Hydraulic, Geomorphic, and Habitat Conditions of the Lake

tickbitewryMechanics

Oct 30, 2013 (3 years and 7 months ago)

101 views


Prepared in c
ooperation

with Colorado Water Conservation Board and Colorado River Water
Conservation District


Hydraulic
, Geomorphic, and Habitat Conditions of the Lake
Fork of the Gunnison River in Hinsdale County, Lake City,
Colorado, Water Years 2010
-
2011

By
Cory A. Williams, Keelin R. Schaffrath, Jennifer L. Moore, and Rodney Richards






Report Series
2012

XXXX


ii

U.S. D
epartment of the Interior

U.S. Geological Survey


iii

U.S. Department of the Interior

DIRK KEMPTHORNE, Secretary

U.S. Geological Survey

Mark D. Myers, Director

U.S. Geological Survey, Reston, Virginia 20
12

Revised and reprinted: 20
xx

For product and ordering
information:

World Wide Web: http://www.usgs.gov/pubprod

Telephone: 1
-
888
-
ASK
-
USGS

For more information on the USGS

the Federal source for science about the Earth,

its natural and living resources, natural hazards, and the environment:

World Wide Web:
htt
p://www.usgs.gov

Telephone: 1
-
888
-
ASK
-
USGS

Suggested citation:

Williams, C.A., Schaffrath, K.R., Moore, J.L., and Richards, R.J.,

20
12
,
Hydraulic
,
g
eomorphic, and
h
abitat
c
onditions
of the Lake Fork of the Gunnison River in Hinsdale County, Lake City, Colorado, Water Years 2010
-
2011: U.S.
Geological Survey
Scientific

Investigations Report
2012

XXXX, X

p.

Any use of trade, product, or firm names is for descriptive purposes on
ly and does not imply

endorsement by the U.S. Government.

Although this report is in the public domain, permission must be secured from the individual

copyright owners to reproduce any copyrighted material contained within this report.


iv

Contents

Abstract

................................
................................
................................
................................
............................

1

Introduction

................................
................................
................................
................................
.......................

4

Purpose a
nd Scope

................................
................................
................................
................................
...........

6

Description of Study Area

................................
................................
................................
................................
...

6

Data Collection Methods

................................
................................
................................
................................
....

9

Topographical Surveying
................................
................................
................................
................................
.

9

Hydraulic

Data Collection

................................
................................
................................
...............................
10

Streambed Sediment Characterization

................................
................................
................................
............
12

Benthic Invertebrates and Vegetation Canopy Assessment

................................
................................
...............
12

Two
-
Dimensional
Hydra
ulic

and Habitat Modeling Methods

................................
................................
..................
13

Two
-
Dimensional
Hydraulic

Model Calibration

................................
................................
..............................
15

Sediment Mobility

................................
................................
................................
................................
......
17

Habitat Assessment

................................
................................
................................
................................
...
20

Two
-
Dimensional
Hydraulic

Model Sensitivity Analysis
................................
................................
.........................
23

Hydraulic

and Geomorphic Conditions of the Mode
led Reaches

................................
................................
...........
24

Sediment
-
Mobility Characteristics

................................
................................
................................
...................
25

Geomorphic
Characteristics

................................
................................
................................
...........................
27

Habitat Conditions of the Modeled Reaches

................................
................................
................................
........
28

Micro
scale Fish Habitat Quantification

................................
................................
................................
.............
28

Mesoscale Habitat Characteristics

................................
................................
................................
..................
29

Macro
-
invertebrate Community Assessment

................................
................................
................................
....
30

Summary

................................
................................
................................
................................
.........................
32

References Cited

................................
................................
................................
................................
..............
37


v


Figures

Figure 1.

Map of study area for the Lake Fork of the
Gunnison River at Lake City, Colorado.

...............................

5

Figure 2.

Hydrograph of the Lake Fork of the Gunnison River at U.S. Geological Survey streamflow
-
gaging station
09124500.


................................
................................
................................
................................
......................

8

Figure 3.

Map of study reach with pressure transducers, grain
-
size sample locations and cross
-
section locations
for the Lake Fork of the Gunnison River at Lake

City, Colorado.

................................
................................
...........
10

Figure 4.

Benthic invertebrate sampling sites and vegetation canopy density assessment locations in the Lake
Fork of the Gunnison River near La
ke City, Colorado, August 2011.
................................
................................
......
13

Figure 5.

Map of MD
-
SWMS grids of the two study reaches.

................................
................................
............
14

Figure 6.

Graph showing water
-
surface elevations for the best
-
fit flow simulations for A)150 cubic feet per second;
B) 260 cubic feet per second; and C) 1,600 cubic feet per second streamflow
s.

................................
.....................
16

Figure 7.

Habitat suitability curves for three life stages of a) cutthroat trout and b) brown trout.
............................
22

Figure 8.

Flow frequency curve for the Lake Fork of the Gunnison River at Lake City, Colorado.

.........................
24

Figure 9.

Sediment
-
size

characteristics of the five sampling sites.

................................
................................
....
25

Figure 10.

Shear stress and water
-
surface elevation plots at cross
-
sections 2, 3, and 4.

................................
...
25

Figure 11.

Shear stress and water
-
surface elevation plots at cross
-
sections 5, 6, 7, and 8.

...............................
25

Fig
ure 12.

Reach
-
scale assessment of sediment mobility in the upper reach (A
-
C) and lower reach (D
-
F) for of
the Lake Fork of the Gunnison River at streamflows of 400; 900; and 1,800 ft
3
/s, respectively.

................................
26

Figure 13.

Plotting positions of two surveyed reaches of the Lake Fork of the Gunnison River near Lake City,
Colorado, in comparison to thresholds for meandering and braided channel patterns based on finding from (A
)
Leopold and Wolman (1957) and (B) Richards (1982).

................................
................................
.........................
27

Figure 14.

Weighted usable area for the adult, juvenile, and fry life stages of the cutthroat and brown tro
ut in the
study reach of the Lake Fork of the Gunnison River.

................................
................................
............................
28


vi

Tables

Table 1.

Summary of the lateral eddy viscosity and model diagnostics for the best
-
fit two
-
dimensional
hydraulic

and habitat model simulations.

................................
................................
................................
...........................
17

Table 2.

Mesosc
ale
-
habitat suitability ranges of habitat type for cutthroat and brown trout.

................................
22

Table 3.

Results from the analysis of the sensitivity of the lateral eddy viscosi
ty parameter for three of the
streamflows simulated in the multidimensional flow model for the upper and lower study reaches on the Lake Fork.

.
23

Table 4.

Results
from the analysis of the sensitivity of the roughness parameter (Z
0
) for three of the streamflows
simulated in the multidimensional flow model for the upper and lower study reaches on the Lake Fork.
....................
23

Table 5.

Sediment particle
-
size characteristics for the five sites sampled in the Lake Fork study area.

................
25

Table 6.

Microscale
-
habitat suitability total weighted
-
useable habitat in a typical year for cutthroat and brown trout.



................................
................................
................................
................................
.....................
29

Table 7.

Mesoscale
-
habitat characteristics during low
-
flow conditions for cutthroat and brown trout.

...................
30

Table 8.

Macro invertebrate community metrics sampled in the Lake Fork of the Gunnison River, Colorado
........
32









1

Hydraulic
, Geomorphic, and Habitat Conditions of the
Lake Fork of the Gunnison River in Hinsdale County, Lake
City, Colorado, Water Years 2010
-
2011

By
Cory A. Williams
,
Keelin R. Schaffrath
,
Jennifer L. Moore,
and
Rod
ney Richards

Abstract

Channel rehabilitation, or reconfiguration, to mitigate a variety of riverine problems has become
a common practice in the western United States.
Numerous private entities and resource
-
management
agencies have attempted to modify stre
am channels by using designs based on different geomorphic
philosophies and classification schemes. However, little work has been done to monitor and assess the
channel response to, and the effectiveness of, these modificati
ons over a long period of time.

The Lake Fork of the Gunnison River has been an area of active channel modification since the
1950s to accommodate the needs of the Lake City community. The Lake Fork Valley Conservancy
District has begun a planning process to assess restoration option for

a reach of the Lake Fork in Lake
City to enhance the
hydraulic

conditions
and ecologic health of the reach.
The U.S. Geological Survey
in cooperation with the Colorado Water Conservation Board and Colorado River Water Conservation
District began

a study in 2010 to quantify existing hydraulic and habitat conditions for a reach of the
Lake Fork of the Gunnison River in Lake City, Colorado.



2

The purpose of this report is to quantify existing Lake Fork conditions and establish a baseline
against whic
h post
-
rehabilitation conditions can be compared.
Th
is

report

uses two
-
dimensional
hydraulic and habitat modeling, and data collection to characterize (1) the existing
hydraulic

and
geomorphic conditions in a 1.1 kilometer section of the Lake For
k that has been proposed as a location
for future channel
-
rehabilitation efforts; (2) the habitat suitability of the reach for two aquatic fishes
based on physical conditions within the stream; and (3) the current riparian canopy cover and benthic
-
inverteb
rate structure of the reach

(
using two metrics:
vegetation canopy cover

and
be
nthic invertebrate
characteristics
)
.

This characterization provides resource managers with information to evaluate the
existing channel conditions as well as a baseline for post
-
restoration comparisons to evaluate restoration
project success.

The
current
morphology

of the channel is affected by land
-
use changes within the basin and
geologic controls within the reach

and the current flow regime
. The historic channel
was defined as a
dynamic, braided channel (multiple, bifurcating channels) with an active floodplain
Comparisons
of the
current channel morphology
to
channel
-
form thresholds
indicate that the current channel characteristics
may favor a braided channel
form.

The FaSTMEC
H computational
flow
-
model within MD_
SWMS was
selected
to
characterize the
the
channel
hydraulics
over a range of flow conditions
and
calculate the resulting
habitat
-
suitability
conditions
for a
study reach o
f

the Lake Fork

of the Gunnison
River
.

Aquatic habitat
was evaluated for
two species of trout, the cutthroat trout (
Oncorhynchus clarkii
) and brown trout (
Salmo trutta

morpha

fario
).
Microscale
habitat

based on habitat preference curves
,
presented
as weighted usable area,
generally increased as streamflow increased for both species and all

life stages. Habitat occurs along the
banks for flows of 900 ft
3
/s and less. Out
-
of
-
bank areas become more substantial contributors to overall
habitat availability for flows of 1,300 ft
3
/s or more. In general, the upper reach provided 2
-
3 times more

3

avail
able habitat than the lower reach.

Mesoscale fish
-
habitat assessment of the Lake Fork was done
based on the conditions present in the 150 ft
3
/s flow simulation as well as field observation. The
presence of pool habitat was limited within the reach and occu
rred along the channel margins. For both
reaches, the pool habitat was less than 5 percent of the total
-
wetted area, a percentage that is
substantially lower than the recommendations for sustainable populations. An exception to this is an
isolated scour po
ol located outside of the modeled area (between the two reaches) below a waterfall
which was not quantified in this analysis.
Areas of cover were adjacent to potential feeding areas in the
lower reach, and often occurred within the same pool habitat. This
may favor
the ratio of energy uptake
vs energy consuption
, wherein little energy is
expended
to acquire adequate food sources
or energy
consumption
.

