigation and logging impacts, as
have the other two regions. Precipitation varies on the basins used in the three
regions from 40-200 in/yr on the Olympic Peninsula, to 37-66 in/yr in the Puget
Lowlands, and to 18-30 in/yr in northeastern Washington.
In this paper, we do not attempt to account for land-use effects on stream
channels in a detailed cause and effect manner for such a broad range of
conditions. Rather, our objective is to determine IF relationships exist among
channel, streamflow and basin characteristics, and IF those relationships can be
of assistance in the investigation of those streams, for whatever purpose.
Applications- Case Studies
Examining channel characteristics on a regional basis should provide a means
whereby a problem (such as degraded fish habitat) can be more effectively
defined, and solutions designed and monitored. The smaller the region, and the
more uniform the climate and geology, the better will be the analysis. The
methods of analysis developed in Part 2 or 3 will be demonstrated in Part 4 to
examine four case studies of habitat improvement on the Olympic Peninsula,
restoration of a gold-dredged stream in Idaho, documentation of increased
sediment load effects on a Kitsap Peninsula stream and the effects of dams and
diversions on fish habitat and channel geometry in a North Olympic stream.
Summary and Conclusions
The results of the analyses as applied to the case studies are summarized, and
compared with the project objectives. There will be no panaceas, but there
should be a better understanding of the relationships between basins and their
stream channels. The basins generate the channels that we have an affinity to
degrade, and aggrade, in response to our development activities on a watershed,
or in response to local, so-called channel "improvements".
As researchers, planners, designers and managers, we work on stream problems
as if they were something new. These are definitely not new problems, but the
problem-solvers are new, and some tend to reinvent the wheel. The
fundamentals of problem definition tend to be set aside in favor of talking about
solutions, issues and stakeholders. These latter emphases are important, but
there is nothing like good data and the application of fundamental principles
to assist in problem definition and the comparison of alternatives.
1-5
-------
EPA Channel Condition Project
By focusing on apparent solutions before conducting a thorough and
thoughtful analysis of the problem, we are doomed to treat the symptoms, and
not the actual causes of the problem. Linking the problem analysis with an
understanding of the fundamental watershed processes that control channel
form is key to project success. In terms of restoring instream habitats that are
critical for the recovery of native salmonids, one must also understand (or at
least appreciate) and anticipate the functional relationship of channel condition
to life history requirements.
A note about the method of allometric analysis:
"(it is the) development of simple or multiple power-function equations
that express the relative rates of change among the variables of a system.
A principal geomorphic utility of the method is to show adjustment
between two variables" (Osterkamp 1979).
These power relationships are used throughout this report. Allometric analysis
is used to relate one variable in a fluvial-geomorphic system to another variable
in that system.
We hope that the methods and examples described in this report will be of
assistance to those persons engaged in stream projects, whatever their
professional position.
"Today natural diversity still baffles us. Even the simplest
natural communities escape our comprehension. We
abstract and simplify them intellectually with energy flow
charts or systems diagrams. When we understand the
pictures and formulae, we delude ourselves into believing
we understand reality." (Dasman 1973)
1-6
-------
EPA Channel Condition Project
References for Part 1
Arnold, C. L., P. J. Boison, and P. C. Patton. 1982. Sawmill Brook: An example
of rapid geomorphic change related to urbanization. Journal of Geology
90:155-156.
Blench, T. 1957. Regime behavior of canals and rivers. Butterworth Scientific
Publications. London, England.
Booth, D. B. 1990. Stream-channel incision following drainage basin
urbanization. Water Resources Bulletin 26:407-417.
Burkham, D. E. 1981. Uncertainties resulting from changes in river form. ASCE
Journal of the Hydraulics Division. Vol. 107, HY5.
Dasman, R.C. 1973. A rationale for preserving natural areas. Journ. soil Water
Conserv., V. 28, No. 3, pp. 114-117.
Hammer, T. R. 1972. Stream channel enlargement due to urbanization. Water
Resources Research 8:1530-1540.
Leopold, L. B. 1973. River channel change with time: an example. Geological
Society of America Bulletin 84:1845-1860.
Mackin, J. H. 1948. Concept of a graded river. Geological Society of American
Bulletin. Vol. 59, pp. 463-512.
Made], M.A. 1978. Response of a stream channel to an increase in sediment load.
MS thesis. Dept. of Geological Sciences. University of Washington,
Seattle, WA.
Miller, J. R. and J. B. Ritter. 1996. An examination of the Rosgen classification of
natural rivers. Catena 27:295-299.
Osterkamp, W.R. 1979. Invariant power functions as applied to fluvial
geomorphology. In Adjustments of the fluvial system. Proceedings of the
10th Annual Geomorphology Symposia Series. SUNY, Binghamton, NY,
Sept. 21-22,1979. Kendall/Hunt Publishing Co., Dubuque, Iowa
Rosgen, D. 1996. Applied river morphology. Printing Media Co., Minneapolis,
MN.
Strahler, A.N. 1958. Dimensional analysis applied to fluvially eroded landforms.
Geological Society of America Bulletin. V. 60, p.p. 279-299.
1-7
-------
EPA Channel Condition Project
2. FUNDAMENTALS
One of our project objectives was to examine the feasibility of estimating channel
top width and other channel characteristics from basin characteristics such as
drainage area. This could be done empirically, but without doing a thorough
analysis of the physical relationships between basin, streamflow and channel
characteristics, the foundation and linkages would be missing.
State of Our Current Knowledge about Width Adjustment
A Task Committee (TC) of the American Society of Civil Engineers prepared the
most comprehensive report on river width adjustment to date (ASCE 1998). In
Part I the TC covered processes and mechanisms and in Part II the TC discussed
modeling. The objectives of the TC efforts were to:
• "Review the current understanding of the fluvial processes and bank
mechanics involved in river width adjustment
• Evaluate methods (including regime analysis, extremal hypothesis and
rational, mechanistic approaches) for predicting equilibrium river
width
• Assess our present capability to quantify and model width adjustment
• Identify current needs to advance both state-of-the-art research and the
solution of real world problems faced by practicing engineers" ASCE
(1998).
The ASCE/TC reports covered the following topics:
• geomorphic context of river width adjustment;
• the regime theory and the power law approach (including hydraulic
geometry by Leopold and Maddock 1953);
• the extremal hypothesis approach which uses sediment transport and
friction combined with stream power (or energy dissipation) to
determine channel width;
• tractive force methods to obtain the geometry of stable channels;
• the near-bank fluvial processes and their interactions with bank
materials;
• the formation of the cross-sectional channel shape;
• longitudinal changes in channel cross sections; and
2-1
-------
EPA Channel Condition Project
• linking fluvial processes to channel-width adjustments through
velocity, boundary shear stress, secondary flows and turbulence
structure.
Under the heading of bank mechanics the TC addressed: bank erosion; reduced
resistance to erosion; mass failure and bank stability; basal endpoint control;
vegetative effects; seepage effects; and the advance of banks. The conclusion
and recommendations of Part I - Processes and Mechanisms, are closely related,
especially the conclusion that civil engineers be aware of the geomorphic aspects
of width adjustment. Likewise, the first recommendation proposed that stream
reconnaissance procedures should be developed that emphasize the
geomorphic context of width adjustments. It is interesting to note that none of
the work by Rosgen (1994) was cited in the two TC reports.
Part 2 of the ASCE/TC report covered modeling and included:
RAPID ASSESSMENT TECHNIQUES: Empirical Models of Channel
Evolution; Channel Stability Diagram;
NUMERICAL WIDTH-ADJUSTMENT MODELS: Hydraulics and
Hydrodynamics (including summaries of 12 models); Sediment
Transport and Continuity: sediment (sand and gravel) is routed
using the 12 models;
RIVER BANK MECHANICS: The types of bank processes and bank
materials are accounted for in the 12 models. None of the models
accounts for the influences of riparian vegetation.
TESTING AND APPLICATIONS: Tests with laboratory data; and field
testing.
Procedure for Approaching Width-Adjustment
An eight-step procedure was outlined by the Task Committee:
1. Problem identification;
2. Reconnaissance and data collection;
3. Desk assessment of equilibrium conditions;
4. Application of empirical channel response or dynamic models;
5. Application of numerical models (if warranted);
6. Validate the model results against field data (if available);
7. Numerical models should be applied to existing conditions and to
assess any known or anticipated future impacts; and
8. Selection of a solution (river management).
2-2
-------
EPA Channel Condition Project
Considering these two comprehensive ASCE articles as a point of departure for
our more general EPA study on "Channel Condition", there are some useful
observations to be made:
1. although the ASCE/TC reports on width adjustment deal mainly with
mechanics and mathematical models, the TC concluded that they can
only make "tentative predictions of width adjustment,"
2. our empirical and fundamental models, derived from dimensional
analysis of basin, flow and channel characteristics, can be expected to
demonstrate both low and high degrees of variability in their
predictive capabilities due to natural and data anomalies;
3. there is a lack of sufficient laboratory and field data for testing the TC
width adjustment models (these models are data-intensive);
4. our simpler models based on parameters such as the expected width,
depth, velocity, flow area and wetted perimeter, are less data-intensive
and are gathered on a regular basis; and
5. we will be representing the "geofluvial" approach as described by the
Task Committee (ASCE 1998) because of the analogous parameters
considered in our basin, flow and channel models.
The eight-step procedure on page 2-2 for approaching channel width adjustment
is not a new approach to problem solution, but it is sound, especially if Step 1
includes problem definition.
Overview of Basin, Streamflow and Channel Characteristic
Models
The fundamental models have been organized along the lines of work done on
the Colville Indian Reservation (Orsborn and Orsborn, 1997). The models were
developed by relating drainage basin, streamflow and channel characteristics to
each other and to themselves. The basin and channel characteristics are linked
physically by streamflow. Changes on the basin cause changes in streamflow
and responsive changes in channel characteristics. Streamflow data can be
highly variable in a region due to priorities for gaging programs by resource
agencies, natural variability in precipitation, geology, soils, elevation and
uncommon periods of record. Superimposed on natural variability are changes
in land use which cause changes in streamflow, debris and sediment loads, and
thus channel geometry. Also, local impacts due to streamside road building,
changes in riparian vegetation and cattle grazing will cause direct changes in the
stream channel without any upstream changes in land use. Diversions and
storage also exert influences on streamflow, and thus channel characteristics.
Natural variability causes wide swings in precipitation over a climatic-
geographic region. Add to this natural variability the influences of diversions,
storage, channel changes and measurement accuracy, and we are forced to model
2-3
-------
EPA Channel Condition Project
"average-condition" relationships, and their variability. Sometimes we have to
remove data from the models for certain gaging stations due to their unusual
influence on the models caused by geologic anomalies, or biases in the periods of
record. Usually some of the station data are left out of models to test their
accuracy. Actually, the variability of the model data points gives a good idea of
how reliable the models are.
Building the models involves the measurement and use of basin, streamflow and
channel characteristics (BC, QC and CC). The basic relationship says that one
characteristic is related to (is a function of (f), or is dependent upon, or is related
to) another set of characteristics. For example we know that basin area, Ab,
catches precipitation and explains 80-90% of the variability in a large number of
streamflow (QC) models. As an example of the logic:
• Flow characteristics are related to basin characteristics
• Q (Any flow, say Max, Peak Flood) = f(Basin Area, Ab)
• QC (characteristic flow) = f(BC, Basin characteristics)
• QPF Max = C (Ab)n (Power Equation) (one application)
• Dependent flow = f (Independent basin area), which in turn is a
function of the maximum, basin-wide precipitation within the flow
period of record.
Characteristic flows can be low, average or flood flows, extreme flows or
monthly flows. This simple starting point does not cover all cases, of course. In
the following diagram (matrix, Figure 2-1) one reads up along the vertical scale
(1,2 or 3) to select a dependent characteristic and then horizontally across a line
to select an independent characteristic (1,2 or 3) to relate to the dependent
characteristic. A few of the possible combinations are listed in Table 2-1.
Figure 2-1. Matrix of Models of Combinations of Basin, Streamflow and Channel
Characteristics (Orsborn and Orsborn 1997). BC = Basin Characteristics, QC =
Flow Characteristics and CC = Channel Characteristics
(3) CC
Dependent
Variables
(2)QC
(1)BC
ORIGIN
3:1
2:1
1:1
3:2
2:2
(na)
3:3
2:3
(na)
BC QC CC
(1) (2) (3)
Independent Variables
2-4
-------
EPA Channel Condition Project
Table 2-1. Combinations of Basin, Streamflow and Channel Characteristics In
Hydrologic and Hydraulic Geometry Models (REFER TO FIGURE 2-1) (Orsborn
and Orsborn 1997).
Combination
Numbers
Notes and Examples of Models Developed for CCT Study
(Start at lower left in Figure 2-1 and go up)
1:1
2:1
3:1
Basin Characteristics (BC) related to BC.
Example: Stream Length (LS) related to Basin Area (Ab): LS = 1.2(Ab)1-0
Flow Characteristics (QC) related to BC.
Example: Average Annual Flow (QAA),
QAA = 0.0025 (P) 1 -64 Aft, where P = Average Annual Precipitation, in/yr.
Channel Characteristics (CC) related to BC.
Example: Water Surface Top Width, W, at a characteristic flow such as QAA,
related to basin area (Ab): W = C(Ab)n ; C = f(Q, Chan. Type) Regional Model
(Move to bottom of center column and go up, Figure 2-1).
1:2
2:2
3:2
(na) BC: QC; logically not physically correct for basin characteristics to be a
function of (dependent on) flow characteristics. The inverse equations are
covered by 2:1 above.
Flow to flow, QC: QC: models can be built either by :
(1) Correlating the same types of flows at two sites such as peak flows; or by
(2) using ratios of various statistical flows to the average annual flow such as:
Q7L2/QAA, QPF2/QAA, etc. at long-term gages in a region.*
CC related to QC, Channel to Flow Characteristics, the basis of Channel
Hydraulic Geometry; one of the models used to check channel geometry
over time using changes in W, D, V Ac and P related to Q; regional models
for common flows such as QAA.
(Move to bottom of last column on right and read up, Figure 2-1).
1:3
2:3
3:3
(na) BC:CC, like 1 :2 not physically logical because basin characteristics are
not dependent on channel characteristics; conditions covered in 3:1, CC: BC.
Flow related to Channel Characteristics, QC = f (CC); this is hydraulic
analysis of flow down a channel, Q = Ac V where; Ac = cross-sectional flow
area; V is the mean velocity over the flow area, Ac. This equation, Q = AV,
is the standard basis for stream gaging and hydraulic geometry; when the
energy equation is used with Q = AV at several cross-sections the water
surface profile in a stream can be calculated for individual flows.
CC : CC, Channel to channel characteristics; W/D, channel shape factor used
in fish habitat; W/D versus P^ / Ac, where P is the wetted perimeter (contact
length of water with the bed of the stream at a cross-section of the channel).
*Note: Q7L2 = 7-day average Low Flow, 2-yr. Recurrence Interval (RI)
QPF2 = Flow (Q), Peak (P), Flood (F), 2-yr RI
QAA = Flow (Q), average annual (long-term)
2-5
-------
EPA Channel Condition Project
In 1970 the U. S. Geological Survey completed a nationally oriented study of
determining streamflow characteristics from drainage-basin characteristics
(Thomas and Benson 1970). The subtitle read, "A study of relations for
estimating streamflow characteristics from drainage-basin characteristics in four
hydrologically differing regions of the conterminous United States." The study
included drainage basins in the: East (Potomac River), Central (subbasins of the
Missouri River in Kansas, Nebraska, and Missouri); South (in Louisiana,
Arkansas and Mississippi); and West (the Sacramento and San Joaquin River
basins in the central valley of California).
The authors tried to select "virtually natural streamflow" for analysis. Because
they used a multiple regression analysis they chose to use the longest periods of
record rather than use a common base period. It is interesting to note the
number of records and their length available in each region: East (41 of 18 years
or more); Central (41 of 12-61 years); South (42 of 15-29 years and West (44 of 16
or more years). This thorough study used 71 streamflow indices and their
statistical characteristics (e.g. standard deviation), and tested them against 30
meterologic and topographic characteristics of the basins, because they "control
the amount of streamflow from the basin and the distribution of this flow in
time" (Thomas and Benson 1970).
This report became the bible of USGS personnel who used it to evaluate the
gaging station programs in the States. Although the basin characteristics were
selected on the basis of hydrologic knowledge, their retention was primarily
statistical. Basin area was the most common parameter in all regions to be found
significantly related to all characteristic streamflows.
It is interesting to note some of the conclusions from this study:
1. "The interrelationships between the basin indices along with the
inability to describe completely a drainage basin, makes tenuous
any assertions about the physical effects of the basin
characteristics on runoff;
2. despite the inability of the relations to describe the fundamental
causes of streamflow variation, the basin indices significant in
the relations are numerical measures that are related to the
flow variations; and
3. low-flow relations are unreliable in all study regions; they can
provide only rough estimates of low-flow characteristics at
ungaged sites." (Thomas and Benson 1970).
Perhaps part of the problem is that some of the basin characteristics selected for
analysis are not physically compatible with the purpose for which they were
chosen, and/or they have unknown interdependence with other characteristics.
For example, "the index of forest cover (F) used in this analysis is the percentage
of total drainable area shown as forested on the topographic maps." The maps
2-6
-------
EPA Channel Condition Project
were updated only once or twice in 15-20 years, and logging was changing the
forested area much more rapidly than that in many basins. Mean basin elevation
was selected to account for variations in precipitation, temperature, wind,
vegetation and ruggedness. But, these factors are not physically true, because
mean basin elevation (E) says that the outlet of the basin is at mean sea level.
This is true for only those basins, which empty into the sea. Physically, the relief
of the basin accounts for all the energy available to cause both surface and
ground water to flow from the basin.
Dimensional Analysis of the Basin, Flow and Channel
Characteristics
Analysis of the Basin
This analysis has been provided by Strahler (1958), who opened his article stating
that geomorphic studies can be founded on sound geometrical and mechanical
bases using dimensional analysis. Dimensional analysis is based on the
dimensions in Newton's second law: F = Ma = ML/T2, or the dimensions of
force, mass, length and time. Details of dimensional analysis, dimensionless
numbers and the Buckingham Pi theorem are available in many textbooks
(Rouse, 1938).
Strahler's analysis focused primarily on drainage density (drainage length
divided by basin area), and a ruggedness number (relief, H, times drainage
density, D). He also considered the Reynolds Number of the basin, (the ratio of
inertia to viscous forces), and the Froude Number of the basin Q2/gH (the ratio
of inertia to gravity forces). Relief (H) is used as the characteristic dimension,
much like the Froude Number for open channel flow uses depth, D. The Q2 term
represents the Froude Number squared, not the true Froude Number.
Strahler used Q as a volume rate of flow per square foot of channel cross-section,
which reduces to a velocity term (L/T). Also, Strahler considered relief (H) to be
the maximum in the basin. If one uses the difference in elevation between the
basin outlet and the highest contour within the basin, a much more consistent
value of H results (Orsborn 1976). The acceleration due to gravity, g, is
considered to be constant, and when regional models are built, g becomes a part
of the coefficients.
The Froude Number of the basin offers us the best of Strahler's relationships for
estimating streamflow in terms of basin characteristics.
Rewriting the Strahler basin Froude Number: Ql/(gH)°'5 (2-1)
where: Ql is the discharge generated from a watershed flowing through
one square foot of river channel cross-section; a "unit" discharge
2-7
-------
EPA Channel Condition Project
with the dimensions of L3/T/L2 or L/T, a velocity term like V =
Q/A; g is the acceleration due to gravity (32.2 ft/sec2, L/T 2); and
H is the basin relief in feet (L).
For this ratio of inertia to gravity forces we have L/T on the top and bottom of
Eq. (2-1), giving the dimensionless Froude Number.
Now, if we assume that the "unit" discharge in Eq. (2-1) comes from each square
mile of watershed area, instead of flowing through each square foot of channel,
and we multiply both top and bottom by Ab, we have not changed conditions,
and
(Ql / (gH) °'50 )(Ab / Ab ) = Q2 / ((gH) °'50 Ab) . (2-2)
We can also rearrange Eq. (2-2) and use Q(x) to denote any statistical flow of
interest:
Q (x) = C (g) °-50 Ab (H) a50 (2-3)
where C is part of a total coefficient and (x) denotes some characteristic flood,
average or low flow which must be regionally calibrated from USGS gage
records. Combining the first coefficient, C and the (g) °"50 gives C' in
Q (x) = C' Ab (H)a50 (2-4)
which is the form of the equation used to develop the relations in Figures 2-2a
and 2-2b. These are the basic regional equations developed from Eq. 2-4 for
characteristic low, average, and flood flows for the Siuslaw National Forest in the
mid- and north-coast regions of Oregon (Orsborn 1981). The basins range in size
from 0.3 to 667 sq. mi. and the relief ranges from 400 ft. to 2,400 ft.
In Figure 2-2, the graphs from top to bottom display, as a function of the "basin
energy" terms A
• the 50-year, peak flood;
• the 2-year, peak flood;
• the range of maximum and minimum average annual flow that has
occurred at gages in the region (range is + 70% of QAA);
• the average annual flow (QAA); and
• the 7-day average low flows with 2- and 20-year recurrence intervals
for the different groups of basins; the low flows become less the
farther south their basins lie in the mid-coast region.
The original Figure 2-2 was drawn on 6 by 5-cycle log-log graph paper, and
when reduced, the lines, data points, stream names, gage numbers and notes
2-8
-------
EPA Channel Condition Project
Figure 2-2a. Siuslaw National Forest Basin Energy Models (Orsborn 1981)
100000
0.01
1000
A(Hrb (mi")
Note: QAAmax = 1.7(QAA) and QAAmin = 0.3(QAA)
2-9
-------
EPA Channel Condition Project
1000
o
u.
1
a
41
en
Q
O
10
0.1
0.01
Figure 2-2b. Detail of Low Flows for Siuslaw National Forest
Basin Energy Models (Orsborn 1981)
J_
Coast & N Basins
Q7L2 a 1.20(AH°-5)°
Coast & N Basins
Q7L20 = 0.80(AH°-5)0-88
Alsea Basin
Q7L2 = 0.80(AH0'5)0 88
0.1
Siuslaw Basin
Q7L2 = 0.45(AH° S)"-9S
Alsea Basin
Q7L20 = 0.43(AH0'5)0-95
y Siuslaw Basin
Q7L20 = 0.25(AH0'S)°-8S
10
100
1000
A(H)°
2-10
-------
EPA Channel Condition Project
became too crowded to show clearly (Orsborn, 1981). Therefore, only the graphs
have been shown for the characteristic flood, average and low flows.
It has been demonstrated that the same average relationships (coefficients and
exponents) apply equally well to the coastal regions in Alaska and Washington
(Orsborn 1983; Amerman and Orsborn 1987). In 1971, Yang related the potential
energy to stream morphology through two laws governing stream systems: (1)
the "ratio of average fall between any two different order streams in the same
river basin is unity"; and (2) "a natural stream chooses its course of flow in such a
manner that the rate of potential energy expenditure is a minimum" (Yang 1971).
Using Horton's (1945) laws of stream order, average stream length and average
stream slope, Yang was able to calculate longitudinal stream profiles. They
agreed with observed data quite well, and define the average rate of energy
expenditure of watersheds (H).
Analysis of the Channel
The channel can be analyzed in plan, profile and cross-section, the three views
being physically interrelated. Plan, or pattern, provides the aerial view of
geographic arrangement of the channel in straight, meandering or braided
patterns, plus other less common patterns.
One of the most comprehensive and current references for determining historical
changes in streams was prepared by Smelser and Schmidt (1998). Although they
limited their investigation to mountainous streams, they provided numerous
examples of historical studies to evaluate geomorphic channel changes in
different geologies. The stream types that Smelser and Schmidt studied included
were B3, B4, C3, C4, F3, F4, G3 and G4 as organized and documented by Rosgen
(1994,1996).
Chitale (1970) used data from 35 rivers inside and 7 rivers outside India, whereas
Ackers and Charlton (1970) studied the development of meanders in a laboratory
flume using four median sand diameters of 0.15, 0.21-0.26, 0.45,0.70 mm and all
sizes mixed together.
Chitale (1970), and Ackers and Charlton (1970) focused on the meander length
and both used dimensional analysis to develop their analytical parameters which
included the Froude Number of the flow F = V/(gD) °'50. F is a ratio of water
velocity to the velocity of a gravity wave superimposed on the water surface, or
the ratio of inertia to gravity forces in the channel flow.
Chitale (1970) used prototype data for streams ranging in discharge from 5000 to
1,500,000 cfs, and bed material mean sizes of 0.01 to 5.0 mm on very mild slopes.
He tied the ratio of river length (LR) to the valley length (LV) to: m/D (mean
grain size/average depth); S (the slope in ft per 10000 ft of channel length); and
W/D (the water surface width to mean depth ratio). With respect to channel
2-11
-------
EPA Channel Condition Project
cross-sectional shape, the tortuosity ratio (LR/LV) varies as (W/D) "°'66. So we
can expect W/D to make a large change from 10 to 100 (10-fold increase) and
only a small, average change in LR/LV of 17%. Channel width is variable
through the meander, so using this water surface width in our cross-sectional
analysis will not be as prudent as using the W/D ratio in a riffle or glide.
Stypula (1986) performed a dimensional analysis of channel cross-sectional and
flow characteristics using mean hydraulic depth (D), the mass density (p) and the
average velocity (V) as the repeating variables. Resulting dimensionless
relationships included (D/W, D/P, D/R, D/kh and V2/gD ), where: P is the
wetted perimeter; R is the hydraulic radius (A/P); and kh is a bed roughness
height, and D/ kh is referred to as a relative smoothness.
The "shape" factor of W/D was used with D/P and D/R to develop a "Shear-
Shape" relationship. The shear component was developed as follows:
[l/(D/P)j (D/R) = (P/D) (D/R) = P/R = P2/A (2-5)
where the substitution of R = A/P has been made.
To develop a theoretical basis for the relationship of W/D versus P2 / A the two
factors were calculated for rectangular channels by varying W/D between 0.01
and 1000 and calculating P2/A. This yielded the equation for natural and
artificial rectangular channels of
W/D = P 2/A - (4 + 4D/W) (2-6)
Next, natural channel data were used from numerous sources listed in Figure 2-
3. Then more natural channels were added for sand channels in Central
Washington and small eroded rills in the loess hills near Pullman, Washington.
All the channels were combined by Orsborn and Stypula (1987) into one set of
two curves in Figure 2-3. Natural (non-rectangular) channels follow the
relationship
W/D = P2/A - (2 + 2D/W) (2-7)
To combine the hydraulic geometry with the shear-shape equation one must
merely substitute W = a (C?) and D = cQd into Eqs. (2-6) and (2-7), which yields
aQb/D = P2/A - (2 + 2D/W) (2-7a)
for natural, non-rectangular channels, and
W/(cQd) = P 2/A - (4 + 4D/W) (2-6a)
for both natural and artificial rectangular channels where W is at, or nearly a
constant. Notice that Manning's resistance coefficient does not appear in these
2-12
-------
EPA Channel Condition Project
Figure 2-3. Shear-Shape Relationships for Natural and Rectangular Stream
Channels (Orsborn and Stypula 1987)
1000 F
TOO
W/D*P /A • (2+2D/W)
~ REAL CHANNELS
Q
10 r
2.0
T.O
0.1 L
0.0 1
W>0
'/
tl
\
LEGEND
0 Barnes
0 Chrostowski
• Copp & Rundquist
A Emmett
A Stypula sand
• channels
T Stypula loess rill
• USGS Alaska
\
RECTANGULAR
_ CHANNELS
W/D* P2/A
D>W
I. iml—
TO
\
\
\
\
\
_.' I M I I 1 I 1
roo
7000
2-13
-------
EPA Channel Condition Project
open-channel flow equations. Their utility and reliability were examined by
Orsborn and Stypula (1987,2000) and an example of the results is shown in Table
2-2.
An analysis of W/D versus P2/A was made for the Lower Elwha River on the
Olympic Peninsula by Orsborn and Orsborn (1999a). This set of graphs in Figure
2-4 shows those from Figure 2-3 plus the straight-line relationships for very wide
and very narrow channels. The left-hand scale has been changed to A/P2 to
show increasing numerical values on both scales. Also, Q is directly
proportional to A, and inversely proportional to P. The Lower Elwha channel
has become starved for gravel below the two dams (Figure 2-4) and now has a
mean grain size of about 6-8 inches. Note that natural channels have a most
efficient section when W/D is 1.5, (A/P2 is a maximum), not 2.0 as it is for
rectangular channels. The low flow measurements gave W/D values of 38 to 221
for the Lower Elwha River, all "wide, shallow channels" used in hydraulic
calculations when R approaches D.
This leads us into a discussion of CC = f (QC) (from Figure 2-1 on page 2-4,
relationship 3 : 2) where the channel cross-sectional characteristics (CC) are
related to discharge (flow) characteristics (QC).
2-14
-------
EPA Channel Condition Project
Table 2-2. Measured and Modeled Values of Average Annual Flow,
Width, Depth and Velocity for Deer, Fall and Flynn Creeks in the
Oregon Mid-coast Region (Orsborn and Stypula 1987).
Averaqe Flow,
USGS NUMBER Gaging Station Name ~L Top Width
(mV1) C")
Average
Depth
(m)
Average
Velocity
(ms"
14306810
14306300
14306800
Deer Creek
Est. eq. (2-6a)a
Est. eq. (2-7a)b
Actual sizes0
Est. sizesd
Fall Creek
Est. eq. (2-6a)a
Est. eq. (2-7a)b
Actual sizes0
Est. sizesd
Flynn Creek
Est. eq. (2-6a)a
Est. eq. (2-7a)b
Actual sizes0
Est. sizesd
0.18
0.19
0.19
4.67
4.14
4.60
0.12
0.18
0.14
3.26
3.20
15.16
16.20
3.14
2.60
0.16
0.17
0.46
0.50
0.13
0.14
0.34
0.34
0.67
0.58
0.30
0.32
NOTES:
a Assumes P = W + 2D, rectangular section.
b Assumes P = W + D in natural channels, and P = W for Flynn Creek.
0 Actual sizes are based on hydraulic geometry at the gaging stations.
d Estimated from equations for W, D and V based on Qa of record at 10 Regional USGS gaging stations.