Macro
-
invertebrate communities at
the Lake Fork stud
y area
were different in comparison to the
control sites upstream of the reach. The
biotic condition index
may indicate that increased sediment
deposition and the resulting highly embedded substrate

in the study reach sites are affecting macro
-
invertebrate habitat. The Ephemeroptera (mayfly), Plecoptera (stonefly), and T
richoptera

(caddisfly),
(EPT) score for the study sites indicate that the study reach sites may be more disturbed or stressed than
t
he control site on Henson Creek. The control site on Lake Fork had a low EPT score and low number
of intolerant taxa in comparison to the other sites. It is unknown if the macro
-
invertebrate community
upstream of the study reach

is affected by stressors fr
om upstream or adjacent land uses.

Sediment mobility is an important process for flushing of
these fine sediments
from within the
gravel frameworks. Evaluations of channel and flow characteristics at
cross
-
section locations 2
-
8
show a
range of streambed mobility. In general, boundary shear stress and streambed mobility increase with
increases in streamflow. At cross sections 2
-
8, sediment is potentially mobile at streamflows of 900
ft
3
/s. The streambed is substantially mobile at
cross sections 4
-
8 at streamflows of 1,300 to 1,600 ft
3
/s

4

with all cross sections being substantially mobile at 1,800 ft
3
/s. Within the cross sections, the greatest
boundary shear stresses occurred towards the center of the channel.
Reach
-
scale assessment
of sediment
mobility

in the lower reach shows increased streambed mobility. This is due in part to smaller grain
-
sizes in the lower reach, but may also reflect the greater extent of channel alterations, specifically the
constructed berms, present in thi
s reach. Generally, sediment is more mobile in the lower reach.

Introduction

The Lake Fork of the Gunnison River
(hereafter, Lake Fork)
has been an area of active channel
modification since the 1950s to
accommodate

the needs of the Lake City community (
Lake
Fork Valley
Conservancy, 2010). Historically, the Lake Fork was characterized by its
braided channel
which
frequently overtopped its banks
creating an

active and
dynamic
flood
plain
(Lake
Fork Valley
Conservancy, 2010). However,
b
eginning in the 1950s, increased rates of land use change
within Lake
City have resulted in the
straighten
ing

and channeliz
ation

of the Lake Fork
to
promote channel stability
and development of
riverfront
properties

(Lake Fork Valley Conservancy, 2010)
.

Reduced channel width, sinuosity, and riparian habitat have been noted in
the Lake Fork as well
as

Henson Creek
, a smaller tributary to the
north
Lake Fork Valley Conservancy, 2010)

(fig.
1
)
.
Failures
of
s
ediment
-
retention structure
s

(cribbing

piles)
in
Henson Creek
during the

1960s and 1970s
contributed large volumes of sediment into the

Lake Fork
.
During
the

late 1980s and early 1990s,
the
Colorado Department of Transportation
(CDOT)
built
temporary berm structures

along the

river on the
north side of town
to protect
the
highway
from channel migration until
permanent

structures could be
completed (Lake Fork Valley Conservancy, 2010)
. The

temporary

berms remain in place

today
, and
may be a
potential source of
channel
instability
and sediment
inputs

since they were

not designed to
withstand high
-
flow

cond
i
tions
.
Additional areas along the north side of town

w
ere

also
straightened by
CDOT

and U.
S. Army Corps of Engineers in the
1980s
(Lake Fork Valley Conservancy, 2010)
.


5

Figure 1.

Map of study area for the Lake Fork of the Gunnison River at Lake City, Colorado
.

The Lake Fork Valley Conservancy District
is currently in the

process
of
assess
ing

restoration
option
s

for a reach of the Lake Fork in Lake City to
increase
the
hydraulic

and ecologic
function
of the
reach (Lake Fork Valley Conservancy, 2010).
The

proposed
channel restoration
options
consider

the
outcome
of similar efforts in other
reconfigured

reaches
of
the
Lake Fork.
Channel reconfiguration
efforts
to
improve aquatic habitat and channel stability

on the Lake Fork
have been completed
by
the
Colorado Division of Wildlife (CDOW) and the Bureau of Land Management (BLM)

in the
last few
decades
.
For example, i
n 1997
,
,

a two mile reach of the Lake Fork at
Gateview

(
22 miles downstream
of Lake City
)

was reconfigured

to improve habitat for Brown Trout and Rainbow Trout (Lake Fork
Valley Conservancy, 2010).
The focus of

this channel restoration was to

improve riparian habitat,
vegetation, and
stream bank

stability and to reduce sediment sources that were
beli
e
ved to be
impairing
the fisheries (Lake Fork Valley Conservancy, 2010)

by
reshaping of the channel
,

increasing

the channel
stability,
p
lace
ing

large boulders to improve fish
habitat
.
In a
ddition
,

riparian vegetation
was planted
where the channel widths were decreased in the reconfigured channels
to increase overall productivity
in the reach
,

increase stability
,

and to provide cover habitat for fish
. Following reconfigured channel
efforts, the total fish biomass (weight of fish across all age classes per acre o
f stream,
lbs
/acre) increased
from approximately 40 to 100
lbs
/acre
, an indication

that the reconfiguration efforts
improved

fish
habitat
(Lake Fork Valley Conservancy, 2010).

Channel
restoration
has become a
common
practice

throughout the United States.
.

The goal of
channel
restoration efforts

is
to
create a
more natural
channel
while

balanc
ing

a variety of
often
competing
factors including:

the channels
historical conditions, improved
connection between the
channel and floodplain while maintaining flood control
,
channel stability
,
and
maintaining a mobile bed.
The
se

physical factors all lead to

enhancement
ment

of
the
aquatic and riparian habitat

function

and

6

water quality,
and
also lead to increas
ed
recreation

opportunities
. Numerous private entities and
resource
-
management agencies have attempted to
restore

stream channels
using designs based on
different geomorphic philosophies and classification schemes. However,
little work has been
done to
monitor and assess the channel response to, and the effectiveness of, these modificati
ons over a long
period of time.

The
success of a project consider
s

the stated goals
,

such as those described above
,

which

include
both stability and habitat
con
siderations
.

The
analysis of quantifiable metrics,
gives

resource managers,
planners, and designers
information to

determine the
effectiveness and durability of channel restoration
techniques
. The U.S. Geological Survey in cooperation with the Colorado Water Conservation Board
and Colorado River Water Conservation District began a

study in 2010 to quantify existing hydraulic
and habitat conditions for a reach of the Lake Fork of the Gunnison River in Lake City, Colorado.

Purpose and Scope

The purpose of this report is to quantify
the
existing
hydraulic and habitat
conditi
ons
in a
1.1
kilometer
reach of the Lake Fork proposed for future channel restoration
to

establish a baseline against
which post
-
res
t
o
r
ation

conditions can be compared.

This report (1) quantifies the existing
hydraulic

and
geomorphic conditions
; (2) characterizes the habitat suitability of the reach for two
species of

fish

based
on physical conditions within the stream
; and (3)

characterizes the current riparian
and aquatic
habitat
conditions (
using two metrics:
vegetation canopy cover

and
be
nthic invertebrate characteristics
)
.


Description of Study Area

The study reach

is located in the town of Lake City, Colorado
,

along the Lake Fork of the
Gunnison River.
Lake City is
a small
mountain

town
with a population of 375 year round residents
(U.S. Census Bureau
, 2011
)
.
The Lake Fork of t
he Gunnison River
originates in the

no
rthern portion of

7

the San Juan Mountain Range and flow
s

in a northern direction

(fig.
1
).

The drainage area of the Lake
Fork of the Gunnison ri
ver

study reach

is 210 squar
e miles of the 442 square miles total basin area

located at
8,671 feet
above sea level

(
U.S. Geological Survey, 2010b
). Most of the land in the Lake Fork
basin is Federally owned and
managed

by the Bureau of Land Management

and

U.S. Forest Service.
The remainder of land is privately owned except for Lake San Cristobal which is owned and m
anaged
by the local government.

Land cover in the basin varies with elevation (
Homer and others, 2004
). The highest elevations
include barr
en and grassland classifications. Treeline in Colorado occurs at an elevation of
approximately 11,500 feet (
Pielke

and others, 2003) with both deciduous and coniferous forests found
throughout the basin along stream margins (
Homer and others, 2004
).

The cl
imate of the area is typical of the mountainous areas of western Colorado. Precipitation
tends to increase with elevation while temperatures decrease with elevation (
Pielke

and others, 2003).
The majority of precipitation falls as snow and is stored as sno
wpack through the winter months. The
Western Regional Climate Center in cooperation with the National Climatic Data Center and National
Weather Service operated a weather station at Lake City from August 1, 1948 through December 21,
2008 (Western Regional
Climate Center, 2011).
The
average
annual precipitation in Lake City for this
time period was 14.1 inches of water
,
and the

average annual snowfall
was

82.8 inches. Average annual
maximum temperature in Lake City is 55
.7 degrees Fahrenheit while the average annual minimum is
22.4 degrees Fahrenheit (Western Regional Climate Center, 2011).

The hydrology of the Lake Fork is governed by snowpack accumulations and spring
temperatures which drive snowmelt
-
runoff. The
annual
streamflow pattern
is characterized by
streamflow
increases
begin
ning

in April
,
peak
streamflow conditions
May
-
June, and
a
decreases
in
streamflow occurring
in July and August

(U.S. Geological Survey, 2010
a
)
.

The

Lake Fork of the

8

Gunnison River

is monitored at U.S. Geological Survey stream
flow
-
gaging station

09124500, Lake
Fork at Gateview
, Colorado (hereafter, Gateview)
.
The drainage area
of Gateview

is 334 square miles
and it is located approximately 22 river miles downstream of the study reac
h.
No major tributaries

or
diversions

occur

in the

Lake Fork between the study reach and the streamflow gage.
A summary of the
daily
-
mean
streamflow conditions
is

shown
in
figure
2

including:
maximum, 75
th

percentile, mean, 2
5
th
percentile, and minimum.
The highest r
ecorded
stream
flow was 2,720 cubic feet per second

(ft
3
/s
)
, July
10, 1983
,

with

the lowest daily mean
stream
f
low
of

21 ft
3
/s
, recorded January 3, 2002 (U.S. Geological
Survey, 2010).
Additional w
ater
-
quality and streamflow data are available from the National Water
Information System (NWIS) website at
http://waterdata.usgs.gov/co/nwis
.

Figure 2.

Hydrograph of the Lake Fork
of the Gunniso
n River
at

U.S. Geological Survey streamflow
-
gaging
station 09124500
.

The dominating geology in and
around Lake City is Tertia
ry volcanic deposits (Tweto,
1979
).
As a result of intense volcanic activity in the region, mineral deposits are common.
The exten
sive
volcanic activity, along with intrusions and mineralization created widespread distributions of ore
bodies in the San Juan Mountains that were deposited during the late
-
Tertiary period (Tweto and Sims,
196
3
; Burbank and others, 1947). Discovery of gol
d and silver deposits near Lake City were the driving
force in historical mining activities in the region.
Relic mining cribbing piles have introduced sediment
into Henson Creek which transported it into the Lake Fork in the study area

(Lake Fork Valley
Co
nservancy, 2010). This sediment

may have an eff
ect on channel form and habitat.

In addition to
anthropogenic sediment into the Lake Fork basin, the region is susceptible to
large
-
scale
landsliding

(Tweto, 1979)
.
Most notably,
the Slumgullion Lands
lide
which occurred
ro
ughly 700 years ago

was
large enough to dam the Lake Fork. This landslide deposit
result
ed

in the creation
and persistence
of
Lake San Cristobal
, the second largest natural lake in Colorado (Guzzi and Parise, 1992).


9

Data Collection Me
thods

Combinations of topographic,
hydraulic
,
and
ecologic

data were

collected to characterize
the
existing hydrologic
,
geomorphic
, and
aquatic and riparian
habitat

conditions in a
1.1 kilometer

section
of the Lake Fork at Lake City
.

Topographical Surveying

Topographical surveying was used to map water
-
surface
and stream channel
elevation
s

the study
reach
of the
Lake F
ork at Lake City

(
fig
s
.
1

and

3
).
The majority of the surveying occurred between
August 2
-
6, 2010, wit
h additional flood
-
plain surveying on May 24, 2011.
Topographical surveying of
water
-
surface elevations
includes conditions
present
during each day of surveying
as well as
preserved
high
-
flow
conditions
(high
-
water marks)
from Spring 2010 and 2011
that wer
e identified
though
interpretation of sediment deposit
s and debris lines
.
Eight monumented cross
-
sections were established
in

the study reach for
long
-
term

channel monitoring (fig.
3
).

Surveying was completed using
survey
-
grade Global Positioning System (GPS) equipment,
operated in a Real
-
Time Kinematic mode (RTK
-
GPS) to determine the elevation and coordinate
locations of

each
surveyed

point

(Trimble Navigation Limited, 1998)
. The RTK
-
GPS utilized a
stationary GPS receiver (base station) positioned over a reference benchmark, and roving GPS receivers
(rovers) operated by separate hand
-
held computers (data collectors). The
RTK
-
GPS system calculates
the locations of the rovers in real time, using the positions of orbiting
NAVigation Signal Timing And
Ranging (
NAVSTAR
)

GPS satellites and
a

known
base
-
station
location.

All surveyed elevations (RTK
-
GPS) were output to a spreadsh
eet program in metric Northing,
Easting, and Elevation format using a Universal Transverse Mercator coordinate system (UTM, Zone 12
North). Horizontal coordinate information was referenced to the North American Horizontal Datum of
1983 (NAD83). Vertical co
ordinate information was referenced to the North American Vertical Datum

10

of 1988 (NAVD 88)

and Geoid 2003 (US Conus)
. Elevation values were visually examined for relative
accuracy graphi
cally

and

th
r
o
ugh
inspecti
o
n
of GP
S positional precision

and diagnosti
c parameters
.

Figure 3.

Map of study reach with pressure transducers, grain
-
size sample locations and cross
-
section locations
for the Lake Fork
of the Gunnison River
at Lake City, Colorado
.

Hydr
aulic

Data Collection

Streamflow w
as

measured using a FlowTracker® Handheld Acoustic Doppler Velocimeter
(ADV) for wadable stream conditions and followed
standard
techniques o
utlined in Turnipseed and
Sauer

(2010).

For flows that were not wadable, an
Acoustic

Doppler Current Profiler (ADCP)
was used
to measure streamflow from the 5
th

S
treet pedestrian bridge

using
a
tethered boat
, followin
g standard
techniques

outlined in Mueller and Wagner

(2009).
Streamflow

measure
ments were made

at 40
;

150
;

270
;

760
;

and 1,400 ft
3
/s
between August 5, 2010
and
August 1, 2011
.

Onset Hobo water level loggers
(model
U20
-
001
-
01
,

unvented,
pressure transducer
)
were used
in conjunction

with

local
vertical
elevation reference markers
(
staff
-
pl
ate
s
)

to monitor water
-
surface
elevation at selected
locations

within the study reach
.

P
ressure transducer
s

w
ere

deployed
at the
upstream
study
-
reach boundary,
at
approximately halfway through the study reach
,

and
at the
downstream study
-
reach boundary

(fig.
3
)
.

The pressure transducers w
ere placed in

protective poly
vinyl
chloride (PVC) casing
s

and attached to
T
-
post
s

which were
secured

to the river bed
.
A forth pressure
transducer

was located outside of the

stream
at cross
-
section 8
to record barometric
-
pressures changes
for post
-
processing

corrections of the unvented, pressure
-
transducer readings
(fig.
3
)
.

This resulted in a
continuous record (30 minute intervals) of water
-
surface elevations at these locations from August 2 to
October

28, 2010; and from March 24 to August 1, 2011.

A stage
-
discharge rating curve based on the relation between the measured streamflow and
water
-
surface elevation pairs was determined for the upstream study reach boundary at cross
-
section 1

11

through least
-
squares regression techniques (fig.
3
)(Rantz,

1982
).
At the remaining pressure
-
transducer

locations water
-
surface elevations
corresponding to
50
;

150
;

260
;

400
;

900
;

1,300;

1
,
6
00
;

and 1,8
00

ft
3
/s

were determined
based upon the date and time of the estimated streamflow at the upstream boundary
(
It
was assumed that the streamflow at the downstream boundary was equal to the streamflow at the
upstream boundary
based on
field observations
of tributarie
s
and diversions within the reach
)
.

High
-
water marks from the water year 2010 peak flow were

identified
and
survey
ed

within the
study reach
;

h
owever, no streamflow measurement
s

within the study reach
relate
d to these

water
-
surface
elevation
s

were
available
.
The
corresponding streamflow for the high
-
water marks was estimated from
the
Gateview

data set
using the

Maintenance

of Variance 1
method
(Hirsh, 1982)
(
hereafter,
MOVE.1)
.

The MOVE.1 equation uses
data from
an adjacent
site

(downstream

in this instance
)

to

predict
streamflow for the

site of interest
.


To determine how often any specific flow typically occurs each year, flow
-
duration curves were
used to determine flow frequencies as percent of year.
The flow frequency for
cross
-
section 1

was
ca
lculated
using 30 arithmetic bins following

the regionalized
-
duration

curve method

for estimating
flow
-
duration
s

for ungaged sites

(U.S. Department of Agriculture, 2009).
The regionalized
-
duration

curve
method
relates flow between two sites (a

gaged site

a
nd an

ungaged site of interest
)

by
transferring a
non
-
dimensional
flow
-
duration
curve. This
technique
preserves the shape of the

flow
-
duration

curve,
and

allows the magnitude of the streamflows to be
scaled

appropriately
between sites

based on standardized

peak
-
flow values
between

sites
.
Peak
-
flow magnitudes
(2
-
, 5
-
, 10
-
year floods)
were estimated based on regional regression
analysis

(Capesius and Stephens, 2009; U.S. Geological
Survey, 2010b).
T
he 2
-
year recurrence

interval
peak
streamflow
was selected for
use in this
technique

12

for consistenc
y

in methodology between the two sites
(Capesius and Stephens, 2009; U.S. Geological
Survey, 2010
b
).

Streambed Sediment Characterization

Sediment

particles were measured at 5 locations in the study reach

to characterize
the grain size

of
different geomorphic surfaces
, as well as
to determine channel roughness

(fig.
3
)
.
Bars and channel
margins
(outside of areas strongly affected by eddies) were chosen
because t
hese areas are inundated
most of the year and represent the coarser material within the river channel

that is transported
.

Grain
-
size
was

determined in the field by using areal techniqu
es (pebble
-
count
s
) described by
Wolman (1954). A minimum of 100 clasts were measured during each pebble count.
The s
ampling
pattern
was
oriented
parallel to the direction of streamflow at
intervals of
one or two paces

along the
streambed or alluv
ial bars
at each sampling site.

The intermediate, or "b
-
axis
,
" of
each

sediment particle was measured to the nearest millimeter
for gravel and small cobbles, and to the nearest
5
millimeters for large cobbles and small boulders.

The
b
-
axis length was recorded in th
e field n
otes and
grain
-
size statistics (
D
50
,

size

at the 50th percentile
;
and
D
84

size

at the 84
th

percentile)

were computed in a spreadsheet from the cumulative
-
frequency
distribution function of sampled sediment particles
.

Benthic Invertebrates and V
egetation Canopy Assessment

Macro
-
invertebrate samples were collected

during a low
-
flow period August 17
-
18, 2011,
following National Water
-
Quality Assessment (NAWQA) Program protocols (
Mou
l
ton and others,
2002
;
Fitzpatrick and others, 1998
). In general, f
ive discrete collections were made from riffle habitats
at each
of the 4
site
s

with
in the reach with

a
s
lack sampler equipped with 500
-
micron mesh and a 0.25
m
2

sampling grid along the reach

(Fitzpatrick and others, 1998) (fig.
4
)
.
Two sites upstream of the reach

13

were also sampled

(HC
-
1 and LF
-
5)
, and are used as control sites outside of the reach (fig.
3
).
The
contents of the five
collections

were composited, elutriated, and poured through a 500
-
micron mesh
sieve, and preserved in the field

with 10
-
percent buffered formalin
following standard techniques
(Fitzpatrick and others, 1998)
.
Samples were transferred to a laboratory (Aquatic Associates, Inc.)
where

taxa
were

identified to the lowest practical resolution (gen
us or species) and enumerated (Klemm
and others, 1990).
Macr
o invertebrate samples
were

processed in accordance with
the State of
Colorado
protocols

using the 300
-
count method with 1
0 percent

of the samples analyzed for
quality assurance and
quality control
purposes.

Aquatic Associates also provided a rapid bio ass
essment using methods
described by
Barbour and others

(1999).

Figure 4.

Benthic invertebrate sampling sites and vegetation canopy density assessment locations in the Lake
Fork of the Gunnison River near Lake City, Colorado, August 2011
.

Vegetative cover was measu
red at each of the benthic sample locations at the time of sampling
using a spherical densiometer (Model A) (Lemmon, 1956).
Differences in canopy density can affect
water temperatures, terrestrial insect and leafy organic inputs into the aquatic system, an
d habitat cover.

Four measurements were taken at each location from the center of the stream with the operator facing:
upstream, downstream, left bank, and right bank. The percentage of the densiometer view
-
field that was
obstructed was recorded for each measurement. This
provides a standardized methodology to
characterize the canopy cover within the stream reach. Observations are presented as rose diagrams
within the study reach (fig.
4
).