2-15
-------
EPA Channel Condition Project
0.1
A/P'
0.01
0.001
Figure 2-4. General Shear-Shape Relationships for Rectangular and Non-rectangular
(natural) Channels (Orsborn and Orsborn 1999a)
0.01
NATURAL CHANNELS (NON-RECTANGULAR)
W/D = P2/A - [2 + 2D/W]
W/D = 4(A/PA2)
W/D = P*2/A
Rectangular Channels
Natural Channels
A MAX A/P*2 Nat Chans W/D=1.5
• MAX A/PA2 Rect Chans W/D=2.0
RECTANGULAR CHANNELS
(Natural & Artificial)
W/D = P2/A - [4 + 4D/W]
Lower Elwha '
Sites W/D =
38 to 221
0.1
10
100
1000
W/D
-------
EPA Channel Condition Project
Channel Hydraulic Geometry
The traditional analysis of hydraulic geometry is applied to streams based on the
continuity equation: Q = AV = WDV, and W = aQfe, D = cQd and V = eQf where:
W is the water surface width; D is the mean depth; and V the mean velocity
(Leopold and Maddock 1953). For ease of understanding we have not used
Leopold's and Maddock's nomenclature for coefficients and exponents; we have
used alphabetical continuity a-j, keeping W, D, V, A and P in the same sequence.
Chezy's and Manning's works showed that V is a function of the hydraulic
radius (R) which is the flow area (A) divided by the wetted perimeter (P).
Including these latter two factors in the suite of hydraulic geometry equations,
we have A = gQh and P = iQ'. Wetted perimeter accounts for two influences,
the resistance to flow (shear), and a measure of available habitat for certain life-
stages of fish.
Williams (1978) examined the at-a-station exponents in the hydraulic geometry
equations for W, D, V, slope (energy gradient) and friction factor at 165 USGS
gaging stations across the country. The cross-sections had ranges of exponents
of: width (b) = 0.00 -0.82; depth (f) = 0.10 - 0.78; and velocity (m) = 0.03 - 0.81 (f
and m are d and f in our report). The Williams' flows varied between 0.01 and
70,000 cfs, widths from 1.0 to 1900 ft, mean depths from 0.1 to 35.0 ft and median
bed material sizes varied from 0.06 to 100 mm (0.0024 to about 4 inches).
Quoting from Williams (1978) to summarize the objectives and results of his
study:
"The original theory was intended to produce only the average hydraulic
exponents for a group of cross sections in a similar type of geologic or
hydraulic environment. The present test shows that the theory does
indeed predict these average exponents, with a reasonable degree of
accuracy.
An attempt to forecast the exponents at any selected cross section was
only moderately successful. Empirical equations are more accurate than
the minimum variance, Gauckler-Manning, or Chezy methods.
Predictions of the exponent of width are most reliable, the exponent of
depth fair, and the exponent of mean velocity poor." (Williams 1978)
Also, in comparing measured and theoretical hydraulic exponents (b, f, and m, or
b, d, and f in this report for W, D, and V), Williams (1978) stated:
"A number of variables, as discussed earlier, might have some influence
on the hydraulic exponents. However, Langbein's papers suggest that
most such factors cannot be taken into account individually in a
minimum-variance analysis because their effects usually cannot be
2-17
-------
EPA Channel Condition Project
determined separately. He believes that in spite of the interaction and net
influence of such variables, there will result in nature a statistical array of
exponent values in which certain values (the averages) are more common.
These most common values of m, f, and b represent a central tendency,
and the correct combination of variables is that for which the
minimization of variances yields the most common exponents." (Williams
1978)
These general guidelines are for regional channel geometry analysis, or as
originally described "in a downstream direction". This indicates increasing
discharge (as a function of increasing drainage area), but it occurs at a decreasing
rate due to decreasing precipitation at lower elevations. Regional hydraulic
geometry analyses can be completed using gages from different basins. These
analyses are used later in this report to connect channel characteristics to basin
characteristics.
We will be using regional hydraulic geometry analyses based on average low,
annual and flood flows at-each-station. The channel cross-section is the
"response variable" that can react to changes in watershed and streamflow
characteristics. Analyses of changes in channel hydraulic geometry for different
periods of record will indicate changes in width, depth and velocity, wetted
perimeter, bankfull flow, sediment size (or bed slope), or all of the above.
Usually, even though width and depth may change, cross-sectional area will
remain about the same for a particular streamflow.
One other comment about analyzing data to determine the hydraulic geometry
at-a-station; Williams (1978) showed two examples, one for the Colorado River
near Grand Canyon, Arizona and the other for Prairie Dog Fork of the Red River
near Childress, Texas. The Colorado River showed a near-perfect plot of W, D,
and V as a function of Q. Conversely, the Prairie Dog Fork showed a high degree
of variability (width varied by up to 110 percent, with a mean of 24 percent).
Lines were drawn parallel to the mean lines for W, D, and V to indicate they
included 90 per cent of the data points (Williams 1978).
As an outgrowth of this data scatter problem, we analyzed the feasibility of using
just three (and increasing numbers) of the W, D, V, A and Q data points. Class
problems in river engineering and a recent analysis for the Colville Tribe showed
that if you select W, D, V, and A data points at just three discharges (low,
medium and high), the graphs and equations will all fall within the 90 percent
lines. These are 95 percent lines (two standard deviations) if you conduct a
statistical analysis of the data (Orsborn and Orsborn 1999b). The variability in
coefficients and exponents as a function of the number of data points used in the
at-a-station hydraulic geometry analysis is shown in Figures 2-5a and 2-5b. In
the graphs the general coefficient C is used to represent the coefficients, while E
is the exponent used in lieu of the exponents for W, A, D, and V, respectively.
2-18
-------
EPA Channel Condition Project
Figure 2-5a. San Poll River in Northeastern Washington (SPR047). Number of Points Test for
"C", used in the Power Relationships for Hydraulic Goemetry.
tft
C
o
jo
Q}
S. 10
>
o
«
s.
0)
Q
? 1
<
.c
13
S
S
O
0 1 .
'
,
'
— I —
'
'
I 1
I , >
1 • '
J — 1 — 1 —
1 '
. .
* ;
1 '
i i
A :
•
| .,._. , .
, , ,
• : •
— i — i — i —
¥ : :
i i i
© ,
A •
j
4
<
i
] — L_
— ' —
1 '
•
,
• •
• i
• '
• '
j — ' —
1
i
<
j — . , ._. —
1
i .
• .
: : »
— . — i — i — i —
I..,
• • • •
, ,
. , , ,
0
j — , — i — , — , —
iii
i i i. ,
, , , -
' i i <
— i — i — i — i —
, , , , |
i i i i
1 . ! 1
I
; : :
j — , — i — , — , —
, , , ,
— i — i — i — i —
i , , , ,
, , , ,
i i i i
, ,
) '
i *
1 _, -L. .^ j —
iii
- i .1
, i , ,
! ! ! !
— 1 — 1 — 1 — 1 —
....
1 I 1 1
,
1
'
1
I
: 4 :
— i — i — i — i —
1 p ' '
i i i ,
: * : ;
, , , ,
1 1 1 !
1 A •
»C(w)
• C(a)
AC(d)
| •C(v)
15 20 25
Number o) Data Points Used In Analysis
30
35
Figure 2-5b. San Foil River in Northeastern Washington (SPR047). Number of Points Test for
"E", used in the Power Relationships for Hydraulic Goemetry.
0 1 -
0
1
4
/
4
1
>
\
>
c
4
i
<
^
*
\
(
i
1
1
1
L
>
0
1
5
1
i.
<
^
1
^
»
^
2
0
1
t
i
4
2
1
»
k
^
5
3
0
3
5
|
^
<
4
|
»
^
4
»E(w)
• E(a)
AE(d)
SE(v)
0
Number of Data Points Used in Analysis
2-19
-------
EPA Channel Condition Project
A Severity Factor Analysis to Determine the Influence of Flow
Reduction on Channel Characteristics
In 1976, Orsborn and Deane, while working on the physical aspects of instream
flow needs, developed a method for evaluating the effects of flow reductions and
other factors on habitat parameters. Called the Severity Factor (SF), the method
allows an individual to select and evaluate any set of factors, as long as one
chooses correct physical, quality or biological relationships. The method is based
on channel geometry under initial flow conditions (Stage 1) compared with
channel geometry under reduced flow conditions (Stage 2,3, etc.). The initial
flow condition can be any desired reference flow, bankfull, average annual, or
the average low flow.
Our 1976 example used five sets of conditions involving:
• flow reduction (Q1/Q2);
• the volume of that reduction (Vol. I/Vol. 2, flow x time);
• the change in width to depth ratio (W2 : D2/W1: Dl);
. • the change in water surface width with respect to the flow area (W2 :
A2) / (Wl: Al) to account for increased potential heating; and
• a depth ratio term raised to an exponent (D1/D2)L33 to account for the
reduction in reaeration in pools based on the increase in reaeration as
depth decreases in riffles where the measurements of channel
geometry are made (Langbein and Durum 1967).
This set of five severity factors (SF5) was developed for linear-sided triangular,
trapezoidal and rectangular channels. These results were compared with the
real-stream data from Chrostowski (1972). All five terms were calculated for a
series of 10 percent flow reductions below bankfull. The generated data for three
shapes of channels are given in Table 2-3. Note that SF5 in the last two columns
of Table 2-3 can take two forms, a summation (SSF5) or a multiple (XSF5), the
latter form being more sensitive. The reader is referred to the original report for
developmental details of the severity factor methodology (Orsborn and Deane
1976).
The five parametric ratios in each of the three shapes of channels were plotted
separately as a percentage of the original bankfull flow. These were combined to
produce the results of multiple XSF5 in Figure 2-6.
2-20
-------
EPA Channel Condition Project
Table 2-3. Components of Severity Factor Analysis of Dimensionless Ratios for
Triangular, Trapezoidal and Rectangular Channels.
Data for Figure 2-6. (Orsborn and Deane 1976)
Stage
%QBF1
01:02 -
01 + .101
02 + .101
W2 : D2
W1 : D1
W2 : A2
W1 : A1
[D1/D2]1'33
ISF5
XSF5
Triangular Section
10
8
6
4
2
0
Trapezoidal
10
8
6
4
2
0
100
55
26
9
1
...
Section
100
66
39
19
6
...
1.00
1.81
3.90
11.51
73.67
...
1.00
1.51
2.55
5.13
16.84
...
1.00
1.69
3.09
5.89
9.69
...
1.00
1.44
2.24
3.73
6.90
...
1.00
1.00
1.00
1.00
1.00
...
1.00
1.09
1.21
1.59
2.55
...
1.00
1.30
1.70
2.50
5.00
...
1.00
1.23
1.54
2.23
4.15
...
1.00
1.35
1.97
3.38
8.50
...
1.00
1.28
1.69
2.85
6.43
...
5.00
6.15
10.66
23.28
96.86
...
5.00
6.55
9.23
15.53
36.87
...
1.00
5.37
40.36
572.86
30339.15
...
1.00
3.73
17.99
193.36
7906.61
...
Rectangular Section
10
8
6
4
2
0
100
71
45
23
8
...
1.00
1.42
2.24
4.30
13.26
...
1.00
1.36
2.01
3.31
6.27
...
1.00
1.25
1.67
2.50
5.00
...
1.00
1.30
1.70
2.50
5.00
...
1.00
1.33
1.97
3.38
8.50
...
5.00
6.66
9.59
15.99
38.03
...
1.00
4.17
25.18
300.67
17667.29
...
2-21
-------
EPA Channel Condition Project
Figure 2-6. Multiple Severity Factor (XSF5) for Flows Less than Bankfull for
Triangular, Trapezoidal and Rectangular Idealized channels. Data In Table 2-3.
(Orsborn and Deane 1976).
10 10* |0»
MULTIPLE SEVERITY FACTOR, XSF5
An example of natural channel geometry W/D ratios is shown in Figure 2-7 for
nine of the channel sections measured by Chrostowski (1972). We found that
some of the W/D ratios plotted versus %Q in the real channels were very close to
"ideal" triangular, trapezoidal and rectangular channels.
CHANNEL
SHAPE
Triangular
Trapezoidal
Rectangular
STREAM NAME
Rock#2 (f)*
L. Brush #1 (c)*
L. Brush #3 (a)*
EQS. FOR REAL
CHANNELS
5.0 / (%Q)035
8.0 / (%Q)0'45
13.4 /(%Q)0'56
EQS. FOR IDEAL
CHANNELS
5.0/(%Q)0'35
10.0/(%Q)0'50
18.0/(%Q)0'60
"Figure letter in Figure 2-7.
Eqs: (W/D) 2: (W/D) 1 = C/(%Q)'
2-22
-------
EPA Channel Condition Project
These same kinds of relationships can be derived from at-a-station hydraulic
geometry equations. For example, Wilmont Creek, on the Colville Indian
Reservation in north central Washington, is triangular in shape at gaging site
WIL028, and W/D is a constant (Orsborn and Orsborn 1999c, Figure 2-8).
Plotting the cross-section in the traditional, distorted fashion is not as effective in
portraying the true channel geometry as is plotting at an undistorted scale
(Potyondy and Schmidt 1999).
Regional Relationships between Basin Characteristics (BC) and
Channel Characteristics (CC) Using Flow Characteristics (QC)
Strahler (1958) developed a Froude Number of the basin, which Orsborn (1981)
expanded into a regional streamflow equation
QX = C (A) (H)°-50 (2-8)
where QX = any characteristic regional flow such as QPF2, Q1F2, QAA or Q7L2
for a series of gaging stations. The regional coefficients have average values of
230,15, and 1.2 for QPF2, QAA and Q7L2 from the mid-coast of Oregon to south
central Alaska along the Pacific Coast.
Channel hydraulic geometry, either regional or at-a-station, gives us
relationships between flow and channel geometry. For example, using one such
relationship for water surface width at the average annual flow:
For the Dungeness River USGS Gage No. 12048000:
At-a-station: W = 59.5 (Q)0'049 (2-9)
Regional Eq.: W = 4.82 (QAA)0'47 (2-10)
For the Dungeness River, with QAA = 383 cfs, by Eq. (2-9), W = 80 ft, and by the
regional Eq. (2-10), W = 79 ft. The low value of the width exponent (b = 0.049)
indicates an essentially rectangular cross section.
If the regional Eq. 2-10 is rearranged
or
= (0.21W)2'13
QAA = 0.035 (W) 2'13 (2-11)
2-23
-------
EPA Channel Condition Project
Figure 2-7. Channel Width to Depth Ratios as a Function of Discharge Reduction
for Different Natural Channel Shapes. (Orsborn and Deane 1976). Data from
Chrostowski (1972).
10
8
6
20 40 60 100
LITTLE BRdSH
CREEK *3
( w/o)21 = l3.4/(%Q}°-"
V
10 20 40 60 100 10 20 40 60 100
10
I I I
UINTA RIVER #1
CURRANT CR. ff\
(W/0),|'7.25/(%0)°-4J
(h)
X
I-
O-
UJ
o
10
8
6
LITTLE BRUSH CR.4H
(O
i i i
I
10 20 40 60 100
RECTANGULAR SECT.
(TRAPEZOIDAL)
CURRANT CR.
4/l%or
lW/0>n « S.5/l%0)0
10
8
6
I
10 20 40 60 100 10 20 40 60 100
TRIANGULAR SECT. MIXED ASYMMETRIC
RECT. AND TRI. SECT.
J REFERENCE FLOV (Q0) RETAINED
2-24
-------
EPA Channel Condition Project
Figure 2-8. Cross-Sections for CCT Priority Low Flow Study
Sites, Wilmont Creek WIL028
(Orsborn and Orsborn 1999c)
Exaggerated Vertical Scale
0.0
-0.5
-1.0
-1.5
-2.0
-2.5 -»•
7\
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Distance Along Cross-Section (ft)
Equal Horizontal and Vertical Scales
0
5 6 7 8 9 1011
Distance Along Cross-Section (ft)
12 13 14 15
2-25
-------
EPA Channel Condition Project
Now we have equations for QAA in terms of basin characteristics (BC, Eq. 2-8
and in Figure 2-2a) and channel characteristics (CC, Eq. 2-11). Setting these two
equations equal to each other:
QAA = 15 [A (H) a50] °'91 = 0.035 (W) 2'13 (2-12)
and reducing to find W (or D, V, A and P by other equations)
W=(428(A)a9I(H)a50)a47
or
W (at QAA) = 17.2 (A) °'43 (H) °'22 (2-13)
This equation is good only for stations (or ungaged watersheds) which have the
coefficient of 15 in Eq. 2-8. The coefficients range from 20 to 1.7 across the
Olympic Peninsula. Inserting average annual precipitation (P) into Eq. 2-8 gives
QAA = 0.0193 (PBE) U4 (2-14)
where BE is the basin energy (AH °'50 ), and Eq. 2-14 is an average line for the
Olympic Peninsula gages. Combining Eq. 2-14 with Eq. 2-11 yields
QAA = 0.035 (W) 2'13 = 0.0193 (PA(H) a50) U4
Reducing these equalities gives
W (at QAA) = 0.75 (P) °'54 (A) °-54 (H) °'27 (2-15)
Also, the equation developed by Amerman and Orsborn (1987) for QAA on the
Olympic Peninsula states
QAA = 0.0032 (P) L6° Ab (2-16)
where P is the average annual precipitation (inches per year) and A is the
watershed area ( square miles). Although this equation was developed for USGS
gages on the Olympic Peninsula, similar equations have been developed for
other regions of Washington, Oregon, Idaho and Alaska. For example, in
northeastern Washington,
Q A A = 0.0025 (P) ' 'M Ab (2-17)
The exponent of 1.60 -1.64 on P allows for changes in QAA as a function of
changes in P.
2-26
-------
EPA Channel Condition Project
Now, if we write Eq. (2-16) equal to Eq. (2-11) for regional channel width, then
QAA = 0.0032 (P) L6° Ab = 0.035 (W)2'13
and W = [0.088 (P) L60Ab] °'47
and W (at QAA) = 0.32 (P) 0'75Ab M7 (2-18)
Rounding the area exponent 0.47 to 0.50 makes about a 6% difference for an area
of 10 sq. mi. and 14% difference at A = 100 sq. mi. This is a simpler expression to
use than Eq. 2-15 to estimate W at QAA. Similar expressions can be developed
for the other channel characteristics of D, V, A and P (wetted perimeter). But the
USGS summary form 9-207 only provides Q, W, D, V and A. To find the
wetted perimeter (P), one must obtain form 9-275 that covers the field
measurements of the discharges used to verify the calibration curve for the
station.
These other expressions for channel dimensions related to basin characteristics
at three characteristic flows (Q1F2, QAA and Q7L2), are developed in Part 3
for three regions in Washington.
The USGS considers gage records as excellent if 95% of the calibration
measurements are within about 5% of the true value. The grading goes to good
(10%), fair (15%) and poor for records greater than 15% from true. The
variability of the flow measurements over time would be a function of land-use
changes, gaging station channel changes, the stability and amount of
precipitation from year to year and whether or not the streamflow was
influenced by upstream storage or diversions.
Accounting for Changes in Channel Geometry
To account for the "condition of a channel" (poor or good) one must consider a
number of scales, or indexes, of evaluation:
• a channel may be "in balance" with its water and debris load, and still
not fit a cross-sectional template for the region due to geologic or
human geometric constraints;
• the main stream channel may be underfit due to excessive diversions
of flow out of the watershed, and the accumulation of sediment in the
mainstem from unaffected tributary sediment flows;
• the channel may be over- or under-sized due to a modified flow
regime caused by either a natural extended increase or decrease in
flow, or a regulated flow regime, or both; and
2-27
-------
EPA Channel Condition Project
• an historical mass wasting may have been deposited in a stream
valley, and the stream is now downcutting (as a function of the
existing flow regime).
It appears that we need a systematic method of analysis that may involve each of
the following steps, but to a varying degree:
• review of historical records of flow;
• a method of classification to put some geomorphic boundaries on the
site being investigated, and to help in the visualization of the site;
• a simple hydrologic analysis to estimate the characteristic flows at a
site (average low, average annual and average flood) Q7L2, QAA and
Q1F2, and major changes in these characteristic flows and in
precipitation over time;
• an abbreviated analysis of the channel hydraulic geometry of the site
to provide relationships of geometric characteristic (W, D, V, A and P)
as a function of discharge;
• regional channel hydraulic geometry models for comparison with the
present site geometry;
• an integrating analysis of how the W/D ratio, and other geometric
dimensionless ratios, change as a function of streamflow reduction; a
type of severity factor analysis which ties flow to geometric
characteristics which serve as analogs to water quantity and quality
parameters; and
• an evaluation of the history of major land-use and water-use changes
on the watershed.
The steps listed above will be explored in Part 3 - Methods of Analysis. We will
be analyzing slices of data about stream conditions, but a series of slices taken
over time should provide a more comprehensive evaluation of stream condition
and/or trends.
2-28
-------
EPA Channel Condition Project
An Example for Evaluating Effects of Land Use Change on
Channel Geometry
As a final example of a descriptive model let us put some numbers on the
problem of land use change and estimate some effects of a clear-cut on
downstream channel size. We can use the rational equation
QP = CIAb (2-19)
Where Qp is the peak flood (cfs) generated from a basin area (Ab) in acres, the
rainfall intensity (I) is in inches per hour and the coefficient (C) depends on the
type of land use and cover. Let us assume that the entire basin is uniformly
timbered and C = 0.10; and for the logged area (in the first couple of years after
logging), C = 0.8. For 1=2 in./hr on saturated ground, most of the rain is
available for runoff. The peak flow (Q ) is in units of either acre-in/hr or cfs,
because 1 acre-in/hr = 1 cfs. Under natural, pre-logging conditions, Ab = 20 sq.
mi. (12800 acres), I = 2 in/hr, and C = 0.10. Therefore, Q = 0.10 (2) (12800) = 2560
cfs, or 128 cfs per sq. mi. which is common for the Olympic Peninsula. The
overall basin and subbasins are shown in Figure 2-9, and the basin characteristics
are in Table 2-4.
Figure 2-9. Sketch of Example Basin with 5% (1 sq. mi.) clear-cut, (not to
scale).
2-29
-------
EPA Channel Condition Project
Table 2-4. Basin Characteristics and Peak Flows for Figure 2-9 under Pre-
Logging Conditions.
Point No.
Area (Ab)
(sq. mi.)
Area (Ab)
(acres)
Natural Q at Ft
(cfs)
1
2
3
4
5
4
8
4
12
20
2560
5120
2560
7680
12800
512
1024
512
1536
2560
After logging 1.0 sq. mi. (640 acres), C = 0.8 on that area and the flows would
adjust about as shown in Table 2-5.
Table 2-5. Comparison of Pre- and Post-Logging Peak Flows at the Check
Points*
Natural Q Post-Log Qp Increase in Percent
Point No. atPt. at Ft. Flow at Pt. Change
(cfs) (cfs) (cfs) (%)
1
2
3
4
5
512
1024
512
1536
2560
1408
1920
512
2432
3456
896
896
0
896
896
175
88
0
58
35
The flood flows are not strictly additive because of storage in the channel.
At the lower end of the logging the channel (assuming near bankfull conditions
for Q ) must now be subjected to 1408 cfs instead of the natural condition flow of
512 cfs, an increase of 175 %. Using the regional Olympic Peninsula equation for
Q1F2, the natural channel width would be about W = 3.44 (Q1F2)0'42 , where
Q1F2 = 0.73 (QPF2) (Amerman and Orsborn 1987)
If we assume our calculated peak flood is the average QPF2, then
W = 3.44(0.73QPF2)°'42
(Eq. 2-20)
The natural and logged channel potential widths would be as shown in Table
2-6.
2-30
-------
EPA Channel Condition Project
Table 2-6. Natural and Post-Logging Channel Widths for Estimated Average
Daily Flood Conditions
Point No.
1
2
3
4
5
Natural
Q1F2
(cfs)
375
750
375
1125
1875
Natural
Width, W
(ft)
41
55
41
66
82
Post-Log
Q1F2
(cfs)
1028
1400
375
1775
2522
Post-Log
Width, W
(ft)
63
72
41
80
92
% Change
inW
(%)
54
31
0
21
12
Based on the following assumptions, we have estimated the percent change in
channel width due to logging one sq. mile out of a 20 sq. mi. watershed on the
Olympic Peninsula:
1. Logging was equally distributed on both sides of this first-order
perennial stream;
2. Changes in runoff due to changes in land use were estimated by
changing the runoff coefficient (C) in the rational equation (Q = CIAb)
for the logged area from 0.1 to 0.8 after logging; and
3. The channels were formed in bank and bed materials that were freely
deformible.
The results of this example have demonstrated that for an increase in flood
runoff due to a land use change, we can expect the following:
1. Most of the channel widening will have the potential to take place in
the reach between Points 1 and 2;
2. There is a potential for about a 50 percent increase in channel width in
this reach;
3. Sediment deposited in the lower, flatter reaches will cause the channel
to widen and become shallower, but the cross-sectional area will stay
about the same. (e.g. S. F. Skokomish River in Amerman and Orsborn
(1987) and in Figure 3-9 on p. 3-15 );
4. Although the percent increase in the potential channel width decreases
as the flow moves downstream (54 to 12%), most of the channel change
will probably take place in the downstream, flatter reaches; and
2-31
-------
EPA Channel Condition Prefect
5. Regional channel geometry equations are useful in conducting
analyses of historical and current channel sizes.
We could have used the same kind of analogy for an urbanizing area. The runoff
coefficient, C, would have increased to about 0.90, and the floods would have
increased. But in this case, there would be less infiltration, and the low flows
would tend to decrease. For the logging operation the low flows may have
actually increased due to reduced transpiration by trees.
"The December 1964 flood on Coffee Creek (a small high-
gradient mountain stream in Trinity County, California)
was of rare frequency and unprecedented in historic time.
Erosion and deposition during the flood were catastrophic
and significantly changed the character of the valley."
"Within the valley, the preflood channel was commonly
filled, and new channels formed at entirely different
locations." (Stewart and LaMarche 1967)
2-32
-------
EPA Channel Condition Project
References for Part 2
Ackers, P. and F. G. Charlton. 1970. Dimensional analysis of alluvial channels
with special reference to meander length. Journal of Hydraulic Research.
8 (No. 3).
Amerman, K. and J.F. Orsborn. 1987. An analysis of streamflows on the Olympic
Peninsula in Washington State. Dept. of Civil and Environmental
Engineering. Washington State University, Pullman, WA (Two Volumes).
ASCE Task Committee on Hydraulics, Bank Mechanics and Modeling of River
Width Adjustment. 1998. River width adjustment. I: Processes and
mechanisms; II: Modeling. Journal of Hydraulic Engineering, Vol. 124,
No. 9. Paper No. 14412, pp. 881 - 902 and 903 - 917. Discussion and
closure: Feb. 2000, pp. 159-164. Quotation of TC Objectives by
permission.
Barnes, H. H. 1967. Roughness characteristics of natural channels. Water-Supply
Paper 1849. U. S. Geological Survey.
Chrostowski, H. P. 1972. Stream habitat studies on the Uinta and Ashley
National Forests. Forest Service Intermountain Region, USDA, Central
Utah Project, Ogden, UT.
Chitale, S. V. 1970. River channel patterns. Journ. Hydraulics Div., Proc. of the
American Society of Civil Engrs., Vol. 96, No. HY1, Jan.
Copp, H. C. and Rundquist, J. N. 1977. Hydraulic characteristics of the Yakima
River for anadromous fish spawning. Technical Report HY-2/77.
Albrook Hydraulics Laboratory, Washington State University, Pullman,
WA.
Emmett, W. W. 1975. The channels and waters of the upper Salmon River area,
Idaho. Professional Paper 870-A. U. S. Geological Survey.
Horton, R.E. 1945. Erosion development of streams and their drainage basins;
hydrophysical approach to quantitative morphology. Bull. Geol. Soc.
Amer., 56(3), pp. 275-370.
Langbein, W. B. and Durum, W. H. 1967. "The Aeration Capacity of Streams,"
U. S. Geological Survey, Circular 542.
Leopold, L. B. and Maddock, T., Jr. 1953. The hydraulic geometry of stream
channels and some physiographic implications. Professional Paper 252.
U. S. Geological Survey.
2-33
-------
EPA Channel Condition Project
Orsborn, J. F and F. D. Deane. 1976. Investigation into methods for developing a
physical analysis for evaluating instream flow needs; development of a
severity factor analysis for changes in stream channel flow and geometry.
Department of Civil and Environmental Engineering, Washington State
University, Pullman, WA.
Orsborn, J. F. 1976. Drainage basin characteristics applied to hydraulic design
and water-resources management. In Proceedings of the 7th
Geomorphology Symposium at SUNY Binghamton, NY; Geomorphology
and Engineering, pp. 141-171.
Orsborn, J. F. 1981. The development of new methods for hydrologic analysis in
the Siuslaw National Forest in the mid-coast of Oregon. File Notes,
Siuslaw National Forest. Corvallis, OR.
Orsborn, 1983. A regional hydrologic model for Southeast Alaska. American
Fisheries Society, Alaska Chapter Meeting, Nov. 14-17, Soldatna, AK.
Orsborn, J. F. and J. M Stypula. 1987. New models of hydrological and stream
channel relationships. Erosion and Sedimentation, Pacific Rim. Corvallis,
OR.
Orsborn, J. F. and M. T. Orsborn. 1997. An operational hydrologic system for the
Colville Indian Reservation. In three volumes: Vol. 1 - Summary,
Descriptive Text and References; Vols. 2 and 3 - Data Appendices.
Orsborn, J. F. and M. T. Orsborn. 1999a. Low flow assessment of the Lower
Elwha River—effects of diversions on channel geometry and fish habitat.
Lower Elwha Tribal Fisheries. Port Angeles, WA.
Orsborn, J. F. and M. T. Orsborn. 1999b. Hydraulic geometry number of data
points test for gage SPR 047. Data analysis file report. 9 p. Colville
Confederated Tribes.
Orsborn, J. F. and M. T. Orsborn. 1999c. Hydrologic aspects of the Colville
Indian Reservation low flow program. Nespelem, WA.
Orsborn, J.F. and J.M. Stypula. 2000. Solving for streamflow without using
Manning's equation. Stream Notes, USDA Forest Service. Stream
Systems Technology Center, Ft. Collins, CO.
Potyondy, J. and L. Schmidt. 1999. Why do we exaggerate stream Channel
Cross-section Plots? The case for true scale plotting. Stream Notes -
October, Stream Systems Technology Center. USDA Forest Service.
Rocky Mountain Research Station. Ft. Collins, CO.
2-34
-------
EPA Channel Condition Project
Rosgen, D. L. 1994. A classification of natural rivers. Catena: 22 (169 -199).
Rosgen, D. L. 1996. Applied river morphology. Printing Media Co.,
Minneapolis, MN.
Rouse, H. 1938. Fluid mechanics for hydraulic engineers. McGraw-Hill Book
Co. New York.
Smelser, M. G. and J. C. Schmidt. 1998. An assessment methodology for
determining historical changes in mountain streams. USDA Forest
Service. General Technical Report RMRS - GTR - 6. Ft. Collins, CO.
Stewart, J.H. and V.C. LaMarche, Jr. 1967. Erosion and deposition produced by
the flood of December 1964 on Coffee Creek, Trinity County, California.