Two
-
Dimensional
Hydrologic and Habitat
Model
ing Methods

The M
ulti
dimensional Surface Water Modeling System (MD_
SWMS) is a graphical user
interface that was developed by the USGS
to aid in
pre
-

and post
-
processi
ng
two
-
dimensional

14

streamflow analysis
(McDonald and others, 2001;
2006; in press
). The FaSTMEC
H computational
flow
-
model within MD_
SWMS was
selected
to
characterize the effects of streamflow on
hydraulic
and
habitat
-
suitability condition
s

for a study reach o
f

the Lake Fork
(N
elson and McDonald, 1997;
Thompson and others, 1998; Lisle and others, 2000; Nelson and others, 2003; McDonald and others
,
2006; in press
).

The FaSTMECH model uses a curvilinear orthogonal grid
built upon a
user
-
defined centerline
and width,
that
fits

the
path of
the modeled reach.
FaSTMECh is a vertically averages two
-
dimensional
model that simulates

spatially distributed
values of depth, ve
locity and total bed shear stress
User
-
defined parameters are calculated for each node of the grid.
The resolution of the grid was defined such
that
no fewe
r than 10 locations al
ong

a
ny

cross
-
section within the wetted
-
channel area

at the lowest
modeled flow

(or a spatial resolution of
1.0 X 1.0

meters)
.
Additional information on the interface and
the model can be found in McDonald and others (2001; 2006; in
press)
, and Nelson and others (
1997;
2003)

respectively
.

Channel geometry
was obtained from the topographical survey of the channel
.
Aerial images
were used in conjunction with the survey data t
o improve the i
nterpolation between measured data
points

a
nd
generate a suitable
data
set

to
best define the channel topography
.
The final elevation dataset
was
a densely spaced set of topographic data from which a final data set for channel geometry
was
created using
the
Triangul
ated

Irregular
Network

(
TIN
)

method.
The grid in the upper reach was
approximately
455
m long and
60

m wide (
fig.
5
).

The grid in the lower reach was approximately 41
1
m
long and
10
0

m wide (fig.
5
).

Figure 5.

Map of MD
-
SWMS grids of the two study reaches.


15

Two
-
Dimensional Hydrologic
Model Calibration

Model calibration was achieved through i
terative adjustments
of

the
roughness
until the
root
mean squared error between

the measured and predicted water
-
surface elevations was minimized.

Multiple techniques are available to the user for selection of the roughness parameter within the model
interface. Z
0

was selected to calculate roughness where

Z
0

is defined as the distance above the
streambed at which the velocity is zero (Julien, 2010).


The
Z
0

parameter

is calculated

as the
quotient
of the grain roughness height and 30 (Julien,
2010).

The

grain roughness height

in gravel bed rivers

has been
characterized as

function of grainsize in
a variety of ways
including
:

3.
5
D
84

and 6.8D
50
(Bray. 1982 )

and 1.25D
35

and3.5D
90

(Millar and Quick,
1994)
.
The D
84

was calculated from the five grain size samples (three samples in the upper reach and
two samples in the lower reach) which are discussed in the Streambed Sediment Characterization
Metho
ds section
. The average Z
0

calculated from the D
84

of the five grain size samples was 0.013
meters and the same average resulted when the two reaches were considered separately.

FaSTMECH uses
a constant
Z
0

to calculate a variable drag coefficient as a function of depth for
the study reach (
Rich McDonald, U.S. Geological Survey, pers. commun., September 2011
),

where the
drag coefficient is a dimensionless quantity that is used to quantify the drag or resista
nce of an object in
a fluid. Z
0

was chosen
to determine channel roughness
because it is based on
the
grain size of the
streambed

(a persistent condition within gravel streams for moderate time steps). It is also advantageous
because it is less subjective t
han other roughness estimates (Manning’s n or Darcy
-
Weisbach friction
factor) that are made based on visual comparisons to reference values which integrate roughness sources
along the entire cross section or reach (Barnes, 1977). Another benefit to using Z
0

over other roughness
estimates is that it doesn’t change with stream depth, requiring a single characterization for the entire
range of flow conditions (Julien, 2010).

However, the characteristic roughness height for a channel is a

16

function of scales tha
t range from as large as the channel
-
form to as small as the characteristic grainsize
of the channel and the streambed is discretized onto a grid that may be courser than some of the
roughness scales of the channel, the model is therefore calibrated by var
ying the roughness height as
described above.

Five
of the
model
ed streamflow
s had one
measured
water
-
surface elevation
available
located
at
the
upstream

boundary of each reach
.

For
streamflow conditions of
150
;
260
;

and 1,600 ft
3
/s

multiple
measured
water
-
surface elevations distributed throughout the reach
allowed for more rigorous
calibration of the
se

flow simulations

(fig.
6
)
.

There were 183 surveyed water
-
surface elevation
s (data
points)

for the 150 ft
3
/s
simulation
; 110 data points for the 260 ft
3
/s
simulatio
n
; and 40 data points for
the 1
,
600 ft
3
/s
simulation

(the
data points

used for the 1,600 ft
3
/s
simulation

were 2010 high
-
water
marks)
.

For all of the streamflow s
imulations

except 1,600 ft
3
/s
(
in the lower reach
)
,
the root mean
squared error of the wate
r
-
surface elevation was 0.09 meters or less. The 1,600 ft
3
/s
flow simulation

in
the upper reach had a root mean squared error of 0.24
meters and the predictions of water
-
surface
elevation w
ere

consistently higher than the
data points
. The largest differences between
the
predict
ions

and
data points

for this
simulation

occurred in the middle of the study reach (maximum was 0.58 m)
while the up
-

and downstream boundaries had smaller differences
(
0.11 m or less
)
(fig
.
6
C).
This may
indicate
additional
streamflow

within that section of the reach

that occurs outside of the predicted flow
area or
over estimation of

channel roughness

based

on estimates of

Z
0
.

Figure 6.

Graph showing water
-
surface elevations for the best
-
fit
flow simulations

for
A)150 cubic feet per
second; B) 260 cubic feet per second; and C) 1,600 cubic feet per second
streamflow
s.

The final root mean squared error
in
the cr
oss
-
sectional mass conservation
provides an
indication of
how well
the model converges
.


For
most

flow simulations
, the root mean squared error

17

was less than
1.0

percent (table
0
).
T
he 50 ft
3
/s
simulation

had the highest
root mean squared
error

in
both study reaches
.

Summary of the lateral eddy viscosity and
model
diagnostics

for the best
-
fit
two
-
dimensional hydrologic and habitat model
simulat
ions
.

Sediment Mobility

Fluvial transport of
sediment particles is an important component
of the assessment of channel stabi
lity
.
In general, sediment transport occurs w
ithin a stream when the
forces acting to move
a particle exceed

the
res
is
t
ing

forces

of the particle
.
In natural river systems, the
fraction

of the

total

boundary shear stress
available to move sediment
particles

(grains)
can be referred to as the grain shear stress.
In natural
river
channels
the
total
flow
resistance
or energy loss

is the sum of
energy losses

at increasing scales from
individual grains to the channel
form
.
E
nergy losses
reduce the shear stress available to move sediment
particles
.
Substantial

losses
of energy i
n
co
arse
-
bed streams
(gravel
-
sized particles and larger)
can be
the result
bedforms
, vegetation, and changes in channel geometry (
channel expansions and contractions
,
al
luvial bars, and meander bends). The equations within FaSTM
ECH in conjunction with the
computational grid density used for this analysis inherently account for these energy losses except those
induced by vegetation. Therefore within this analysis, the
total
boundary shear stress estimates within
the wetted channel (free of vegetation) are equal to the grain shear stress of those areas.
FaSTMECH
calculate
s

the total
b
oundary

shear stress based on the following equa
tion (
Barton and others, 2005
):




(2)

where



is the boundary shear stress, in Newtons per square meter;


is the fluid density, in kilograms per cubic meter;


is the non
-
dimensional drag c
oefficient;


18


is the vertically averaged stream
-
wise velocity, in meters per second; and


is the vertically averaged cross
-
stream velocity, in meters per second.

In order to determine the precise conditio
ns that will result in the initiation of motion for more
than one particle of interest, reasonable generalizations of the particle shape, orientation, submerged
weight, and protrusion into flow must be assessed
.

The critical shear stress (
) for a particle is the
minimum shear stress needed for general movement of the particle to begin. The critical shear stress
should be considered the minimum value for motion of the
streambed because

only a small fraction of
the sedime
nt
s

will be in motion

over short time periods
(Milhous, 1982).
Substantial

mobilization of the
sediments

has been shown to occur at roughly twice the value of the
critical shear stress of
particles of
that size (Wilcock and McArdell, 1993).

Within this rep
ort, a comparison of the
simulated total
boundary shear stress to

the critical shear
stress
of
t
he median

grain size

(D
50
)

is used to determine when sediments
within the channel
are mobile
.
This accounts for the continuum of possible interactions of partic
les including ‘hiding’ (conditions when
mixtures of coarse and fine particles are positioned such that the finer material is sheltered by larger
particles, thereby decreasing sediment mobility of the finer particles) or increased mobility (when larger
part
icles are surrounded in a matrix of finer material, thereby increasing sediment mobility of the la
rger
particles) (Julien, 2010).

Multiple

critical shear stress
values were selected for the
sediment mobility

analyses in this
report

to cover the range of conditions within the reach
.

B
ased on
the
D
50
, t
he dimensionless critical
shear stress was calculated
to be 0.047
using equations

3
and 4

from

Julien (2010):



(
3
)

Where,


19


is the
sedime
nt
particle angle of repose (
assumed to be
38 degrees); and




(
4
)

Where,


is the dimensionless
sediment
particle diameter;


is the median
grain size

(D
50
)
, in meters;


is the specific
gravity

of sediment (2
.65, dimensionless
);


is
gravitational acceleration

(
9.81 meters per square seconds
); and


is the kinematic viscosity
at

20 degrees
C
elsius

(1.0 x 10
-
6

square meters per

second).

A

dimensionless critical shear value of

0.060

was selected for areas w
here the
stream
bed is tightly
structured or where the particles are strongly imbricated

(Parker and others, 1982; Andrews, 1983;
Komar, 1987; Powell and Ashworth, 1995).
These two values
cover

a range of potential transport
conditions within the stream based on field observations of the streambed conditions.

These values will
be used to present sediment mobility at the surveyed cross
-
sections.

The
dimensional
critical shea
r stress need
ed

to initiate motion of the streambed can be calculated
relative to
the D
50

using equation
6
. In this report comparisons are made to the D
50

within each reach.
As previously mentioned, this method represents the general condition of sedime
nt motion within the
channel by accounting for occurrence of particle hiding and increased mobility of differently
-
sized
particles.







(
5
)

W
here



is the critical shear stress, in Newtons per square meter (Pascal);


20



is the dimensionless critical shear stress or Shields parameter;



is the specific weight of sediment (25,996.5 Newtons per cubic met
er);



is the specific weight of water (9,810 Newtons per cubic meter); and


is the
particle size

on interest (
D
50
)
, in meters.


Reach scale
assessment of s
ediment mobility

was done through an assessm
ent of transport
strength

u
sing FaSTMECH
.
This method
will be most sensitive to the potential for sediment mobility
because it is using the lowest threshold for motion

(dimensionless critical shear stress value of 0.03)
.
This is a reasonable threshold that

accounts for conditions that may be present that were not observed in
the cross section locations.
Sediment mobility is
report
ed

us
ing

equation
6

to calculate the transport
strength

(T), where T=0 is the threshold of motion and T=1 is the threshold for substantial motion

for
the D
50
:







(
6
)

Where,


is the transport strength, dimensionless,


is the b
oundary

s
hear stress or grain shear stress, in Pascals; and


is the critical shear stress
, in Pascals.

Habitat Assessment

Aquatic habitat was evaluated for two species of trout, the cutthroat trout (
Oncorhynchus clarkii
)
and brown trout (
Salmo

trutta

morpha

fario
)
. These species

were chosen because they represent the two
extremes of trout found within these types of hydrologic systems and also because they represe
nt a
comparison between native (cutthroat trout) and introduced (brown trout) salmonids in Colorado.
In

21

stream systems where multiple species of trout are present, a longitudinal stratification o
f these species
is observed.
C
utthroat trout are found

i
n

the

headwater areas

with
cooler water temperatures, lower
sediment concentrations, and less biological production
. B
rown trout
are found
at the lowest reaches,
where

warmer water temperatures, higher sediment concentration,
and
higher biological production

ar
e
typical
.

Microscale and mesoscale
habitat was assessed using the combination of field observations,
measurements, and hydrologic simulations within the study reach of the Lake Fork.
This provides
resource managers with tools to assess the effects of hydrologic conditions (wet, average, and dry years)
and flow alterations on habitat availability for specific streamflows or over annual timescales.
To assess
how habitat availability chan
ges with
stream
flow, microscale
-
habitat quantification was done for each
modeled
stream
flow and
also reported as a time
-
weighted annual total

(
based on typical flow
frequencies
)
.

H
abitat builder in MD_SWMS was used to quantify
microscale
habitat availabili
ty following
methodologies similar to the Habitat Suitability Index method (
Hickman and Raleigh, 1982; and
Raleigh and others, 1986
). Suitability criteria were defined based on published values for three life
stages of the two fish species: 1) adult, 2) ju
venile, and 3) fry, based on depth and velocity ranges (fig.
7
). Habitat suitability curves, scaled from 0 to 1, were developed for each of the
three life stages

based
on depth and velocity conditions. A value of zero means that the velocity or depth condition is not
acceptable; a value of 1 means that the condition is optimal. Between 0 and 1, habitat suitability values
were interpolated linearly to indicate incre
asing or decreasing acceptability of the condition. For each
streamflow and life stage, the habitat builder provides a map of the calculated geometric mean of the
suitability curves and total weighted usable area which is the
sum of the area of each grid c
ell multiplied
by the geometric mean of the grid node
.


22

Figure 7.

Habitat suitability curves for three life stages of a) cutthroat trout and b) brown trout.

M
eso
scale habitat quantification was done through the separation of each reach into features

of
specific habit
at type (pool or

riffle
/
run habitat) based on the geomorphic and hydraulic characteristics of
larger, continuous features within each reach. Separation of these habitat types
was done in a geographic
information system (GIS) based on
velocity
,

depth,
and v
is
ual inspection of morphological form.

Separation

of the reaches into two categories provides
a
compar
ison between the observed proportions
of habitat and

general guidelines derived from observations
of

reaches with sustained populations of
each species

(
Hickman and Raleigh, 1982; and Raleigh and others, 1986
)
(table
1
)
.

Table 1.

Mesoscale
-
habitat suitability ranges of habitat type for cutthroat and brown
trout.

A

vital component of
fish
habitat
includes an evaluation of cover.

Differences exist in the type of
cover fish species seek, and it is dependent on availability of cover

type
, location of cover relative to
other important habitat fe
atures (such as feeding habitat or refuge), time of year, and life stage.

Cover
can vary from rubble bed substrate, aquatic vegetation or plant debris, or deep pools and surface
disturbances that obscure visibility.

Determination of the abundance of cover was limited in this
analysis
to adult life stages and was quantified as a function of the distance to cover from feeding
locations
within each reach. Feeding
locations
were
defined
as
those
areas

in the river
that
provide
access to food sources while requiring little exertion. These were typically lower velocity areas adjacent
to or downstream from higher velocity areas that can funnel benthic invertebrates and terrestrial insects
to these f
ish from the current. Cover was identified as pool areas or vegetated channel banks within each
reach.


23

Two
-
Dimensional Hydrologic Model Sensitivity Analysis

Model sensitivity was analyzed for two parameters: the
lateral eddy viscosity

and
channel
roughness

based on
Z
0
, for the 150; 260; and 1,600 ft
3
/s
flow simulations

because of the water
-
surface
elevation data available along the entire length of each reach. Sensitivity of the two parameters was
evaluated by
holding all parameters constant in the best fit

flow simulations

and then
individually
adjusting
the lateral eddy viscosity and
Z
0

by
-
25 and +25
percent
.

Model results from the sensitivity
analysis were compared to the original best
-
fit
flow simulations

(tables
2

and

3
).
The water
-
surface
elevations, velocity, and shear stress are calculated from the means of all wetted nodes in each
flow
simulation
. Differences between the two
flow simulations

were reporte
d from means of the differences
in values between wetted node
-
pairs from each
flow simulation
. The percent difference is calculated as
the average difference between the two
flow simulations
, relative to the mean value of the best
-
fit
flow
simulation
.

Larg
er differences suggest greater sens
itivity of the model parameter.

Table 2.

Results from the analysis of the sensitivity of the lateral eddy viscosity parameter for three of the
streamflows simulated in the multidimensional flow model for the upper and lower

study
reaches on the Lake
Fork.

Table 3.

Results from the analysis of the sensitivity of the roughness parameter (Z
0
) for three of the streamflows
simulated in the multidimensional flow model for the upper and lower study reaches on the Lake Fork
.

None of the
flow simula
tions

showed sensitivity to changes in lateral eddy viscosity (table
2
).
The percent difference between
flow simulations

ranged from
-
0.25 perce
nt to +0.52 percent. Both of
the extremes occurred in
boundary

shear stress in the upper reach for the 150 and 1,600 ft
3
/s
simulations
, respectively. In the 1,600 ft
3
/s s
imulations
, the +0.52 percent for
mean boundary
shear
stress corresponded to a +0.20
Pascals. The largest change in mean velocity was +0.01 m
eters per

24

second

and the largest chang
e in mean depth (and mean water
-
surface elevation) was +0.01 m
eters

(table
2
).

The flow
simulations

showed
greater
sensitivity to changes in
Z
0

(table
3
).
Most model
outputs

changed by less than

10 percent.
Shear stress was generally the most sensitive of the model outputs, the
difference from the mean
boundary
shear stress ranged from
-
10.0 percent to +8.0 percent. The percent
difference from the mean velocity ranged from
-
1.6 to +1.7 percent and the percent difference of the
mean water
-
surface elevation ranged from
-
6.2 to +5.6 percent (table
3
).

Hydrologic
and Geomorphic
Conditions

of the Modeled Reaches

The
existing
geomorphic and hydrologic

conditions for
a 1
.1

kilometer

section

of the Lake Fork
at

Lake City
are

summarized as a function

of the current
stream
flow regime

based on model simulations
for

8 streamflows that cover the range of
stream
flow conditions
typically
observed at the site (fig
s
.
2

and
8
)
.

Comparisons

and evaluation of the differences in sediment mobility and habitat suitability will
provide resource managers with information to
evaluat
e

the existing cha
nnel stability and habitat
conditions of the reach

as well as provide a baseline to compare future c
ond
i
tions against.

Within these systems
,

removal of fine sediments (sizes less than 2 mm)
within framework grain
interstitial

spaces
enhances habitat
function
.

G
ravel substrates
relatively
free
of find sediments are
required
for
spawning redds

(spawning habitat)

(Hickman and Raleigh, 1982; and
Raleigh and others,
1986
). Benthic macro
-
invertebrates

productivity and abundance, as well as larval and fry refugia are
optimized when fine sediments are limited to less than 5 percent (
Hickman and Raleigh, 1982; and
Raleigh and others, 1986
).

Sediment mobility is an important process for flushing of these f
ine
sediments from within the gravel frameworks.

Figure 8.

F
low
frequency

curve for
the
Lake Fork of the Gunnison River at Lake City, Colorado.


25

Sediment
-
Mobility

Characteristics

Boundary s
hear stress calculated from each
flow simulation

is compared to the critical s
hear
stress
of the mean D
50

in each reach
to determine the mobility of the streambed
(fig.
9

and

t
able
4
).
In
this report, where the
boundary
shear stress is equal to the critical shear
, sediment transport conditions
will be reported as potential
ly

mobile; w
he
re

the
boundary
shear

stress exceed
two
times

the critical
shear

stress
,
sediment transport conditions will be reported as
substan
tial
ly

mobile
.

Figure 9.

Sediment
-
size characteristics of the five sampling sites.

Table 4.

Sediment particle
-
size characteristics for the five sites sampled in the Lake Fork study
area.

Evaluation
s

of channel and flow characteristics at cross
-
section locations 2
-
8 show a range of
streambed
mobility (figs.
10

and
11
). In general,
boundary
shear stress and
streambed

mobility increase
with increases in streamflow.
At
stream
flows
of
50
;

150
;

and 260 ft
3
/s
the
streambed

is not mobile

(fig
s.

10

and
11
)
.
The streambed is
potentially
mobile
for the 400 ft
3
/s
stream
flow

at
cross
-
section
s

4
-
8
depending on which dimensionless
critical
-
shear stress value is referenced (0.03, 0.047, or 0.06)
(figs.

10

and
11
)
.
At cross sections 2
-
8, sediment i
s

potentially mobile
at streamflows
of

900 ft
3
/s.

The
streambed is s
ubstantial
ly mobile

at cross sections 4
-
8 at streamflows of 1,300 to 1,
6
00 ft
3
/s
with all
cross sections
being substantially mobile
at 1,800 ft
3
/s
(figs.
10

and
11
)
.

Within the cross sections, the
greatest
boundary
shear stresses

occurred towards the cent
er of the channel. Sediments were generally
less mobile near the channel margins
for

all
flow simulation
s.

Figure 10.

Shear stress and water
-
surface elevation plots at cross
-
section
s
2,

3, and
4.

Figure 11.

Shear stress and water
-
surface elevation plots at cross
-
sections 5, 6,
7, and 8.


26

Reach
-
scale assessment of sediment mobility

i
n the upper reach

show


patches

of

the streambed
are
potentially mobile

at al
l
simulated
streamflow
s

Boundary
shear stress and streambed
mobility
increase as streamflow increases

(fig.
12
, A
-
C)
.