USGS Prof. Paper 422-K. Washington, D.C.
Strahler, A. N. 1958. Dimensional analysis applied to fluvially eroded
landforms. Geological Society of America Bulletin. V. 60, p.p. 279-299.
Stypula, J. M. 1986. An investigation of several streamflow and channel form
relationships. M. S. thesis. Department of Civil and Environmental
Engineering, Washington State University, Pullman, WA.
Thomas, D. M and M. A. Benson. 1970. Generalization of streamflow
characteristics from drainage-basin characteristics. U. S. Geological
Survey Water-Supply Paper 1975. USGPO, Washington, DC.
U. S. Geological Survey. 1974-1980. Measurement summary sheets, Susitna
River near Gold Creek, Alaska. Gaging station No. 15292000.
Williams, G. P. 1978. Hydraulic geometry of river cross sections - theory of
minimum variance. USGS Professional Paper 1029. USGPO, Washington,
DC.
Yang, C.T. 1971. Potential energy and stream morphology. Water Resources
Research. 7(2), April.
2-35
-------
EPA Channel Condition Project
3. METHODS OF ANALYSIS
Introduction
In the evaluation of "channel condition" there are various levels of evaluation
that can be conducted, but all levels need a frame of reference, a benchmark, a
template, a basis of comparison. For the condition of a stream we need a
comparative reach of stream that is natural. Better yet, we need a series of
natural reaches from which more comprehensive regional models can be
developed. Although USGS gaging stations provide the best and most complete
flow and geometry data, some of the data has been distorted by either nature or
humans, or both. And, USGS sites are selected for their stability.
In Part 2, a series of example models were developed that related channel
characteristic (CC) width (W) to basin characteristics (BC) at average annual flow
(QAA). In Part 3, this analysis will be expanded to include: (1) the channel
cross-sectional dimensions (W, D, Ac and P); (2) at the three characteristic flows
(Q1F2, QAA and Q7L2); and (3) for three regions in Washington State: the
Olympic Peninsula (Amerman and Orsborn 1987); a region north, and east of
Lake Washington (Johnson and Orsborn 1997; Moscrip and Montgomery 1997);
and a region in northeastern Washington (Orsborn and Orsborn 1997,1999).
General Analytical Methods
In each region we use the following steps:
(1) develop a table of USGS gaging stations with their gage numbers,
basin characteristics, and their combined parameters (basin input,
PA; basin energy (BE) = A (H) aso; and P-BE); (the reliefs (H) were
not measured in the Puget Lowland region, because they were not
needed for that project);
(2) prepare a table of the width, depth, area and wetted perimeter
(where available) for each gage site at the three characteristic flows
(Q1F2, QAA and Q7L2);
(3) plot the regional hydraulic geometry graphs of channel
characteristics as a function of each characteristic flow; this step is
preceded by development of the at-a-station hydraulic geometry
equations for each USGS gage site;
3-1
-------
EPA Channel Condition Project
(4) develop and select the best regional models of characteristic flows
related to basin characteristics;
(5) equate the regional hydraulic geometry models to the best basin
model for each of the characteristic flows, and for each channel
geometric property (W, D, Ac and P); (note that velocity is not
included because it is not a geometric characteristic of the channel);
(6) check for the applicability of empirical relations between channel
dimensions in the field and basin characteristics;
(7) check to see if the channel dimensions can be estimated within
reasonable limits by developing regional models of channel
dimensions as a function of basin characteristics; and
(8) compare measured versus modeled channel dimensions; and
(9) decide on the project design approach.
Examples of using these nine steps towards evaluating channel conditions are
presented next for three regions in Washington State.
OLYMPIC PENINSULA REGION
The information to develop the analyses for the Olympic Peninsula gages is
given in:
Table 3-1. Basin characteristics for gaging stations;
Table 3-2. Calculated values of at-a-station hydraulic geometry for three
characteristic flows at USGS gaging stations;
Table 3-3. Channel properties at Q1F2, QAA and Q7L2 for USGS Gaging
Stations including W/D values;
Figure 3-1. Regional hydraulic geometry at Q1F2;
Figure 3-2. Regional hydraulic geometry at QAA;
Figure 3-3. Regional hydraulic geometry at Q7L2;
Figure 3-4. Regional hydraulic geometry: cross-sectional area versus Q7L2,
QAAandQlF2; and
Figure 3-5. USGS stream gaging stations on the Olympic Peninsula.
In the interest of brevity the basin, flow and channel characteristics are not
included in such detail for the Puget Lowland and Northeast Washington
regions. Only summary data, graphical relationships and regional equations that
were developed from the databases are presented.
Empirical relationships between channel and basin characteristics are examined
first, followed by the combination of basin characteristics with hydraulic
geometry, examples of which were developed in Part 2.
3-2
-------
EPA Channel Condition Project
Table 3-1. Basin Characteristics for Gaging Stations on the Olympic Peninsula
Province/
Stream Gage
Code
1.3
1.5
3.1
3.5
3.7
4.1
4.2
5.2
6.1
6.2
6.3
8.2
8.3
8.8
9.1
9.2
Station Name
Satsop River
Humptulips River
N.F. Quinault River
Hoh River
Soleduck River
Hoko River
East Twin River
Dungeness River
Siebert Creek
Snow Creek
L. Quilcene River
Duckabush River
Hamma Hamma River
S.F. Skokomish River
Goldsborough Creek
Kennedy Creek
USGS Gage
No.
12035000
12039000
12039300
12041000
12041500
12043300
12043430
12048000
12047500
12050500
12052000
12054000
12054500
12060500
12076500
12078400
Basin Relief,
H
(mi)
0.47
0.58
0.64
0.79
0.59
0.22
0.22
0.84
0.33
0.60
0.88
0.90
0.66
0.63
0.030
0.055
Drainage
Area, Ab
(sq. mi.)
299.0
130.0
74.1
208.0
83.8
51.2
14.0
156.0
15.5
11.2
19.6
66.5
51.3
76.3
39.3
17.4
Average
Annual
Precip., P
(in/yr)
128
155
200
167
99
124
90
62
41
43
51
113
110
153
84
59
COMBINED PARAMETERS
Basin Input
(PA)
(sq. mi-in/yr)
38272
20150
14820
34736
8296
6349
1260
9672
636
482
1000
7515
5643
11674
3301
1027
Basin Energy
(A)(H)°-5
(mi)2'5
205.0
99.0
59.3
184.9
64.4
24.0
6.6
143.0
8.9
8.7
18.4
63.1
41.7
60.6
6.8
4.1
P-BE
(in/yr)(mi)2'5
26238
15346
11856
30874
6372
2978
591
8865
365
373
938
7129
4584
9266
572
241
3-3
-------
EPA Channel Condition Project
Table 3-2. Calculated Values of At-a-Station Hydraulic Geometry for Three
Characteristic Flows at USGS Gaging Stations on the Olympic Peninsula.
Province/
Stream Gage
Code
1.3
1.5
3.1
3.5
3.7
4.1
4.2
5.2
6.1
6.2
6.3
8.2
8.3
8.8
9.1
9.2
Station Name
Satsop River
Humptulips River
N.F. Quinault River
Hoh River
Soleduck River
Hoko River
East Twin River
Dungeness River
Siebert Creek
Snow Creek
L. Quilcene River
Duckabush River
Hamma Hamma River
S.F. Skokomish River
Goldsborough Creek
Kennedy Creek
USGS Gage
No.
12035000
12039000
12039300
12041000
12041500
12043300
12043430
12048000
12047500
12050500
12052000
12054000
12054500
12060500
12076500
12078400
Q7L2
(cfs)
238.7
146.7
161.1
610.0
79.3
19.5
3.7
113.6
2.6
2.2
9.4
73.4
59.9
88.8
20.6
2.7
QAA
(cfs)
2035.0
1337.0
887.0
2028.0
621.0
408.0
64.7
393.0
17.1
16.2
48.6
422.0
364.0
741.0
116.0
61.3
Q1F2
(cfs)
18307
13393
6182
13053
6021
4739
595
1903
249
151
365
2965
2576
7083
778
563
W @
Q7L2
(ft)
212.6
160.1
110.2
106.4
80.1
52.2
14.8
75.4
12.8
15.4
19.9
65.4
79.2
168.7
33.5
11.4
D@
Q7L2
(ft)
1.20
0.96
2.17
2.48
1.82
0.63
0.55
1.31
0.48
0.36
0.62
1.01
0.83
1.00
0.80
0.35
V @
Q7L2
(fps)
0.93
0.95
0.67
2.30
0.54
0.60
0.46
1.14
0.42
0.40
0.75
1.11
0.91
0.53
0.75
0.62
A@
Q7L2
(ft2)
255.1
153.7
239.1
263.9
145.8
32.9
8.1
98.8
6.1
5.5
12.3
66.1
65.7
168.7
26.8
4.0
W @
QAA
(ft)
252.3
186.9
133.0
128.9
85.2
93.0
33.1
80.2
17.7
21.7
25.9
72.6
88.3
213.1
38.2
29.0
D@
QAA
(ft)
2.33
2.70
3.56
3.65
3.74
1.93
1.00
2.08
0.75
0.63
0.97
2.14
1.68
1.55
1.62
1.05
V @
QAA
(fps)
3.44
2.63
1.87
4.30
1.95
2.28
1.96
2.35
1.29
1.19
1.92
2.71
2.45
2.24
1.81
1.86
A @
QAA
(ft2)
587.8
504.6
473.5
470.5
318.6
179.5
33.1
166.8
13.3
13.7
25.1
155.4
148.3
330.3
61.9
30.4
W@
Q1F2
(ft)
300.8
219.6
164.6
173.7
91.2
148.2
61.6
86.8
27.8
31.6
35.7
81.6
99.3
273.1
44.1
56.4
D@
Q1F2
(ft)
4.61
7.99
6.25
6.62
8.29
4.79
1.59
3.73
1.38
1.17
1.67
4.95
3.59
2.50
3.52
2.28
V @
Q1F2
(fps)
13.14
7.60
5.98
11.31
7.98
6.71
6.08
5.87
6.44
4.06
6.06
7.31
7.20
10.38
4.77
4.05
A @
Q1F2
(ft2)
1386.7
1754.6
1028.8
1149.9
756.0
709.9
97.9
323.8
38.4
37.0
59.6
403.9
356.5
682.8
155.2
128.6
Notes:
Water surface width (W)
Mean hydraulic depth (D)
Mean velocity (V)
Cross-sectional area (Ac) = (WxD)
-------
EPA Channel Condition Project
Table 3-3. Channel Properties at Q1F2 (Average Flood), QAA (Average Annual Flow) and
Q7L2 (Average Low Flow) for Olympic Peninsula USGS Gaging Stations, including W/D values.
Province/
Stream Gage
Code
1.3
1.5
3.1
3.5
3.7
4.1
4.2
5.2
6.1
6.2
6.3
8.2
8.3
8.8
9.1
9.2
Station Name
Satsop River
Humptulips River
N.F. Quinault River
Hoh River
Soleduck River
Hoko River
East Twin River
Dungeness River
Siebert Creek
Snow Creek
L Quilcene River
Duckabush River
Hamma Hamma River
S.F. Skokomish River
Goldsborough Creek
Kennedy Creek
USGS Gage
No.
12035000
12039000
12039300
12041000
12041500
12043300
12043430
12048000
12047500
12050500
12052000
12054000
12054500
12060500
12076500
12078400
W@
Q1F2
(ft)
300
220
165
173
91
148
62
87
28
22
36
82
99
273
44
56
W@QAA
(ft)
252
187
133
129
85
93
33
80
18
22
26
73
88
213
38
29
W@
Q7L2
(ft)
212.6
160.1
110.2
106.4
80.1
52.2
14.8
75.4
12.8
15.4
19.9
65.4
79.2
168.7
33.5
11.4
D @ Q1F2
(ft)
4.6
8.0
6.2
6.6
8.3
4.8
1.6
3.7
1.4
1.2
1.7
5.0
3.6
2.5
3.5
2.3
D @
QAA
(ft)
2.3
2.7
3.6
3.6
3.7
1.9
1.0
2.1
0.8
0.6
1.0
2.1
1.7
1.6
1.6
1.0
D@
Q7L2
(ft)
1.20
0.96
2.17
2.48
1.82
0.63
0.55
1.31
0.48
0.36
0.62
1.01
0.83
1.00
0.80
0.35
W/D @
Q1F2
(--)
65.2
27.5
26.6
26.2
11.0
30.8
38.8
23.5
20.0
18.3
21.2
16.4
27.5
109.2
12.6
24.3
W/D @
QAA
(--)
109.6
69.3
36.9
35.8
23.0
48.9
33.0
38.1
22.5
36.7
26.0
34.8
51.8
133.1
23.8
29.0
W/D @
Q7L2
(--)
177.2
166.8
50.8
42.9
44.0
82.9
26.9
57.6
26.7
42.8
32.1
64.8
95.4
168.7
41.9
32.6
3-5
-------
EPA Channel Condition Project
1000
100
Q
i
100
1000
10000
100000
Two-Year, One-Day
Average Flood Flow (01F2), (cfs)
100
u
3
IU
10 •-
100
1000
10000
100000
Two-Year, One-Day
Average Flood Flow (Q1F2), (cfs)
0.1 J
100 1000 10000
Two-Year, One-Day
Average Flood Flow (Q1F2), (cfs)
100000
Figure 3-1. Regional Hydraulic Geometry: Width, Velocity and Depth
Versus the Two-Year, One-Day Average Flood Flows
for Olympic Peninsula Streams.
3-6
-------
EPA Channel Condition Project
1000
100
100 1000 10000
Average Annual Flow (QAA), (cfs)
10 100 1000 10000
Average Annual Flow (QAA), (cfs)
100 1000
Average Annual Flow (QAA), (cfs)
10000
Figure 3-2. Regional Hydraulic Geometry: Width, Velocity and Depth
Versus Average Annual Flows for Olympic Peninsula Streams.
3-7
-------
EPA Channel Condition Project
1000
100
10 100
Two-Year, Seven-Day
Average Low Flow (Q7L2), (cfs)
1000
10 100
Two-Year, Seven-Day
Average Low Flow (Q7L2), (cfs)
1000
10 100
Two-Year, Seven-Day
Average Low Flow (Q7L2), (cfs)
1000
Figure 3-3. Regional Hydraulic Geometry: Width, Velocity and Depth
Versus the Two-Year, Seven-Day Average Low Flows
for Olympic Peninsula Streams.
3-8
-------
EPA Channel Condition Project
10000
1000
UJ
cr
<
o 100
UJ
00
ob
oo
o
cc
o
Figure 3-4. Regional Hydraulic Geometry: Cross-Sectional Area
Versus Q7L2, QAA and Q1F2.
Ac = 2.56(Q7L2)°'82
R2 = 0.96
Ac = 0.52(Q1F2)°'83
R2 = 0.98
Q7L2
• QAA
A Q1F2
Power (Q7L2)
Power (QAA)
Power (Q1F2)
100 1000
Q7L2, QAA or Q1F2 (cfs)
10000
100000
3-9
-------
EPA Channel Condition Project
Figure 3-5. USGS Stream Gaging Stations on the Olympic Peninsula. Stations
used in hydraulic geometry analysis are listed in Table 3-1. USGS Gage Number,
and Province/Stream Gage Code (USGS Gage No. has prefix of 12-) (Amerman
and Orsborn 1987).
Provlne»/Str»cm Ooo* Cod*
Hydrotoaiod Provlne*
Number
Corridor Through
Original Boundary
3-10
-------
EPA Channel Condition Project
Width, Depth and Channel Area at Q1F2
Width, depth and channel area data from Table 3-2 for 16 USGS gaging stations
have been plotted in Figures 3-6, 3-7 and 3-8 versus the average annual amount
of water entering the basins (PAb). Other graphs of these channel characteristics
were plotted against just basin area (Ab). Figure 3-9 is an example of W vs. Ab, at
Q7L2, which was one of the better graphs of W vs. Ab. The data points were too
widely scattered to be of use. Therefore, PAb was chosen as the common
independent variables for comparison of W, D and Ac at Q1F2, QAA and Q7L2.
3-11
-------
EPA Channel Condition Project
Figure 3-6. Channel Characteristics versus PAb for Olympic Peninsula Streams at Q1F2.
10000
1000
Ac (ft2)
W (ft)
D (ft)
100
10
100
I I I I rpi
,0.879
y = 0.177x
R2 = 0.922
.0.493
y = 1.383X
R2 = 0.820
0.386
y = 0.128x
R2 = 0.737
1000 10000
PAb (in/yr • mi2)
100000
* W
• D
A Ac
1 Power (W)
• Power (D)
•Power (Ac)
-------
EPA Channel Condition Project
Figure 3-7. Channel Characteristics versus PAb for Olympic Peninsula Streams at QAA.
1000
Ac (ft2)
W (ft)
D (ft)
100
10
0.1
100
7*
i
1000 10000
PAb (in/yr • mi2)
,0.931
y = 0.043x
R2 = 0.961
.0.573
y = 0.524x
R2 = 0.902
.0.358
y = 0.082x
R2 = 0.810
100000
* W
• D
A Ac
1 Power (W)
• Power(D)
•Power (Ac)
3-13
-------
EPA Channel Condition Project
Figure 3-8. Channel Characteristics versus PAb for Olympic Peninsula Streams at Q7L2.
1000
AC (ft2)
W (ft)
100
100
1000
10000
PAb (in/yr • mi2)
= 0.007X1'028
R2 = 0.926
0671
y = 0.171X1
R2 = 0.906
y = 0.043x°'355
R2 = 0.718
100000
• W
• D
A Ac
Power (W)
Power (D)
Power (Ac)
-------
EPA Channel Condition Project
Figure 3-9. Channel Width versus Ab for Olympic Peninsula Streams at Q7L2.
1000
100
Width, W (ft)
10
\
SFS
10
100
Ab (in/yr • mi2)
3-15
.0.90
y = 1.46x
R2 = 0.84
SFS = S. Fork Skokomish
1000
-------
EPA Channel Condition Project
Combined Relationships of Channel and Basin Characteristics for the
Olympic Peninsula
In this section, channel geometric characteristics of W, D and Ac developed in the
regional hydraulic geometry analyses (Figures 3-1,3-2,3-3 and 3-4), are
combined with equations for Q1F2, QAA and Q7L2 as a function of basin
characteristics.
FLOOD FLOWS
The regional hydraulic geometry equations for W, D, and Ac at Q1F2 are:
W = 2.40 (Q1F2) °'47 (3-1)
D = 0.22 (Q1F2) °'36 (3-2)
Ac = 0.52 (Q1F2) °'83 (3-3)
For the mean daily flood as a function of basin characteristics we developed two
relationships:
In Figure 3-10:
Q1F2 = 2.89 (Ab) L74 (3-4)
That is a relationship in which the dimensions are not the same on both sides of
Eq. 3-4..
In Figure 3-11:
Q1F2 = 0.27 (PAb) L05 (3-5)
in which the dimensions on both sides of the equation are the same
(L 3 / T) (assuming the 1.05 exponent is really 1.00).
Setting Eq. 3-5 equal to Eqs. 3-1, 3-2, and 3-3 for W, D, and Ac gives for the
average daily flood (Q1F2) for these 16 Olympic Peninsula streams:
Width: W = 1.30 (PA) " (3-6)
Depth: D = 0.14 (PA) °'38 (3-7)
Area: Ac = 0.18 (PA) °'87 (3-8)
The values of W, D and Ac estimated by Eqs. 3-6, 3-7 and 3-8 are listed in Table 3-
4, with the values calculated from the at-a station hydraulic (transferred from
Table 3-2). The combined hydraulic geometry - basin characteristic values of W,
D and Ac are compared with the estimated values in Figure 3-12.
3-16
-------
EPA Channel Condition Project
Figure 3-10. Q1F2 versus Ab for Olympic Peninsula Streams.
100000
10000
Q1F2 (cfs)
1000
100
1 1
<~»j
./
A
— i
-f-
f
f
k
/
— /
i
1
f
I
y
\
t
{
F
>
f
\2
A
te
/•
/
c
>
f
n -t r
0.92
/ !
r
X
>
< H
Dun
C
oh
ger
Sa
es
ts
s
o
3
^n -i ^
A Q1F2
X Stations Not Used
Power (Q1F2)
Ab (mr
3-17
-------
EPA Channel Condition Project
Figure 3-11. Q1F2 versus PAb for Olympic Peninsula Streams.
100000
10000
Q1F2 (cfs)
1000
100
100
1000
10000
1.05
y = 0.27x
R2 = 0.94
100000
A Q1F2
• Power (Q1F2)
PAb (in/yr • mi2)
-------
EPA Channel Condition Project
Table 3-4. Channel Width, Depth, and Area Comparison at Q1F2 for Olympic Peninsula Streams
C:
E:
W = C(PA)E
1.3
0.5
C:
E:
D = C(PA)b
0.14
0.38
C:
E:
Ac = C(PA)b
0.18
0.87
Table 3-2
Hyd. Geom.
Equation 3-6
Estimated
Province/
Stream Gage
Code
1.3
1.5
3.1
3.5
3.7
4.1
4.2
5.2
6.1
6.2
6.3
8.2
8.3
8.8
9.1
9.2
Station Name
Satsop River
Humptulips River
N.F. Quinault River
Hoh River
Soleduck River
Hoko River
East Twin River
Dungeness River
Siebert Creek
Snow Creek
L Quilcene River
Duckabush River
Hamma Hamma River
S.F. Skokomish River
Goldsborough Creek
Kennedy Creek
USGS Gage
No.
12035000
12039000
12039300
12041000
12041500
12043300
12043430
12048000
12047500
12050500
12052000
12054000
12054500
12060500
12076500
12078400
Q1F2
(cfs)
18307
13393
6182
13053
6021
4739
595
1903
249
151
365
2965
2576
7083
778
563
Basin Input
(PAb)
(sq. mi-in/yr)
38272
20150
14820
34736
8296
6349
1260
9672
636
482
1000
7515
5643
11674
3301
1027
W @ Q1F2
(ft)
301
220
165
174
91
148
62
87
28
32
36
82
99
273
44
56
WPred
(ft)
254
185
158
242
118
104
46
128
33
29
41
1 13
98
140
75
42
Table 3-2
Hyd. Geom.
Equation 3-7
Estimated
D @ Q1F2
(ft)
4.61
7.99
6.25
6.62
8.29
4.79
1.59
3.73
1.38
1.17
1.67
4.95
3.59
2.50
3.52
2.28
DPred
(ft)
7.72
6.05
5.38
7.44
4.32
3.90
2.11
4.58
1.63
1.46
1.93
4.16
3.73
4.92
3.04
1.95
Table 3-2
Hyd. Geom.
Equation 3-8
Estimated
A @ Q1F2
(ft2)
1387
1755
1029
1150
756
710
98
324
38
37
60
404
357
683
155
129
A Pred
(ft)
1747
1000
765
1606
462
366
90
528
49
39
73
424
330
622
207
75
3-19
-------
EPA Channel Condition Project
Figure 3-12. Q1F2 Predictions versus USGS Values for Width, Depth and Channel Area
for Olympic Peninsula Streams
10000
1000
Predicted Values
Ac (ft2)
W (ft) 10°
D (ft)
10
* W
• D
A Ac
— 1:1 Line
100
1000
10000
Hyd. Geom. Values W (ft), D (ft), Ac (ft2)
-------
EPA Channel Condition Project
AVERAGE ANNUAL FLOWS
Example equations for channel width at average annual flows were developed in
Part 2. Equation 2-18 is
W (at QAA) = 0.32 (P) °'75 (Ab) °'47 (2-18, 3-9)
For depth we combine the regional hydraulic geometry equations from Figure 3-
2 with the Olympic Peninsula average annual flow equation (Eq. 2-16) so that
D = 0.29 (QAA) °'32 and
QAA = 0.0032 (P) L6° Ab (2-16, 3-10)
which reduces to
D = 0.29 (0.0032 (P) ^A,,)032
or
D = 0.046 (P)°-51(Ab)032 (3-11)
For channel area, Ac, using the regional geometry equation from Figure 3-4 at
QAA, in combination with Eq. 3-10, yields
Ac = 1.16 (QAA) °'83 (From Figure 3-4)
QAA = 0.0032 (P) ^A,, (substitute above) (3-10)
and
Ac = 0.0098 (P)1>33 (Ab) °'83 (3-12)
For comparison, the hydraulic geometry at-a-station values of W, D and Ac at
QAA and the estimated values from Eqs. 3-9,3-11 and 3-12, are listed in Table 3-5
and plotted in Figure 3-13.
3-21
-------
EPA Channel Condition Project
Table 3-5. Channel Width, Depth, and Area Comparison at QAA for Olympic Peninsula Streams
v\
C:
E1:
E2:
Table 3-2
Hyd. Geom.
' = C(P)E1(Ab)Ei
0.32
0.75
0.47
Equation 3-9
Estimated
D = C(P)E1(Ab)"
C: 0.046
E1: 0.51
E2: 0.32
Table 3-2 Equation 3-11
Hyd. Geom. Estimated
A = C(P)E1(Ab )E2
C: 0.0098
E1: 1.33
E2: 0.83
Table 3-2 Equation 3-12
Hyd. Geom. Estimated
Province/
Stream Gage
Code
1.3
1.5
3.1
3.5
3.7
4.1
4.2
5.2
6.1
6.2
6.3
8.2
8.3
8.8
9.1
9.2
Station Name
Satsop River
Humptulips River
N.F. Quinault River
Hoh River
Soleduck River
Hoko River
East Twin River
Dungeness River
Siebert Creek
Snow Creek
L. Quilcene River
Duckabush River
Hamma Hamma River
S.F. Skokomish River
Goldsborough Creek
Kennedy Creek
USGS Gage
No.
12035000
12039000
12039300
12041000
12041500
12043300
12043430
12048000
12047500
12050500
12052000
12054000
12054500
12060500
12076500
12078400
QAA
(cfs)
2035.0
1337.0
887.0
2028.0
621.0
408.0
64.7
393.0
17.1
16.2
48.6
422.0
364.0
741.0
116.0
61.3
Average
Annual
Preclo.. P
(in/yr)
128
155
200
167
99
124
90
62
41
43
51
113
110
153
84
59
Drainage
Area, Ab
(sq. mi.)
299.0
130.0
74.1
208.0
83.8
51.2
14.0
156.0
15.5
11.2
19.6
66.5
51.3
76.3
39.3
17.4
we QAA
(ft)
252.3
186.9
133.0
128.9
85.2
93.0
33.1
80.2
17.7
21.7
25.9
72.6
88.3
213.1
38.2
29.0
WPred
(ft)
177.5
138.5
128.7
182.7
80.5
75.6
32.3
75.9
18.8
16.7
24.7
79.7
69.2
106.8
49.9
26.1
D @ QAA
(ft)
2.33
2.70
3.56
3.65
3.74
1.93
1.00
2.08
0.75
0.63
0.97
2.14
1.68
1.55
1.62
1.05
DPred
(ft)
3.39
2.86
2.72
3.45
1.98
1.89
1.06
1.90
0.73
0.68
0.89
1.96
1.78
2.40
1.43
0.92
A @ QAA
(ft2)
587.8
504.6
473.5
470.5
318.6
179.5
33.1
166.8
13.3
13.7
25.1
155.4
148.3
330.3
61.9
30.4
A Pred
(ft)
705.7
456.0
401.4
743.8
174.5
156.4
34.8
156.8
13.3
10.8
21.6
171.7
133.6
288.0
74.8
23.8
-------
EPA Channel Condition Project
Figure 3-13. QAA Predictions versus USGS Values for Width, Depth and Channel Area
for Olympic Peninsula Streams
100
Predicted Values
Ac (ft2)
W (ft) 1 °
D (ft)
1
0.1
/
/
/
/
-i r
7
•
/-
H
•
/
p
-
A
•
-
/
7
/
/A
*
4
»
(/
/
— i
V
, h
>
/
•
fr
/
Ts^
&
-
^
/)
5
-/
-^
X
^
>
W
• 0
A Ac
1:1 Line
0.1
1 10 100
Hyd. Geom. Values W (ft), D (ft), Ac (ft2)
1000
3-23
-------
EPA Channel Condition Project
7-DAY AVERAGE LOW FLOWS
As shown in Figure 3-14, with data from Tables 3-1 and 3-2:
Q7L2 = 0.0067 (PBE)a'06 (3-13)
and from Figures 3-3 and 3-4
W = 8.37 (Q7L2) °'52 (3-14)
D = 0.31 (Q7L2) °'30 (3-15)
Ac = 2.56 (Q7L2) °'82 (3-16)
Substituting Eq. 3-13 for Q7L2 in the three equations just above yields
W = 8.37 (0.0067 (PBE) L06) °'52
or
W = 0.62 (PBE) °'55 (3-17)
where PBE is average annual precipitation (P) multiplied by Basin Energy = Ab
(H)°-50.
Substituting these terms into Eq. 3-17 gives
W = 0.62 (P) °'55 (Ab) °'55 (H) °'29 (3-18)
The depth equation becomes
D = 0.31 (0.0067 (PBE) L06) °'30
or
D = 0.069 (P) °'32 (Ab) °'32 (H) al6 (3-19)
For the channel area (Ac) Eq. 3-16 combines with Eq. 3-13 to give
Ac = 2.56 (0.0067 (PBE) W6) °'82
or
Ac = 0.042 (P) °'87 (Ab) °'87 (H) °'46 (3-20)
The calculated and estimated values of W, D and Ac for Q7L2 are listed in Table
3-6 and plotted in Figure 3-15.
3-24
-------
EPA Channel Condition Project
Figure 3-14. Q7L2 versus PBE for Olympic Peninsula Streams.
1000
100
Q7L2 (cfs)
10
DO
A>
^
/
^
/
,ft
/
/
/
i
?
/
1
K
n
/
nn
/,
/
L
/
Ix
\
A
^
^
/
^
-1
?
>
r
^
J
r
/
jr&
inn
^
/
/A
V
_i —
1
-
-}
f
1
J
3
.vJOb/X
2 _ f) QQQR
A Q7L2
•»«••.••»••• P/-\\A/Qr /T^7l O\
rOWei (LJ/ L^j
PBE (in/yr • mi2 5)
3-25
-------
EPA Channel Condition Project
Table 3-6. Channel Width, Depth, and Area Comparison at Q7L2 for Olympic Peninsula Streams
W = C(PA)E'(H)EJ
C: 0.62
E1: 0.55
E2: 0.29
Table 3-2 Equation 3-18
Hyd. Geom. Estimated
D = C(PA)E1(H)E2
C: 0.069
E1: 0.32
E2: 0.16
Table 3-2 Equation 3-19
Hyd. Geom. Estimated
A = C(PA)E1(H)E2
C: 0.042
E1: 0.87
E2: 0.46
Table 3-2 Equation 3-20
Hyd. Geom. Estimated
Province/
Stream Gage
Code
1.3
1.5
3.1
3.5
3.7
4.1
4.2
5.2
6.1
6.2
6.3
8.2
8.3
8.8
9.1
9.2
Station Name
Satsop River
Humptulips River
N.F. Quinault River
Hoh River
Soleduck River
Hoko River
East Twin River
Dungeness River
Siebert Creek
Snow Creek
L. Quilcene River
Duckabush River
Hamma Hamma River
S.F. Skokomish River
Goldsborough Creek
Kennedy Creek
USGS Gage
No.