For
streamflows of
400 ft
3
/s or less,
patches

of
the
streambed
are
p
otentially mobile
.
of
The
se patchs

become
substantial
ly

mobil
e

at

flow simulations

of
900 ft
3
/s and greater
,

and
increase
in size
as streamflow increases
.
At streamflows of
1,300 ft
3
/s or
greater, the
majority of the
channel is potentially mobile except
along the channel margi
ns.

F
low
simulations
of
1,
8
00 ft
3
/s indicate
areas of
potential deposition
between cross sections 2 and 3 based on

the differences in relative mobility between this area and areas upstream
.

Figure 12.

Reach
-
scale assessment of sediment mobility

in the upper reach
(A
-
C) and lower reach (D
-
F) for
of the
Lake Fork of the Gunnison River

at streamflows of 400; 900; and 1,800 ft
3
/s, respectively.

Reach
-
scale assessment of sediment mobility

in the lower reach show
s

increased

s
treambed
mobility

(fig.
12
, D
-
F)
. This is

due in part to smaller
grain
-
size
s

in the lower reach
(mean D
50

is
11 mm
smaller

than the upper reach mean D
50
,
table
4
)

but may also reflect the
greater extent of
channel
alterations, specifically the constructed berms, present in this reach
. Small
patches
of
streambed that are
potential
l
y mobil
e

occur downstream of cross section 5 beginning at 50 ft
3
/s

and

increas
e

in size as
streamflow increases
.
In
the
400

ft
3
/s

flow simulation,

some locations within the channel

bec
o
me
substantial
ly
. In the 900 ft
3
/s
flow simulation

the entire length of the channel bec
o
me
s

potentially
mobile
with
small patches

of the streambed
that are
s
ubstantial
ly

mobil
e
. In the 1,800 ft
3
/s
flow
simulation,

the
majority of the channel along its entire length is

substantial
ly mobile

except
along the
channel
margins

which are potentially mobile
.


27

Geomorphic Characteristics

The geomorphic form of the channel is affected by land
-
use changes
within the basin
and geologic
controls within the reach (
Lake Fork Valley Conservancy, 2010
).
The historic channel was defined

as a
dynamic, braided channel (multiple, bifurcating channels)
with an active floodplain

(
Lake Fork Valley
Conservancy, 2010
).
Co
mparisons of the channel slope, high
-
flow conditions

or
floods

(Capesius and
Stephens, 2009; U.S. Geological Survey, 2010b)
, and D
50

of each reach

to
channel
-
form
thresholds
(meandering and braided channel)
indicate
that the current channel
characteristics
may
favor a

braid
ed

channel
form
(fig.
13
)

(Richards, 1982; Leopold and Wolman, 1957)
.

This can result in a natural
tendency of

the channel to braid. This tendency can affect the channel stability of
reconfigured reaches
when a single
-
thread meandering

channel is imposed on the stream (
Elliott and Capesius, 2009
).

Figure 13.

Plotting
positions

of two surveyed reaches of the Lake Fork of the
Gunnison River near Lake City,
Colorado, in comparison to thresholds for meandering and braided channel patterns based on finding from (A)
Leopold and Wolman (1957) and (B) Richards (1982)
.

Additional constraints on the channel form include natural and
artificial
impingement

of the active
channel occurring from placement of boulder revetments, cobble
-
boulder berms, and bedrock outcrops
along the channel margins.
Boulder revetments are common between cross
-
sections 1 and 5 along river
left, with limited u
se along river right.
Cobble
-
boulder berms were observed between cross
-
sections 4
and 8 along both banks.
Bedrock constraints occur along both banks for much of the channel between
cross
-
sections 4 and 5, and along river left between cross
-
sections 6 and 8
.

These features constrain the
channel from lateral migration and channel adjustments
, potentially limiting geomorphic response of the
channel to alteration in this area.


28

Habitat
Conditions of the Modeled Reaches

Habitat
suitability
analysis
can
include

an evaluation of several inter
-
related stream conditions
(Bovee, 1982).
Dissolved oxygen, water temperature, pH, and other water
-
chemistry constituents
(nutrients, metals, and total dissolved solids) are important to overall aquatic habitat; however,
t
his

analysis
will focus
o
n
the
physical habitat elements

of the reach

such as depth,

velocity, and distance to
cover
.

Model output for each
stream
flow was used to quantify microscale and meso
scale habitat
abundance.

Microscale
Fish
Habitat Quantification

Micr
oscale habitat,
presented
as weighted usable area
, generally in
creased as streamflow
increased
for both species and all life stages
(fig.
14
)
.

Habitat
occurs

along the banks for flows of 900
ft
3
/s and less. Out
-
of
-
bank areas become more substantial contributors to overall habitat availability for
flows of 1,300 ft
3
/s or more.
Adult habitat
,

for both trout species
,

was the most abundant habitat fo
r
nearly all streamflows.
T
he least abundant habitat type
f
or cutthroat trout
was the

juvenile

life stage and

the least abundant habitat for
brown trout

was fry
.

In general, the upper reach provided 2
-
3 times more
available habitat than the lower reach.

Figure 14.

We
ighted usable area for the adult, juvenile, and fry life stages of the cutthroat and brown trout in
the
study reach of the Lake Fork of the Gunnison River
.

In order to determine the interaction of flow frequency and habitat abundance within the current
flow regime, a combination of the available habitat was related to the probability of that flow occurring
each year.

This provides a measure of the overall habitat for each life stage over the course of an
average flow year.

The results are summari
zed by s
pecies and life stage
in table

5
.


29

There are differences in the timing of
life stages
betwe
en the two trout species for fry. Both t
rout
species
construct redds in small gravelly areas with good hyporheic flows, and minimal fine sediments.

However, c
utthroat trout typically spawn in early spring (February or later, based on water
temperatures) and the larvae typically hatch within 28
-
49 days depend
ing on water temperatures. Brown
trout spawn in late summer to early fall but the larvae remain in the spawning gravels until spring when
they emerge as fry. This makes them more susceptible to over
-
wintering conditions within the gravels
and possible smot
hering due to fine sediment deposition; however, when they emerge in the spring they
are larger and typically better able to out
-
compete the cutthroat fry if resources are limited.

The flows
that typically occur during these two different time

periods were

used to calculate the total weighted
-
usable habitat in a typical year for the fry of both species

(
table

5
)
.

Table 5.

M
icr
oscale
-
habitat suitability
to
tal weighted
-
useable habitat in a typical year

for cutthroat and brown trout.

Mesoscale Habitat Characteristics

Mesoscale fish
-
habitat
assessment of the Lake Fork was done based on the conditions present in
the 150 ft
3
/s
flow simulation

as well as field ob
servation. Both the upper and lower reach
is

primarily
characterized as riffle/run habitat
; however, in the upper reach the total available pool habitat is greater
(table
6
).

The presence of pool habitat was limited within the reach and occurred along the channel
margins. For both reaches, the pool habitat was less than 5 percent of the total
-
wetted area, a percentage
that is substantially lower t
han the recommendations for sustainable populations of 40
-
70 percent (tables
1

and
6
)

(
Hickman and Raleigh, 1982; and Raleigh and others, 1986
)
.
An exception to this is a
n

isolated

scour pool located outside of the modeled area (between the two reaches) below a waterfall whic
h was
not quantified in this analysis.

Vegetative canopy study results

show differences between the upper and lower reach
es
.
Vegetative
canopy

density measurements were consistent in the upper reach, with approximately 20
-

30

percent coverage occurring along b
oth banks
throughout
(fig.
4
). Vegetation patterns
in the lower reach
were less consistent
, and demonstrate local effects of

topography
. The lar
ge
hillslope

along the left bank
between cross
-
sections 7 and 8 produced greater canopy density and shading than other areas (fig.
4
).

Assessmen
t of distance to cover from potential feeding areas shows differences between the
characteristics of the two reaches.
Areas of c
over
were adjacent to potential
feeding areas in the
low
er
reach
, and often occurr
ed within the same pool habitat. This may

favor energy expenditure ratios of both
fish species
, wherein little energy is need
ed

to acquire adequate food sources

(table
6
).

In the
upp
er
reach distances from feeding areas to cover were much greater, a condition that is less conducive to
Cutthroat Trout
which tend to occupy
feeding
ar
eas where cover is located near
by
(
Hickman
and
Raleigh, 1982
).

Table 6.

Mesoscale
-
habitat characteristics during low
-
flow conditions for cutthroat and brown trout
.

Macro
-
invertebrate Community Assessment

Using
macro
-
invertebrate community data as an index of
channel
disturbance is an effective way
to
characterize

aquatic

habitat conditions.

M
acro invertebrates

can be goo
d indicators of
habitat
because
these organisms
are continuously exposed to water and sediment in streams

(
Barbour and others
, 1999).

Sampling locations within the study reach provide a longitudinal dissection of the stream channel in the
vicinity of Lake
City.

In order to compare the La
ke Fork study area

sites

(LF1

LF4)
(Fig)
to
sites outside the study
area
, samples were compared to
upstream
control sites

(
Lake Fork
,

LF
-
5
;

and
Henson Creek
,

HC
-
1)
(Fig)
.
These control sites
provide the basis for repeat
surveys t
o assess
if any observed
changes within
the system

occurred outside of the reconfigures reach
es
. Any changes at these control sites
can be
interpreted as the normal variability within the system or may
indicate effects of channel or watershed
chan
ges
upstream

of the study reach.
The upstream site on the Lake Fork (LF
-
5) was selected
as a

31

control site
because it was upstream of disturbances associated with

channel modifications and

influences from
the highway.
Henson Creek

(HC
-
1)

was used as a
contr
ol

site
based on site
reconnaissance
which indicates
it represent
s

conditions in the same spatial region as the Lake Fork
study area
that was

minimally disturbed
by
channel stabilization and straightening
.

The
HC
-
1

site
is
downstream

of
a decommissioned

si
lver
mine and may have been

subject to changes in water chemistry
associated with historic mining

and contaminated sediments
,
conditions
which may also affect sites
LF1

LF4
.

Taxa diversity in addition to presence/absence of specific macro
-
invertebrate species can be a
good indicator of the relative quality of the aquatic habitat and may indicate
that
specific aquatic
stressors are present
in the system
. The U.S. EPA indicated
in a report that riparian vegetated cover
disturbance as one of the main stressors increasing the risk of macro
-
invertebrate impairment and taxa
loss in the Western United States (
Barbour and others
, 1999). Other stressors that increased the risk of
biolo
g
ical impairment in the Western

United States included metals, nitrogen, phosp
horous, and
suspended sediment.

M
acro
-
invertebrate

communities at the Lake Fork study area sites
(LF
-
1

LF
-
4)
had similar
macro
-
invertebrate communities which
were
different

in com
parison to
the control sites

(LF
-
5 and HC
-
1)
. The Diptera family, primarily
Simulium

species (black flies), made up more than 45 percent of
macro
-
invertebrate community sites
at
LF
-
1

LF
-
4.
Results from the macro
-
invertebrate survey showed
that control site
s HC
-
1 had the greatest diversity or var
iety of taxa of sampled sites (t
able

7
).
LF
-
5 had a
different macro
-
invertebrate community in comparison

to
LF
-
1

LF
-
4

and HC
-
1.