12035000
12039000
12039300
12041000
12041500
12043300
12043430
12048000
12047500
12050500
12052000
12054000
12054500
12060500
12076500
12078400
Q7L2
(cfs)
238.7
146.7
161.1
610.0
79.3
19.5
3.7
113.6
2.6
2.2
9.4
73.4
59.9
88.8
20.6
2.7
Basin Input
(PA)
(sq. mi-in/yr)
38272
20150
14820
34736
8296
6349
1260
9672
636
482
1000
7515
5643
11674
3301
1027
Basin Relief, H
(mi)
0.47
0.58
0.64
0.79
0.59
0.22
0.22
0.84
0.33
0.60
0.88
0.90
0.66
0.63
0.030
0.055
W @ Q7L2
(ft)
212.6
160.1
110.2
106.4
80.1
52.2
14.8
75.4
12.8
15.4
19.9
65.4
79.2
168.7
33.5
11.4
WPred
(ft)
165
123
107
182
76
49
20
92
16
16
27
81
64
94
19
12
D @ Q7L2
(ft)
1.20
0.96
2.17
2.48
1.82
0.63
0.55
1.31
0.48
0.36
0.62
1.01
0.83
1.00
0.80
0.35
DPred
(ft)
1.79
1.51
1.39
1.89
1.14
0.89
0.53
1.27
0.46
0.46
0.62
1.18
1.02
1,28
0.53
0.40
A @ Q7L2
(«2)
255.1
153.7
239.1
263.9
145.8
32.9
8.1
98.8
6.1
5.5
12.3
66.1
65.7
168.7
26.8
4.0
APred
(ft)
288
182
145
336
85
43
10
114
7
7
16
94
64
117
10
C
-------
EPA Channel Condition Project
Figure 3-15. Q7L2 Predictions versus USGS Hydraulic Geometry Values for Width, Depth and
Channel Area for Olympic Peninsula Streams
100
Predicted Values
Ac(ft2)
W (ft) 10
D (ft)
1
0.1
/
/
—.
A
H
d
r
ii
n
,
r
/
•
k
/
/
A
r
-j
^
L
n
4
^
/
A
^
/
f
4,
i^^
IX
A
iT
>A
^
>
/
^
A \A/
• D
A Ac
I: I Line
0.1
1 10 100
Hyd. Geom. Values W (ft), D (ft), Ac (ft2)
1000
3-27
-------
EPA Channel Condition Project
Discussion of Olympic Peninsula Empirical and Combined Relationships of
Channel, Flow and Basin Characteristics
Figures 3-6, 3-7, 3-8 and 3-9: (Data in Tables 3-1 and 3-2) The values of
Ac, W, and D are plotted against PAb (the average annual amount of
precipitation entering the basins) in the first three figures for the three
characteristic flows. For an example of simpler graphs, the W at Q7L2 is plotted
against just Ab in Figure 3-9. It compares with W versus PAb in Figure 3-8:
Figure 3-9: W = 1.46 (Ab)a90; R2 = 0.84
Figure 3-8: W = 0.17 (PAb) °'67; R2 = 0.91
The graph in Figure 3-8 improves the correlation by adding (P) and spreads the
data along the PAb axis.
The Olympic Peninsula gages represent a wide range of stream hydrology and
geomorphology, and variable periods of record (Amerman and Orsborn 1987).
Average annual precipitation (P) ranges from 40 to 200 inches per year on the
USGS-gaged basins, and the accuracy of these values is limited by the low
number of precipitation gages and snow courses (Williams et al 1985a, 1985b,
and Williams and Pearson 1985).
Many of the stream gages on the Olympic Peninsula are situated in bedrock or
large boulder cross-sections. The large rock and bedrock conditions, for
example, exist at the Soleduck, Dungeness, Duckabush and Hoh gages, all of
which plot below the width graph in Figure 3-6.
Also, some readily deformible gaging station cross-sections exist, such as the S. F.
Skokomish, which has a width of 273 ft at Q1F2 (farthest point from the W vs. PA
line in Figure 3-6). As a result of the excess sediment caused by logging the
depth D has reduced to 1.0 ft (in Figure 3-8 for Q7L2). The low flow width has
increased to 168 ft from a pre-logging width of about 90 ft.
Therefore, even though the Olympic Peninsula gages display quite a bit of
variation as a function of (PAb ) in Figures 3-6,3-7 and 3-8, knowledge of the
geomorphology of gage sites assists in examining their plotting positions with
respect to the average equations. Note the plotting position of the S.F.
Skokomish in Figure 3-9 (W vs. Ab).
3-28
-------
EPA Channel Condition Project
Comparing COMBINED equations and the EMPIRICAL relations for Q1F2:
At Q1F2
COMBINED
Eq. No.
EMPIRICAL
Fig. No.
Width: W = 1.30 (PA) °-50
Depth: D = 0.14 (PA) °'38
Area: Ac = 0.18 (PA) °'87
3-6
3-7
3-8
W = 1.38 (PA) °'49
D= 0.13 (PA)0'39
Ac = 0.18 (PA) °'88
3-6
3-6
3-6
These equations are all very similar because Q1F2 was equal to 0.27(PAb)l'05
from Fig. 3-11, and this equation of 0.27 (PAt,)1'05 was substituted into the
hydraulic geometry equations. The comparison of the values of W, D and Ac at
Q1F2 are given in Table 3-4 the last six columns, and in Figure 3-12. Some values
of W and Ac are quite close, but depth, as Williams (1978, Part 2) discussed, "the
exponent of depth (in hydraulic geometry) (is) fair." One would expect the W, D
and Ac values at Q1F2 to display quite a bit of scatter over such a large
hydrologically diverse region with highly variable geology and variable periods
of record.
For average annual flow (QAA) the COMBINED and EMPIRICAL relations
are:
At QAA
COMBINED
Eq. No.
EMPIRICAL
Fig. No.
W = 0.32(P)°-75Aba47
D = 0.046 (P)a51(Ab)a32
Ac = 0.0098 (P) 133 (Ab) °'83
3-9
3-11
3-12
W = 0.52 (PA) a57
D= 0.08 (PA)0'36
Ac = 0.04 (PA) LM
3-12
3-12
3-12
For QAA, the combined equations account for the variation in runoff as a
function of (P) L6° , whereas for Q1F2 (and for Q7L2) the equations do not
account for this. But for low flow Q7L2 = 0.067 (PBE)1'06, which says the low
flow is a function (almost to the first power) of average annual precipitation (P),
the drainage area of the basin (Ab) and the relief (H)a5°. The area and relief
(AH°'5°) were developed early in Part 3, Methods of Analysis, and as part of the
Froude No. of the watershed (Eq. 1-2 and Eq. 2-4).
3-29
-------
EPA Channel Condition Project
Comparing the COMBINED and EMPIRICAL EQUATIONS FOR Q7L2:
At Q7L2
COMBINED
Eq. No.
EMPIRICAL
Fig. No.
W = 0.62(P)°-55(Ab)°-55(H)0-29
D = 0.069 (P) °'32 (Ab) °'32 (H) °'16
Ac = 0.042 (P)a87(Ab)°-87(H)°-46
3-18
3-19
3-20
W = 0.17 (PA) °'55
D = 0.04 (PA) °'32
Ac = 0.01 (PA) °'87
3-8
3-8
3-8
The influence of the geology at the gaging sites on the estimating capability of
the combined equations has been discussed. A comparison of common periods
of record and site visits to all the gages not mentioned might improve our
reasoning. But, the combined equations do improve our estimates of channel
characteristics, even in this diverse region of about 8000 sq. mi.
The COMBINED equations are compared with the hydraulic geometry values of
W, D, and Ac at Q7L2 in Table 3-6 and Figure 3-15.
3-30
-------
EPA Channel Condition Project
PUGET LOWLAND REGION
Database and Empirical Relationships
There is no doubt that the streams in this region are responding to the
urbanization of their watersheds (Moscrip and Montgomery 1997). These
authors examined the influences of urbanization on: increases in flood flows
(during a period of gradual decline in annual precipitation), and the attendant
decrease in fish production, probably due to the increase in floods, and the more
frequent and deeper scour of spawning beds.
Johnson and Orsborn (1997) used the following USGS-gaged streams for their
preliminary design of the restoration of a natural, meandering channel in North
Creek at the new University of Washington campus in Bothell: Quilceda,
Woods, North, Swamp, Mercer, Griffin and two sites on Issaquah Creek. Mercer
and Swamp Creeks, plus four others, were used by Moscrip and Montgomery
(1997).
Most of the basin, channel and streamflow data for the eight Puget Lowland
gages used in the North Creek restoration design are in Table 3-7. Data for Table
3-7 came from the USGS records for the gages (Form 9-207 and Williams, Pearson
and Wilson 1985b). The data were arranged in common periods of record so that
any changes in channel dimensions could be noted. For a preliminary regional
analysis, channel area (Ac) was plotted as a function of basin area (Ab) for
average daily floods (Q1F2) in Figure 3-16. The letters denote the stream name
from Table 3-7, with IU denoting the Issaquah Creek upstream gage, and ID the
downstream gage. There seemed to be a fairly good relationship among the
upper data points, but Upper Issaquah, Griffin and North Creek fell well below
the upper line.
This is made more obvious in Figure 3-17 where the graph has been drawn using
only the upper five data points, and R2 has increased from 0.70 to 0.97. This
relationship, with three undersized cross-sectional channel areas, was found to
hold true at the average annual flow (QAA) in Figure 3-18. This was still true
when channel area (Ac) was plotted as a function of PAb , the average annual
basin inputs in Figure 3-19.
Now the question became, are the undersized channels narrow and deep, or
shallow and wide, or a mixture? Next, the depth at Q1F2 was plotted versus
basin area in Figure 3-20. Now Quilceda Creek appears as a deep channel
(which may mean it is incised), but North, Griffin and Issaquah (U) are still
below the main graph. In Figure 3-21, the width at Q1F2 was plotted against
basin area, and shows Quilceda Creek has a narrower width to go with its
greater depth to give an average area at Q1F2.
3-31
-------
EPA Channel Condition Project
Table 3-7. Basic Streamflow, Channel and Basin Data for the Puget Lowland Region
Station
Units:
Mercer
fssaquah-U
Issaquah-D
North Cr
Swamp Cr
Woods Cr
Griffin Cr
Quilceda Cr
Q1F2
cfs
164.7
493.0
1103.6
255.8
290.7
1075.7
336.1
144.2
QAA
cfs
21.9
69.7
143.9
36.4
33.8
154.5
40.3
25.6
Q1 F2/QAA
Ratio
7.5
7.1
7.7
7.0
8.6
7.0
8.3
5.6
Q7L2
cfs
5.2
14.9
27.7
6.3
4.0
18.7
3.2
4.1
Ab P
sq mi in/yr
12 43
27 66
55 53
25 38
23 39
56 48
17 53
15 37
P*A
smi/y
516
1782
2915
950
897
2688
901
555
c
Years
55-58
68-71
87-90
55-58
68-71
87-90
55-58
68-71
87-90
55-58
68-71
55-58
55-58
W
ft
31
27
21
54
47
47
20
25
40
47
44
35
20
Q1F2
D W/D
ft
2.3
2.9
2.2
1.5
4.0
4.3
2.3
2.4
2.8
3.7
3.7
1.5
3.7
13.5
9.3
9.5
36.0
11.8
10.9
8.7
10.4
14.3
12.7
11.9
23.3
5.4
I
Ac
ft'ft
71.0
75.1
47.6
85.0
180.1
196.4
47.0
62.3
103.9
165.6
157.6
53.4
73.7
W
ft
18
15
17
33
43
36
18
21
27
42
37
23
16
QAA
D
ft
0.9
1.3
1.0
0.9
1.4
1.6
1.0
1.0
1.1
1.5
1.5
0.8
1.2
W/D Ac
ft'ft
20.0 16.6
11.5 18.7
17.0 17.7
36.7 28.4
30.7 57.5
22.5 55.5
18.0 19.2
21.0 22.2
24.5 29.8
28.0 61.6
24.7 54.2
Ac Ratio
4.3
4.0
2.7
3.0
3.1
3.5
2.4
2.8
3.5
2.7
2.9
28.8 18.9 2.8
13.3 20.1
3.7
Nomenclature:
Q1F2 one-day average flood flow with 2-yr recurrence interval (Rl)
QAA average annual flow for period of record (FOR)
Q7L2 7-day average low flow, 2-yr Rl
Ab basin area
P average annual precipitation on basin
PA average inflow to basin (sq mi-in/yr)
W channel water surface width
D mean flow depth, Ac/W
Ac cross-sectional flow area of channel
-------
EPA Channel Condition Project
Figure 3-16. NORTH CREEK REGIONAL ANALYSIS:
Channel area at Q1F2 vs. Basin Area (All Points)
1000
cr
•!£,
CM
(0
o
n
o>
v
c
(0
£
O
100
y = 10.78X
R2 = 0.70
.0.66
Ac Ave
Power (Ac Ave)
100
BASIN AREA, Ab (sq mi)
3-33
-------
EPA Channel Condition Project
Figure 3-17. NORTH CREEK REGIONAL ANALYSIS:
Channel area at Q1F2 vs. Basin Area (Selected Points)
1 WU
£"
a-
u.
1—
a
0
o 100
-
0)
i_
<
"aJ
c
(0
0
10 -
lx
p
^
Q ^,
H _a^*
^ ++*
P
0
G
*^
X
0
N
x
X*
IU
rf
X
ID
•tl
W
~*
'
^>
»^
*
y = 15.95X0'59
R2 = 0.97
* Ac Ave
- .
v AC Ave-Low
~~ rowei (Au Ave)
10
BASIN AREA, Ab (sq mi)
100
-------
EPA Channel Condition Project
Figure 3-18. NORTH CREEK REGIONAL ANALYSIS:
Channel area at QAA vs. Basin Area (Selected Points)
i uuu -
^
cr
v^
0
+rf
(0
< 100 H
(0
o>
<
"55
c
c
(0
0
10 -
/
i
p
|\
/
< vx
G
S8^
s
O
O
N
*
X
IU
V
ix
V
^
1
X
D
s
r-
r
^
* Ac Ave
O Ac Ave-Low
n i n ft N
rOWei ^AC AV6)
-. * . 0 77
y = 2.61x
R2 = 1.00
10
BASIN AREA, Ab (sq mi)
100
3-35
-------
EPA Channel Condition Project
Figure 3-19. North Creek Regional Analysis, Channel Area at QAA for P*A
1000
s
O
4-1
(0
o
o
0)
0)
c
c
re
O
100
100
w
1000
P»A (sq mi-in/yr)
10000
0.65
y = 0.32x'
R2 = 0.99
* Ac @QAA
A Ac @QAA-Low
Power (Ac @QAA)
-------
EPA Channel Condition Project
Figure 3-20: North Creek Regional Analysis of Depth at Q1F2.
CM
n
Q
Q.
HI
Q
10
100
.0.365
y = 0.916x
R2 = 0.944
* DOQ1F2
A D@Q1F2-Low
Power (DOQ1F2)
BASIN AREA, Ab (sq mi)
3-37
-------
EPA Channel Condition Project
100
CM
LJL
5
13
10
Figure 3-21: North Creek Regional Analysis of Channel Width
at Bankfull Flow (Q1F2)
10
y = 3.814x0781
R2 = 0.937
U
z:
,x_
A
V
-s
ID (68-71)
W (68-71)
y = 3.31 5x0649
R2 = 0.982
N (68-71)
Q (55-58)
W@Q1F2
W@Q1F2-Low
• Power (W@Q1F2)
Power (W@Q1 F2-Low)
100
BASIN AREA, Ab (sq mi)
-------
EPA Channel Condition Project
But Issaquah (U) and Griffin Creeks in Figure 3-21 show wider channels at Q1F2
whereas North Creek still shows as narrow. A field inspection of these three
sites showed the North Creek upstream channel is riprapped with parallel
vertical walls. The gaging station is located just upstream of a 90-degree bend as
the creek turns and goes through a constricting bridge. Several hundred feet
downstream of the bridge would be a better place to measure unrestricted
channel characteristics. It was interesting to note that at the next street crossing
upstream on North Creek, the stream channel had been rerouted in the same
manner, parallel to the street, to make room for a new mall parking lot. The new
channel was laden with large rock and LWD, and it turned 90° to go through a
new bridge.
Issaquah (U) and Griffin Creeks were also checked and Issaquah (U) was
confined in an almost rectangular channel (riprapped) just upstream of a bridge.
The channel top does widen just before it reaches bankfull conditions. The
Griffin Creek gage was located just downstream of a Tolt River pipeline trestle
with it's numerous columns, and had unstable sediment deposits in multiple
channels downstream of the trestle in several low bank channels.
Regional Hydraulic Geometry
Using the width, depth and area data in Table 3-7, plus the data for channel sizes
at Q7L2, regional models of hydraulic geometry were developed.
For the mean daily flood, Q1F2:
W = 4.95 (Q1F2) °33 (3-21)
D = 0.98 (Q1F2) °'20 (3-22)
Ac = 4.85 (Q1F2) °'53 (3-23)
It was found that the average floods at each station (except North Creek) fit the
model in Figure 3-22, which says
Q1F2 = 0.12 (PA) U4 (3-24)
Substituting Eq. 3-24 into Eqs. 3-21,3-22 and 3-23 we find that
width equals
W = 4.95 (0.12
which reduces (similarly for depth and area)
3-39
-------
EPA Channel Condition Project
Figure 3-22. Q1F2 versus PAb for Eight Puget Lowland USGS Gages for their Periods of Record (Data
from Table 3-7).
10000
Q1F2 (cfs)
1000
100
1.14
y = O.l2x
R2 = 0.97
100
1000
10000
A Q1F2
•Power (Q1F2)
PAb (in/yr • mi )
-------
EPA Channel Condition Project
for Q1F2 in this sample of Puget Lowland streams :
Width: W = 2.46 (PA) °'38 (3-25)
Depth: D = 0.64 (PA)0'23 (3-26)
Area: Ac = 1.58 (PA)0'60 (3-27)
The values of W, D and Ac at Q1F2 calculated by Eqs. 3-25,3-26 and 3-27, and the
values determined by at-a-station hydraulic geometries, are listed in Table 3-8
and plotted against each other in Figure 3-23.
Once again floods show the most variability in depth, width and area in Figure
3-23. Most of the data points lie close to the line, but for Griffin, North and
Upper Issaquah show the largest departures. This was probably due in large
part to the confinement of the channels.
For average annual flow, QAA:
The regional average hydraulic geometry equations are:
W = 4.39 (QAA) °'44 (3-28)
D = 0.59 (QAA) a17 (3-29)
Ac = 2.77 (QAA) a60 (3-30)
The regional equation for QAA, as a function of basin characteristics, was
developed for the eight USGS gages in Table 3-7, and is similar to the Olympic
Peninsula equation (Eq. 3-10) where QAA = 0.0032 (P)a'62 Ab , which allows for
variations in runoff as a function of precipitation.
For the Puget Lowland streams
QAA = 0.0040 (P)162 Ab (3-31)
Inserting Eq. 3-31 for QAA into Eqs. 3-28, 3-29 and 3-30 gives
W = 4.39 (0.0040 (P)l-62 Ab) °'44
which reduces to (along with the depth and area relationships)
for QAA in this sample of Puget Lowland streams:
Width W= 0.39 (P)0'71 (Ab) °'44 (3-32)
Depth: D = 0.23 (P) °'28 (Ab) °'17 (3-33)
Area: Ac = 0.10 (P) °'97 (Ab) °'60 (3-34)
3-41
-------
EPA Channel Condition Project
Table 3-8. Channel Width, Depth, and Area Comparison at Q1F2 for Puget Lowlands
W = C(PA)E
C: 2.46
E: 0.38
Table 3-7
Hyd. Geom.
Equation 3-25
Estimated
Station Name
Mercer
Issaquah-U
Issaquah-D
North Cr
Swamp Cr
Woods Cr
Griffin Cr
Quilceda Cr
USGS Gage
No.
Q1F2
(cfs)
165
493
1104
256
291
1076
336
144
Basin Input
(PAb)
(sq. mi-in/yr)
516
1782
2915
950
897
2688
901
555
W @ Q1F2
(ft)
26.3
54.0
47.0
22.5
40.0
45.5
35.0
20.0
WEst
(ft)
26.4
42.3
51.0
33.3
32.6
49.4
32.6
27.1
D = C(PA)E
C: 0.64
E: 0.23
Table 3-7
Hyd. Geom.
Equation 3-26
Estimated
D @ Q1F2
(ft)
2.47
1.50
4.15
2.35
2.80
3.70
1.50
3.70
DEst
(ft)
2.69
3.58
4.01
3.10
3.06
3.94
3.06
2.74
Ac = C(PA)E
C: 1.58
E: 0.60
Table 3-7
Hyd. Geom.
Equation 3-27
Estimated
A @ Q1F2
(ft2)
65
85
188
55
104
162
53
74
AEst
(ft)
67
141
189
97
93
180
94
70
-------
EPA Channel Condition Project
Figure 3-23. Channel Width, Depth and Area Estimations versus Hydraulic Geometry Values at Q1F2
for Puget Lowland Streams
1000
100
Estimated Values
Ac (ft2)
W (ft)
D (ft)
10
10
100
Hyd. Geom. Values W (ft), D (ft), Ac (ft2)
1000
* W
• D
A Ac
•1:1 Line
3-43
-------
EPA Channel Condition Project
The W, D and Ac values calculated by Eqs. 3-32, 3-33 and 3-34 and those
determined from hydraulic geometry are in Table 3-9. They are also plotted in
Figure 3-24.
As usual, channel relationships at QAA tend to have less scatter than at floods or
low flows. The combined relationships in Figure 3-24 (as developed from Eqs. 3-
32,3-33 and 3-34 above), are much better than the empirical relationships in
Figures 3-18 and 3-19.
For the 7-day average low flow, Q7L2, the regional hydraulic geometry is:
W = 6.46 (Q7L2) a51 (3-35)
D = 0.36 (Q7L2) a16 (3-36)
Ac = 2.33 (Q7L2) a67 (3-37)
Using the data in Table 3-7 for Q7L2 and (PA), the following equation was
graphed in Figure 3-25,
Q7L2 = 0.0033 (PAb) ™° (3-38)
which is very similar to the flood equation (3-24) of Q1F2 = 0.12 (PA) l"u
Substituting Eq. 3-38 into Eqs. 3-35,3-36, and 3-37 yields,
for Q7L2 in this sample of Puget Lowland Streams:
Width: W = 0.35 (PA) °'56 (3-39)
Depth: D = 0.14 (PA)a18 (3-40)
Area: Ac = 0.051 (PA) °'74 (3-41)
The comparison of predicted and measured values of W, D, and Ac
at Q7L2 for Puget Lowland streams is shown in Table 3-10 and Figure 3-26.
Surprisingly, the combined low flow equations give some of the best results.
Only some estimated areas are large or small, and the depth values are
exceptionally close (see Table 3-10 in Columns 7 and 8).
3-44
-------
EPA Channel Condition Project
Table 3-9. Channel Width, Depth, and Area Comparison at QAA for the Puget Lowlands
C:
E1:
E2:
= C(P)E'(Ab)E2
0.39
0.71
0.44
D = C(P)E1(A,, )E2
C: 0.23
El: 0.28
E2: 0.17
C:
E1:
E2:
= C(P)E1(Ab)E2
0.10
0.97
0.60
Table 3-7 Equation 3-32
Hyd. Geom. Estimated
Table 3-7 Equation 3-33 Table 3-7 Equation 3-34
Hyd. Geom. Estimated Hyd. Geom. Estimated
Station Name
Mercer
Issaquah-U
Issaquah-D
North Cr
Swamp Cr
Woods Cr
Griffin Cr
Quilceda Cr
USGS Gage
No.
QAA
(cfs)
21.9
69.7
143.9
36.4
33.8
154.5
40.3
25.6
Average
Annual
Precip., P
(in/yr)
43
66
53
38
39
48
53
37
Drainage Area,
Ab
(sq. mi.)
12.0
27.0
55.0
25.0
23.0
56.0
17.0
15.0
W@ QAA
(ft)
16.7
33.0
39.5
19.5
27.0
39.5
23.0
16.0
W Est
(ft)
16.8
32.6
38.1
21.3
20.9
35.8
22.7
16.7
D@ QAA
(ft)
1.07
0.90
1.50
1.00
1.10
1.50
0.80
1.20
DEst
(ft)
1.01
1.30
1.38
1.10
1.09
1.35
1.13
1.00
A @ QAA
(ft2)
17.7
28.4
56.5
20.7
29.8
57.9
18.9
20.1
A Est
(ft)
17.1
42.1
52.1
23.5
22.9
47.8
25.8
16.9
3-45
-------
EPA Channel Condition Project
Figure 3-24. Channel Width, Depth and Area Estimations versus Hydraulic Geometry Values at QAA
for Puget Lowland Streams
1 UUU
100
PctimatpH Valnoe
AC (ft2)
W (ft) 1 °
D (ft)
1
0.1
/
/
•
/
/
i
/
^
3
?'
Ff=
L^
/
f
y
f
?*
\\ 1
j
/I
—
^
,
^¥
i
— I
/
-
/
r
^
A
_j
^
_j
i
.
/
/
f
y
^
^
,
*
w
• D
A Ac
"~~~~~~" 1 . 1 Liny
0.1
10
100
1000
Hyd. Geom. Values W (ft), D (ft), Ac (ft2)
-------
EPA Channel Condition Project
Figure 3-25. Q7L2 versus PBE for Eight Puget Lowland USGS Stations for Their Periods of Record
(Data from Table 3-7).
100
Q7L2 (cfs)
10
100
A/
I
1000
PAb (in/yr • mi2)
1.1009
y = 0.0033X
R2 = 0.8097
A Q7L2
Power (Q7L2)
10000
3-47
-------
EPA Channel Condition Project
Table 3-10. Channel Width, Depth, and Area Comparison at Q7L2 for the Puget Lowlands
W = C(PA)E
C: 0.35
E: 0.56
C:
E:
D = C(PA)E
0.14
0.18
Ac = C(PA)E
C: 0.051
E: 0.74
Table 3-7 Equation 3-39 Table 3-7 Equation 3-40 Table 3-7 Equation 3-41
Hyd. Geom. Estimated
Station Name
Mercer
Issaquah-U
Issaquah-D
North Cr
Swamp Cr
Woods Cr
Griffin Cr
Quilceda Cr
USGS Gage
No.
Q7L2
(cfs)
5.2
14.9
27.7
6.3
4.0
18.7
3.2
4.1
Basin Input
(PAb )
(sq. mi-in/yr)
516
1782
2915
950
897
2688
901
555
W @ Q7L2
(ft)
15.0
25.6
35.1
16.5
13.1
28.8
11.7
13.3
WEst
(ft)
11.6
23.2
30.5
- 16.3
15.8
29.1
15.8
12.0
Hyd. Geom. Estimated
D @ Q7L2
(ft)
0.47
0.55
0.61
0.48
0.45
0.58
0.43
0.45
DEst
(ft)
0.43
0.54
0.59
0.48
0.48
0.58
0.48
0.44
Hyd. Geom. Estimated
A @ Q7L2
(ft2)
7.0
14.2
21.6
8.0
5.9
16.6
5.1
6.0
AEst
(ft)
5.2
13.0
18.7
8.1
7.8
17.6
7.8
5.5
3-48
-------
EPA Channel Condition Project
Figure 3-26. Channel Width, Depth and Area Estimations versus Hydraulic Geometry Values at Q7L2
for Puget Lowland Streams
1000
100
Estimated Values
Ac (ft2)
W (ft) 10
D (ft)
0.1
* W
• D
a AC
" 1:1 Line
1 10 100
Hyd. Geom. Values W (ft), D (ft), Ac (ft2)
1000
3-49
-------
EPA Channel Condition Project
NORTHEASTERN WASHINGTON REGIONAL STREAMS
Database and Empirical Relationships
The database for the streams used to develop regional models for the analysis of
the hydrology and stream channels on the Colville Indian Reservation were
reported initially in Orsborn and Orsborn (1997). The regional models for
average floods and low flows were developed further in a later report by
Orsborn and Orsborn (1999).
As was shown in Figure 2-1, flow characteristics (QC) are related to basin
characteristics (BC); and flow characteristics (QC) are related also to channel
characteristics (CC) through the analysis of channel hydraulic geometry. By
setting the equations for Q1F2, QAA and Q7L2, in terms of BC and CC, equal to
each other for each characteristic flow, combined solutions, as have been done
for the Olympic Peninsula and Puget Lowland Regions, were developed.
But, in some regions, channel dimensions demonstrate strong empirical
relationships, such as were seen for the Puget Lowland Region for channel are
(Ac) related to basin area (Ab). Some of the empirical relationships for the
Northeastern Washington Region will be examined next.
Width and Channel Area at QAA
The relationships of channel width (W) and channel area (Ac) to basin area (Ab)
were explored at QAA to determine if this type of analysis should be pursued
with Q7L2 and Q1F2. These extreme flows usually have poorer relationships to
BC's than does QAA.
The data for the QAA test is given in Table 3-11. Water surface channel width
(W) and channel cross-sectional area (Ac) are plotted against basin area (Ab) in
Figure 3-27. Note that the regional USGS gage basins range in size from 36
(Deer) to 2200 (Kettle) square miles.
In Figure 3-28, the channel characteristics are plotted against basin energy
(BE = A (H) °'50) at QAA. The plotting points for the larger basins are improved,
but some of the smaller basins still do not fit the relationships, especially Haller
Creek. It and Sheep Creek have not been examined in the field, but they display
channel areas (Ac) that are too small. Perhaps Haller Creek is cutting through
deposition. We found this to be true for Hall Creek in the Northeast part of the
CCT Reservation. But, none of the USGS records for stations on the Reservation
were of long enough duration for use in this analysis.
3-50
-------
EPA Channel Condtion Project
Table 3-11. USGS Stations, Basin Area, Basin Energy, and Channel Width (W) and
Area (Ac) for Developing CC:BC Preliminary Models at Average
Annual Flow (QAA) in NE Washington.