The biotic condition index (BCI) is useful for evaluating the macro
-
invertebrate community
response to sedimentation (
Barbour and others
, 1999). HC
-
1 had the smallest BCI values and LF
-
1

LF
-
4 had the
largest

BCI values. This may
in
dicate that increased sediment deposition and the resulting

32

highly embedded substrate in the study reach sites (LF
-
1

LF
-
4) are affecting macro
-
invertebrate
habitat
.

Specialist feeders, such as scrapers and shredders usually represent sensitive taxa and
their
abundance can indicate

healthy stream

conditions
. HC
-
1 had a greater number of both scrapers and
shredders compared to LF
-
1

LF
-
4. The number of metal intolerant species was similar between HC
-
1
and LF
-
1

LF
-
4. Ore mining for silver was a historic l
and use upstream on Henson Creek and could
affect both HC
-
1 and LF
-
1

LF
-
4.

Ephemeroptera (mayfly), Plecoptera (stonefly), and T
richoptera

(caddisfly)
,

(EPT)

can be used
as an indicator of the quality of the aquatic environment since they commonly
dec
rease in number as
stressors in the system increase.
The Ephemeroptera
Baetis

s
pecies

(mayfly) were common

(greater than
10 percent of the
macro
-
invertebrate

community)
at all the sampled sites.
There were approximately 87
percent
EPT at HC
-
1

in comparison

to approximately 36 percent
at LF
-
1

LF
-
4
. The dominant
taxon
,
which increases with disturbance, was approximately 33 percent
EPT
at
HC
-
1
, and 57 percent at LF
-
1

LF
-
4
.

The EPT indicates that the sites within the study reach (LF
-
1

LF
-
4) may be more disturbe
d or
stressed than the control site (HC
-
1).
LF
-
5 had a low EPT score and low number of intolerant taxa in
comparison to the other sites. It is unknown if the macro
-
invertebrate community
at LF
-
5
is affected by
stressors

from upstream or adjacent land uses.

Table 7.

Macro invertebrate

community metrics sampled in the Lake Fork of the Gunnison River, Colorado
.

Summary

The Lake Fork of the Gunnison River has been an area of active channel modification since the
1950s to accommodate the needs of the Lake City communi
ty
.

The Lake Fork Valley Conservancy

33

District has begun a planning process to assess restoration option for a reach of the Lake Fork in Lake
City to enhance the hydrologic and ecologic health of the reach
.


Channel
restoration

to mitigate a variety of riverine problems has become a common practice in
the western United States
.

Numerous private entities and resource
-
management agencies have attempted
to modify stream channels by using designs based on different geo
morphic philosophies and
classification schemes
.

However, little work has been done to monitor and assess the channel response
to, and the effectiveness of, these modificati
ons over a long period of time
.

The basis for determining
the ‘success’ of a
project is often not made through analysis of quantifiable metrics, leaving little
information for resource managers, planners, and designers to use to determine the effectiveness and
durability of channel restoration techniques used for stream rehabilitat
ion and channel stability
.

The
U.S. Geological Survey in cooperation with the Colorado Water Conservation Board and Colorado
River Water Conservation District began a study in 2010 to quantify existing hydraulic and habitat
conditions for a reach of
the Lake Fork of the Gunnison River in Lake City, Colorado
.


The purpose of this report is to quantify existing Lake Fork conditions and establish a baseline
against which post
-
rehabilitation conditions can be compared
.

Th
is

report

uses two
-
dimensional
hydraulic and habitat modeling, and data collection to characterize (1) the existing hydrologic and
geomorphic conditions in a 1.1 kilometer section of the Lake Fork that has been proposed as a location
for future channel
-
rehabilitati
on efforts; (2) the habitat suitability of the reach for two aquatic fishes
based on physical conditions within the stream; and (3) the current riparian canopy cover and benthic
-
invertebrate structure of the reach

(
using two metrics:
vegetation canopy cove
r

and
be
nthic invertebrate
characteristics
)
.

This characterization provides resource managers with information to evaluate the
existing channel conditions as well as a baseline for post
-
restoration comparisons to evaluate restoration
project success
.


34

Sediment

particles were measured at 5 locations in the study reach

to characterize
the grain size
of
different geomorphic surfaces
, as well as to determine channel roughness
.

Bars and channel margins
(outside of areas strongly affected by eddies) were
chosen
because t
hese areas are inundated most of the
year and represent the
coarser material
within the river channel

that is transported
.

Macro
-
invertebrate samples were collected

during a low
-
flow period August 17
-
18, 2011,
following National Water
-
Quality Assessment (NAWQA) Program protocols
.

In general, five discrete
collections were made from riffle habitats at each
of the 4
site
s

with
in the reach with

a
s
lack sampler
equipped with 500
-
micron mesh and a 0.25 m
2

sampling grid along the reach
.

Two sites upstream of the
reach were also sampled, and are used as control sites outside of the reach
.


Vegetative cover was measured at each of the benthic sample locations at the time of sampling
using a spherical densiometer
.

Four measurements w
ere taken at each location from the center of the
stream with the operator facing: upstream, downstream, left bank, and right bank
.

The FaSTMEC
H computational model within MD_
SWMS was
selected
to
characterize the
effects of streamflow on hydrologic and
habitat
-
suitability conditions
for a study reach o
f

the Lake Fork
.

Channel geometry was obtained from the topographical survey of the channel
.

Model calibration was
achieved through iterative adjustments of the lateral eddy viscosity, relaxation para
meters, and initial
conditions (
including streamflow, water surface elevation, and drag coefficient) until the optimal fit of
each flow simulation
was found
.

Within this report, a comparison of the boundary shear stress to

the critical shear
stress
of
t
he
median grain size (D
50
)

is used to determine when sediments within the channel are mobile
.

Multiple

critical shear stress values were selected for the
sediment mobility

analyses in this
report to cover the
range of conditions within the reach
.

Comparisons are made along surveyed cross sections as well as in
reach scale assessments
.


35

Aquatic habitat was evaluated for two species of trout, the cutthroat trout (
Oncorhynchus clarkii
)
and brown trout (
Salmo trutta

morpha

fario
)
.

This provides resource managers with tools to assess the
effects of hydrologic conditions (wet, average, and dry years) and flow alterations on habitat availability
for specific streamflows

or over annual timescales
.

M
eso
scale habitat quantification was done through
the separation of each reach into features

of specific habitat type (pool or

riffle
/
run habitat) based on the
geomorphic and hydraulic characteristics of larger, continuous fe
atures within each reach
.

Separation

of
the reaches into two categories provides
a
compar
ison between the observed proportions of habitat and

general guidelines derived from observations
of

reaches with sustained populations of each species
.

Model se
nsitivity was analyzed for two parameters: the
lateral eddy viscosity

and channel
roughness based on Z
0
, for the 150; 260; and 1,600 ft
3
/s flow
.

Model results from the sensitivity analysis
were compared to the original best
-
fit flow simulations
.

Larger differences suggest greater sensitivity of
the model parameter
.

The
existing geomorphic and hydrologic

conditions for
a 1
.1

kilometer

section

of the Lake Fork
at Lake City
are summarized as a function of the current streamflow regime based on mo
del simulations
for 8 streamflows that cover the range of streamflow conditions typically observed at the site
.

Comparisons and evaluation of the differences in sediment mobility and habitat suitability will provide
resource managers with information to

evaluate the existing channel stability and habitat conditions of
the reach as well as provide a baseline to compare future conditions against
.

Within these systems
,

limiting of
fine sediment
deposition and removal of fine sediments (sizes
less than 2
mm)
within framework grain interstial spaces
enhances habitat suitability
.

Sediment mobility
is an important process for flushing of these fine sediments from within the gravel frameworks
.

Evaluations of channel and flow characteristics at cross
-
sect
ion locations 2
-
8 show a range of streambed
mobility
.

In general, boundary shear stress and streambed mobility increase with increases in

36

streamflow
.

At cross sections 2
-
8, sediment is potentially mobile at streamflows of 900 ft
3
/s
.

The
streambed
is substantially mobile at cross sections 4
-
8 at streamflows of 1,300 to 1,600 ft
3
/s with all
cross sections being substantially mobile at 1,800 ft
3
/s
.

Within the cross sections, the greatest boundary
shear stresses occurred towards the center of the channel
.

Reach
-
scale assessment of sediment mobility

in the lower reach shows increased streambed mobility
.

This is due in part to smaller grain
-
sizes

in the
lower reach, but may also reflect the greater extent of channel alterations, specifically the constructed
berms, present in this reach
.

The geomorphic form of the channel is affected by land
-
use changes within the basin and
geologic controls wit
hin the reach
.

The historic channel was defined as a dynamic, braided channel
(multiple, bifurcating channels) with an active floodplain
.

Comparisons to channel
-
form thresholds
indicate that the current channel characteristics may favor a braided cha
nnel form
.

Microscale habitat, as weighted usable area, generally increased as streamflow increased for
both species and all life stages
.

Habitat occurs along the banks for flows of 900 ft
3
/s and less
.

Out
-
of
-
bank areas become more substantial contributors to overall habitat availability for flows of 1,300 ft
3
/s or
more
.

In general, the upper reach provided 2
-
3 times more available habitat than the lower reach
.

Mesoscale fish
-
habitat assessment of

the Lake Fork was done based on the conditions present in
the 150 ft
3
/s flow simulation as well as field observation. The presence of pool habitat was limited
within the reach and occurred along the channel margins
.

For both reaches, the pool habitat w
as less
than 5 percent of the total
-
wetted area, a percentage that is substantially lower than the
recommendations for sustainable populations
.

An exception to this is an isolated scour pool located
outside of the modeled area (between the two reaches)
below a waterfall which was not quantified in
this analysis
.

Areas of cover were adjacent to potential feeding areas in the lower reach, and often

37

occurred within the same pool habitat
.

This may favor energy expenditure ratios of both fish species,
wherein little energy is needed to acquire adequate food sources
.

Macro
-
invertebrate communities at the Lake Fork study area sites (LF
-
1

LF
-
4) had similar
macro
-
invertebrate communities whi
ch were different in comparison to the control sites upstream of the
reach
.

The biotic condition index may indicate that increased sediment deposition and the resulting
highly embedded substrate in the study reach
sites
are affecting macro
-
invertebrate
habitat
.

The
Ephemeroptera (mayfly), Plecoptera (stonefly), and T
richoptera

(caddisfly)
, (EPT)

s
core

for the study
sites indicate that the study reach sites may be more disturbed or stressed than the control site on
Henson Creek
.

The control site on
Lake Fork had a low EPT score and low number of intolerant taxa in
comparison to the other sites
.

It is unknown if the macro
-
invertebrate community at LF
-
5 is affected by
stressors from upstream or adjacent land uses
.