Channel Characteristics at QAA
Station Area Basin Energy Width Area
Name Ab BE= Ab(H)°-50 W Ac
(sq. mi.) (mi.)2'50 (ft) (sq. ft.)
Kettle
Sheep
Deer
L Pend. O.
Haller
Mill
Hangman
2200
48
36
132
37
83
689
1691
28
27
96
25
59
441
180.1
11.3
12.0
25.3
7.0
25.5
62.5
488.8
7.7
9.5
35.0
4.1
27.4
145.0
3-51
-------
EPA Channel Condition Project
Figure 3-27. W and Ac versus Ab at QAA in NE Washington
1 \J\J\J -1
<
<* ° 100 -
< 15
S w
+•» o
£ <
> ^>
^ C
£ 5
!2 -c
> _ 10-
o
1
V*
^^
/
C
J
X.
^
C
j^H
* —
_
MA
^
•
I
4
N
X
dL
/C
1
,,
H
/
/
^
^
/
•K
*X
-
"^LPO
X
x
X
--7X
r
x
x
X
X
4
H
/i
^
<
ar
1
«
.
in
^
X
x^
x
S
^>X
^
4
^
KR
* Channel Width
^B «• I .
Ac = 0.36(Ab)A0.93
^~" W = 0.56(Ab)A0.75
10
100 1000
Basin Area, Ab (sq. mi.)
10000
-------
EPA Channel Condition Project
Figure 3-28. W and Ac versus Basin Energy at QAA in NE Washington
i uuu -
<
< ° 100 -
^ *•>
s ra
£ <
6 i
TJ £
> 0
* s 10
<
1 -
[
— ^
Xyr
X
DC
-«•
P
HA
^
—
£C
I
MC
^:
|
,
^
1
k*
'
/
f
f
.
P
/
/
S '
if
LPO
^
X —
j
S
/
f
f
h
^
•v
Ian
V
n
X
^
^
^
s
^x*
X
^
^ KR
X^
4
X
f
'
* Channel Width
• Channel Area
Ac = 0.44(BE)A0.95
— — — - \A/ — n QfURF^An 7n
10
100 1000
Basin Energy, Ab(H)°-50 (mi2)
3-53
10000
-------
EPA Channel Condition Project
Combined Relationships of Channel and Basin Characteristics in
Northeastern Washington
The at-a-station equations for the hydraulic geometry analyses are summarized
in Table 3-12. The characteristic flows for the seven gages are listed in Table 3-13.
The regional models for W, D, V, and (Ac) are graphed and listed in Figures 3-29,
3-30 and 3-31 for Q1F2, QAA and Q7L2, respectively.
The regional hydraulic geometry equations are:
FOR AVERAGE DAILY FLOOD FLOWS (Q1F2):
W = 1.67 (Q1F2) °'53 (3-42)
D = 0.28 (Q1F2)034 (3-43)
Ac = 0.47 (Q1F2) °'87 (3-44)
FOR AVERAGE ANNUAL FLOWS (QAA):
W = 2.27 (QAA) °'60 (3-45)
D = 0.34 (QAA) °'31 (3-46)
Ac = 0.76 (QAA) °'91 (3-47)
FOR 7-DAY AVERAGE LOW FLOWS (Q7L2):
W = 4.00 (Q7L2) °'69 (3-48)
D = 0.35 (Q7L2) °'29 . (3-49)
Ac = 1.40 (Q7L2) °'98 (3-50)
The models for the characteristic flows related to basin characteristics are
presented next.
3-54
-------
EPA Channel Condition Project
Table 3-12. AT-A-STATION CHANNEL GEOMETRY SUMMARY: USGS REGIONAL STATIONS, NORTHEASTERN WASHINGTON
Constant and exponent from power relation y = C(Q)"^ where y = W, D, V or Ac
Station No. Station Name
12401500
12407500
12407520
12408300
12408420
12408500
12424000
Kettle R nr Ferry
Sheep Creek
Deer Creek
L Pend Orielle
Haller Creek
Mill Creek
Hangman Creek
WY
1993-95
1970-73
1970-72
1973-75
1968-77
1977-80
1994-96
Width, W
C exp
49.017
10.849
6.100
11.867
3.465
18.554
14.601
0.178
0.015
0.234
0.186
0.352
0.083
0.267
Depth, D
C exp
0.142
0.309
0.460
0.541
0.277
0.295
0.471
0.404
0.320
0.189
0.233
0.376
0.336
0.294
Mean Velocity, V
C exp
0.143
0.297
0.358
0.157
1.041
0.182
0.146
0.418
0.667
0.576
0.581
0.272
0.584
0.439
Area Channel, Ac
C exp
6.940
3.350
2.809
6.421
0.961
5.465
6.870
0.582
0.335
0.422
0.418
0.728
0.419
0.560
Check
Product Sum
C(W*D'V) expI(W,D,V)
0.995
0.996
1.005
1.008
0.999
0.996
1.004
1.000
1.002
0.999
1.000
1.000
1.003
1.000
3-55
-------
EPA Channel Conditions Project
Table 3-13. CHARACTERISTIC FLOWS, FOR NE WASHINGTON USGS GAGES
No.
(12--)
USGS STATION
Name
Q7L2
(cfs)
QAA
(cfs)
Q1F2
(cfs)
401500
407500
407520
408300
408420
408500
424000
Kettle
Sheep
Deer
L. Pend O.
Haller
Mill
Hangman
120.0
7.2
3.6
14.0
0.6
8.5
10.1
1496
12
18
58
7
47
250
11560
37
105
289
37
286
5710
3-56
-------
EPA Channel Condition Project
Figure 3-29. Regional Models of Width, Depth, Velocity and Channel Area Related to Q1F2 at USGS
Stations in NE Washington.
10000
1000
10
100 1000
FLOW, Q (cfs)
10000 100000
w
Ac
°'527
y = 1.672x
R2 = 0.995
°-134
y = 2.115x
R2 = 0.801
* w
• D
A V
X Ac
Power (W)
Power (D)
Power (V)
Power (Ac)
3-57
-------
EPA Channel Condition Project
Figure 3-30. Regional Models of Width, Depth, Velocity and Channel Area Related to QAA at USGS
Stations in NE Washington.
1000
100
o
<
>
Q"
3T
10
0.1
V
10
100
FLOW, Q (cfs)
1000
10000
w
Ac
y = 2.270X0'603
R2 = 0.995
y = 1.320X0'084
R2 = 0.460
* W
• D
A V
X Ac
Power (W)
• Power (D)
Power (V)
Power (Ac)
-------
EPA Channel Condition Project
Figure 3-31. Regional Models of Width, Depth, Velocity and Channel Area Related to Q7L2 at USGS
Stations in NE Washington.
1000
100
u
10
0.1
o.-
,/s
10
FLOW, Q (cfs)
100
1000
w
Ac
°'e9°
y = 3.996x
R2 = 0.954
y = 0.71 9x
R2 = 0.011
0022
D
Power (W)
Power (D)
Power (V)
Power (Ac)
3-59
-------
EPA Channel Condition Project
FOR AVERAGE FLOOD FLOWS:
The regional equation of average flood flows as a function of basin energy
(BE = A (H) 0'50 ) was developed for the CCT low flow report (Orsborn and
Orsborn 1999). The data is in Table 3-14 and the graphical relation is in Figure 3-
32.
The equation for average flood flow is
Q1F2 = 1.87 (BE) U5 from Figure 3-32
In terms of its basic elements
Q1F2 = 1.87 (Ab) L15(H) a58 (3-51)
This equation is substituted into Eqs. 3-42, 3-43, and 3-44 for Q1F2.
For Q1F2 at these six stations in Northeastern Washington:
Width: W= 2.33 (A) °'61 (H) °'30 (3-52)
Depth: D = 0.35 (A) °'39 (H) °'20 (3-53)
Area: Ac = 0.83 (A)1M (H)0'50 (3-54)
The values of W, D and Ac at Q1F2 for the hydraulic geometry and from the
above three equations are summarized in Table 3-15 and compared graphically
in Figure 3-33.
FOR AVERAGE ANNUAL FLOWS
The regional model for QAA was developed by Orsborn and Orsborn (1997)
along the lines of those for the Olympic Peninsula and Puget Lowlands
equations.
QAA = 0.0025 (P) L64 (Ab) (3-55)
Next it is substituted into Eqs. 3-45, 3-46 and 3-47 to yield W, D and Ac at QAA
in terms of basin characteristics.
For QAA in Northeastern Washington:
Width W= 0.062 (P)0-98 (Ab)a60 (3-56)
Depth: D = 0.053 (P)0'50 (Ab) °'31 (3-57)
Area: Ac = 0.0032 (P)1M (Ab) °'91 (3-58)
The values of W, D and Ac are compared in Table 3-16 and in Figure 3-34.
3-60
-------
EPA Channel Condition Project
Table 3-14.
Data From Table 3-9, BASIN CHARACTERISTICS
(Orsborn & Orsborn, 1997)
End POR
Current
1972
1975
1986
1973
1929
STATION
No. Name
(12--)
401500
407520
408300
408500
409500
437500
Kettle
Deer
L. Pend O.
[Mill
Hall (Res)
Nespelem (Res)
Upper Lower Basin Basin
Basin Area Elev. Elev. Relief Energy QAA Q1F2
Ab H A(H)050
(sq. mi.) fft.) (ft.) (mi.) (mi.)2'50 (cfs) (cfs)
2220
36
132
83
160
122
4920
4920
4760
4590
5410
4600
1837
1970
1983
1950
1420
1790
0.58
0.56
0.53
0.50
0.76
0.53
1691
27
96
59
139
89
1496
18
58
47
73
45
11560
105
289
286
402
234
Figure 3-32. Ql
US
100000
10000
"»"
e 100° •
a
100 -
m
F2 as a function of Basin Energy for Selected
>GS Stations in NE Washington.
p
-LJt
r
/
/
::t=
I
~7^
Jrm
1*1
LJ
*g
/
^f
/
_M
~f-
, '
'
"-\
... Q1F2 = 1.87(BE)1'15
R2 - 0 97
10 100 1000 10000
BASIN ENERGY, BE (mi)2"50
3-61
-------
EPA Channel Condition Project
Table 3-15. Channel Width, Depth and Area Comparison at Q1F2 for NE Washington
V
C:
E1:
E2:
/V = C(Ab)E1(H)E2
2.33
0.61
0.30
I
C:
E1:
E2:
D = C(Ab)E'(H)E2
0.35
0.39
0.20
/
C:
E1:
E2:
^c = C(Ab)E'(H)E2
0.83
1.00
0.50
Equation 3-52 Equation 3-53 Equation 3-54
Hyd. Geom. Estimated Hyd. Geom. Estimated Hyd. Geom. Estimated
Station Name
Kettle
Sheep
Deer
L. Pend O.
Mailer
Mill
Hangman
USGS Gage
No.
12401500
12407500
12407520
12408300
12408420
12408500
12424000
Q1F2
(cfs)
11560
37
105
289
37
286
5710
Drainage Area,
Ab
(sq. mi.)
2220
48
36
132
37
83
689
Relief, H
(mi.)
0.58
0.34
0.56
0.53
0.44
0.50
0.41
W @ Q1F2
(ft)
231.4
11.2
19.4
33.1
11.2
32.9
159.6
WEst
(ft)
217.6
17.9
17.4
37.9
16.5
28.0
96.1
D @ Q1F2
(ft)
6.77
0.97
1.38
1.94
0.97
1.93
5.33
DEst
(ft)
6.34
1.28
1.26
2.07
1.21
1.71
3.75
A @ Q1 F2
(ft2)
1559
10.8
26.7
64.1
10.8
63.6
847
AEst
(ft)
1403.3
23.2
22.4
79.8
20.4
48.7
366.2
-------
EPA Channel Condition Project
Figure 3-33. Channel Width, Depth and Area Estimations versus Hydraulic Geometry Values at Q1F2
for NE Washington
1 UUUU
1000
100
Estimated Values
Ac (ft2)
W (1t\
vv iiij
Dfftl
V'v
10
1
0.1
/
/
_J
/
/
i«-
1
T/
I1
/
1
/
^
?
1
•
1
»
^
X
.
^
»
/
-ai
71
/
/
/
/
U/
^
X
.
/
/
-?'
171
!..
::: * W
• D
A Ac
::: 1:1 Line
0.1
1 10 100 1000
Hyd. Geom. Values W (ft), D (ft), Ac (ft2)
10000
3-63
-------
EPA Channel Condition Project
Table 3-16. Channel Width, Depth and Area Comparison at QAA for for NE Washington
\
C:
E1:
E2:
/V = C(P)E'(At,)E2
0.062
0.98
0.6
C:
E1:
E2:
D = C(P)E1(Ab)E2
0.053
0.5
0.31
Equation 3-56
Hyd. Geom. Estimated
Equation 3-57
Hyd. Geom. Estimated
A = C(P)E1(Ab )E2
C: 0.0032
E1: 1.49
E2: 0.91
Equation 3-58
Hyd. Geom. Estimated
Station Name
Kettle
Sheep
Deer
L Pend O.
Haller
Mill
Hangman
USGS Gage
No.
12401500
12407500
12407520
12408300
12408420
12408500
12424000
QAA
(cfs)
1496.0
12.0
18.0
58.0
7.0
47.0
250.0
Average
Annual
Preclp., P
(in/yr)
27
18
20
29
20
26
20
Drainage Area,
Ab
(sq. mi.)
2220
48
36
132
37
83
689
W@QAA
(ft)
186.4
10.2
13.0
26.3
7.3
23.1
63.4
WEst
(ft)
159.6
10.7
10.0
31.5
10.2
21.4
58.9
D@QAA
(ft)
3.28
0.73
0.83
1.19
0.61
1.11
1.88
DEst
(ft)
3.00
0.75
0.72
1.30
0.73
1.06
1.80
A @ QAA
(ft2)
612.4
7.4
10.7
31.3
4.5
25.8
119.1
AEst
(ft)
482.0
8.0
7.2
41.1
7.4
22.9
106.3
-------
EPA Channel Condition Project
Figure 3-34. Channel Width, Depth and Area Estimations versus Hydraulic Geometry Values at QAA
for NE Washington
100
Estimated Values
AC (ft2)
W (ft) 1 0
D (ft)
1
0.1
/
/
,
/
/
-4
..
fl
. —
lU
/
\
,
>
J
?
/I
,/
\
,
4
/
y
/
~^
A
• T
/
/
*—
_j
t
-f
r
W
• D
A Ac
1:1 Line
~
0.1
1 10 100
Hyd. Geom. Values W (ft), D (ft), Ac (ft2)
1000
3-65
-------
EPA Channel Condition Project
FOR 7-DAY AVERAGE LOW FLOWS
In the report on the low flow program for the CCT Reservation, Orsborn and
Orsborn (1999) separated the annual Q7L2 values in the USGS records (Williams
et al 1985) into winter and fall events using Internet records for the Kettle River,
Deer Creek, Little Fend Oreille River and Mill Creek (Table 3-17).
The regional equations for these four USGS stations, shown in Figure 3-35, are:
For fall: Q7L2 = 0.021 (PBE) °'83 (3-59)
For winter: Q7L2 = 0.051 (PBE) °73 (3-60)
Using just the fall Eq. 3-59 for low flows and substituting this equation Eqs. 3-48,
3-49 and 3-50, gives the following combined equations:
For Q7L2 in the fall in NE Washington::
W = 0.28 (P)0-57 (Ab)°-57(H)0'29 (3-61)
D = 0.11 (P) °'24 (Ab) °'24 (H) °'12 (3-62)
Ac = 0.032 (P) ' (Ab) ' (H) ' (3-63)
The Q7L2 estimated values of W, D and Ac , and those from the regional
hydraulic geometry equations, are in Table 3-18 and are compared graphically in
Figure 3-36.
Discussion of NE Washington Results
The NE Washington region, considering the range in basin size, had fairly good
empirical relations between channel and basin characteristics, except for a few of
the gaging stations (Figures 3-27 and 3-28). These inconsistencies for Hangman,
Haller and Sheep Creeks are repeated in the combined relationships shown in
Figure 3-32 (Q1F2), Figure 3-34 (QAA) and Figure 3-36 (Q7L2).
The Hangman Creek channel size has been strongly affected by heavy flooding
form its watershed (shallow bedrock and agricultural land). It has extremely low
flows due to poor groundwater supply and over-appropriated water rights.
Sheep Creek and Haller Creek are the only other sites, which do not "fit" the
relationships at Q7L2 (Figure 3-36).
3-66
-------
EPA Channel Condition Project
TABLE 3-17.
SUMMARY OF CHARACTERISTIC SEASONAL Q7L2 LOW FLOWS
(Orsborn & Orsborn, 1999)
USGS Gage:
USGS No:
POR (WY):
BE:
PBE:
Kettle R
12401500
(1929-97)
Deer Creek
12407520
(1960-72)
LPOR
12408300
(1959-75)
Mill Creek
12408500
(1941-86)
1691
45648
27
540
96
2784
59
1534
Q1L2
Q7L2
Q7L10
Q7L20
Q30L2
Q60L2
Fall
Winter
Fall
Winter
Fall
Winter
Fall
Winter
Fall
Winter
Fall
Winter
157.5
100.0
162.4
130.9
83.4
70.0
77.0
57.9
189.2
169.9
213.3
186.5
4.0
4.6
4.1
5.4
1.7*
4.6
Extr 1.2*
Extr 4.3
4.4
6.4
5.0
6.8
14.0
14.0
14.1
15.0
9.9
10.4
Extr 8.6
Extr 9.0
15.4
19.3
16.3
21.1
9.3
9.0
9.5
10.2
5.2
6.3
5.1
5.6
10.7
12.1
11.4
13.4
Notes: Extr - Extrapolated graphically from Q7L2 and Q7L10, cannot be calculated from data;
period of record too short.
* Unusually low values compared to other gages; maybe due to diversions.
BE = Basin Energy; PBE = Annual Precipitation times Basin Energy
These seasonal low flows were calculated from USGS daily flow records on the Internet for each POR.
3-67
-------
EPA Channel Condition Project
Figure 3-35. Q7L2 versus Annual Preciptiation times Basin Energy (PBE) for
Four Northeast Washington USGS Gages
1000
100
CM
r-
O
10
0.1
— y
—
WINTER
= 0.051x
R2 = 0.9£
S
/
'
-^
s
*
>
~7
3.728-
17
1 1
4
f
s
s
'
u
-
/
^
1
\
V
'
FALL
)7L2 = 0.021 (PBE)0'83
R2 = 0.999
j
J-
j*
H
-^
X
s
H
i
Q7L2 FALL
Q7L2 WINTER
Power (Q7L2 FALL)
Power (Q7L2 WINTER)
10
100 1000 10000
PBE (in/yr • mi2'5)
100000
-------
EPA Channel Condition Project
Table 3-18. Channel Width, Depth and Area Comparison at Q7L2 for for NE Washington
W =
C:
E1:
E2:
b)E1(Hp D = C(P*Ab)E1(H)E2
0.28 C: 0.11 C:
0.57 E1: 0.24 E1:
0.29 E2: 0.12 E2:
A = C(P*Ab)E'(H)E
0.032
0.81
0.41
Equation 3-61 Equation 3-62 Equation 3-63
Hyd. Geom. Estimated Hyd. Geom. Estimated Hyd. Geom. Estimated
Station
Name
Kettle
Sheep
Deer
L. Pend O.
Haller
Mill
Hangman
USGS Gage
No.
12401500
12407500
12407520
12408300
12408420
12408500
12424000
Q7L2
(cfs)
120.0
7.2
3.6
14.0
0.6
8.5
10.1
Average
Annual
Preclp., P
(in/yr)
27
18
20
29
20
26
20
Drainage
Area, Ab
(sq. mi.)
2220
48
36
132
37
83
689
Relief, H
(mi.)
0.58
0.34
0.56
0.53
0.44
0.50
0.41
W © Q7L2
(ft)
108.7
15.6
9.7
24.7
2.8
17.5
19.7
W Est
(ft)
126.4
9.7
10.1
25.7
9.5
18.2
49.5
D © Q7L2
(ft)
1.39
0.62
0.50
0.75
0.30
0.65
0.68
D Est
(ft)
1.44
0.49
0.50
0.74
0.49
0.64
0.97
A © Q7L2
(ft2)
150.1
9.6
4.9
18.4
0.8
11.3
13.4
A Est
(ft)
189.7
4.9
5.2
19.7
4.8
12.1
50.0
3-69
-------
EPA Channel Condition Project
Figure 3-36. Channel Width, Depth and Area Estimations versus Hydraulic Geometry Values at Q7L2
for NE Washington
1000
100
Estimated Values
AC (ft2)
W (ft)
D (ft)
10
0.1
0.1
^—<
7
1 10 100
Hyd. Geom. Values W (ft), D (ft), Ac (ft2)
1000
w
D
Ac
•1:1 Line
-------
EPA Channel Condition Project
Summary Comparisons of Regional Analyses
The various phases of analysis covered in Part 3 are summarized for the three
regions in Washington, beginning with regional hydraulic geometries. Table 3-
19 summarizes the hydraulic geometry equations for the Olympic Peninsula,
Puget Lowland and NE Washington Regions.
Table 3-19. Comparison of Three Regional Sets of HYDRAULIC
GEOMETRY Equations for Three Characteristic Flows.
Region
At Flood Flow At Average Flow
At Low Flow
Olympic
Peninsula
Puget
Lowlands
NE
Washington
W = 2.40(Q1F2)°'47
D = 0.22(QlF2)a36
Ac = 0.52(Q1F2)°'83
W = 4.95(QlF2)a33
D = 0.98(Q1F2)0'20
Ac = 4.85(Q1F2)°'53
W = 1.67(Q1F2)°'53
D = 0.28(Q1F2)0'34
Ac = 0.47(Q1F2)°'87
W = 4.02(QAA)0'51
D = 0.29(QAA)°'32
Ac = 1.16(QAA)a83
W = 4.39(QAA)°'44
D = 0.59(QAA)°'17
Ac = 2.27(QAA)a6°
W = 2.27(QAA)a6°
D = 0.34(QAA)°'31
Ac = 0.76(QAA)°'91
W = 8.37(Q7L2)a52
D = 0.31(Q7L2)8'30
Ac = 2.56(Q7L2)°'82
W = 6.46(Q7L2)asi
D = 0.36(Q7L2)ai6
Ac = 2.33(Q7L2)°'67
W = 4.00(Q7L2)°'69
D = 0.35(Q7L2)°'29
Ac = 1.40(Q7L2)0'98
If one is planning to develop an analysis of "channel condition" there are "office
steps" which can be done before going to the field. Regional hydraulic geometry
estimates at three characteristic flow stages can be done IF one has estimates of
the characteristic flows. These flow estimates can be made from regional models
of the flows related to basin characteristics.
The ranges of the characteristic flows used in the analysis are given in Table 3-20.
3-71
-------
EPA Channel Condition Project
Table 3-20. Ranges of Flows and Average Annual Basin Precipitation (P) in
the Three Regions of Washington Used in Regional Models of
Hydraulic Geometry.
„ . P Q1F2 QAA Q7L2
ReS1on (inlyr) (cfe) (cfs) (cfs)
Olympic
Peninsula
Puget
Lowland
NE
Washington
40 to 200
37 to 66
18 to 30
150 - 18300
144 - 1076
37 - 11560
16 - 2035
22 - 154
7 - 1496
2.2 - 610
3.2 - 28
0.6 - 120
Although the flows given in Table 3-20 represent a very broad range from 0.6 to
18,300, a better comparative way to look at the flows is in terms of cfs/mi2, or
"unit flows". These are the net flows released from the watersheds based on the
form of precipitation, which caused those flows. For example, on the West and
Southwest sides of the Olympic Peninsula heavy rains on top of an already
elevated stream stages result in large floods. Puget Lowland streams usually
have rain combined with snowmelt. Northeastern Washington floods are
usually a result of snowmelt. The ranges of unit flows in the three regions are
summarized in Table 3-21.
The unit flow floods range from just 0.8 to 83.4 cfs/mi2 (ratio 104), average flows
from 0.19 to 12.0 (ratio 63) and low flows from 0.014 to 2.90 (ratio 207). These
unit flood values represent the rate of precipitation, or snowmelt, or both, and
the valley morphology and slope. Average flow values include the flood and
low flow events of record. The low flows are most strongly influenced by the
available groundwater storage and/or glacial supply (e.g. the Hoh at 2.90
cfs/mi2 and Hangman Creek near Spokane at 0.014 cfs/mi2, a huge watershed
with low precipitation, poor ground water storage and is over-appropriated).
An "office step" for estimating characteristic flow for channel condition analysis
can be done from: (1) good gaging records; (2) extending short records by
correlating them with the same-day flows at a long-term gage; or (3) by using
regional models of the types given in Table 3-22. The combined equations for W,
D and Ac for the three regions are given for the three characteristic flows on
pages 3-29 and 3-30 for the Olympic Peninsula; on pages 3-41 and 3-44 for the
Puget Lowlands; and on pages 3-60 and 3-66 for NE Washington.
3-72
-------
EPA Channel Condition Project
Table 3-21. Comparison of Maximum and Minimum Unit Values in cfs per
square mile for Q1F2, QAA and Q7L2 in the Three Regions for
the Ranges of Flow in Table 3-20.
Region
Olympic
Peninsula1
Puget
Lowland2
NE
Washington3
Stream
Name
NF
Quinalt
Snow
Woods
Quilceda
Hangman
Sheep
QlF2/Ab
(cfs/mi2)
83.4
13.4
19.2
9.6
8.3
0.8
Stream
Name
NF
Quinalt
Snow
Woods
Mercer
Kettle
Hatter
QAA/Ab
(cfs/mi2)
12.0
1.4
2.8
1.8
0.68
0.19
Stream
Name
Hoh
Snow
Issacj-U
Swamp
Sheep
Hangman
Q7L2/Ab
(cfs/mi2)
2.90
0.20
0.55
0.17
0.15
0.014
1. Data in Tables 3-1 and 3-2.
2. Data in Table 3-7.
3. Data in Tables 3-11 and 3-13.
Table 3-22. Comparison of Regional Equations for Estimating
CHARACTERISTIC FLOWS in the Three Regions (Streamflow
Equations).
Region
At Flood Flow
(cfs)
At Average Flow
(cfs)
At Low Flow
(cfs)
Olympic
Peninsula
Puget
Lowland
NE
Washington
Q1F2 =
0.27(PAb)L05
Q1F2 =
0.12(PAb)L14
Q1F2 =
1.87(Ab)L15(H)a58
QAA =
0.0032(P)1'60(Ab)
QAA =
0.0040(P)L62(Ab)
QAA =
0.0025(P)L64(Ab)
Q7L2 =
0.0067[PAb(H)a50]1-06
Q7L2 =
O.OOSSCPAJ1'10
Q7L2 (Fall) =
0.021[PAb(H)a50]0-83
3-73
-------
EPA Channel Condition Project
Another surveyor in 1918 "...found the course of the river
radically different from that shown in Curry's Survey of
1882, his measurements ranging from 330 to 550ft in the
same stretch of stream channel where Curry (1882) found
widths of 12 to 49ft."
(From Burkham 1981, page 594 in a discussion of the Rio
Salado near Santa Rita, NM).
3-74
-------
EPA Channel Condition Project
References for Part 3
Amerman and Orsborn. 1987. An analysis of streamflows on the Olympic
Peninsula in Washington State. Dept. of Civil and Environmental
Engineering. Washington State University, Pullman, WA. (Two
Volumes).
Burkham, D.E. 1981. Uncertainties resulting form changes in river form. ASCE
Journal of Hydraulics Division. Vol. 107, HY5.
Dasman, R.C., 1973. A rationale for preserving natural areas. Jour. Soil Water
Conserv, v. 28, no. 3, p. 114-117.
Johnson, A. W. and J. F. Orsborn. 1997. North Creek Channel Design: regional
analysis of channel size related to basin characteristics. OTAK, Kirkland,
WA.
Moscrip, A. L. and D. R. Montgomery. 1997. Urbanization, flood frequency and
salmon in Puget Lowland streams. Journal AWRA. V36 (6).
Orsborn, J. F. and M. T. Orsborn. 1997. An operational hydrologic system for the
Colville Indian Reservation. In three volumes: Vol. 1 - Summary,
Descriptive Text and References; Vol. 2- Database Appendices; Vol. 3 -
Hydrologic Analysis Appendices. Environmental Trust, CCT. Nespelem,
WA.
Orsborn, J. F. and M. T. Orsborn. 1999. Hydrologic aspects of the Colville Indian
Reservation low flow program. Environmental Trust, CCT. Nespelem,
WA.
Williams, J. R., H. E. Pearson and J. D. Wilson, 1985a. Streamflow statistics and
drainage basin characteristics for the Puget Sound region, Washington.
Vol. I. Western and Southern Puget Sound. USGS Open-file Report 84-
144-A. Tacoma, WA.
Williams, J. R., H. E. Pearson and J. D. Wilson. 1985b. Streamflow statistics and
drainage basin characteristics for the Puget Sound region, Washington.
Vol. II. Eastern Puget Sound from Seattle to the Canadian Border. USGS
Open-File Report 84-144-B. Tacoma, WA.
Williams, J. R., and H. E. Pearson. 1985. Streamflow statistics and drainage basin
characteristics for the Southwestern and Eastern Regions, Washington.
Vol. I. Southwestern Washington. USGS Open-file Report 84-145-A.
Tacoma, WA.
3-75
-------
EPA Channel Condition Project
4. APPLICATIONS
Introduction
Flow characteristics and their interactions with the channel boundaries are
central to all river management problems. Alluvial streams develop an average
geometry that reflects the load of flow and sediment. Because most natural
stream channels exist in erodible soils, they alternately aggrade and degrade,
depending on the load in the channel. The resultant channel dimensions reflect
average values for width and depth imposed by water and sediment discharge,
bed sediment size, bank vegetation, and average bed slope. Recognizing the
natural channel relationships of a stream thus becomes a basic step in
understanding a stream's behavior and characteristics.
Channel condition studies, when coupled with stream hydrology, lead to the
following general categories of applications:
Reconnaissance: inventories/analysis and planning.
Restoration: projects and activities that modify existing channel.
Reconstruction: design leading to construction of new channels.
There is not always a strict, clear difference between these applications. Because
of the many facets that exist in any given project, overlap most likely will occur.
Following is further discussion and examples of these general categories.
Reconnaissance
Reconnaissance is the general term for studies gathering information on historic
and/or present channel conditions. This information is used to plan, design, and
monitor projects. A common question in these studies is how much have stream
channel widths and depths changed with changes in land use? This is
particularly true in preparing watershed analyses or basin plans, and exploring
the land use effects of urbanization, logging, or agriculture.
Considerable effort is often expended in these studies to identify the natural
dimensions of the stream channel under pre-disturbance conditions and
following a change in land use, how did the channel respond? It is known that
increases in the amount of impervious surface increases the amount and rate and
runoff (Leopold 1990). Such changes may trigger channel erosion or stream
incision (Booth 1990), and cause significant increases in channel capacity (Knight
1979; Mosley 1975). This information is also central to availability and suitability
4-1
-------
EPA Channel Condition Project
investigations. Examples include habitat availability studies of instream flow
related to water diversion/flow reservation studies.
Fish habitat in its simplest physical terms can be described as hydraulic diversity.
Specifically, the basic elements of instream habitat are water depth and velocity.
All stream fish have adapted to a particular range of depths and velocities. Even
body shapes of fish have adapted. Habitat availability relates to width, depth
and velocity at various seasonal flows.
In other reconnaissance studies, channel geometry relationships could serve as
preliminary estimates of channel capacity, flood flow characteristics and
floodplain inundation. While these estimates have to be confirmed with local
topographic data, they would provide the starting point to initiate the
investigation. For both planning and design the sizing of bridges and culverts is
obviously related to anticipated flow characteristics. New design criteria for
sizing culverts in Washington and Oregon, for example, now require the culvert
to contain the bankfull width plus a safety factor, as one design alternative.
Channel dimension estimates are valuable at a programmatic level in defining
culvert size and location.
Recent listings of several salmonids under the Endangered Species Act (ESA) has
brought with it new expectations in project analysis. Section 7 of the ESA
requires an effects analysis for any proposed action that could modify fish
habitat. Channel geometry would be integral in the analysis of channel
modifications, dredging for flood control, or gravel and gold mining, or flow
reduction. It would also be useful in the analysis of institutional programs such
as river management. These programs routinely affect many miles of channel
through dredging, straightening, and bank protection projects.
Restoration
Identifying natural channel geometric and flow relationships for a stream is an
important step towards understanding the stream's behavior and characteristics.
Based on drainage area and other basin characteristics, the channel geometry
measurements can be linked to the channel pattern and profile, and used to size
stream rehabilitation works that mimic natural conditions.
The geometry of meanders and pool/riffle profile for all river patterns in
credible materials can be related to the bankfull width. Meander radius, wave
length, amplitude, belt width, channel entrenchment also relate to bankfull
width. Flood prone areas have empirical relationships to bank full width and the
50-year flood flow. Even a preliminary estimate of the hydraulic geometry based
on an abbreviated field survey in which only the bankfull width and depth are
measured will provide useful guidelines (Rosgen 1996). In planning/design of
projects to recreate meander geometry, to what dimensions will we design?
4-2
-------
EPA Channel Condition Project
Width to depth relationships lead to other relationships such as the radius of
curvature.
Some past projects have achieved undesired results. How can we undo errors of
the past (channel dredging, straightening)? There are many examples of stable
channel design given in Newbury and Gaboury (1993), Brookes and Shields
(1996) and Thorne, Hey and Newson (1996).
Reconstruction
In the course of completing projects for flood alleviation and channel
stabilization, many rivers have been considerably modified. River engineering
and mining works involving dredging, widening, straightening and diversions
have affected hundreds of miles of rivers. These changes have adversely affected
the stability of the engineered and adjacent reaches and destroyed the
conservation and amenity value of riverine areas (Brookes, 1988; Purseglove,
1988). Consequently there is an urgent need to use more sympathetic
engineering design procedures which will preserve the natural stability of the
river, its habitat diversity and its amenity values. By designing with nature
rather than imposing on nature, such approaches are more cost-effective, require
less maintenance and, above all, minimize environmental impacts.
Increasingly, the demands to restore and rehabilitate stream reaches requires the
adoption of solutions to recreate channel features that are enduring and in
harmony with local flow conditions. Meandering channels with pools, riffles,
glides, dead zones and point bars need to be recreated to restore the habitat
features destroyed by previous works or natural disasters. These features cannot
be installed at random, and badly designed schemes will quickly be made
dysfunctional as the river reacts to the unnatural imposed conditions. This
emphasizes the need for the development of sympathetic design procedures that
are in harmony with local river fluvial geomorphology.
In the remainder of Part 4, four project examples (case studies) have been
summarized. Project 1 covers instream habitat and basin improvements made at
LeBar Creek, a tributary to the S.F. Skokomish River in the southeast part of the
Olympic Peninsula. Project 2 deals with planning for the restoration of channel
meanders in Crooked River, a gold-dredged tributary to the South Fork of the
Clearwater River in Idaho. Project 3 presents the reconnaissance and
comprehensive documentation and analysis of the effects of road building,
logging and urbanization on the sediment load, channel geometry and the
decline of coho runs in Big Beef Creek west of Bremerton, Washington on Hood
Canal. The fourth project examines the effects of dams and diversions on
instream channel geometry and habitat in the Lower Elwha River on the north
coast of the Olympic Peninsula.
4-3
-------
EPA Channel Condition Project
REFERENCES FOR INTRODUCTION TO PART 4- APPLICATIONS
Booth, D.B. 1990. Stream-channel incision following drainage basin
urbanization. Water Resources Bulletin 26:407-17.
Brookes, A. 1988. Channelized Rivers: Perspectives for Environmental
Management. John Wiley & Sons. Chichester, England.
Brookes, A. and F.D. Shields. 1996. River Channel Restoration: Guiding
Principles for Sustainable Projects. John Wiley & Sons. Chichester,
England.
Leopold, L.B. 1990. Lag times for small drainage basins. Catena 18:157-171.
Knight, C. 1979. Urbanisation and natural stream channel morphology: the case
of two English towns. In: Hollis, G.E. (ed.). Man's Impact on the
Hydrological Cycle in the United Kingdom. Geobooks, Norwich. 181-
198.
Mosley, M.P. 1975. Channel changes on the river Bollin, Cheshire, 1872-1973.
East Midland Geographer 6: 185-199.
Newbury, R.W. and M.N. Gaboury. 1993. Stream Analysis and Fish Habitat
Design. Newbury Hydraulics Ltd., Gibsons, British Columbia.
Purseglove, G. 1988. Taming the Flood. Oxford University Press. Oxford.
Rosgen, D. 1996. Applied River Morphology. Wildland Hydrology. Pagosa
Springs, Colorado.
Thorne, C.R., R.D. Hey and M.D. Newson. 1997. Applied Fluvial
Geomorphology for River engineering and Management. John Wiley &
sons. Chichester, England.
4-4
-------
EPA Channel Condition Project
CASE STUDIES
CASE STUDY 1. HABITAT IMPROVEMENT PROJECTS IN LOWER LEBAR
CREEK BASIN
LOCATION: LeBar Creek, Tributary to the S. F. Skokomish River, a tributary
at the South End of Hood Canal; Water Resource Inventory Area (WRIA) 16;
Project Located in S 1/2, Sec 4, T22N, R5W. (See location map in Figure 4-1).
Figure 4-1. Location Map for the LeBar Creek Project (Not to Scale)
fi.F.
HOOD
CANAL
4-5
-------
EPA Channel Condition Project
MAIN REFERENCE: Orsborn, J. F. 1993. Habitat improvement projects in
Lower LeBar Creek Basin. USDA Forest Service, Hood Canal Ranger District,
Hoodsport, WA.
OBJECTIVES:
• Restoration of fish habitat and passage in the 1.0 mi. of LeBar Creek below
the barrier falls;
• Reconnect the off-channel, point-bar ponds (remnant flood channels) to
the main channel in the lower 0.3 mi; and
• Stabilize eroding areas, including the road that crosses the creek and
enters the basin.
SUMMARY:
LeBar Creek (Figure 4-1) is a tributary to the South Fork Skokomish River. The
South Fork joins the North Fork at RM 9.0 and then flows into Hood Canal near
Union, Washington. As is typical of the tributaries to the South fork, a bedrock
outcrop forms a hanging valley about 1.0 mile above the confluence of LeBar
Creek with the South fork. These hanging valleys form waterfalls and high
velocity chutes, which are complete barriers to upstream migration by fish.
The loss in anadromous fish runs in the basin can be attributed to impacts on
instream habitat caused by road building, logging activities and associated
landslides. The LeBar Creek drainage lies within the boundaries of the Shelton
Cooperative Sustained Yield Unit, which was intensively logged between 1955
and 1989. Increased flood flows and sediment loads, and the loss of woody
debris from riparian areas, have combined to degrade the fisheries habitat in
LeBar Creek.
The Forest Service has undertaken corrective activities on the watershed and in
the lower one-mile project reach. On the watershed, hill slopes were replanted,
abandoned roads pulled back and stabilized, and unneeded culverts removed.
To help restore the habitat in the anadromous reach and to increase the
productivity of the fishery below the falls, the following tasks were undertaken
in this project:
(1) habitat survey of the lower one-mile reach of LeBar Creek, identifying
potential fish habitat improvements within the reaches;
(2) in the lower 0.3 miles fish habitat modification structures were designed,
which will: help stabilize the reach, reduce road fill erosion, improve fish
passage and habitat diversity, and
4-6
-------
EPA Channel Condition Project
complement the development of off-channel rearing habitat on the large
adjacent point bar;
(3) survey and design the off-channel rearing habitat on the adjacent point
bar; and
(4) design fish habitat improvements for the upper 0.7 mile where there is no
heavy equipment access.
A preliminary planning schedule called for:
• completion of the lower 0.3 mile of instream habitat improvements in
1993;
• installation of the off-channel rearing area water supply pipeline (and
internal supply channels to ponds), and the lower connecting channel to
LeBar Creek in old remnant channels in 1993;
• expansion and refinements in the off-channel rearing site in 1994 after a
year of observation and operation; and
• installation of habitat improvement structures in the upper 0.7 mile reach,
depending on the results of project monitoring and evaluation in 1993-
1995.
The body of the project report was supplemented with six appendices covering:
geomorphic analysis of the subbasins and tributaries, hydrologic analysis,
topographic survey notes of the lower 0.3 mi, habitat survey notes of the entire
1.0 mi project reach, the project photographic record and drawings.
PHYSICAL SETTING:
This section of the report includes: (1) the physical characteristics of the basin
and stream system which are used to characterize basin morphology and to
estimate streamflows; (2) recent land use activities (logging and road building)
which caused downstream channel adjustments and habitat degradation; (3) an
evaluation of watershed conditions [U. S. Forest Service, 1991]; (4) an evaluation
of land use changes on channel size; (5) a summary of estimated streamflows at
the project site in Lower LeBar Creek; and (7) fisheries information including a
life-stage periodicity chart.
Physical Characteristics of the Basin-Stream System and Estimated Stream
Flows (See Table 4-1 and Figure 4-2).
Drainage Area (A): 9.7 sq. mi. (Cols. 5 and 6)
4-7
-------
EPA Channel Condition Project
Stream Length (LST): 14.0 mi. (Cols. 3 and 4)
1st Order (LI): 7.3 mi.
2nd Order (L2): 6.7 mi.
Total Stream Density (SD) 14.0/9.7 = 1.4 mi/sq. mi. (Col. 8, cumulative)
Average Annual Precipitation (P): 130 in./yr. (Col. 15)
Average Annual Water Input to the Basin (PA): 1265 sq. mi. - in./yr. (Col. 16)
Basin Relief (H): 0.56 mi. (Col 12)
Basin Energy (A) (H)05: 7.28 mi.2'5 (Col. 14)
The hydrologic analysis used several modeling approaches to estimate floods,
low flows, average annual flow and its extremes, and maximum, minimum and
mean monthly flows. The average flood flow was related to channel size on a
regional basis, as was the average annual flow. Low flows and monthly flows
were used to estimate seasonal fisheries flows and minimum flow conditions.
Table 4-1. Geomorphic Characteristics of LeBar Creek Basin
BASIN
NO.
(-)
(-1
Col (1)
1
2
3
4
5
•6
7
8
0
10
11
TOTALS.
STREAM
ORDER
(•)
(•)
(2)
1
1
2
1
2
1
2
1
:
i
2
Wkolt
Basin
STREAM
LENGTH
LS
(mi)
m
1.54
1.14
1.80
1.54
0.75
0.69
1.37
1.30
0.57
1.07
2.25
14.02
CUMUL.
LENGTH
LST
(mi)
«>
1.54
2.68
4.48
6.02
6.77
7.46
8.63
10.13
10.70
11.77
14.02
14.02
BASIN
AREA
A
(mi)"2
(5)
1.11
0.77
1.22
0.92
0.60
0.36
1.12
1.11
0.41
0.70
1.41
9.73
CUMUL.
AREA
A
«.|
(7)
1.38
1.48
1.48
1.67
1.25
1.92
1.22
1.17
1.39
1.52
1.60
CUMUL.
STREAM
.DENS.
SD
(mi)«.|
(>)
1.38
1.42
t.44
1.50
1.46
1.50
1.45
1.40
1.40
1.41
1.44
HEAD
WATER
ELEV.
EH
(fl)
(9)
4300
4200
3000
4400
2500
2600
2500
3000
2500
3000
2000
3500
OUTLET
ELEV
ED
(fl>
(10)
2000
2000
1400
1400
1300
1300
1050
1050
950
950
500
RELIEF
BASIN CUMUL.
H H
(mi) (mi)
(H)
0.44
0.42
0.30
0.57
0.23
0.26
0.27
0.37
0.29
0.29
0.26
(12)
0.44
0.44
0.49
0.49
0.53
0.52
0.52
0.52
0.48
0.46
0.47
0.56
BASIN ENERGY
BASIN CUMUL
AH'0.50 AH"0.50
(mi)"2.5 (mi)"2.5
(13)
0.74
0.51
0.67
0.69
0.26
0.19
0.56
0.68
0.22
0.36
0.75
(14)
0.74
1.23
2.17
2.81
3.36
3.62
4.40
5.19
5.26
5.76
6.67
7.26
AVER. BASIN (P'A)
PRECIP INPUT (LT'H"2
P (P'A)
(in/yr) tsq.mi-in/vr' (in/mi)
(IS)
140
138
138
133
136
133
134
131
132
127
130
(1«) (17)
155 520
259 499
428 398
628 335
817 343
1005 408
1265 268
4-8
-------
EPA Channel Condition Project
Figure 4-2. Project Basin and Stream Map for LeBar Creek.
S.F. Skokomlah River
4-9
-------
EPA Channel Condition Project
Recent Land Use Activities Which Caused downstream Changes in
Channels and Habitat.
LeBar Creek was heavily roaded and logged from 1955 to 1989. Annual data and
cumulative totals are summarized in Table 4-2.
Table 4-2: Percent of LeBar Creek Basin Logged, and Estimated Annual
Miles of Road Constructed (Based on % cut).
YEAR
ACRES
CUT
%BASIN
CUT
CUMUL.
% CUT
EST. ROADS
(mi/yr)
1955
1956
1964
1965
1966
1967
1968
1969
1971
1972
1974
1975
1976
1977
1978
1980
1981
1982
1983
1984
1985
1988
1989
TOTALS
2.9
94.7
112.4
196.6
223.1
145.0
164.8
2.0
47.1
1.8
123,0
89.4
214.4
535.5
131.3
73.5
89.8
134.4
85.6
45.1
185.6
58.3
45.2
2800.00
0.005
1.52
1.80
3.16
3.58
2.33
2.64
0.003
0.76
0.003
1.98
1.44
3.44
8.60
2.10
1.18
1.44
2.16
1.37
0.72
2.92
0.90
0.70
44.75%
0.005
1.53
3.33
6.49
10.07
12.40
15.04
15.04
15.80
15.80
17.78
19.22
22.26
31.26
33.36
34.54
35.98
38.14
39.51
40.23
43.15
44.05
44.75
44.75%
0.59
0.59
0.70
1.23
1.39
0.90
1.02
-
0.29
-
0.77
0.56
1.33
3.33
0.81
0.45
0.50
0.84
0.53
0.28
1.13
0.35
0.27
38. 8 mi.
Evaluation of Watershed Conditions (U. S. Forest Service, 1991)
In 1991 the Olympic National Forest made an examination of a series of impacted
watersheds on the Forest to determine their relative "condition".
4-10
-------
EPA Channel Condition Project
The ONF criteria used threshold values as developed by the interdisciplinary
team.
STANDARD CRITERIA TTt
VALUES
(1) % of basin with elevation between 1500 - 3000 ft
to account for rain on snow events, and geologic „„„ _ ,
* • u- u* j* u i.r 50% of the area
formations which tend to have pockets of
unconsolidated material (bed-rock hollows).
(2) Tree Stand < 35 years. Accounts for logging back
to 1955, hillslope instability due to loss of root 40% of the area
structure and hydrologic balance (changes).
(3) Soil classes (C, D and/or E) which are STEEP, cn(V , ,
EROSIVE SOILS. 50% of the area
(4) ROAD DENSITY - to account for increases in
surface runoff, runoff concentration and 2.5 mi. per sq. mi.
silt/sediment load.
MULTIPLE STANDARD THRESHOLD
VALUE (MSTV) = 0.25 Criteria (1x2x3x4)
The Multiple Standard Threshold Value (MSTV), an extension of the Forest
Service Thresholds, provides a way to assign a relative total "condition" rating to
a basin, to compare basins with baseline conditions and to compare between
basins. Multiplying the threshold values (0.50 x 0.40 x 0.50 x 2.5) gives a baseline
MSTV of 0.25. individual thresholds in a test basin might indicate only one
severe value with the other three values being less than the threshold values.
Even though the MSTV might be less for the test basin than the threshold MSTV
of 0.25, the one or two severe values should not be neglected.
For LeBar Creek the threshold values of the factors, the ratios of LeBar Creek
values to standard values (a Severity Ratio), and the multiple of those ratios are
listed in Table 4-3. The multiple of the Severity Ratios gives a more descriptive
measure of the basin values to the threshold values than does the MSTV. The
severity ratios demonstrate how far above (or below) the LeBar values are to the
thresholds. The baseline multiple severity factor for the standard thresholds
would be 1.00
4-11
-------
EPA Channel Condition Project
Table 4-3: LeBar Creek Basin Threshold Rating and Their Severity
Compared to the Standard Thresholds Listed Above .
THRESHOLDS
LeBar Cr.
Standard
Severity Ratios
LeBar/Standard
ELEV.
63.5%
50%
1.27
TREES
44.9%
40%
1.12
SOILS
66%
50%
1.32
ROADS
(mi/mi2 )
3.9
2.5
1.56
MULTIPLE
VALUES
(MSTV)
0.73
0.25
2.92
Roads are seen to provide the most severe index at 56% (1.56 Ratio) above the
standard. Actual stream densities in the LeBar Creek basin average 1.4 mi/mi2,
only about one-third (36%) of the road density.
Evaluation of Changes in Stream Channels Due to Changes in Land Use
Two basic questions needed to be answered in order to make the most direct
evaluation of upstream land use activities on possible downstream channel
changes:
(1) What would be the expected impacts on the stream channel in the
reach just upstream of the confluence of LeBar Creek and South Fork
Skokomish River (widening); and
(2) How can the site channel be checked as to whether or not this impact
has occurred?
• by using regional channel geometry models (from USGS gage
calibration records) to estimate top width, depth and area at
bank full flow (at average 2-year daily flood and at average
annual flow); and
• by comparing the regional equation estimated values with
actual measurements in LeBar Creek.
Channel measurements were made in a straightened reach of channel between
1600 -1900 feet upstream of the mouth of LeBar Creek and beside the point bar.
It appears that someone straightened this channel, pushed up fill and debris
along the left bank, and the channel has steepened resulting in a cobble-bedded
4-12
-------
EPA Channel Condition Project
channel. During low flow there are places where fish would be hard pressed to
have successful migration due to depth and velocity constraints. Floods cannot
overtop the left bank.
The regional equations for channel geometry in terms of the daily mean annual
flood (1-day average, 2-year recurrence interval flow) are:
Top Width:
Mean Depth:
Mean Velocity:
Flow Area:
\0.46
W = 2.38 (Q1F2)U-46 (ft)
D = 0.51 (Q1F2)0'28
(ft)
0.26
V = 0.82 (Q1F2)U^ (fps)
0.74
Ac = 1.18 (QlF2)a/4 (ft)
The regional channel geometry equations were developed from USGS gaging
calibration records of the stations on the Little Quilcene, Duckabush and
Dungeness Rivers, and Goldsborough and Kennedy Creeks. The S. F.
Skokomish River and Skokomish River gages could not be used because of
dramatic changes in their channel geometries due to logging-generated sediment
aggradation and associated channel widening.
Using 600 cfs as the bankfull flow in the above set of equations says the LeBar
Creek channel should have these characteristics:
W =
D =
V =
Ac =
2.38 (600) °'46
0.51 (600) °'28
0.82 (600) °'26
1.18 (600) °74
= 45ft
= 3.0ft
= 4.2 fps
= 135 ft2
The existing channel measured (average of three cross-sections)
W = 60 ft; D = 3.2 ft; and Ac = 192 ft2
The LeBar Creek ratios of channel size components are:
Values
LeBar
Model
Ratios
W
(ft)
60
45
1.33
D
(ft)
3.2
3.0
1.07
Ac
(ft)2
192
135
1.42
4-13
-------
EPA Channel Condition Project
Fisheries Information
Assuming the coho and steelhead will continue to use LeBar Creek in the lower
mile it is important to be able to compare the timing of fish utilization with
streamflow. Fisheries requirements were displayed in a periodicity chart for the
Skokomish-Dosewallips WRIA. Considering the seasonal life stages of upstream
migration, spawning, incubation and juvenile outmigration, the monthly flows
were modeled and catalogued.
The rest of this report covers these topics: project guidelines and planning;
descriptions of the lower instream and the offstream projects; upstream
potential project design recommendations; and appendices.
EXAMPLES OF LOWER CREEK PROJECTS
As shown in Figure 4-3, (Drawing LC-1), the lower LeBar Creek habitat
modification project involved two components, instream and offstream. The
project baseline drawing shows a total of 7 Habitat Improvement Units (HIU), 5
in LeBar Creek and 2 in the off-channel area. The units were selected based on
channel reach geometric and physical characteristics and function. An example
of the details in one HIU is presented in Figure 4-4, which shows HIU4 in the
channel bend where the toe of the access road has been eroding.
The problems in HIU 4 were: steep cut bank and eroding slope (earlier attempts
at dumping riprap from the road above have been only partially successful);
Unit contains the best pool habitat in the project reach; cover lacking; and
passage problems just upstream due to wide, shallow, steep channel.
Proposed solutions included: (Fig. 4-4) start at upstream end with two rock or log
deflectors on right bank; add two rows of turning rocks from the downstream
deflector across to the point bar; move corner of point bar to opposite cut bank;
rearrange existing boulders in a series of deflectors; cable logs along toe between
deflectors; need 10 new 2-ft and 10 new 3-ft rocks at this site; and vegetation was
planted in bare earth exposures.
The location of HIU 4 within the whole lower creek habitat project is shown in
Figure 4-3.
CONCLUSION
To prepare for this habitat project the road and slide conditions on the watershed
were addressed. Besides improving habitat in the project reach, the recurring
problem of road fill erosion was included in the project and the local problem
was corrected mainly with habitat structures.
By using regional hydraulic geometry analysis, the riffle (passage-limiting) reach
problems were analyzed and addressed. Channel narrowing log deflectors were
4-14
-------
EPA Channel Condition Project
installed at 135° to the left downstream bank. These structures trapped gravels
and restored bank vegetation. Also, installing boulders and log structures
periodically along the right side of the reach deepened the thalweg.
Sediment deposition problems arose at the water intake but these have been
corrected. Some maintenance and fine-tuning were required in the first years of
the project, but now it has hardened. The alders on the point bar were girdled
and cedars have been planted to accelerate the succession.
Figure 4-3. LeBar Creek Habitat Project Map and Baselines.
IE.BAR. (T)
e.it BRwu.—1-» gunnel ufsmuM rwt JML
BASELINE FOR. It
CHANNEL STRUCTURES
LOCATION MAP
BASELINE. FOR OFF
CHANNEL STRUCTURES
— (MM m.HM)
MltUMTWn nuTMIT) IH.IM4
.DETAILED BRA WINGS
FOR EACH HIU ARC
ATTACHED
LEBAR CREEK.
HABITAT PROTECT
__ . |S»taJ«.C»«NP(tL
-HAtltXT IMfROVE.UIIITlcll<**CIEIl"i"<:i
J. F. fflCiVHtH. fl \SHHt LC-l
4-15
-------
EPA Channel Condition Project
Figure 4-4. LeBar Creek Habitat Project Details of HIU 4 Modifications
LEBAR0
TYPICAL CONSTRUCTION DETAILS
0-1") i«.Loaa, »ccn,Atmi AH» outs
4-16
-------
EPA Channel Condition Project
CASE STUDY2. PLANNING AND DESIGN FOR THE RECONSTRUCTION OF
A GOLD-DREDGED STREAM, Crooked River (Idaho) Habitat Improvement
Project
LOCATION: Crooked River, a tributary of the South Fork Clearwater River,
Southwest of Elk City, Idaho, in the SW 1/4, NW 1/4, Sec. 30, T28N, RGB. (See
Figure 4-5)
Figure 4-5. Crooked River Study Site and Regional Location
Culvert
Propottd
M«ondtr
Location*
South Fork Cltarwaltr River
to
-»04) Elk
City
Crooked Rivtr
Fornt Route
233
Preltet Area
T28N, R8E
SW 1/4, NW 1/4
Section 30
Runway Crooked
River
I I
Orogrondt
Regional Location
Culvert
Not to Seel*
AHL/WSU, 10/84
4-17
-------
EPA Channel Condition Project
MAIN REFERENCE: Orsborn, J. F., K. Amerman, B. Clark, K. Coulton, B. Naik
and J. Stypula. 1985. Planning for the restoration of meanders on a trial basis;
Crooked River habitat improvement project. Prepared for the USDA Forest
Service, Nez Perce National Forest, Elk City Ranger District, Elk City, Idaho.
Department of Civil and Environmental Engineering, Washington State
University, Pullman, WA.
MONITORING REFERENCE: Keifer, R. B. and J. N. Lockhart. 1994. Intensive
evaluation and monitoring of chinook salmon and steelhead trout production,
Crooked River and Upper Salmon River sites. Annual Progress Report for 1992.
Fisheries Research Section, IDFG. Prepared for USDOE, BPA, Portland, OR.
Project No. 91-73.
ABSTRACT:
Long reaches of Crooked River, southwest of Elk City, Idaho, were heavily
impacted by gold dredging in the 1940's and 1950's. Some reaches have been
pushed to one side of the valley, straightened and steepened. Vegetation, woody
debris, shade, overhanging banks, and pools ... components of diverse fish
habitat, are in extremely short supply. High velocity riffles and large substrate
are predominant. In essence, Crooked River has been turned upside down by
gold dredging.
Some preliminary habitat improvements were completed by the Forest Service
on several reaches of Crooked River. The project discussed in this report covers
the hydrologic, geomorphic, river mechanics and bio-engineering aspects of
considering the reconstruction of "pilot meanders" in a reach of Crooked River
about three miles north of Orogrande, Idaho (see Figure 4-5).
Consideration was given to several alternatives including: (1) installing habitat
structures and building a flood plain in existing, altered reaches of Crooked
River; (2) adding recessed backwater areas to the existing channel for rearing
habitat; (3) building one or two pilot meanders just north of the emergency
airstrip (the project reach); (4) cutting a more random channel through the
dredge tailings in a less-constricted valley area upstream (south) of the airstrip;
and (5) letting the stream continue to work towards its former natural state (do
nothing).
The last alternative is not reasonable in light of the time required for natural
restoration in this completely altered environment. The major risk in meander
restoration is the possible loss of water through the highly porous bed and
banks. These would seal over time, but can be corrected with the addition of
gravels, sands and fines during an initial, low-water diversion period. The
design of the meanders is based on similar channels in the region and calls for
lower than normal floodplains to encourage overbank flow, riparian vegetation
and bank stabilization.
4-18
-------
EPA Channel Condition Project
COMPONENTS:
In order to evaluate the alternatives for this study, the following functions were
completed:
• data collection (survey) and interpretation of existing land surface
conditions;
• relations of surface and groundwater elevations;
• sampling of surface water quality;
• a regional hydrologic analysis;
• a regional channel geometry analysis
• a hydraulic design of the pilot meanders including stability of the
channels, bed material size and the habitat characteristics of the
meanders;
• the new channels were sized by (1) comparing its plan view with
Tenmile Creek in the next valley to the West; (2) developing a channel
design based on regional models of channel size at bankfull flow; and
(3) fine tuning the design to fit the project site constraints such as
existing contours, swales and ponds; the elevation of the new
meanders with respect to the road; and the location of the north end of
the emergency airstrip (Figure 4-5).
An important component of this study was the discussion of methods for habitat
improvement. The five alternatives and the factors to be considered are in Table
4-4 with an explanation of terms. The Alternative(s) Matrix in Table 4-5 has been
completed to show an example of how the method was used to plan for the
alternatives. Further details on the matrix methodology are discussed in the
project report.
4-19
-------
EPA Channel Condition Project
Table 4-4: Conditions and Explanation of Terms (see text for details),
Crooked River habitat Improvements— Alternatives Matrix
10/34
AHL/WSU 10/84
Indict! Factors
To Consider
A, Earth
Moving
B, Add Habitat
Improvement
C. Surface and
GrourxNatar
Conditions
0. Stability
of channel
ind Hablut
improvement
E. Relative
TIM 10
Achieve
High
Productivity
F. Risk - or
Probability
Of SUCCeSS.
G. Other , . .
such as
Habitat
Diversity
To til Index
AH, (I):
Do Nothing
None
None
- FAST -
'oor poo];
— » o.
Poor -In
transition to
•eander.
Very lono.
High risk; or
CM proba-
blllty of
uccets.
Alt. (*): l«pro«
HibtUt In
Existing CNaiwel
Sow. to Instill
structures and
nodtfy Channel.
Only shade and a
few poor »tructur*j
ivillable ....
requires tot*)
flow control.
River on very steep
slope at higher
west ponds. G. W.
drops rapidly.
Run risk of high
•itntenance . . .
strew still steep
it high flows.
Hgch Quicker thin
Alt. (1) depending
OR hibftit iBprove-
•enti used.
Hlghtr risk of
dwHge to Mbltit
lHprove«cnt
structures.
Alt. (3)'. Construct M
-------
EPA Channel Condition Project
Table 4-5: Sample Analysis-- Use with Table 1-- Explanation of Terms
A
B
C
D
E
"
G
CROOKED RIVER HABITAT IMPROVEMENTS ALTERNATIVE MATRIX
F«r Mrw fl«*th ).»( »orth *t tktUM ta *.o. 30, TJiN. ME.