References Cited

Andrews, E.D.,
1983, Entrainment of gravel from naturally sorted riverbed material: Geological Society
of America Bulletin, v. 94, p. 1225

1231.

Barbour
, M. T., J. Gerritsen, B. D. Snyder, and J. B. Stribling. 1999. Rapid bioassessment protocols for
use in streams and
wadeable rivers: periphyton, benthic macroinvertebrates and fish, second edition.
EPA 841
-
B
-
99
-
022. U.S. Environmental Protection Agency; Office of Water; Washington, D. C.

Barnes, H.H. Jr., 1977, Roughness characteristics of natural channels
: U.S. Geol
ogical Survey
Water
-
Supply Paper 1849
,
213
p.

Barton, G.J., McDonald, R.R., Nelson, J.M., and Dinehart, R.L., 2005, Simulation of flow and sediment
mobility using a multidimensional flow model for the white sturgeon critical
-
habitat reach, Kootenai
River ne
ar Bonners Ferry, Idaho: U.S. Geological Survey Scientific Investigations Report 2005

5230,
54 p.


38

Bovee, K.D., 1982, A guide to stream habitat analysis using the Instream Flow Incremental

Methodology, Instream flow information Paper 12: U.S.D.I. Fish and Wildlife Service,

Office
of Biological Services, FS/OBS
-
82/26, 249 p.

Bray, D.I., 1982, Flow resistance in gravel
-
bed rivers, p. 109
-
137 in Hey, R.D., Bathurst, J.C., and
Thorne, C.R., Gra
vel
-
Bed Rivers, Fluvial Processes, Engineering
and Management: New York,
Wiley, 892 p.



Burbank, W. S., Eckel, E. B., and Varnes, D. J., 1947, The San Juan region, p. 396
-
446 in V
andenvilt,



J. W., Mineral resources of Colorado: Denver, Colo. Min. Res. Board, 547 p.

Capesius, J.P., and Stephens, V.C., 2009, Regional regression equations for estimation of natural
streamflow
statistics in Colorado: U.S. Geological Survey Scientific Investigations Report 2009

5136, 46 p.

Pielke, Sr., R.A., N. Doesken, and O. Bliss, 2003, Climate of Colorado: Climatology Report 60 Dept. of
Atmos. Sci., CSU, Fort Collins, CO. January, 2003.


Elli
ott, J.G., and Capesius, J.P., 2009, Geomorphic changes resulting from floods in reconfigured
gravel
-
bed river channels in Colorado, USA, in James, L.A., Rathburn, S.L., and Whittecar, G.R.,
eds., Management and Restoration of Fluvial Systems with Broad Hi
storical Changes and Human
Impacts: Boulder, Colorado, Geological Society of America Special Paper 451, p. 173

198.

Fitzpatrick, F.A., Waite, I.R., D’Arconte, P.J., Meador, M.R., Maupin, M.A., and Gurtz, M.E., 1998,
Revised methods for characterizing strea
m habitat in the National Water
-
Quality Assessment
Program: U.S. Geological Survey Water
-
Resource Investigations Report 98

4052, 67 p.

Guzzi, R. and Parise, M., 1992, Surface features and kinematics of the Slumgullion
la
n
dslide
, near Lake
City, Colorado: U
.D. Geological Survey Open
-
File Report 92
-
252, 45 p.


39

Hickman, T., and Raleigh, R.F., 1982, Habitat suitability index models: Cutthroat trout:

U.S. Fish and
Wildlife Service Biology Report, vol. 82, no. 10.5, 38 p.

Homer, C., Huang, C., Yang, L., Wylie, B.
K.,

and Coan, M., 2004, Development of a 2001 National
Land Cover Database for the United States: Photogrammetric Engineering and Remote Sensing, v. 70,
no. 7, p. 829
-
840.

Julien, P.Y.,
2010
, Erosion and sedimentation

second edition
: Cambridge, United
Kingdom, The
Cambridge University Press, 3
71

p.

Klemm, D.J., Lewis, P.A., Fulk, F., Lazorchak, J.M., 1990, Macroinvertebrate field and laboratory
methods for evaluating the biological integrity of surface waters: Cincinnati, Ohio, U. S.
Environmental Prote
ction Agency, Environmental Monitoring and Support Laboratory, EPA
-
600
-
4
-
90
-
030.

Komar, P.D, 1987, Selective gravel entrainment and the empirical evaluation of flow competence
:

Sedimentology, v. 34, p. 1165

1176.

Lake Fork Valle
y Conservancy, 2010, Lake Fo
rk R
iver restoration project,
4p.

Lemmon, P.E., 1956, A spherical denisometer for estimating
forest overstory density, Forest Services,
vol. 2, issue 4, 314
-
320 p.

Leopold, L.
B
.
, and Wolmann,
M.G.
, 195
7
, River channel patterns

Braided, meandering and stra
ight:
U.S.
Geological

Survey Professional Paper 282
-
B, p. 39
-
84.

Lisle, T.E., Nelson, J.M., Pitlick, J., Madej, M.A., and Barkett, B.L., 2000, Variability of bed mobility
in natural, gravel
-
bed channels and adjustments to sediment load at local and reach
scales: Water
Resources Research, v. 26, no. 12, p. 3743

3755.

McDonald, R.R., Bennett, J.P., and Nelson, J.M., 2001, The USGS Multi
-
Dimensional Surface Water
Modeling System,
in

Proceedings of the Seventh Federal Interagency Sedimentation Conference,

40

Marc
h 25

29, 2001, Reno, Nev.: p.

I
-
161

I
-
167, accessed April 5, 2011, at
http://pubs.usgs.gov/misc/FISC_1947
-
2006/pdf/1st
-
7thFISCs
-
CD/7thFISC/7Fisc
-
V1/7
FISC1
-
1.pdf
.

McDonald, R.R., Nelson, J.M., and Bennett, J.P., in press, Multi
-
dimensional surface
-
water modeling
system user’s guide: U.S. Geological Survey Techniques and Methods, book 6, section B, chap. 6,
156 p.

McDonald, R.R., Nelson, J.M., Kinzel, P
.J., and Conaway, J., 2006, Modeling surface
-
water flow and
sediment mobility with the Multi
-
Dimensional Surface Water Modeling System (MD_SWMS):
U.S.

Geological Survey Fact Sheet 2005

3078, 6 p.

Milhous, R.T., 1982, Effect of sediment transport and flow regulation on the ecology of gravel
-
bed
rivers,
in

Hey, R.D., Bathhurst, J.C., and Thorne, C.R., eds., Gravel
-
bed rivers:
Chichester, England,
Wiley
, p. 819

842.

Millar, R.G. and Quick, M.C., 1994.

Flow resistance of high
-
gradient gravel channels. In:

G.V. Cotroneo and R.R. Rumer (Editors), Hydraulic Engineering '94. ASCE,

Buffalo, NY, pp. 717
-
721.
Moulton, S.R.II.,
Kennen, J.G., Goldstein, R.M., Hambrook, J.A., 2002,
Revised protocols for samplin
g
algal
, invertebrate, and fish communities as part of the National
Water
-
Quality Assessment Program; Open File Report 2002
-
150

Mueller, D.S., and Wagner, C.R., 2009, Measuring discharge with acoustic Doppler current profilers
from a moving boat: U.S. Geol
ogical Survey Techniques and Methods 3A
-
22, 72 p.

Nelson, J.M., Bennett, J.P., and Wiele, S.M., 2003, Flow and sediment
-
transport modeling,
in

Kondolf,
G.M., and Piegay, H., eds., Tools in fluvial geomorphology: Chichester, England, Wiley, p. 539

576.

Nels
on, J.M., and McDonald, R.R., 1997, Mechanics and modeling of flow and bed evolution in lateral
separation eddies: Bureau of Reclamation, Glen Canyon Environmental Studies Report, 69 p.


41

Parker, G., Klingman, P.C., and McLean, D.G., 1982
,

Bedload and size distribution in paved gravel
-
bed
streams: American Society of Civil Engineers, Journal of the Hydraulics Division 108(HY4)
,

p. 544

571.

Powell, D.M., and Ashworth, P.J., 1995, Spatial pattern of flow competence in bed load transport in a
divided gravel bed river: Water Resources Resear
ch,

v. 31, no. 3, p. 741

752.

Raleigh, R.F., Zuckerman, L.D., and Nelson, P.C., 1986, Habitat suitability index models and
in stream

flow suitability curves: Brown trout, revised. U.S. Fish Wildlife Service B
iology Report, vol. 82, no.
10.124,

65 p.

Rantz, S.E., 1982, Measurement and computation of streamflow

Volume 1, M
easurement of stage and
discharge: U.S. Geological Survey Water Supply Paper 2175, v. 1, p. 1

284.

Richards, K., 1982, Rivers

Form and process

in alluvial channels: London, Methuen, 358 p.

Thompson, D. M., Nelson, J.M., and Wohl, E.E., 1998, Interactions between pool geometry and
hydraulics: Water Resources Research, v. 34, no. 12, p. 3673

3681.

Turnipseed, D.P., and Sauer, V.B., 2010, Discharg
e measurements at gaging stations: U.S. Geological
Survey Techniques and Methods book 3, chap. A8, 87 p.

Tweto, Ogden, compiler, 1979, Geologic map of Colorado: U.S. Geological Survey Map, scale
1:500,000.

Tweto and Sims, 1963, Pr
e
cambrian
ancestry

of the
Colorado Mineral Belt:

Geological Society of
America Bulletin, v. 74, p. 991
-
1014.

U.S. Census Bureau, 2011, Population Finder, accessed June 26, 2011,
http://www.census.gov/
.

U.S. Department of Agriculture, 2009, Par
t 654: Stream restoration design, National Engineering
Handbook, 714 p.


42

U.S. Geological Survey, 2010
a
, 091
24500 Lake Fork at Gateview, CO:

U.S. Geological Survey Water
Data Report, Accessed August 9, 2011,
http://wdr.water.usgs.gov/wy2010/pdfs/09124500.2010.pdf

U.S. Geological Survey, 2010
b
, StreamStats: Colorado, Accessed August 9, 2011,
http://water.usgs.gov/osw/streamstats/colorado.html
.

Western Regional Climate Center, 2011, Colorado climate summaries: Information available on Web,
accessed June 26, 2011,
http://www.wrcc.dri.edu/climsum.html/
.

Wilcock, P.R., 1992, Experimental invest
igation of the effect of mixture properties on transport
dynamics,
in

Billi, P., Hey,

R.D., Thorne, C.R., and Tacconi, P., eds.,
Dynamics of gravel
-
bed r
ivers:
Chichester,
England, Wiley, p. 109

139.

Wilcock, P.R., and McArdell, B.W., 1993, Surface
-
based f
ractional transport rates

M
obilization
thresholds and partial transport of a sand
-
gravel sediment
:

Water Resou
rces Research, v. 29, no. 4,
p.

1297

1312.

Wolman, M.G., 1954, A method of sampling coarse river
-
bed material: American Geophysical Union
Transact
ions, v. 35, p. 951

956.