FACTORS
Ewth Uevtof
AM Habitat
ImprovMitMt
W.l.r Ce*4IHam
»wt*e» M»d t»<*»4
I utntty »f cr>MM*>
M»4 iMk. tar*.
fl«l«l(T* |lm»
to »cMav«
ht|h pf«4wcllv»tr
Probabllttv
• f IHCCMI
Habitat
Diversity
TOTAL INDEX
ALT. (11-
D» MOlhl*
10
10
1
1
1
1
1
25
ALT. If): !•*•*•
•itaUHf tftMM4
8
Z
7
6
7
8
9
47
ALT. 13): CMI
JL M»*n4«r 1
1
3
8
9
8
8
8
45
r»cl •••ntfcKi)
1. M«»fitfar 1
3
4
9
10
9
8
8
51
chnwvta
0
3
2
7
7
4
9C
32
•.t.H i..- ilFfl f>.>.. 8/84
(C»M*M* BA
7
5
5
7"
8
2
6d
40
•ftlr with IBt
5
10
8
8"
10
9
7*
57
a. 1 River; 9 Side Channel Combined.
b. 1 River; 10 Side Channel Combined.
c. Assumes Habitat Improvement'in Both Channels.
d. Hole: Factor D above assumes no improvements in river.
e. Ho River improvement.
AHL/WSU, 10/84
4-21
-------
EPA Channel Condition Project
ACTIVITIES:
• Topographic surveys of the project area were conducted over a period of
eight days to tie the proposed meander channels and the existing river
channel together.
• Stream flows were measured near existing USFS log weirs and at other
cross-sections along Crooked River to establish the existing channel
hydraulic geometry and the accretion or depletion flows.
• Cross-sections were measured along the two planned two meander lines
so the cuts and fills could be estimated.
• The interrelationships of surface and ground-water were measured by
surveying pond elevations and establishing staff gages in the ponds,
feeder streams and river (Figure 4-6).
• Electrical conductivity and temperature measurements were made in the
stream and in the ponds.
• In the hydrologic analysis, regional models were built relating
characteristic flows (low, average and flood flows) to basin characteristics
(area, average annual precipitation and basin relief).
• The flows estimated by the regional basin characteristics models
compared favorably with flows estimated by correlation of Crooked River
flows with same-day flows at USGS gages on the Lochsa and the S. F.
Clearwater Rivers.
• Using the three characteristic daily flows (low, average and flood) the low,
mean and high annual duration curves for Crooked River were estimated.
• For the channel design the following analyses were completed:
The Crooked River was compared with Tenmile Creek to determine
the stream gradient and the meander length for the restored meanders;
- regional models were developed for channel hydraulic geometries at
seven USGS gages for bankfull width, depth and cross-sectional area
(see Figure 4-7 for an example of the region model results);
- trial values of bankfull flows were estimated for different bottom
widths and bed slopes;
- values of W, D, V, shear stress, critical shear stress and channel slope
were estimated using equations developed for gravel-cobble streams
by Kellerhals (1967);
4-22
-------
EPA Channel Condition Project
Figure 4-6. Staff Gage Locations
Culvert
Crib 4 with
Strtom Goof "5
aw.
Survived S«cona° P-Lin«
KEY
© S. G. Staff Gog*
G. W. Groundwattr Flow
Oirtctlon
-Runway
AHL/WSU, 10/84
Not to Seal*
4-23
-------
EPA Channel Condition Project
Figure 4-7. Bankfull How Area (AB), Top Width (WB), and Mean Depth
(DB), Related to Bankfull Flow (QB) in the Crooked River Study
Region
IL iuui*
d
w
~ 600
CD
< 40°
UJ
cc
5 200
3
u.
i 100
t 60
03 Af\
> tw
I
| 20
0.
o
*" 10
Ll_ K
•_» D
CD E
0 4S
i" • F
, ______
1 .-- .;
S=r=
^
r- -..-.- 4
WB.,)=.
' . ! ; =
2.5f
i
_____
QB
)°_!_L=ry
— i
A
if
i?\. _.
I i i-j ; — — i — yt — 1 j j j*j
— hr
m
Zp— i
£ ^-~
— i ) •>• i .1 — i
-4— -1 . 1 ! 1 1
DB=0.2(QB}°-^-^
| i j \
— -i-| i,. -..-,11. r.., _. j.
-f^\
i
^ *i ' l 1
g 40 60 100 200
It:
_4—_,
WB(
r|rt-J
x. • 1 _Ji» 1 '& '
M M,^-- J^rJ
2)=1.8CQB)0'50 =
; — , — | 1 — !
F *— I i • . I
1 J
4006001000 2000 600C
4-24
-------
EPA Channel Condition Project
the Kellerhals (1967) bed material mean sizes were compared with
methods used by Jackson and Van Haveren (1984) and showed good
agreement;
the Jackson and Van Haveren (1984) methods involved solving for D50
in terms of the channel slope (S) and also in terms of unit stream
power (VS); and
habitat features given consideration were: pool to riffle ratio,
spawning gravel sizes, pool depth, revegetation of stream banks,
boulders for cover and flow deflection, stream shading, undercut
banks, slackwater and backwater areas.
RESULTS:
As it turned out the meanders in this reach were not built. A large area of
dredge spoils was crushed and stockpiled for USFS roads. Another spoils area
was set aside for historical preservation. Near the crushing operation, the Forest
Service installed a variety of habitat instream structures and revegetated the
stream banks. A concrete rearing pond was built on a leveled part of the pilot
meander project area to offset part of the impacts of the dams on the Snake River.
REFERENCES CITED IN THIS SUMMARY:
Jackson, W. L. and B. P. Van Haveren, 1984. Design of a stable channel in coarse
alluvium for riparian zone restoration. AWRA. Water Resources Bulletin
20(5). October.
Kellerhals, R. 1967. Stable channels with gravel-paved beds. Journal Waterways
and Harbors Division, ASCE, 93 (63-84).
4-25
-------
EPA Channel Condition Project
CASE STUDY 3. EVALUATION OF LAND USE IMPACTS ON BIG BEEF
CREEK
LOCATION:
South East Side of Hood Canal, the East water Boundary of the Olympic
Peninsula; on the West side of the Kitsap Peninsula in Western Washington;
Water Resource Inventory Area (WRIA) 15. T24N and T25N, R1W, near Seabeck;
Basin area: 38 km2
Figure 4-8. Location of USGS surface-water stations in the Hood Canal
Watershed (USGS 1995)
122° 30'
123" IT
l7Skokomi«h
EXPLANATION
™~ ••"— ButnboundAry
n.K.(||l( ,,.UOD
0 2 MILES
IVH
0 2 KILOMETERS
4-26
-------
EPA Channel Condition Project
MAIN REFERENCE: Made], Mary Ann. 1978. Response of a stream channel to
an increase in sediment load. MS Thesis. Department of Geological Sciences,
University of Washington. Seattle, WA (111 pages). (Summary of thesis
reviewed and approved by M.A. Made], 3/01).
OBJECTIVES:
• Assess the impacts of land use changes on the sediment load in Big Beef
Creek; and
• determine the effects of the increased sediment load on the geometry, and
thus the fish habitat, in the Big Beef Creek channel.
COMPONENTS:
A comprehensive base-line reconnaissance survey and monitoring of the impacts
of land-use changes on the channel geometry of a salmon-bearing stream.
Components of the study included:
• Review of previous studies;
• Background assessment of vegetation, climate, soils, geology, hydrology,
and land use;
• Description of stream reaches above and below man-made Lake
Symington;
• Channel cross-section surveys;
• Regional channel geometry surveys;
• Spatial distribution of sediment;
• Sediment sampling;
• Calculation of sediment budgets;
• Monitoring of sediment movements;
• Channel changes;
• Storage of sediment in the channel; and
• Relations of channel changes to land use.
ACTIVITIES:
• Resurveyed (in 1976-77) the cross-sections established by Cederholm
(1969) in lower Big Beef creek and at five other monitoring sections in the
lower 11 km (6.8 mi).
• Evaluated channel changes at the USGS gaging station (No. 12069550) for
the period of 1969-77 using USGS forms 9-207 and 9-275.
• Surveyed channel cross-sections in fifteen nearby streams (drainage areas
0.52-52.0 km2), measuring bankfull width, depth, slope and sediment size.
4-27
-------
EPA Channel Condition Project
• These measurements provided the regional channel geometry equations
that were used to estimate the undisturbed (pre-logging) conditions in Big
Beef Creek.
• The developed equations were: Eq. Nos.
Top Width: W= 1.60(Ab)°'42 (meters) (1)
Bankfull Depth: D = 0.14(Ab)°'24 (meters) (2)
Mean Sediment Diameter: D50 = 17.8/(Ab)0'10 (mm) (3)
Channel Gradient: Sc = 0.037/(Ab) °'18 (4)
where: Ab is the basin area in km2 .
• Suspended sediment was measured at several stations on Big Beef Creek
and sediment rating curves were constructed for three sites.
• Sediment transport rates were estimated by three methods.
• A painted rock experiment was used to determine the size of the largest
bed particle that was moved during high flows.
RESULTS:
• The cross-sectional surveys were conducted at 38 stations and 15 control
sites, and the results were used to evaluate the relative erosion or deposition
between reaches.
• The plot of the changes in the thalweg elevation and channel cross-
sectional area, as a function of drainage area between 1970 and 1977, was one of
the most revealing graphics.
• The exponent for width (W = aQb), b, changed from 0.30 to 0.17 between
1970 and 1977 indicating a shift to a more rectangular channel.
• Results of the regional channel geometry survey are shown in the
following table:
4-28
-------
EPA Channel Condition Project
Table 4-6. Predicted and Actual Dimensions of the Big Beef Creek Channel
at Station 50+50
Dimension
wb *
db*
D50**
S
Predicted
(Eqs 1 to 4)
8.5
0.43
15.0
0.0085
Predicted
(Kellerhals,
1966)
10.0
0.41
—
0.0017
Actual
15.5
0.40
22.0
0.0085
Units
m
m
mm
(-)
wb and db stand for width and depth at bankfull flow
refers to armor layer
• By plotting D50 against drainage area, the typical decrease in particle size
was observed, except in the middle watershed where intensive road
construction and logging increased D50 from about 45 to 60 mm.
• During high flows the suspended sediment concentrations increased from
3 ppm above the lake to 600 ppm six km below the lake.
• From the three suspended sediment rating curves (at Stas. 1 + 00,19 + 00
and 50 + 50 ft) the average sediment load was about 2325 t/yr, with a
higher contribution from the lower watershed.
• Log jams temporarily store sediment, and there were 4 jams in the 2 km
below the lake, and 14 jams to 18 jams (larger jams) in the lower reaches.
• Three approaches were used to estimate the source and volume of
sediment entering the stream, and its rate of movement; a sediment
budget; sediment transport equations; and survey measurements.
• The sediment analyses suggest that the sediment load to the stream
increased from about 525 t/yr for undisturbed conditions to about 4100
t/yr in 1977.
• The sediment transport rate equations did not agree with the volumes
measured by cross-section surveys.
The results were interpreted by the author in terms of actual modifications to the
channel. The deposition in the estuary increased, the lower channel widened
4-29
-------
EPA Channel Condition Project
about 25%, and the depth decreased by a similar amount. (This is typical for
widened, shallower channels; flow area stays about the same).
A decrease in Manning's resistance coefficient (n) was noted associated with a
small increase in sinuosity and the loss of bank vegetation due to widening.
Several methods were tested by the author using a computer program of
Einstein's bed load function, to evaluate the increase transport capacity as a
function of increased W/D values. Other factors considered were: armoring of
the channel bed; changes in Manning's (n); the width to depth ratios; limitations
on channel width; critical shear stresses; size distributions of the channel bed
materials; storage of sediment in the channel; and channel changes related to
land use.
Current conditions (year 2000) in the Big Beef Creek basin show decreased
logging, but increased, low-density urbanization around the basin perimeter.
The sediment wedge of deposition in the upper estuary is pronounced. The road
fill constructed across the mouth of the estuary, with about a 15% bridge
opening, has severely constrained the natural functions in the estuary.
The author's conclusions are quoted as statements of the impacts of land-use
changes on channel geometry and fish habitat.
CONCLUSION:
"In a forested watershed, sediment is supplied to the stream channel by the
processes of soil creep and mass movements. The channel form is adjusted to the
amount of the supplied sediment. Logging, road construction, and urban
development remove vegetation and cause accelerated soil erosion due to
sheetwash,. mass movements and rainsplash erosion, and hence increase the
sediment load of the stream. Under undisturbed conditions Big Beef Creek
would receive 525 t of sediment per year. Land use changes have caused an
increase to 4100 t/yr. Construction of a weir at the mouth of the stream has
caused the bedload fraction to be caught above the weir, where an average of
2100 t of sand and gravel are deposited annually.
"Dimensions of the stream channel have changed in order to adapt to the
increase in sediment load. The channel is wider and shallower than the 1970
channel, more gravel bars are present, and sinuosity has decreased slightly in
areas of high sediment transport. Mean flow velocity has remained
approximately constant. The result of the changes has been an increase in the
shear stress along the bed and banks, which in turn results in a higher rate of
sediment transport. Before disturbance, the sediment transport rate of bedload
was probably 230 t/yr, whereas it is 970 t/yr at the present.
"The adaptations of the channel are restricted by hydraulic constraints described
in the continuity and Manning equations. Velocity and slope are relatively
conservative parameters, and much of the change is taken up by the parameters
of width and depth.
4-30
-------
EPA Channel Condition Project
"The stream channel presently has 141,000 t of sediment in storage. Active
sediment, 49,000 t, moves an average of 200 m/yr in the main channel. Thus
sediment placed in the channel by present disturbances will take an average of
20-40 years to be removed." (M. A. M., 1978).
REFERENCES CITED IN THIS SUMMARY:
Cederholm, C. J. 1972. The short-term physical and biological effects of
stream channelization at Big Beef Creek, Kitsap County, Washington.
Masters Thesis, University of Washington. 80 pp.
Kellerhals, R. 1966. Stable channels with gravel-paved beds. American
Society of Civil Engineers, Water Resources Engineering Conference.
Preprint 330. Denver, CO. 28pp.
4-31
-------
EPA Channel Condition Project
CASE STUDY 4. LOWER ELWHA RIVER LOW FLOW RECONNAISSANCE
STUDY
MAIN REFERENCE:
Orsborn, J.F. and M.T. Orsborn, 1999. Low flow assessment of the Lower Elwha
River; effects of diversions on channel geometry and fish habitat. Lower
Elwha Tribal Fisheries. Port Angeles, WA.
BACKGROUND REFERENCES:
USDI National Park Service, 1994. The Elwha Report: Restoration of the Elwha
River Ecosystem & Native Anadromous Fisheries. National Park Service,
Department of Commerce and Lower Elwha S'Klallam Tribe.
USDI National Park Service, 1996. Final environmental impact statement: Elwha
River ecosystem restoration, Olympic National Park, Washington.
LOCATION:
Northern Olympic Peninsula, 4 miles west of Port Angeles, Sec. 3, T30N,
R7W and Sec. 34, T31N, R7W. See Figure 4-9.
OBJECTIVE:
The main objective of this study was to determine the influences of municipal
and industrial diversions from the Elwha River at RM 3.4 on the channel
hydraulic geometry (and thus on fish habitat) in the lower river, below the
diversion dam (Figure 4-10).
COMPONENTS:
The parts of the study included: the existing and developed databases; historical
hydrology; flow measurement sites; streamflow measuring procedures; data
management and analysis; results; and references. Examples of graphs, tables
and summary tables are interspersed throughout the project report and the
balance of the database and analytical work is contained in seven appendices.
4-32
-------
EPA Channel Condition Project
Figure 4-9. Location Map for the Lower Elwha Project (from USD I, NFS
1996).
Port Angeles
Elwha
River
Glines Canyon
Dam
Olympic National
Park Boundary
Tribal Hatchery
Slat* Hatchvry
4-33
-------
EPA Channel Condition Project
Figure 4-10. Schematic Representation of Measurement Sites and Facilities,
Lower Elwha Low Flow Study (Figure 1-1 in Orsborn 1999).
Site C2W
SG2
LETIndex
Channel
SG4
Site COW
Tribal Hatchery
Infiltration
Gallery
SG5
Site COE
State
Hatchery
Diversion Dam
Elwha Dam
USGS Gage #12045500
at McDonald Bridge
N
^industry
4-34
-------
EPA Channel Condition Project
BACKGROUND:
Some environmental factors, which have adversely affected the Lower Elwha
River geometry and fish habitat, include:
• the cessation of the upstream gravel supply caused by the dams;
• the continuation of large flood flows due to a lack of reservoir storage; and
• past storage and release operations which ramped flows sharply up and
down during the migration and spawning seasons.
As a result of the first two factors, the Lower Elwha River channel is extremely
short of bed material in the gravel size-range. Also, the channel is armored with 6
to 12-inch rock. The only gravels are old deposits in vegetated bars, and those
derived from local bank erosion. This combination of armoring and reduced
gravel supply, and a lack of LWD, have significantly depleted spawning habitat.
SITE CONDITIONS:
The main channel below the Municipal & Industrial (Figure 4-10) diversion dam
and the State hatchery is typically very wide and shallow with width to depth
ratios ranging from 40-220 for a lower flow range of 100-1000 cfs. (The average
annual flow is about 1500 cfs). Also, LWD is mostly locked in jams at the
entrances to secondary channels, the upstream supply having been curtailed by
the dams. The LWD supply is limited to the jams, and trees (mostly alders)
being undercut by bank erosion.
STUDY TASKS:
In order to evaluate the effects of the M & I diversions on downstream flow
conditions, channel geometry and habitat, the following information was
developed.
DATABASES:
Data for USGS stream gages; description of the M & I diversion system; water
rights total 205 cfs, but only about 40-70 cfs were diverted during the monitoring
period. Data gathered for the project included: historical hydrology; existing
and new cross-sections; site maps; channel elevation surveys; depth and
velocity measurements; and photographs. These data were analyzed to provide
information on flow amounts related to water surface elevations at selected sites
spaced throughout the Lower Elwha River.
HISTORICAL HYDROLOGY
Pre-and post dam natural and regulated flows were evaluated using the USGS
gages on the Lower Elwha River (12045500) and the N. F. Skokomish River
(12056500). Strong and irregular regulation of low flows by the dams until about
1955 caused poor, unstable habitat conditions. Since then low flow releases have
more closely followed natural inflows. Examples of poor and natural regulation
are shown in figures 4-11 and 4-12.
4-35
-------
EPA Channel Condition Project
10000
1000 -
01
Discharge
(cfs)
100 -
Figure 4-11. August to November, 1938 Hydrograph,
Recession Flows
Elwha Gage #12045500 & NF Skokomlsh Gage #12056500
1
-H-
i
'I
— o — ^^
¥S
...
4— •/
to
~\ P
!
* ~4
A At
1
0 10
8/1/38
[I
m
u •••
1
l_Jh r\ i
i
in
4p
~rt~
sU
v~
ll
""•
1 |11 j — UJft J ,
I 1) / 1 III 1
v 1 V\ /
1 V 'I V
I ^ ^
J
1
A #12056500 Min
0 #12045500 Min
Expon. (#12056500 Min)
Recession Curves for Low Flows
Elwha #12045500
y = 56.2e-°-018x
R2 = 0.55
NF Skokomish #12056500
y = 100.9e-°-012x
R2 = 0.96
20 30 40 50 60 70 80 90 100 110 120 130
9/1/38 10/1/38 11/1/38 Day After August 1st
10000 -
1000
01
Discharge
(cfs)
Figure 4-12. August to November, 1962 Hydrograph,
Recession Flows
Elwha Gage #12045500 & NF Skokomlsh Gage #12056500
a
r
\
i * \
n kn
I nil
"^••••Ot-vr^ :
'®--. \ / \
H i /" ' ^ -
A
A (i
^ V
v^ V
-^>^ /J >v
*— £l — "^^
w\4\ — ^^-vv-M^ — '
I n \ i \
1 / V . J \
11 \ r "
I VI
\l N
#12056500
#12045500
A #12056500 Min
0 #12045500 Min
Expon. (#12056500 Min)
Recession Curves for Low Flows
Elwha #12045500
y = 937.9e-°022x
R2 = 1.00
NF Skokomlsh #12056500
R2 = 1.00
0 10 20 30 40 50 60 70 80 90 100 110 120 130
8/1/62 9/1/62 10/1/62 11/1/62 Day After August 1st
4-36
-------
EPA Channel Condition Project
The mid-day of the average historical 7-day low flow with a 2-year recurrence
interval (Q7L2) occurs around October 7th, but it has occurred as early as August
17th and as late as November 28th. The 7-day annual low flows range between
202 and 606 cfs with an average of 404 cfs.
The Elwha River gage at McDonald Bridge was discontinued in 1997, but its
automatic telephone readout was supported by Daishowa America and was
functioning in 1999. The N. F. Skokomish gage was used as an index for flows in
the Elwha. By relating historical flows in the Elwha to those in the N. F.
Skokomish, we were able to correlate the 1998 measured project flows with
same-day flows in the North Fork Skokomish River while the Elwha gage at
McDonald Bridge was not in operation. Estimated flows from the performance
curves of the Lower Elwha Dam turbine-generators were obtained for
comparison with the correlated estimates.
STUDY SITES
The study sites were selected in two categories: (1) staff gage sites (SGS); and
(2) flow measuring sites (FMS), which also had staff gages for calibration with
the measured flows. These gaging sites are shown with respect to other project
features in Figure 4-10. Sites Al and B were used to measure the net amount of
diversion.
STREAMFLOW MEASURING PROCEDURES
We modified USGS standard forms to record staff stage readings and depth and
velocity measurements. Federal standards for streamflow measurements were
followed. All elevations were referenced to a local benchmark on top of the pin
at Station 0+00 on the left bank of the baseline transect at each FMS.
We selected measurement dates in September and October and were fortunate to
bracket the lowest flow of the year on October 1st. Velocity meters were
calibrated in ponds at the Tribal hatchery and were checked against each other in
the field. All flow measurements were made by wading, and two sites were
measured simultaneously by two teams of three people each. Paired
measurements were made between sites: (Al and B; C2E and C2W). We used
radios to transmit depth, velocity and baseline station data from the reader to the
recorder. Some later flow measurements were recorded directly into a computer
so the measured flow could be compared with the flow determined by reading
the staff gage, and entering this elevation into the rating curve. All staff gages
were read on days between flow measurement days.
DATA MANAGEMENT AND ANALYSIS
After the field data were acquired at each site, we: made backup copies of the
data sheets; developed rating curves; and calculated relationships between
streamflow and channel geometric characteristics.
Analytical results included: hydrographs of water surface elevation versus flow;
plots of flow versus the dates of measurements; staff gage rating curves;
surveyed channel cross-sections; depth and velocity profiles; channel cross-
sections for expanding the rating curves and the hydraulic geometry analyses;
4-37
-------
EPA Channel Condition Project
the measured hydraulic geometries; and the modeled (expanded) hydraulic
geometry, for lower and higher flows than were measured, to expand the rating
curves.
RESULTS:
Based on the study objective, and data acquisition and analyses, several methods
of presenting the results were selected to demonstrate how channel geometric
characteristics change as a function of flow reductions due to diversions. We
were dealing with "low flow" conditions and the effects of flow reductions on
channel geometry (width, depth, area and wetted perimeter), and the methods
selected are summarized below.
Using the four best measuring sites (Al, B, C2E and C2W; see Figure 4-10), the
flow geometry was analyzed by:
(1) tabulating incoming flows at Al, and net flows below the diversion
at B, C2E and C2W; the tabulated incoming flows ranged from 100-
1000 cfs in 100 cfs increments, and the diversions ranged from 20-200
cfs in 20 cfs increments; the results are in Table 4-7 and show how
the channel geometry characteristics change as a function of the net
flow at each site (only one page of the original eleven is included in
this summary);
(2) dividing the water surface width (W) by the mean depth (D), is a
commonly used habitat parameter; smaller W/D values (10-20)
usually indicate better fish habitat, but the Elwha values range from
about 40-220; this indicates very wide, shallow conditions (refer to
Figure 3-4 on page 2.16 for the Lower Elwha graphs of W/D vs.
P2/A, and the general relationships);
(3) dividing the water surface width(W) by the flow area (A) gives an
index of the solar heating surface divided by the volume of water
available to absorb heat; smaller ratios are better (such as 0.10-0.20),
but the Lower Elwha ratios range from about 0.40-1.40;
(4) when the wetted perimeter (P) is plotted versus flow (Q), P initially
increases rapidly (above Q = 0); then the graph flattens so that a large
change in Q makes only a relatively small change in P. This method
is called the "wetted perimeter" or "toe-width" method. Because of
the shallow conditions at all four channel sites, P begins to be
reduced more rapidly below 300 cfs; a larger reduction in P occurs
below 100 cfs.
The application of the study results will be governed by the management
objectives in place at the time of their application. The balance of the report
expands the information presented in the summary. The appendices contain the
background information, databases and analytical tools.
4-38
-------
EPA Channel Condition Project
Since the fieldwork for this study was completed in October 1998, high flows have
enlarged the west channel. It is now carrying closer to 60% of the total flow as opposed
to 40% during the low flow period. The west channel is being carved through the base of
a large forested gravel bar (Figure 4-10).
Table 4-7. Example of Modeled Hydraulic Geometry Parameters for Various Flows at Site Al and Different Rates of Diversion
Power Relation Rating Curves for Sites
Q_A1 Q_DIV ID_SITE Q_SITE
Ids) (ds) (els)
W
(It)
D
(ft)
W/D
A
(Ifi)
W/A
P
(ft)
A/P"2 P*2/A
V
(Ips)
R
(It)
1 00
1 00
1 00
1 00
2 0
2 0
2 0
2 0
1 00
1 00
1 00
1 00
1 00
1 00
1 00
1 00
1 00
1 00
1 00
1 00
4 0
4 0
4 0
4 0
Elwha A1
Elwha B
Elwha C2E
Elwha C2W
100.0
80.0
44.0
31.3
Elwha A1
Elwha B
Elwha C2E
Elwha C2W
6 0
6 0
6 0
6 0
8 0
8 0
8 0
8 0
ffElwha A1
Elwha B
Elwha C2E
Elwha C2W
Elwha A1
Elwha B
Elwha C2E
Elwha C2W
100.0
60.0
33.0
23.5
154.0
113.8
112.2
54.1
154.0
106.7
109.2
53.3
0.90
1.20
0.91
1.02
0.90
1.08
0.83
0.90
100.0
40.0
22.0
..15.6
100.0
20.0
11.0
7.8
154.0
93.8
104.5
51.1
154.0
76.2
91.6
40.5
0.90
0.96
0.72
0.75
0.90
0.74
0.59
0.62
170.75
94.92
123. 6C
53.07
138.9
136.5
101.9
55.2
1.11
0.83
1.1C
0.98
154.3
114.0
112.4
55.1
0.0058
0.0105
0.0081
0.0182
170.75
98.77
131.94
59.33
170.75
97.85
144.75
68.06
170.75
103.49
156.00
65.15
138.9
115.3
90.4
47.9
138.9
89.8
75.4
38.3
138.9
56.0
53.8
25.2
1.11
0.93
1.21
1.11
1.11
1.04
1.39
1.33
1.11
1.36
1.70
1.61
154.3
106.9
109.3
54.2
154.3
93.9
104.5
51.8
154.3
76.3
91.7
41.1
0.0058
0.0101
0.0076
0.0163
0.0058
0.0102
0.0069
0.0143
0.0058
0.0096
0.0064
0.0149
171.5
95.2
123.9
55.0
0.72
0.59
0.43
0.57
171.5
99.1
132.3
61.3
171.5
98.1
144.9
70.1
171.5
103.7
156.3
67.1
0.72
0.52
0.37
0.49
0.72
0.45
0.29
0.41
0.72
0.36
_ 0.20
0.31
0.90
1.20
0 91
1.00
0.90
1.08
0.83
0.88
0.90
0.96
0.72
0.74
0.90
0.73
0.59
0.61
Definition ot Terms tor Table 4-7
Term Definition
Q_A1 Flow at Site A1 Used in Model
Q_DIV Diversion Flow Used in Model
ID_SITE Site Name
Q_SITE Corresponding Flow at each Site
(from Correlations App. VII)
Term
W
D
W/D
A
W/A
Definition
Channel Width
Channel Depth
Width to Depth Ratio
Channel Area
Width to Area Ratio
Term Definition
P Wetted Perimeter
A/P"2 Area to Wetted Perimeter Ratio
P«2/A Wetted Perimeter to Area Ratio
V Mean Velocity
R Hydraulic Radius
4-39
-------
EPA Channel Condition Project
Comparative Notes on the Four Example Projects
The LeBar Creek, Crooked River, Big Beef Creek and Lower Elwha River projects
had many tasks in common, and each project had some unique tasks. The tasks
and objectives are summarized in Table 4-6. The ratings from 0-10 are assigned
values primarily based on the relative emphases within each project, and some
secondary comparative analysis between projects.
Table 4 -8. Comparative Emphasis on Tasks and Objectives for Four
Example Projects (0-10 High in Relative Amount of Activity, or
Importance to Each Project)
TASKS and
OBJECTIVES
LeBar Creek
WA
Crooked
River
ID
Big Beef
Creek
WA
Lower Elwha
River
WA
Tasks
Hydrology
Channel Geometry
Channel Morphology
Geomorphology/Soils
Sediment Effects
Basin Land Use
Stream Impacts
Groundwater
Stream Improvement
Habitat Improvement
Basin Improvement
Reconnaissance,
Design, Habitat
Improvement,
Restoration
10
8
6
8
8
10
7
0
8
8
9
Reconnaissance,
Design
Alternatives,
Reconstruction
8
10
9
8
9
1
10
8
10
10
0
Channel Impact, Diversion
Analysis of Impacts on
Land Use, Habitat,
Reconnaissance Reconnaissance
5
10
10
10
10
8
9
0
6
4
6
10
10
10
8
10
0
10
8*
8*
8*
8*
Objectives
Planning Alternatives
Design
Estimate Habitat
Documentation
Analysis
Build Habitat
Arrest Erosion
6
10
10
10
8
10
8
10
10
7
10
10
9
7
4
0
5
10
10
6
8
6
0
8
10
10
0
0
"Lower Elwha deals with impacts of dams on downstream channel beds.
4-40
-------
EPA Channel Condition Project
For example, compare the Hydrology Task ratings in Table 4-7 for each project:
10 for LeBar Creek, 8 for Crooked River, 5 for Big Beef Creek, and 10 for Lower
Elwha. In LeBar, the hydrology was used to determine bankfull floods, low
fish passage flows and channel size. Hydrologic analysis was not as
comprehensive for the Crooked River (8) because only ranges of high flows were
needed to size alternative channels. In the case of Big Beef Creek, there were 8
years of USGS records, and analyses were done to verify average annual and
average peak flows, as well as the channel size. The hydrologic analysis for the
Lower Elwha River project dealt with reservoir storage effects, calibration of
channel sections, correlation with another USGS gage and net flow after
diversion.
Whereas the LeBar Creek project emphasized the improvement of basin stability
and instream habitat, it had an equal component of off-stream, juvenile coho
rearing habitat on the point bar. The point bar channels and ponds had been
disconnected from the main channel by construction activities associated with
logging. Crooked River had to contend with a complete overturning of the
valley floor deposits, as well as channel straightening. In the lower impacted
reach the natural meanders were reformed where the dredges cut a regular,
unnatural zigzag pattern with 90 percent pool and 10 percent riffle. The pools
became huge sand traps and the gravel-sized transport became nil. This was
mainly due to the sorting and redeposit of the gold-bearing sands and gravels
under mounds of cobble.
In the case of Big Beef Creek, road building, logging and suburbanization caused
a rapid increase in sediment load, and impacts on channel geometry and coho
habitat from which the stream has not recovered after 30 years.
The Lower Elwha case study dealt with the downstream impacts of two dams,
plus the effects of M & I diversions below the dams, on low flows, channel
geometry and habitat. All of the projects emphasized the documentation of
stream channel condition through stream surveys, aerial surveys, maps, regional
channel geometry models and/or other measures of present and past of channel
conditions such as local geology.
Although each project had a somewhat unique history, they all had the common
thread of man's infinite capacity to modify the finite natural environment in the
name of progress and/or profit. Also, the projects reflect a lack of application of
county, state and federal regulations that has resulted in a situation that is now
upon us, the ESA.
4-41
-------
EPA Channel Condition Project
5. SUMMARY AND CONCLUSIONS
In order to evaluate methods for determining the physical condition of a channel,
we arranged this study around the physical relationships between basin
characteristics (BC), channel characteristics (CC) and the link between basins and
channels, the stream flow characteristics (QC). These are the combined fluvial-
geomorphic relationships based on available streamflow records. These
relationships can be used to estimate historical and future channel conditions.
To introduce these concepts we began by posing some questions about natural
and unnatural conditions in a stream, and asked, "What is natural?" A series of
settings was listed for which channel condition studies are conducted, ranging
from channel capacity to more detailed instream flow habitat analyses.
In Part 2, our discussion of fundamentals which can be applied to channel
condition studies, we summarized topics from: the current state of our
knowledge about river width adjustment; dimensional analysis and its
importance to geomorphology, hydrology and channel hydraulics; channel
hydraulic geometry (channel dimensions related to flow) at-a-site and regionally
at a series of sites; the influence of flow reductions on channel characteristics and
thus habitat; and steps for using channel indices as tools to protect and recover
stream habitat.
In Part 3 we reviewed some old methods and introduced some new ones for
estimating channel characteristics (CC) as a function of both basin (BC) and
stream flow (QC) characteristics. To examine these relationships we chose three
regions in Washington for which the necessary data bases had already been
developed: the Olympic Peninsula, part of the Puget Lowlands; and northeastern
Washington. We could not avoid streams that have already been impacted by
logging, urbanization and agricultural activities. But these effects showed up in
the analyses as widened streams, those in bedrock, streams where the banks had
been armored and those in which the hydrologic regimes had been altered. So,
there is a mixture of natural and altered data in our analyses, but this is one of
the problems with which we are all faced, as long as we recognize the alterations.
By using combined solutions of channel characteristics (such as width, depth,
area and wetted perimeter) as functions of both basin characteristics and flow
characteristics, we were usually able to improve on the strictly empirical graphs
of say just width related to drainage area. However, in northeastern Washington
there were some instances where channel width to basin area relationships were
very strong. In all of these regional relationships they were done for: (1) a
particular flow such as the average flood, average annual flow and average low
flow; and (2) they depended on USGS gaging station calibration records for the
measured channel characteristics. These could have been changing over time,
and so we used average power equations in all relationships. We devoted Part 3
5-1
-------
EPA Channel Condition Project
in its entirety to the analysis of regional channel stream cross-sectional geometry
(W, D, Ac) as functions of basin characteristics (area, Ab; relief, H; and average
annual precipitation, P). Part 3 concludes with a comparison of the three
regional sets of equations and ranges of flow values for each region.
In Part 4 we selected four example projects to demonstrate the application of
channel condition studies to streams in Washington and Idaho. Reconnaissance,
restoration and reconstruction were three terms we used to compare these
example projects. LeBar Creek, a tributary to the South Fork Skokomish River
(SE Olympic Peninsula) was a habitat restoration project. Crooked River, a gold-
dredged tributary to the S. F. Clearwater in north central Idaho, was a
reconnaissance and planning study of stream and habitat reconstruction
alternatives. Big Beef Creek, a tributary to SE Hood Canal, was a reconnaissance
and analytical study to determine pre-logging and post-logging conditions and
sediment loads in the stream. On the Lower Elwha River, the fourth example
project was a detailed reconnaissance study. The effects of the diversions on
channel geometry and habitat were documented. The calibrations were
expanded to any combination of streamflow and diversion flow. The net flow in
the Elwha River just below the diversion was projected to the downstream
monitoring transects.
There are a variety of circumstances, which are fundamental to any channel
condition study. Some of these are listed here:
• In a recent document Reid and Furniss (2000) discussed the use of
regional channel-based indicators for monitoring purposes. Their
conclusion was that there is no general solution to "the monitoring
problem", and that no single set of indicators is applicable everywhere.
"In channel physical parameters often are the most useful monitoring
variables for such applications (e.g. for cause-effect, or hypothesized,
relationships), but in each case the variables used are selected to be
relevant to the specific application" (Reid and Furniss 2000). This is
why we chose to use dimensional analysis for each component we
explored whether it was basin, flow or channel characteristics.
Dimensional analysis in each case relates the solution for the
dependent variable in terms of dimensionless numbers, thus reducing
scale effects.
• The ASCE Task Committee (TC) on Hydraulics, Bank Mechanics and
Modeling of River Width Adjustment in 1998 reviewed and evaluated
the current methods of predicting equilibrium (channel) width
adjustments. The TC's first recommendation proposed that stream
reconnaissance procedures should be developed that emphasizes the
geomorphic context of width adjustments.
• We demonstrated that using channel hydraulic geometry in the
geomorphic context we could relate channel to basin characteristics
more comprehensively.
5-2
-------
EPA Channel Condition Project
The graded stream and regime theory concepts both use average
conditions and the variability in those average conditions, as we have
done in this report.
A new approach for calculating channel flow (without using
Manning's equation) as a function of regional hydraulic geometry and
a shear-shape relationship in the stream channel has been
demonstrated (Table 2-2, page 2-15). This relationship could be
combined with basin characteristics to estimate W/D and P2/AC values
for channels, but this phase has not yet been accomplished.
Williams (1978) found that predictions of channel size using the
exponents of width in hydraulic geometry are more reliable than the
exponents of depth and velocity.
A simple hydraulic geometry analysis of three data points at low, near
average and high (within the banks) flows will fall within the range of
any hydraulic geometry analysis done with more data points (Figure
2-5, page 2-19).
A first-level channel geometry analysis can be conducted with a
minimum amount of information: a cross-section (or series of cross-
sections in riffles) from tops of banks or high water marks; measure the
flow while doing the transect; make hydrologic model estimates of the
three characteristic flows (Q7L2, QAA, Q1F2); insert the estimated
flows into regional hydraulic geometry to obtain W, D, Ac and P
estimates; and conduct a graphical comparison of the estimated values
(by regional hydraulic geometry models) versus the field measured
values. This will tell you which parameters are within "reasonable"
accuracy based on the combined accuracy of the regional gaging
records, the channel cross-sections and the flow measurements you
made.
The variability in streamflow periods of record sometimes causes
errors in regional hydraulic geometry models, especially at the average
flood flows and average low flows; dry or wet spells during the shorter
periods of record may skew the analysis.
Short periods of record should be compared with a regional, long-term
"base" USGS gaging station to determine coincidence with wet or dry
cycles (Orsborn and Orsborn 2000).
The Severity Factor Analysis (page 2-19) is a flexible, straightforward
way to evaluate the influences of flow reductions on channel
geometry, and habitat features.
5-3
-------
EPA Channel Condition Project
• Plotting undistorted channel cross-sections will help in the
visualization of the true channel shape.
• Some examples of stream reaches that are in a natural or an unnatural
condition are listed below:
Natural Stream Reaches
a meandering meadow stream with grassy banks and very little, if
any, in-channel LWD;
a stream in old growth timber with numerous pieces of LWD and
good habitat diversity;
a stream with no buffer strips in a rocky geologic environment with no
flood plain;
a stream with buffer strips in which some blowdown has occurred;
a braided stream with large amounts of LWD on the bars at the outlet
of a canyon;
a stream flowing in a forced meander pattern in a canyon between
side controls of rock outcrops, and with no LWD;
a stream with a large mass-wasting deposit from a hill slope failure;
the stream immediately begins to store water upstream of the slide,
wash out fines, and downcut through the fill until it reaches an armor
layer and a graded (equilibrium) state; and
streams that have been heavily impacted by extreme floods or
droughts in natural environments.
Altered Stream Reaches
any stream reach which has had its natural flow regime changed by
reducing floods, or diverting flows, and increasing or decreasing low
flows;
urbanizing basins result in similar alterations of the natural stream
flow regime as do storage and diversion projects;
5-4
-------
EPA Channel Condition Project
streams that have been heavily impacted by large floods due to
unnatural flow releases from dams;
a reach near a clear-cut;
a reach in a clear-cut with no buffers;
a stream reach with a hardened side (or sides) that limit the capability
of the stream to deform naturally;
streams whose valleys are blocked by road fills for 80-90 per cent of
their valley width leaving only a culvert or bridge opening;
these valley constrictions cause contractions that dam the flow, raise
flood levels upstream, change channel patterns up- and downstream,
and
even more importantly, road fills at the upstream ends of estuaries,
keep the estuaries from fully functioning to their natural potential (e.g.
the Skokomish River estuary operates while being choked (throttled
down) by two road fills, each having 15% of the valley width left open
at bridges; requiring that all new and replacement bridges be designed
to have their approaches built on pilings would restore the estuaries to
a much improved, near-natural state.
The usual reaction to the estuary-road fill problem is to say that pilings would be
too expensive—compared to what? These reaches of streams upstream and
downstream of any road fill have been impacted and opportunities for
improvement have been foregone. Reconstruction and restoration of estuary
roads and estuary functions will be crucial to ESA opportunities. Not all road fill
across valleys need to be placed on pilings, but their hydraulic competencies
need to be checked and improved.
In conclusion, let us review the primary purpose of the project: "to evaluate the
concept of regional indices of channel morphology and to determine if they (the
regional indices) can provide a useful diagnostic and predictive tool to help
evaluate existing and potential channel characteristics" (page 1-1). The answer to
both of these questions is a qualified "yes", qualified in that the quality of the
answers depends on the quality of the data base (regional streamflow, basin,
land-use, precipitation, in-channel and stream corridor records). The steps
involved in any evaluation of channel condition have been summarized in
various places throughout the report and are repeated here in conclusion.
The general procedure for approaching width-adjustment analyses was outlined
by the ASCE Task Committee on page 2-2 and is modified below:
(1) Problem identification (including careful definition);
(2) Reconnaissance and data collection;
5-5
-------
EPA Channel Condition Project
(3) Desk assessment of channel conditions;
(4) Application of empirical channel regional models;
(5) Application of numerical models (if warranted);
(6) Validate the model results against field data;
(7) Numerical models should be applied to existing conditions and to
assess any known or anticipated future impacts;
(8) comparison and assessment of alternatives; and
(9) Selection of a solution.
Note in Steps 5 and 6 that value judgements have to be made as to whether or
not numerical models should be applied and if enough field data is available for
validation. Recall that simpler models are better in that they are less data
intensive. Note also that key words in each step have been made bold for
emphasis.
Near the end of Part 2 (page 2-27) we listed some conditions to consider in
accounting for changes in channel geometry dealing with scales of indices,
followed by a systematic method of analysis.
Evaluation Conditions
(1) A channel may be "in balance" with its water and debris load, and
still not fit a cross-sectional template for the region due to geologic or
human geometric constraints;
(2) the main stream channel may be underfit due to excessive diversions
of flow out of the watershed, and the accumulation of sediment in
the mainstem from unaffected tributary sediment flows;
(3) the channel may be over- or under-sized due to a modified flow
regime caused by either a natural extended increase of decrease in
flow, or a regulated flow regime, or both; and
(4) an historical mass wasting may have been deposited in a stream
valley, and the stream is now downcutting (as a function of the
existing flow regime) with a narrower, deeper channel than
"normal".
Analytical Steps
Review of historical records of flow (database):
(1) a method of classification is used to put some geomorphic
boundaries on the site being investigated, and to help in the
visualization of the site;
(2) a simple hydrologic analysis to estimate the characteristic flows at a
site (average low (Q7L2), average annual (QAA) and average flood
5-6
-------
EPA Channel Condition Project
(Q1F2)), and major changes in these characteristic flows and in
precipitation over time;
(3) an abbreviated analysis of the channel hydraulic geometry of the site
to provide relationships of geometric characteristic (W, D, V, A and
P) as a function of discharge;
(4) regional channel hydraulic geometry models for comparison with the
present site geometry, and with similar, template streams;
(5) an integrating analysis of how the W/D ratio, and other geometric
dimensionless ratios, change as a function of streamflow reduction; a
type of severity factor analysis which ties flow to geometric
characteristics which serve as analogs to water quantity and quality
parameters; and
(6) an assessment of the history of major land-use and water-use
changes on the watershed.
Can you imagine the condition that our streams and fish stocks would be in now
if, as was proposed in the early 1970's, buffers had been mandated by the Forest
Practices Board on all streams, clear to the drainage divides?
Ruling on a controversy over logging in a redwood forest in California,
Judge R. H. Kroniger wrote the following: "While numerous expert
witnesses in the field of geology, forestry, engineering, and biology were
presented, their conclusions and the opinions they derived from them are
hopelessly irreconcilable in such critical questions as how much and how
far solid particles will be moved by any given flow of surface water. They
were able to agree only that sediment will not be transported upstream'
[State of California, Mann County versus E. Richetti and others, 1969].
(Wolman 1977 in Burkham 1981).
And as for Rivers, I believe it is evident, that they are furnished by a
superior circulation of Vapours drawn from the Sea by the heat of the sun,
which by Calculation are abundantly sufficient for such a supply. For it is
certain that nature never provides two distinct ways to produce the same
effect, when one will serve. But the increase and decrease of Rivers,
according to wet and dry Seasons of the year, do sufficiently show their
Origination from a Superior circulation of Rains and Vapours (From
John Keill 1698, in White 1968)
5-1
-------
EPA Channel Condition Project
References for Part 5
ASCE Task Committee on Hydraulics, Bank Mechanics and Modeling of River
Width Adjustment. 1998. River width adjustment. I: Processes and
mechanisms; II: Modeling. Journal of Hydraulic Engineering, Vol. 124,
No. 9. Paper No. 14412, pp. 881 - 902 and 903 - 917. Discussion and
closure: Feb. 2000, pp. 159-164. Quotation of TC Objectives by
permission.
Burkham, D. E. 1981. Uncertainties resulting from changes in river form.
Proceedings of ASCE, Vol. 197, No. HY5, May.
Orsborn, J.F. and M.T. Orsborn 2000. Streamflow Characteristics of the Big and
Little Quilcene Rivers. Prepared for the City of Port Townsend,
Washington.
Reid, L. M. and M. J. Furniss. 2000. On the use of regional channel-based
indicators for monitoring. In press.
White, G.W. 1968. John Keill's view of the hydrologic cycle, 1698. Amer.
Geophys. Union Water Resources Res. v. 4, no. 6, p. 1361-1374.
Williams, G. P. 1978. Hydraulic geometry of river cross sections - theory of
minimum variance. USGS Professional Paper 1029. USGPO, Washington,
DC.
Wolman, M. G. 1977. Changing needs and opportunities in the sediment field.
Water Resources Research. Vol. 13, No. 1, pp. 50-54.
5-8
-------
EPA Channel Condition Project
6. NOTATION
English Gravitational System (EGS) of Units followed by
Dimensions of Force (F), Mass (M), Length (L) and Time (T).
Symbol Description
EGS Units Dimensions
a
a-j
Ab
Ac
BC
BE
B3, B4, C3, etc.
C
CC
D
Dma*
E
acceleration
empirical coefficients and
exponents in hydraulic geometry
equations (e.g. W = aQb )
basin area
channel flow area
basin characteristics
basin energy, AH °-50
channel types
general notation for coefficients
channel characteristics
mean depth of flow, Ac AV
maximum flow depth
general notation for exponents
ft/sec2
(-)
mi2
ft2
(-)
mi2'50
(-)
(-)
(-)
ft
ft
(-)
L/T2
(-)
L2
L2
(-)
L2'50
(-)
(-)
(-)
L
L
(-)
6-1
-------
EPA Channel Condition Project
Symbol Description
EGS Units Dimensions
E
F
F
g
H
LR/LV
LS
M
NF
n
P
PAb
PBE
mean basin elevation above mean
sea level
forest cover
force
acceleration due to gravity
basin relief (potential energy)
River Length / Valley Length
length of perennial streams (solid
blue lines on USGS topographic
maps)
mass from matter
Froude No., dimensionless ratio
of water velocity to gravity wave
velocity, or inertia to gravity
forces.
Examples: NF (Channel),
NF(Model), NF(Prototype),
NF(Watershed)
Manning's resistance coefficient
average annual precipitation
average annual inflow to the
basin
average annual precipitation
times basin energy
ft
%
lbsF
ft/sec2
mi
(-)
mi
lbsM
(-)
sec/fta33
in/yr
mi2 - in/yr
mi Z50 • in/yr
L
(-)
F
L/T2
L
(-)
L
M
(-)
T/L0.33
L/T
L3/T
L3'5/T
6-2
-------
EPA Channel Condition Project
Symbol
Description
EGS Units Dimensions
p
Q
QAA
QAD
QC
Q1/Q2
QPFMax
QPF2
QPF25
QPF50
QPF100
Q1F2
Q7L2
QX
wetted perimeter of channel or
conduit
stream flow
average annual flow
(also called QAD )
average daily flow
(see QAA)
characteristic flows (Q1F2, QAA,
Q7L2, ...) at a gage or site
dimensionless flow reduction
ratio
maximum instantaneous peak
flow of record
average peak flood
(with a 2-yr RI)
average peak flood
(with a 25-yr RI)
average peak flood
(with a 50-yr RI)
average peak flood
(with a 100-yrRI)
average daily flood (1-day) with a
2-yrRI;
also Q1F25, Q1F50 and Q1F100
seven-day average low flow with
2yrRI;
also Q7L20, Q30L2, etc.
any characteristic flow
ft
ft3/sec
(or cfs)
ft3/sec
(or cfs)
ft3/sec
(or cfs)
ft3/sec
(or cfs)
(-)
ftVsec
(or cfs)
ftVsec
(or cfs)
ftVsec
(or cfs)
ftVsec
(or cfs)
ftVsec
(or cfs)
ftVsec
(or cfs)
ftVsec
(or cfs)
ftVsec
(or cfs)
L
L3/T
L3/T
L3/T
L3/T
(-)
L3/T
L3/T
L3/T
L3/T
L3/T
L3/T
L3/T
L3/T
6-3
-------
EPA Channel Condition Project
Symbol Description
EG S Units Dimensions
R2
RI
V
w
W/D
W2:D2/W1:D1
XSF
ISF
X
y
correlation coefficient
recurrence interval
mean flow velocity
water surface width
dimensionless width to depth
ratio of channel at a particular
flow
dimensionless width to depth
reduction, ratio used in Severity
Factor Analysis
multiple Severity Factor
summation Severity Factor
horizontal axis on graphs;
independent variable
vertical axis on graphs;
dependent variable
(-)
years
ft/sec
ft
(-)
(-)
(-)
(-)
(-)
T
L/T
L
(-)
(-)
(-)
(-)
6-4
-------
EPA Channel Condition Project
7. INDEX OF TERMS AND AUTHORS
acceleration due to gravity 1-3, 2-7,2-8
Ackers 2-11
altered state 1-1
Amerman 2-11, 2-26, 2-30,2-31,3-1, 3-10,
3-28
analysis
allometric 1-6
multiple regression 2-6
Arnold 1-1
ASCE 2-1,2-2,2-3,5-2,5-5
average behavior 1-3
basin energy 2-8,2-26,3-1,3-50,3-60
basin input 3-1
bed material mean size 2-11,4-25
Benson 2-6
Big Beef Creek 1-1,4-3,4-27,4-28, 4-29,
4-30,4-40,4-41,5-2
Blench ,1-3
Booth , ...1-1,4-1
Brookes.. 4-3
Buckingham Pi theorem 2-7
Burkharn 1-3,3-74,5-7
case studies .1-5,4-3
CCT Reservation 3-50,3-66
channel
condition of a 2-27
in balance 2-27
non-rectangular 2-12
rectangular 1-4,2-12, 2-14,2-20,2-22
trapezoidal 2-22
triangular 2-22
underfit. 2-27
wide, shallow 2-14
channel condition 1-2,1-6,2-3,3-1,3-71,
3-72,4-41,5-1, 5-2,5-5
channel dimensions 2-27,3-2,3-31,3-50,
4-1,5-1
channel geometry 4-2
channel hydraulic geometry 1-2,2-17,
2-18,2-23,2-28,3-50,4-22,4-32,5-1,5-2,5-
7
at-a-station 2-23
regional 2-23,2-28
channel morphology 1-1,4-4,5-5
channel width 2-1,2-3,2-27,2-30,2-31,
3-21, 3-50, 4-30, 5-1
channel width to depth ratios 2-24
characteristic flows 2-4,2-28,3-1, 3-2,3-28,
3-54,3-71,3-72,4-22,5-3,5-6,5-7
characteristics
basin (BC) 1-4,1-5,2-1,2-4,2-5,2-6,
2-7,2-18, 2-23,2-26,2-27,2-29,2-30,3-
1,3-2,3-16,3-28,3-41,3-50,3-54,3-60,
3-66,3-71,4-2,4-22,5-1,5-2,5-3
channel (CC) 1-1,1-2,1-3,1-4,1-5,2-1,
2-3,2-4, 2-5,2-14,2-18,2-23, 2-26, 2-27,
3-1,3-2,3-11,3-30,3-39, 3-50,5-1,5-2,
5-5
flow (QC) 2-4,2-5,2-6, 2-7,2-12,2-14,2-
20,2-23,3-50,4-2,5-1
Charlton 2-11
Chitale... ...2-11
Chrostowski 2-20,2-22,2-24
dear-cut 2-29,5-5
Clearwater River 4-3, 4-17, 4-22,5-2
coefficient
runoff 2-31,2-32
Colville Indian Reservation 2-3,2-23, 3-50
Colville Tribe 2-18
condition of the stream 1-1
conditions
existing 1-2, 2-2, 5-6
pre-logging 2-30
unnatural 1-1
Crooked River 4-3,4-17,4-18,4-20,4-22,
4-24,4-40,4-41,5-2
cross-section
bedrock 3-28
large boulder 3-28
Dasman 1-6
Deane 2-20,2-22,2-24
Definitions 1-2
dimensional analysis.. 2-7
dimensionless ratios 1-3,1-4,5-7
geometric 2-28
downcutting 2-28,5-6
Duckabush River ....3-28,4-13
Dungeness River 2-23, 3-28,4-13
Elwha River 4-32, 4-37,4-38, 5-2
Lower 2-14,4-3,4-33,4-34,4-35,4-37,
4-38,4-40,4-41,5-2
Endangered Species Act (ESA) 1-1,4-2, 4-
41,5-5
equation
combined 3-30
fish habitat 1-1,1-2,1-5,2-5,4-2,4-6,4-7,
4-18,4-27,4-30
flood
average daily 2-31
mean daily (Q1F2) 3-16, 3-39
peak 2-8, 2-29,2-30
flow reductions 2-20
flow regime 1-3, 2-27,2-28,5-4,5-6
flows
average annual (QAA) 3-21
peak, pre- and post-logging 2-30
7-1
-------
EPA Channel Condition Project
form
9-207 2-27,3-31
9-275 2-27
Froude number 1-4,2-7,2-11
channel 1-3
dimensionless 2-8
of the basin 2-7, 2-23
Strahler basin 2-7
watershed 1-3
fundamentals 1-2,1-4
Furniss 5-2
Gaboury 4-3
geology 3-30
geomorphology
fluvial 4-3
gold dredging 4-18
Griffin Creek 3-31, 3-39, 3-41
habitat assessment 1-1
Hall Creek 3-50
Haller Creek 3-50,3-66
Hammer 1-1
Hangman Creek 3-66, 3-72, 3-73
Hoh River 3-28,3-72,3-73
Horton 2-11
hydraulic geometry 3-16,3-71
at-a-station 2-18,2-23,3-1,3-2,3-21,3-41
regional 2-18,3-1, 3-2,3-16,3-21,3-39,
3-41,3-44, 3-54, 3-66, 3-71,4-14,5-3
hydrologic analysis 2-28,4-7,4-8,4-19,
4-22,4-41,5-6
impervious surface 4-1
index of forest cover (F) 2-6
Issaquah Creek 3-31
Johnson 3-1,3-31
Keifer 4-18
Knight 4-1
LaMarche 2-32
land use
activities 1-1, 4-7,4-10,4-12
effects of 2-3
laws governing stream systems 2-11
laws of stream order 2-11
LeBar Creek 4-3,4-5, 4-6,4-7,4-8, 4-9, 4-10,
4-11,4-12, 4-13,4-14,4-15,4-16,4-40, 4-41,
5-2
Leopold 1-1,2-1, 2-17, 4-1
log-log graph 2-8
Mackin 1-3
Maddock 2-1,2-17
Made] 1-1,4-27
Manning's resistance coefficient 2-12, 4-30
mass wasting 2-28,5-6
mean basin elevation (E) 2-7
meander length 2-11,4-22
Mercer Creek 3-31,3-73
methods of analysis 1-4,3-29
Miller 1-3
models 2-4
descriptive.. 2-29
examples of 2-5
regional 2-5,2-7,3-1, 3-2,3-39,3-50,
3-54,3-71, 3-72, 4-19, 4-22
Montgomery 3-1,3-31
Moscrip 3-1, 3-31
Mosley 4-1
Multiple Severity Factor 2-22
N. F. Skokomish River 4-35, 4-37
natural diversity 1-6
natural variability 2-3
Newbury 4-3
Newton's second law 2-7
North Creek 3-31,3-33,3-34,3-35,3-36,
3-37,3-39
Northeast Washington 1-4,1-5, 2-26,3-1,
3-2,5-1
Northeastern Washington....3-54,3-55,3-56,
3-57,3-58,3-60
NORTHEASTERN WASHINGTON
REGIONAL STREAMS 3-50
Olympic Peninsula 1-4,1-5,2-14,2-26,
2-29, 2-30, 2-31, 2-33,3-1,3-2,3-10,3-16,3-
21,3-28,3-41, 3-50, 3-60,3-71, 3-72, 3-73,
4-3,4-26,5-1,5-2
OLYMPIC PENINSULA REGION 3-2
open channel flow 1-4
Orsborn 2-3,2-4,2-5,2-7,2-8,2-11,2-12,
2-13, 2-14,2-18,2-20,2-22,2-23,2-24, 2-26,
2-30,2-31,3-1,3-10,3-28,3-31,3-50,3-60,
3-66,4-6,4-18,4-32,4-34,5-3
Osterkamp 1-6
Pearson 3-28,3-31
precipitation
average annual 1-5,2-5,3-72,4-8
PUGET LOWLAND REGION 3-31
Puget Lowlands 3-1,3-2,3-31,3-40,3-41,
3-42,3-43, 3-44,3-45,3-46,3-48,3-49,3-50,
3-60,3-71,3-72,3-73
Purseglove 4-3
Q1F2 2-23,2-27, 2-28,2-30, 2-31,3-1,3-2,
3-11,3-16,3-28,3-29,3-31,3-39,3-41,3-44,
3-50, 3-54, 3-60,3-66,3-71,3-72,3-73,4-13,
5-3,5-7
Q7L2 2-5, 2-23,2-27,2-28,3-1,3-2,3-11,
3-16,3-24,3-28, 3-29,3-30,3-39,3-44,3-50,
3-54, 3-66,3-71, 3-72, 3-73, 4-37,5-3,5-6
QAA....2-5, 2-8,2-23,2-26, 2-27,2-28,3-1, 3-2,
3-11,3-16,3-21,3-29,3-31,3-41,3-44,3-50,
3-54,3-60,3-66,3-71,3-72,3-73,5-3,5-6
Quilceda Creek 3-31,3-73
ratio of average fall 2-11
Reconnaissance 4-1
Reconstruction 4-1,4-3
regional channel geometry analysis 2-18,
4-19
regional geometry equation 3-21
regional indices 1-1, 5-5
7-2
-------
EPA Channel Condition Project
Reid 5-2
relationships
combined 3-16, 3-28, 3-44, 3-66
empirical 3-2,3-28, 3-44,3-50,4-2
response variable 2-18
restoration 1-1, 3-31,4-1,4-2,4-3,4-18,5-2,
5-5
Ritter 1-3
Rosgen 1-3,1-7,2-2,2-11,4-2
Rouse 2-7
S. F. Skokomish River 2-31, 3-28,4-3,4-5,
4-13
Schmidt 2-11,2-23
severity factor analysis 2-20, 2-28
shear-shape relationship 2-12, 2-13
Sheep Creek 3-50, 3-66
Shelton Cooperative Sustained Yield Unit
4-6
Shields 4-3
Siuslaw National Forest 2-8
Smelser 2-11
Soleduck River 3-28
Stewart 2-32
Strahler 1-3, 2-7, 2-23
stream
graded 1-3,5-3
lowland 1-4,1-5
stream models 1-4
stream types 1-1, 2-11
Stypula i, 2-12,2-13, 2-14
Swamp Creek 3-31, 3-73
Task Committee 2-1,2-2, 2-3,5-2,5-5
template 1-1, 2-27,3-1,5-6,5-7
theory
in-regime 1-3
Thomas 2-6
Thorne 4-3
tortuosity ratio 2-12
urbanizing area 2-32
USGS
gage records 2-8
stream gage records 2-27
stream gaging stations 2-17, 2-26,3-1,
3-2,3-10, 3-11, 3-41,4-22,4-35
watershed
land-use 2-28
water-use 2-28
wetted perimeter 2-17
White 5-7
width to depth ratio (W/D) 1-4, 2-5,2-11,
2-12, 2-14,2-22,2-23,2-28,3-2,4-30,4-38,
5-3,5-7
Williams. .2-17,2-18, 3-28,3-29,3-31, 3-66,5-3
Wilmont Creek 2-23
Wilson 3-31
Wolman 5-7
Woods Creek 3-31,3-73
Yang 2-11
7-3
------- |